Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
“Softly, one can shake the world” Mahatma Gandhi Aan mijn ouders aan Wouter Equity in Primary Health Care use among vulnerable populations in Belgium: an exploration PhD-Thesis, Department of Family Medicine and Primary Health Care Ghent University, Belgium Please cite as: Evelyn Verlinde. (2012). Primary health care use among vulnerable populations in Belgium: an exploration. Thesis submitted in fulfilment of the requirements for the degree of Doctor in Medical Sciences. Ghent University ISBN: 9789491125034 NUR: 882 Copyright © Evelyn Verlinde 2012 All rights reserved. No part of this publication may be reproduced in any form, by print, photocopying, recording or otherwise, without prior written permission of the author. Layout: Anja Peleman Equity in Primary Health Care use among vulnerable populations in Belgium: an exploration Evelyn Verlinde 2012 Promotor: Prof. Dr. Sara Willems, Department of Family Medicine and Primary Health Care, Ghent University Co-promotor: Prof. Dr. Jan De Maeseneer, Department of Family Medicine and Primary Health Care, Ghent University Supervisors Prof. Dr. Sara Willems, supervisor, Ghent University Prof. Dr. Jan De Maeseneer, co-supervisor, Ghent University Head of examination board Prof. Dr. Marleen Temmerman, Ghent University Examination board Prof. Dr. Peter De Paepe, Ghent University Prof. Dr. Koen Van Herck, Ghent University Prof. Dr. Peter Groenewegen, Nivel Prof. Dr. Kristin Hendrickx, Antwerp University Prof. Dr. Lea Maes, Ghent University Prof. Dr. Dirk Avonts, Ghent University Table of Content Summary/Samenvatting 4 List of abbreviations 8 Preface 9 Chapter 1: Introduction 11 1.Background 12 2.Determinants of health inequities 18 3. Tackling health inequities 20 3.1. European policy actions 20 3.2. Proportionate universalism 22 4. The importance of primary care in tackling health inequities 24 4.1 Defining primary health care 24 4.2. Defining (in)equity in (primary) health care 25 4.3. A role for primary health care in tackling health inequities 31 Chapter 2: Research aim 43 Chapter 3: Research papers 51 Chapter 4: Discussion 131 4.1. What this thesis adds: summary of the results 132 4.2. Conclusion 137 4.3. What this thesis could not add: suggestions for future research 140 4.4. Policy and practice implications 142 Chapter 5: Dankwoord 173 Chapter 6: CurrisulumVitae 177 5 6 Summary There is extensive literature and evidence on the existing inequities in health. These health inequities are a result of social inequities. It takes a wide range of interventions to tackle these social inequities. One of the interventions is building a sustainable, strong Primary Health Care system in order to contribute to more health equity. The aim of this doctoral thesis is to focus on the equity in primary health care in Belgium in terms of equal access, equal treatment and equal outcomes for vulnerable population groups in equal need. To do so, five studies were set up: one qualitative study, one literature review and three quantitative studies. Five hypotheses were formulated, each described in a paper: Hypothesis 1: The specialist is the doctor of the higher-income people and the GP is the doctor of the lower-income people (paper 1). Hypothesis 2: Homeless people have a higher likelihood to consult the emergency department than the general population. A universal primary health care system with active guidance for homeless people towards this system might counter this (paper 2). Hypothesis 3: People who are not able to cope with their available income have a higher likelihood to postpone a visit to a GP (paper 3). Hypothesis 4: The social gradient in doctor-patient communication persists over time, where patients from lower socio-economic classes are less involved during the consultation, receive less control and a more paternalistic consultation style (paper 4). Hypothesis 5: Physicians have a negative attitude towards patients with a low SES and as a result make changes in their clinical decision which do not favor the outcomes for the patient (paper 5). This thesis shows us that: The specialist is not the doctor of the higher-income people but the doctor of the higher educated people. While the GP is not the doctor of the lower-income people, but the doctor of the lower-educated people. These findings indicate that the 7 Belgian health care system is financially accessible, but still has other access issues related to the educational level of the patient. Low-income people suffering from bad health, those suffering from a severe depression and those without complete trust in their GP have a higher risk of postponing a visit to a GP. People not able to cope with their monthly income do not have a higher risk of postponing a visit to a GP. A universal health care approach for homeless people, with active guidance by social workers seems to be far more effective than a selective setting focusing solely on that specific population. This universal approach leads to less stigmatization and labeling than a selective approach, leading to a higher use of these services. The social gradient in doctor-patient communication persists over time, where patients from lower social classes receive less socio-emotional talk, a more directive and less participatory consultation style characterized by less involvement in treatment decisions, lower patient control over communication, less diagnostic and treatment information and more physical examinations. 8 The attitude and decisions of GPs toward patients from lower social classes depend on the number of contacts physicians have with this population group. GPs with occasional contact with patients from low SES take more time for these patients, invest more energy and see it as a positive challenge. They make changes in their clinical decisions, but these changes are made in order to enhance the patients outcomes. However, when the number of low SES patients is too large, GPs seem not always capable to keep providing good qualitative care. The changes they make in their decisions are not always in favor of the outcomes for the patient. It seems that the Belgian primary health care system does not create huge financial barriers, however, there still seems to be some inequity in the other aspects of access to health care (e.g. cultural, geographical) as well as in the treatment and outcomes patients in equal need perceive. Therefore, action is needed on practice level (e.g. intersectoral approach) as well as on policy level (e.g. additional resources for practices working with low SES patients). Samenvatting Er bestaat reeds een uitgebreide hoeveelheid aan literatuur omtrent de bestaande ongelijkheden in gezondheid. Deze ongelijkheden zijn het resultaat van sociale ongelijkheden. Om deze sociale ongelijkheden aan te pakken, is een breed spectrum aan interventies noodzakelijk. Een mogelijke interventie is het uitbouwen van een goed georganiseerd eerstelijnsgezondheidszorg systeem. Het doel van deze doctoraatsthesis is het onderzoeken van de ongelijkheden in het gebruik van de eerstelijnsgezondheidszorg in België in termen van gelijke toegang, gelijke behandeling en gelijke gezondheidsuitkomsten voor kwetsbare groepen met gelijke noden. Hiertoe werden vijf studies opgezet: een kwalitatieve studie, een literatuur-review en drie kwantitatieve studies. Vijf hypotheses werden geformuleerd, die elk behandeld worden in een paper: Hypothese 1: De specialist is de arts van de ‘rijkere’ mensen, de huisarts is de arts van de ‘armere’ mensen’ (paper 1). Hypothese 2: Daklozen zijn vaker geneigd om de spoeddienst te raadplegen dan de gemiddelde niet-dakloze populatie. Een universeel eerstelijnsgezondheidszorgsysteem met actieve toeleiding voor daklozen kan dit fenomeen tegen gaan (paper 2). Hypothese 3: Personen die niet in staat zijn rond te komen met hun maandelijks inkomen stellen vaker een bezoek aan de huisarts uit (paper 3). Hypothese 4: De sociale gradiënt in arts-patiënt communicatie blijft bestaan doorheen de tijd. Hierbij worden patiënten met een lage sociale status minder betrokken tijdens de consultatie, hebben minder controle en ontvangen een paternalistische communicatiestijl (paper 4). Hypothese 5: Artsen hebben een negatieve attitude tegenover patiënten met een lage sociale status. Als gevolg van deze attitude maken ze wijzigingen in hun klinische beslissingen. De uitkomsten van deze beslissingen zijn niet steeds in het voordeel van de patiënt (paper 5). De resultaten van deze thesis tonen aan dat: De specialist niet de arts is van de ‘rijkere’ mensen, maar de arts van de hoger opgeleiden. De huisarts is eveneens niet de arts van de ‘armere’ mensen, maar de 9 arts van de lager opgeleiden. Deze resultaten tonen aan dat het Belgische gezondheidszorgsysteem financieel toegankelijk is, maar toch nog barrières vertoont die gerelateerd zijn aan het opleidingsniveau van de patiënt. Mensen met een laag inkomen die een slechte gezondheid hebben, een zware depressie of zij zonder volledig vertrouwen in hun huisarts hebben een hoger risico om een bezoek aan de huisarts uit te stellen. Mensen die niet in staat zijn rond te komen met hun maandelijks inkomen daarentegen stellen niet vaker een bezoek aan de huisarts uit. Een universeel eerstelijnsgezondheidszorg systeem met actieve toeleiding door maatschappelijk werkers voor daklozen draagt bij tot toegankelijkere vorm van zorg voor daklozen. Deze aanpak leidt tevens tot minder stigmatisatie dan een selectieve aanpak. 10 De sociale gradiënt in arts-patiënt communicatie blijft bestaan doorheen de tijd. Hierbij ontvangen patiënten van een lagere sociale klasse minder socio-emotionele gesprekken, een directieve consultatie-stijl, ervaren ze minder controle over de consultatie, worden ze minder betrokken en ontvangen ze minder behandelingsinformatie en meer fysieke onderzoeken dan patiënten met een hogere sociale status. De attitude van artsen ten opzichte van patiënten met een lage SES varieert naargelang het aantal contacten artsen hebben met deze patiëntengroep. Artsen die slechts zorgen voor een relatief kleine hoeveelheid patiënten met een lage sociale status investeren meer tijd en energie in deze patiënten en zien het werken met hen als een positieve uitdaging. Deze artsen wijzigen hun beslissingsproces, maar deze wijzigingen worden gemaakt in het voordeel van de patiënt. Echter, wanneer het aantal patiënten met een lage SES te groot wordt in een praktijk, zijn artsen niet meer in staat om dezelfde kwalitatieve zorg te garanderen. De wijzingen in hun klinische beslissingen zijn daardoor niet altijd in het voordeel van de patiënt. De resultaten van deze thesis tonen aan dat het eerstelijnsgezondheidszorgsysteem in België geen al te grote financiële barrières creëert. Toch blijken er nog ongelijkheden te bestaan in de andere facetten van de toegang tot zorg (waaronder culturele en geografische toegang). Ook in de behandeling en de uitkomsten van zorg blijken er enkele ongelijkheden te zijn. Om deze ongelijkheden aan te pakken is er zowel actie nodig op het niveau van de praktijk (vb intersectorale samenwerking) als op beleidsniveau (vb extra steun voor praktijken die werken met een grote hoeveelheid lage SES patiënten). List of abbreviations SES Socio-economic Status WHO World Health Organization EFTA European Free Trade Association EU European Union PHC Primary Health Care AML Acute Myeloid Leukemia PCI Percutaneous Coronary Intervention OECD Organization for Economic Cooperation and Development CI Confidence Interval HIS Health Interview Survey ED Emergency Department GP General Practitioner FP Family Physician CRC ColoRectal Cancer CSDH Commission on Social Determinants of Health US United States UK United Kingdom DES Drug Electing Stents COPC Community Orientated Primary Care DWA Data Warehouse Arbeidsmarkt IMA InterMutualistisch Agentschap 11 Preface The Department of Family Medicine and Primary Health Care of Ghent University supports, respects and is guided by the overall mission statement of the Faculty of Medicine and Health Sciences. The department aims to contribute to the development of teaching, research and health policy development. It wants to promote and provide opportunities for participation in research and development of new forms of primary health care delivery in the city of Ghent, in the country of Belgium, in Europe and in developing countries. 12 Within the Department of Family Medicine and Primary Health Care, the research group “Health Equity” focuses its research activities on the (in)equity in health care. Within the scope of this commitment, the research group participated in the ‘Focal point on Welfare, Public Health and Family’ (Steunpunt Welzijn, Volksgezondheid en Gezin), a policy related research consortium funded by the Flemish Minister of Welfare, Public Health and Family (2007-2011). The research group in health equity was, in cooperation with LUCAS (KULeuven), responsible for a research project on the utilization of health care and welfare services of socially vulnerable people. The foundation of this thesis was laid in this Steunpunt and one of the papers included in this thesis uses data from the work of this project. It formed the basis of further research on (in)equity in health care use, reported in the additional papers. In concordance with the mission statement of the faculty, this doctoral thesis wants to give a better understanding on inequities in health care in Belgium and aims to identify key concepts for actions to reduce inequities in health care use. Ghent, 03.10.2012 Chapter 1 Introduction 13 14 Chapter 1: Introduction 1. Background Increase in life expectancy… Over the last centuries there is a marked improvement in life expectancy across countries. In 1843, a man living in Belgium had a life expectancy of 39.06 years, a woman of 39.59 years. In 2009, Belgian men had a life expectancy of 77.15 years and women of 82.43 years [1, 2]. Figure 1 illustrates that the life expectancy in Belgium has nearly doubled during the past century. Drops in life expectancy can be noticed in 1866 and 1871, due to the cholera- and smallpox-epidemics, and during the two World Wars. Between 1880 and 1940, life expectancy increased with two or three years per decade. In the last decade there is still a rise in life expectancy, however less steeply (table 1). Compared to 1998, in 2009 men lived nearly three years longer and women nearly two years longer [1]. 15 Source: Devos I. De evolutie van de levensverwachting in België, 18de-20ste eeuw. Gent: Academia Press; 2006 Figure 1: the evolution of life expectancy at birth in Belgium (1843-1996) Chapter 1: Introduction Figure 1 and table 1 show that women live longer than men. It is assumed that biological and sociological factors are important determinants [1]. Yet, in the last 10 years this gap in life expectancy between men and women seems to narrow (table 1). Source: FPS economie: http://statbel.fgov.be/nl/statistieken/cijfers/bevolking/sterfte_leven/tafels/ Table 1: the evolution of life expectancy at birth in Belgium: 1998-2009 … but not for all to the same extent 16 Despite this marked improvement in life expectancy, not all population groups enjoy this to the same extent. For example in Belgium, a 25 year old man with a higher educational level has on average another 55.03 years to live, a man with a higher secondary education degree has on average another 52.52 years to live, a man with a lower secondary education degree 51.33 years, a man with a primary school degree 49.29 years and one without any degree 47.56 years. For women, a similar gradient can be noticed (figure 2) [3]. 70 60 50 number of years to live 40 30 Men Women 20 10 0 Higher education Higher secondary education Lower secondary education Primary school degree No degree Educational level Source: Van oyen et al, 2010 Figure 2: life expectancy for men and women aged 25 in Belgium according to education Chapter 1: Introduction Also healthy life expectancy – the number of years a person can expect to live in good health- shows a similar pattern. A 25 year old Belgian woman without a diploma can expect to live another 29 years in good health, a woman with a primary school degree 32 years, a woman with lower secondary education degree 42 years, a woman with higher secondary education degree 41 years and a woman with a higher educational level 47 years [3]. Identical patterns are found in most industrialized countries. For example, in England, people living in the poorest neighborhoods (based on income level), will on average die seven years earlier than people living in the richest neighborhoods [4]. Even more striking is the difference in disability free life expectancy, which is in England 17 years between people in the richest and poorest neighborhood. This means that people living in the poorest neighborhoods not only die younger, they also spend more of their shorter lives with a disability [4]. Lower socio-economic status (SES) is not only associated with lower (healthy) life expectancy, and higher mortality rates, but with all kinds of health problems including psychological and mental disorders. For example in Europe, smoking and obesity are more common among people with a lower educational level [5]. In Belgium, lower educated persons suffer more often of chronic diseases such as diabetes, asthma, arthrosis, etc. than higher educated persons. Lower educated persons also suffer more from functional diseases such as migraine and irritable bowel syndrome and perceive higher levels of pain than higher educated persons [5-7]. These differences in health follow a social gradient. The social differences are not only between the rich and the poor, but there is a stepwise or linear decrease in health expectancy with decreasing social position [4, 8, 9] (see figure 2). Social justice Not all differences between population groups can be considered as inequities. Sometimes genetic and constitutional variations can be responsible for differences in health between population groups. For example the difference in health between younger and older people and some differences between ethnic groups (e.g. higher prevalence of sickle cell anemia in a black population) are purely the effect of genetics or natural aging. Next to these genetic and constitutional variations, there is also the factor chance or luck in people’s lives. (e.g. a car accident, having an infectious disease,…). For these unavoidable differences in health, usually the term “health inequalities” is used. The term “health inequities” is used to refer to systematic dif- 17 Chapter 1: Introduction ferences in health between social groups which are socially produced and therefore considered as unfair [4, 10, 11], preventable, avoidable and unjustifiable [12]. The concept of health inequity fits in the broader context of social justice. This refers to the organization of society in which a common good, to which all are expected to contribute in different ways, is available to all its members. To promote and respect social justice means to be part of a society where all members, regardless of their background, have basic human rights and equal access to their community’s wealth and resources [13]. Creating a social democracy comprises two normative ideas: personal accountability for personal choice and a sense of social commitment in economic behavior. These ideas should shape society’s institutions and the way economic agents behave within those institutions. It calls upon the talented and the powerful, as much as it calls upon the poor and the powerless [14]. A universal phenomenon 18 Variations in health according to social status is a global phenomenon, seen in low, middle and high income countries [9]. For example: in Bolivia, babies born to women without any education have an infant mortality greater than 10 per 100 live births, while the infant mortality rate - the risk of a baby dying between birth and one year of age- of babies born to mothers with at least secondary education is under 4 per 100 births [9]. Life expectancy at birth among indigenous Australians is substantially lower (59.4 for males and 64.8 for females) than that of non-indigenous Australians (76.6 and 82.0 respectively) [9]. The prevalence of long-term disabilities among European men aged 80+ is 58.8% among the lower educated versus 40.2% among the higher educated [9]. Yet, the extent of health inequities tends to vary importantly between countries. According to differences in mortality regarding to educational level, Mackenbach et al. systematically compared gradients in mortality inequity among men and women according to educational level by using individual information obtained by the Eurothine project from studies in 16 countries in the EU and European Free Trade Association (EFTA). The results of this project indicate that there is a great variation among these countries in levels of inequity in mortality, based on the length of education (figure 3). Inequity was greatest in the countries in central and eastern Europe, and least in Italy, Spain and Sweden [11]. Chapter 1: Introduction 19 Source: Mackenbach et al. Figure 3: absolute inequity (slope index of inequality) in male death rate by level of edu- cation in selected EU and EFTA countries The importance of equality in the distribution of income Important is that, when comparing countries, not the level of welfare but the extent of income inequality tends to associate with health inequity. Wilkinson states that in countries which reached a certain level of welfare, a smaller gap between rich and poor means a happier, healthier, and more successful population. Figure 4 shows that countries as the US, the UK, Portugal and New Zealand in the top right of the graph (being countries with greater income inequality) do much worse than Japan, Sweden or Norway in the bottom left. Countries with more equitable societies have among others better child well-being, higher levels of trust, lower levels of mental illness, longer life-expectancy, lower infant mortality rates and lower levels of obesity [15]. Chapter 1: Introduction 20 Source: Wilkinson & Pickett, The Spirit Level (2010) Figure 4: health and social problems are worse in more unequal countries Social determinants of health To understand what causes health inequity and how the phenomenon can be tackled, understanding what determines health is an important prerequisite. Many models have been developed to explain the processes and elements that contribute to the health of our populations. The model developed by Dahlgren and Whitehead is one of the most frequently used models to explain the determinants of health and is recognized by the WHO [16]. Based on a social ecological theory to map the relationship between the individual, their environment and disease, Dahlgren and Whitehead created a conceptual framework of the most important determinants of health (figure 5). Chapter 1: Introduction Source: Dahlgren and Whitehead, 1993 Figure 5: the determinants of health The centre of the figure comprises the fixed factors that influence a person’s health such as age, sex and constitutional characteristics. Surrounding them are the influences that are theoretically modifiable: the individual lifestyle factors such as smoking, physical activity, diet, and the social and community networks such as the interactions with peers and community members. The living and working conditions include aspects such as food supply, access to essential goods and services, housing, education,… Finally, general socio-economic, cultural and environmental conditions mediate population health. A strength of this model is that it emphasizes interactions: individual lifestyles are embedded in social norms and networks, and in living and working conditions, which in turn are related to the wider socioeconomic and cultural environment. These determinants can be positive health factors, protective factors or risk factors [16]. 21 Chapter 1: Introduction 2. Determinants of health inequities Knowledge of the social determinants of health is necessary but not sufficiently for identifying, analyzing and tackling the social inequities in health. One way to understand the determinants of health inequities is to focus on the distinct pathways and mechanisms by which the known health risk factors and risk conditions bring about the social gradients in health that are observed within different countries [17]. Five mechanisms or pathways are generally recognized to explain social inequities in health [16]. 1. Different levels of power and resources 22 A person’s social position in society has a powerful influence on the health risks experienced within different socio-economic groups. The better-off have more power and opportunities to live a healthy life than their worse-off counterparts. This implicates that social position itself is an important determinant of social inequities in health [16, 18]. People in a low socio-economic position experience more psychosocial stress, high job demands with low control, dire housing, financial difficulties,… These factors can lead to multiple disease outcomes by multiple risk-factor mechanisms. As an example, stressful life events have been linked to higher ratios of heart disease, stroke, diabetes, cancer, depression, fetal death, and low birth-weight in offspring [7, 16, 18, 19]. Also the psychosocial determinants of health such as lack of control in the workplace, low social capital and social exclusion are related to social position: these are more common among people from lower social classes compared to people from the higher social classes. Low social status is stressful because it reduces people’s control over their lives and work. As Wilkinson states “ It has a huge impact on health whether people feel valued, appreciated and needed or on the other hand looked down on, treated as insignificant, disrespected, stigmatized and humiliated” [16, 19, 20]. These forms of psychosocial stress can in their turn lead to ill health either through biological pathways (eg. affecting the immune system) or through behavioral pathways (eg. risk behavior) [16, 21]. 2. Different levels of exposure to health hazards Different exposure to risk factors can lead to the observed differences for most major diseases among socio-economic groups. Studies show that, for almost all risk factors, Chapter 1: Introduction exposure is inversely related to social position. The lower a person’s social position, the greater the exposure to different health hazards. People with most access to important social and economic resources are best able to avoid risks, diseases and the consequences of ill health [16]. As time changed, risk factors such as bad sanitation and infectious diseases were eradicated, but instead they were replaced by factors that include smoking, low exercise and poor diet. People of higher socio-economic status are less likely to start smoking and more likely to quit when they smoke [18]. Additionally, as the risk factors tend to increase with decreasing social class, healthy behaviors (such as breast-feeding, exercise, healthy diet…) tend to decrease with lower social class. Also environmental conditions have their impact. A report on environmental pollution in England and Wales showed that people living in deprived neighborhoods (based on income) are more often exposed to air pollution due to the higher number of polluting factories in their neighborhood [16, 22]. 3. The same level of exposure leading to differential impacts The same level of exposure to a certain risk factor may have different effects on different socio-economic groups. A Swedish study showed that similar levels of alcohol misuse cause two to three times more alcohol-related diseases and problems (such as psychological problems, divorce, low emotional control, contact with police and injuries) among lower social classes than among the higher social classes. One of the explanations is the difference in drinking patterns between the different groups, but also the level of social support people perceive -which functions as a protective factor-, was important [16, 23]. A 2002 study from Pope et al. showed increased risk for both cardiopulmonary and lung cancer mortality with less educated participants at the same level of air pollution [24]. The differences in impact of risk factors could be due to the fact that low socio-economic groups are more likely to be exposed simultaneously to several risk factors that reinforce each other. For example, smoking, bad housing, air pollution, suppressed immune system due to prolonged stress,…[19]. 4. Life-course effects The life-course effects include the accumulation of all the above mentioned pathways over a person’s life-time. Many life-events generate poor health later on, and material circumstances in early life are stronger predictors of health status in later life than social position during adulthood [16, 25, 26]. Life-course effects may be passed from generation to generation as they are closely related to social position. For example, 23 Chapter 1: Introduction the social position of parents has an influence on the education children receive, which in turn influences their job opportunities, working condition, and the salary of the children when they are grown [16, 26]. These life-course effects start from the womb on. A Belgian consumer organization studied the smoking behavior of 22.000 pregnant women. Smoking during the pregnancy can lead to a lower birth weight and a higher risk for chronic conditions (such as heart conditions) during adult life. Despite this fact 18% of the women kept smoking. There was a clear social gradient noticeable: 35% of lower socio-economic groups kept smoking compared to 7% of the women of higher social groups [25]. 5. Different social and economic effects of being sick 24 Poor health may have many adverse consequences for the life of individuals: job loss, loss of income, restrictions, social isolation,… At the same time, people with ill health often face additional financial problems due to high out of pocket payments for health care, medication, household help,… [16]. A Swedish survey shows that people from higher socio-economic groups with a limiting longstanding illness have better chances keeping their job than those from the lower socio-economic group with the same health problems [27]. The impact of this pathway depends partly on the health care system and welfare regime (e.g. social security) within a country [16]. For example, when a person in Belgium gets sick, he/she does not automatically lose his/her income thanks to the social security system in Belgium. Chapter 1: Introduction 3. Tackling health inequities As stated above, health inequities are a result of social inequities. Action on health inequities requires action across all the social determinants of health. Also, focusing solely on the most disadvantaged population groups will not reduce health inequities. To tackle these inequities, universal strategies for the entire population are needed, as well as strategies targeting certain subgroups [28]. This is known as proportionate universalism [4, 29]. 3.1 European policy actions Several important organizations addressed the importance of taking action across all the social determinants of health. The actions are global as well as within countries and between countries. In 2005, in the spirit of social justice, the Commission on Social Determinants of Health (CSDH) was set up by the World Health Organization (WHO) to marshal the evidence on what can be done to promote health equity and to foster a global movement to achieve it. The commission aims at supporting health policy changes by assembling effective practices that address the social determinants of health, making health equity a shared goal and building a sustainable global movement for action on health equity and social determinants. The Commission on Social Determinants of Health’s report, Closing the Gap in a Generation, published in August 2008, identified three main principles to reduce health inequalities: improve daily living conditions; tackle the unequal distribution of power, money, and resources; and measure and understand the problem and assess the results of action [13]. Professor Sir Michael Marmot was challenged to convert these three principles into practical policies that are appropriate for England and the UK in general. The report was published in 2010 as Fair society healthy lives, also known as the Marmot review, and emphasizes the importance of inequalities in daily life conditions and the fundamental drivers that give rise to them: inequities in power, money and resources [13]. The review states that health equity can become reality but only when health equity becomes a fundamental principle underlying national and local policies for health delivery. Six main policy objectives are outlined in the Marmot Review as crucial to achieving health equity: give every child the best start in life; enable all children, young people, and adults to maximise their capabilities and have control 25 Chapter 1: Introduction over their lives; create fair employment and good work for all; ensure a healthy standard of living for all; create and develop healthy and sustainable places and communities; and strengthen the role and impact of ill-health prevention [4] [30]. Although this report was written from a UK context, its message inspired many European policy makers. Based on this work and other work of the Commission on Social Determinants of Health, the WHO Regional Office for Europe commissioned a review for the European Region (review of the health divide and inequality in health in the WHO European Region) [31]. In the light of this review, Health 2020 was introduced as a collaborative initiative between EU Member States and their health-related institutions to strengthen existing evidence, expertise and support for action on achieving better health for the European Region at the year 2020 [11]. 26 The goals of Health 2020 are: to achieve better health for the European Region and its people; to increase equity in health and accelerate progress on achieving the right to health; to make health an endeavor for all in society; to enhance regional and global awareness of and action for health and the determinants of health; to develop suggested solutions, tools, evidence, guidance and partnerships that support health ministries, together with other stakeholders, in putting in place national policies, services and governance arrangements that realize their societies’ health potential on an equal basis [11]. Each EU member state must produce a National Reform Program each year in April, explaining what they are doing to move closer to the Europe 2020 targets. Flanders also produced its own Flemish Reform Program. Flanders in Action, the Flemish long term strategy and the Pact 2020, served as a broader policy framework. Furthermore, the Strategic Advisory Board on Welfare, Health and Family created in the light of 2020 a report on socially accountable care. Important concepts are the quality, performance, justice, relevance and access of the delivered care [32]. Chapter 1: Introduction Box 1: Health 2020 in Flanders The Flanders’ health care 2020 objectives are: • In 2020, Flanders will be able to boast of an adapted, accessible, affordable, and high-quality aid and care system. • Efficacy, efficiency, and quality of care from the perspective of the patient are the central points. • Primary health care and homecare are improved in 2020. • Establishing group-practices is being promoted. • Child care is available for minimum half of the children aged 3 years or younger. • There is a sufficient supply of care for the disabled, the elderly and people with mental health problems. In 2011 there was the first follow up measurement on some of these objec- tives. There was a decline in the accessibility and affordability of the medical and social care: data from the HIS 2008 showed that 11% of the households postponed a medical care visit due to financial reasons. This number has dou- bled since the HIS 2004. The percentage of group-practices in Flanders is stable: 33% of all practices in Flanders are group-practices. No statement can be made on the evolution of this number [33]. 3.2 Proportionate universalism To reduce the steepness of the social gradient in health, actions must be universal, but with a scale and intensity that is proportionate to the level of disadvantage. This is called proportionate universalism. Where policies have been designed with a focus on health inequities, they are often aimed at the lower end of the social gradient. Even if they are effective, these policies fail to tackle all the inequalities that exist for other socio-economic groups and even create “new vulnerable groups” [4]. Figure 6 shows that interventions focused on the lower social groups can reduce the differences between the lower and higher social groups. However, when these 27 Chapter 1: Introduction actions are not focused on all social groups in society, a new vulnerable population group is created. I II III IV V Figure 6: action is needed on all levels of society 28 Many of the key health behaviours significant to the development of chronic diseases follow a social gradient: smoking, obesity, lack of physical activity, unhealthy nutrition. Reducing health inequalities requires an amplified focus on these health behaviours in a way that ensures proportional universalism is being achieved. What will be required are population-wide interventions on smoking, alcohol and obesity to reduce the social gradient but targeted interventions may be needed to target particular groups. Often, the most advantaged groups are often better resourced to take advantage of population-wide interventions. Indeed, without attention to distributional impact and the underlying causes of behaviours, interventions to improve health may increase inequities. Wherever ill health prevention and health promotion take place, whether in primary care, a school or in a leisure centre, evaluations need to assess effectiveness across the social gradient. The impact of interventions need to be assessed by measurable indicators of reductions in health inequalities and separately assessed for cost-efficiency [34]. Chapter 1: Introduction 4. The importance of equitable primary health care This thesis focuses on equity in health care and was written with a special interest for primary health care. Therefore, in this part of the intro, we will emphasize on primary health care. 4.1 Defining primary health care Primary Health Care as defined by the World Health Organization on the Alma Ata declaration in 1978 is: “Essential health care; based on practical, scientifically sound, and socially acceptable method and technology; universally accessible to all in the community through their full participation; at an affordable cost; and geared toward self-reliance and self-determination. It forms an integral part both of the country’s health system, of which it is the central function and main focus, and of the overall social and economic development of the community. It is the first level of contact of individuals, the family and community with the national health system bringing health care as close as possible to where people live and work, and constitutes the first element of a continuing health care process” [35, 36]. Primary care addresses the most common problems in the community by providing preventive, curative and rehabilitative services to maximize health and well-being. It integrates care when more than one health problem exists, deals with the context in which illness exists and influences people’s responses to their health problems. It is care that organizes and rationalizes the deployment of basic and specialized resources directed at promoting, maintaining and improving health [37]. These definitions have been used to identify the four main features of primary care services: first-contact access for each new need; long-term person- (not disease) focused care; comprehensive care for most health needs; and coordinated care when it must be sought elsewhere [38]. 29 Chapter 1: Introduction Box 2: Primary care in Belgium In Belgium, GPs do not serve a gatekeeping role. Patients have free choice concerning the first physician to contact, can change physician at any time, get a second opinion, or even consult several physicians at a time. Moreover, they can directly access specialists or enter hospitals. The free choice of phy- sician is an important right granted to patients. Nevertheless, a large number of patients have a regular GP. The average number of physician contacts per person in Belgium is in line with the EU27 average (6.9 outpatient contacts per person in 2008, compared to the average of 6.86 in the EU27 [39, 40]. In Flanders 97.1% declares to have a regular GP, in the Walloon region this percentage is decreasing from 96% in 2004 to 94.4% in 2008 [41]. Physicians operate in solo- or group-practices. Apart from these solo- or group- practices there are integrated primary health care centers, such as me30 dical houses (maison médicale or Community Health Centres), which operate a multidisciplinary team, including several GPs, administrative and reception staff and nurses and sometimes also a physiotherapist and psychotherapists, social workers,…. The number of such practices is growing, although there is still only a small minority of people affiliated to them [39, 40]. According to the satisfaction with the delivered care, 96% of the patients in Flanders is satisfied with the delivered primary care of the GP, in the Walloon region, 92% is satisfied with the delivered primary care and in the Brussels region 91% is satisfied [41]. 4.2 Defining (in)equity in (primary) health care “No one should be denied access to life-saving or health-promoting interventions for unfair reasons, including those with economic or social causes” Margaret Chan, WHO Director-General [42]. There is a higher prevalence of chronic diseases such as diabetes, asthma, chronic obstructive pulmonary disease, heart disease,… among persons from lower socioeconomic groups, leading to higher health care need [43-45]. Tudor Hart captured Chapter 1: Introduction the idea of inequity in health care use in the Inverse Care Law, stating that “the availability of good medical (primary and specialist) care tends to vary inversely with the need for it in the population served”. So those with the greatest need for care, often have the poorest access [46, 47]. Sometimes this inequity in health care is very clear (e.g. low social status is associated with worse quality of diabetes care) [48], but in other cases, it can be very complex. For example, when health status is taken into account, the utilization of GP is equitable (does not vary across income or socio-economic groups) or propoor (favoring the lower socio-economic groups), whereas specialist care tends to favor the better off (higher income and better educated groups) [44, 49, 50]. This phenomenon appears to be universal, but is reinforced when private insurance or private care options are offered [44, 49]. Also social differences in postponing health care have been described. Studies suggest that disadvantaged people are at a higher risk for delaying care for medical conditions. These untreated conditions may eventually lead to hospitalization or emergency department use [51]. 4.2.1 Equity in health care Striving for equity in health care is one of the primary goals of the WHO [52]. Equity in health care means that patients who are alike should be treated in the same way (horizontal equity) and that patients who are unlike, should be treated in the same unlike way (vertical equity) [53-56]. Health care systems contribute most to improving health and equity when the institutions and services are organized around the principle of universal coverage (extending the same scope of quality services to the whole population, according to needs and preferences, regardless of ability to pay) [13]. Essential in an equitable health care system are: (1) equal access for equal need, (2) equal treatment for equal need and (3) equal outcomes for equal need [10]. 4.2.2. Equal access for equal need It is a human right to have access to effective health care. But around the world, access is inadequate for some groups or may even be unattainable for all practical purposes [10, 51]. According to Goddard et al. (2001), ‘access’ refers –at the most general level- to “the ability to secure a specified range of services, at a specified 31 Chapter 1: Introduction level of quality, subject to a specified maximum level of personal inconvenience and costs, whilst in the possession of a specified level of information” [56]. This definition indicates that access to health services has many aspects. Based on literature we can distinguish geographical access, cultural access and economic access to care. Geographical access Geographical accessibility enholds the availability and distance to health care providers (in terms of distance and time). The location of the GP to the place of residence has a major influence on the accessibility to the doctor for the patient, both directly (e.g. time to get to the doctor) and indirectly (by affecting travel mode and through interactions with factors such as family and work commitments). Aside from the distance to the doctor, also personal mobility such as car ownership, access to a car and use of alternative methods of transport (e.g. public transportation) is important. Non-car owners more often make use of home visits, the same results appear for not having access to a car [57, 58]. 32 The study of Hart in the coal-mining valleys of Wales is a good illustration of geographic access. The miners suffered from occupational disabilities (such as pulmonary diseases). However, the health care facilities and the quality of the provided care were much worse in the areas where these miners lived compared to other regions with lower morbidity rates [46]. But then there is also the doctor’s side. Doctors in deprived areas, where there are higher sickness and death rates have to work harder: more out-of-hour contacts and more same day consultation with less support (hospitals far away, transportation problems…) and a higher load of longer-distance house-visits [46, 47, 57-59]. Cultural access Cultural access relates to acceptability and respect. Language barriers, miscommunication between health care provider and patient due to different priorities, receiving an unequal amount of attention or differences in treatment outcomes are often the result of cultural access [10]. Physicians may have a harder time understanding symptoms or they might even misunderstand the symptoms of vulnerable patients due to language barriers or miscommunication. Sometimes physicians may hold unconscious racial or ethnic biases that influence their interactions with minority patients [10, 60-63]. Studies have shown that minority patients often benefit less from treatment and as a result may demand less care [64-66]. A study Chapter 1: Introduction by Campbell et al. reported that non-white-ethnic minority respondents reported less favorable assessments on almost all of the 13 dimensions of care examined. Differences were highest in the performance of the reception staff, accessibility of care and trust in the GP [67]. Shortcomings in respect and dignity to certain groups is another example of cultural access. Several studies showed that homeless people often postpone health care because they feel labeled, stigmatized or considered as “problem patients” [68, 69]. But this is not only the case for homeless people. Van Ryn et al. found that physicians tend to perceive members of low and middle SES groups more negatively on personality characteristics (lack of self-control, irrationality) and level of intelligence. They also perceive low SES patients as people with little compliance, less commitment, no career demands and no desire for a physically active lifestyle [61]. A common issue in these patient groups are the competing priorities. Homeless people postpone more often their health care because of these competing priorities: e.g. finding housing or employment, getting through the day, feeding their children etc. is more important than caring about their own health [70-72]. Furthermore, there is a gap of awareness and understanding of the day-to-day restrictions of low-income or homeless people by health care workers [10, 65]. People from deprived areas in Scotland prefer holistic GPs who understand the realities of life in such areas and whom they can trust as both competent and genuinely caring. Without this, they may judge their GP as socially distant and emotionally detached [73]. Also, there may be differences in the attention that patients from less advantaged backgrounds receive. This can lead to differences in the quality of care [74]. Economic access What use is there in having health care services on your doorstep when you cannot afford to use them? The strongest example of the problem of economic access is when people in need of emergency care are turned away from a clinic or hospital and left to die because they cannot afford to pay [10]. Although this extreme practice is very rare in Europe, there are an increasing number of people postponing medical care due to financial reasons [10, 75]. In Europe, 4.5% of the low income people cannot afford medical care compared to 0.5% of the high income people. In Belgium, for 2008, 1.5% of the 20% low income people cannot afford medical care, compared to 0.1% of the 20% high income people [76]. Data from the Belgian Health survey show that in 2008, 13.7% of the households postponed medical care 33 Chapter 1: Introduction due to financial reasons. There is a clear social gradient: 18.1% of the households with a low educational level postponed care, compared to 17.6% of those with a lower secondary school education, 13.4% of those with a higher secondary education and 9.4% with a high educational level [6]. In Belgium, 7% of the total monthly household income is spent on medical care. As a consequence, 35% of the Belgian households report that health care costs are a large burden on their income [6, 77]. Another aspect of economic access is, when people have to pay the out-of-pocket fees, and get into financial troubles or debts as a result of it. Whitehead and Dahlgren call this the medical poverty trap: impoverishment caused by paying for medical care. This is most frequent in the developing world, but the medical poverty trap is also present in Eastern Europe and the United States [10, 78]. Box 3: Financing of the Belgian health care system Belgium has a nearly universal health care system. Patients can choose their physicians freely and have direct access to any GP or specialist at any time, for 34 any problem. Almost 99% of the Belgian population is covered by the compulsory health insurance [39]. The main payment system used in Belgium is the fee-for-service system. In this system, payment is made for units of service: physicians are paid an amount of money for each type of service they provide. This payment system has the intention to allow physicians to respond in a flexible way to their patient’s needs and demands [79, 80]. The patient pays the set fee for the consultation directly to the GP, and patients are then partly reimbursed by their sickness funds. Co-payments vary from service to service. Co-payments rates are about 25% for GP consultations, 35% for GP home visits and 40% for specialist consultations. For certain groups of patients, among the lower socio-economic groups, there is a reduced co-payment of about 10% for the GP consultation and about 15% for specialist consultation [39]. The total amount of co-payment per year is limited by an income-related thres- hold called the maximum billing: as soon as expenses reach the set ceiling, any further health care costs are fully covered by the health insurance fund for the remaining part of the year [39]. Furthermore, for people with financial dif- ficulties, there is the third-party system. Physicians using the third-party sys- tem, get paid by the sickness fund and the patient only has to pay user charges Chapter 1: Introduction (co-payments) to the physician [39]. Since 1982, the legal possibility to work in a capitation system in Belgium was put into practice. The patient has to be registered with a GP or a group of primary care providers and does not have to pay for a consultation. This method of payment is also meant to eliminate the financial barrier in visiting primary care [79]. Currently 2% of the Belgian population is cared for in the capitation system [39]. In the capitation system, the unit that is paid for is the patient. The health care provider is paid a specified sum of money by the insurance companies for the ongoing care of a patient for a particular period over time no matter how often this patient uses this service [79]. The total number of group practices using a capitation system increased from 53 in 2003 to 99 in 2009 [39, 40]. A Belgian cohort study showed that patients visiting group practices working with capitation system are generally more vulnerable (low income, low social class). Patients visiting these group practices have lower specialist referrals and therefore lower expenses for specialist care and received a better follow up on preventive interventions. The quality of care delivered from practices working with capitation compared to practices working on a fee-for-service base was the same [81]. 4.2.3. Equal treatment for equal need In an equitable health care system, patients should receive the same health care treatment independent from their age, sex, income,.. [54, 82]. However evidence shows that higher-income groups receive higher quality-services than lower-income people [5, 50, 83]. A systematic literature review found that low socio-economic position, measured by income or education, is associated with receiving fewer childhood and influenza immunizations and diabetic eye examinations, later enrollment in prenatal care, lower quality ambulatory and hospital care and fewer cervical smear tests, mammograms [84] For example, in the US, almost 80% of women below the poverty level never had a mammogram, compared with 50% of women in the highest income category. For ever having had a Pap smear, the percentage varies from less than 1% in well-served groups to more than 40% in very poorly served groups [85]. A US study found that low income people received less chronic and preventive care than their more-affluent counterparts, however they did not find 35 Chapter 1: Introduction differences in treatment according to educational level [86]. Le et al found that patients from lower social class, suffering from colorectal cancer (CRC) were more often refused surgery than patients from higher social classes [87]. ‘Treatment’ is not only about the medical procedures followed but also the attitude and the interaction with the health care provider is essential in giving equal treatment. A systematic review of literature found that patients from lower social class received less socio-emotional talk, a more directive and a less participatory consulting style characterized by for example less involvement in treatment decisions, lower patient control over communication, not explaining things in a way patients understand, less diagnostic and treatment information and more physical examination [88, 89]. Bernheim et al. found, by means of qualitative research that physicians undertook changes in their management decisions (for example, referral to hospitals, additional tests,…) depending on the social class of the patient [74]. 4.2.4. Equal outcomes for equal need 36 When people receive the same access and same treatment in health care, they should have the same outcomes when they are in similar needs. However, differences in treatment outcomes between different socio-economic groups may be observed despite the fact that the same treatment was given [8, 10, 90]. A Norwegian study found an overall negative socio-economic gradient in cancer survival [91]. A UK population based study found differences in incidence and mortality from Acute Myeloid Leukemia (AML): poorer survival in AML patients from low socio-economic classes despite the same treatment as patients from higher socio-economic classes [92]. Shimony et al investigated the impact of SES on the outcomes following percutaneous coronary intervention (PCI). Previous studies found no association between SES and cardiovascular outcomes [93] or found that unemployment and a deprived status were associated with poorer quality of life indices one year after PCI, however, clinical outcomes were not taken into account [94]. In their study they found that low SES was associated with less use of drug eluting stents (DES), lower adherence to aspirin and to clopidogrel and higher rates of repeat revascularization and recurrent myocardial infarction [95]. A US study on the health (care) outcomes of epilepsy patients of different socio-economic populations found that low SES patients had a greater likelihood of having uncontrolled seizures, drugrelated side effects, to be stigmatized and have a lower overall quality of life [96]. Chapter 1: Introduction 4.3 A role for Primary Health Care in tackling health inequities While equity marks one of PHC’s boldest features, it is one of the areas where results have been most uneven and where the premium for more effective reforms is perhaps the greatest. Out-of-pocket payments for health care are but one of the sources of health inequity. Deeply unequal opportunities for health combined with endemic inequalities in health care provision lead to pervasive inequities in health outcomes [52]. It takes a wide range of interventions to tackle the social determinants of health and make health systems contribute to more health equity. One of the interventions is building a sustainable strong Primary Health Care system [38, 97]. Improving population health International studies show that the strength of a country’s primary care system is associated with improved population health outcomes for all-cause mortality, all-cause premature mortality and cause-specific premature mortality from major respiratory and cardiovascular diseases [98]. This relationship is significant after controlling for determinants of population health at the macro-level (GDP per capita, total physicians per one thousand population, percentage of elderly) and micro-level (average number of ambulatory care visits, per capita income, alcohol and tobacco consumption) [36, 98]. Furthermore, PHC characteristics such as geographic regulation, longitudinality, coordination and community orientation were associated with improved population health. This reinforces findings of an earlier international comparison involving 11 developed countries which demonstrated that a higher primary care orientation of a health system was more likely to produce better population health outcomes, at lower cost, and with greater user satisfaction [36-38] . But also the relationship between a primary care provider and a patient has an impact on the health outcomes. People who identify a primary care physician as their usual source of care are healthier, regardless of their initial health or various demographic characteristic [38, 99]. Disparities in health outcomes When reviewing the impact of primary care on reducing the inequities in health, Starfield et al. found that higher ratios of primary care physicians to population are associated with relatively greater effects on various aspect of health in more socially 37 Chapter 1: Introduction deprived areas (measured by high levels of income inequity). In the United States a study on county level showed that the adverse impact of income inequity on all-cause mortality, heart disease mortality and cancer mortality was considerably diminished in counties where the number of primary care physicians in county-level analyses was high [100]. The equity-related effect of having a good primary care source was also found in a study that examined the degree of primary care-oriented services that people received. Good primary care experiences were associated with reductions in the adverse effects of income inequality on health, with fewer differences in self-rated health between higher and lower income-inequality areas where primary care experiences were stronger [101]. Cost-effectiveness 38 Health systems dominated by specialists, such as that of the United States, have higher total costs and reduced access to health care by vulnerable populations [36, 102, 103]. The high cost is attributed to proportionately low number of primary care physicians and consequent impairment of the gate-keeping function [104, 105]. Shifting care across specialist-general practice and secondary-primary care boundaries is possible and has been shown to be cost effective without an adverse effect on outcomes. For example: GP-led hospitals in Norway provided health care at lower cost compared to alternative modes of care, due to averted hospital costs [106]. Several studies show that care delivered by GPs, compared to hospital specialists, in hospital-based accident and emergency departments is more cost effective with lower use of diagnostic investigations, lower referral rates to secondary services, lower prescription levels and no significant difference in patient satisfaction or health outcomes [107-109]. Quality, efficiency and satisfaction of care Primary care physicians are more likely than specialists to provide continuity and comprehensive care resulting in improved health outcomes [110]. Better access to primary care physicians leads to less hospitalization, less utilization of specialist and emergency centres and less chance of being subjected to inappropriate health interventions [36]. In contrast, when direct access to specialists is possible without a controlling mechanism by primary care physicians, the quality of care, as measured by appropriateness, worsens and health care costs increase [111]. Chapter 1: Introduction However, it is also important to consider the possible adverse effects of a strong primary health care system. First of all, the expansion of primary health care services may not always reduce costs because it may end up identifying previously unmet needs, improves access, and tends to expand service utilization [36]. In a complex healthcare system, every organizational structure, like gatekeeping, may have unwanted effects [112]. One of the adverse effects of the gatekeeping system are the long waiting times. For the UK, it has been stated that 500-10.000 patients with cancer die each year due to delays in diagnosis resulting from the long waiting times [113]. Further, a gatekeeper system may delay advanced diagnostic tests. GPs are often not licensed to refer patients to the appropriate diagnostic tests, so the patients first need to be referred to a specialist before receiving the appropriate tests (such as imaging technologies) [112]. Finally it is important not to define the role of primary care in isolation but in relation to the other aspects of the health care system. Primary and secondary care all have important roles in the health system. They are not mutually exclusive, but rather necessary ingredients for a good working system [36]. 4.3.1 The mechanisms through which primary care can tackle the inequities in health We have described above that people from lower socio-economic groups are disproportionately affected by many of the various barriers that may hinder health service access, treatment and outcomes. A lot of health may be gained by tackling these inequities. Health professionals are well placed as health advocates to influence social determinants of health and to tackle health inequalities: yet their role is often under-utilised. They are involved in direct patient care, are in contact with people over longer periods in their life- e.g. general practitioners and dentists-, at key points across their life course- e.g. midwives, health visitors, school nurses, nursing home staff-, or at times of illness- primary, secondary or even tertiary care teams [97]. In primary care, shifting service provision away from vertical, disease oriented programs towards horizontal community-oriented programs promises to yield results. There is also great potential for healthcare professionals to support and empower communities and individuals taking more control over their health and well-being, especially when they join forces with local health trainers, community health champions and community development work [97]. 39 Chapter 1: Introduction Literature shows that people from lower social classes receive lower levels of qualitative health care, fewer preventive tests and tend to postpone health care [19, 84, 114]. Primary health care can help to tackle these problems. Based on the available evidence, Starfield et al. proposed several mechanisms that may account for the beneficial impact of primary care on tackling the health inequities. • • 40 • Primary care adds to a better quality of care Primary care physicians do at least as well as specialists in caring for common diseases, and they do better overall when the measures of quality are generic (e.g. not focused on specific diseases). For less common conditions, the care provided by primary care physicians with appropriate backup from specialists may be the best; for rare conditions, appropriate specialist care is needed because primary care physicians do not see these conditions frequently enough [38]. Primary care has a greater focus on prevention Evidence shows that “generic” preventive interventions – these are preventions not related to a disease or an organic system- are best implemented in primary care. Interventions on breast-feeding, smoking cessation, diet, physical exercise,…seem to have the best impact when they are implemented in primary care [38, 115]. Furthermore a greater supply of family physicians is associated with an earlier detection of breast cancer, colon cancer, cervical cancer and melanoma [38, 116-118]. Primary care adds to the early management of health problems Primary care is able to manage health problems before they become too serious and require hospitalization or emergency services [38]. In the UK, each 15-20% increase in the supply of primary care physicians was associated with a decrease in hospital admission rate of about 14 per 100.000 for acute illnesses and about 11 per 100.000 for chronic illnesses, even after controlling for the degree of social deprivation, ethnicity and limiting long-term illness [38, 119]. Furthermore, low social groups have a higher tendency to visit the emergency department and perceive several barriers towards the health care system [50, 120]. Also here, primary care can make a contribution: • Primary care increases access to health services for relatively deprived population groups Chapter 1: Introduction • Primary care functions as the point of first contact with health services and facilitates the entry to the rest of the system. With the exception of the United States, most industrialized countries have achieved universal and equitable access to primary health services [121]. As a results, in the United States, socially deprived population subgroups are more likely than more advantaged people to lack a regular source of care [38]. Primary care reduces unnecessary and potentially harmful specialist care Studies from the United States show that increasing the supply of specialist care either has no effect or an adverse effect on major health outcomes. The adverse effects of seeking specialist care as a point of first contact can be explained by a strong theoretical basis. Specialists are trained in a hospital setting and the patients they see are not representative for patients in community settings. Patients consulting a specialist often have a more specific and more serious disease that needs additional diagnostic tests. The result is that specialist practices as a point of first contact overestimate the likelihood of illness in the patients they see, with the consequently inappropriate use of diagnostic and therapeutic modalities, both of which raise the likelihood of adverse effects [38]. Finally, there is a positive effect of the accumulation of the main primary care characteristics. The beneficial effects of primary care on mortality and morbidity can be attributed, at least in part, to the person-focused aspect of primary care rather than on the management of particular diseases. Primary care focuses on achieving better outcomes for health in all of its aspects rather than on the procedures directed at improving the processes or outcomes of care for a particular condition or disease. Another important characteristic of primary care is its role as the first contact for patients when a problem develops. Having a relationship with a GP who can function as an initial point of contact is strongly associated with less use of specialists and emergency rooms. Furthermore, continuity of care, which means that a person uses his/her GP as a source of care over time for most of his/her health care needs, is associated with greater satisfaction, better compliance and lower hospitalization and use of emergency department. On the other hand, very short-term relationships with physicians are associated with poor health outcomes [38]. 41 Chapter 1: Introduction References 1. Devos I: De evolutie van de levensverwachting in België, 18de-20ste eeuw. Gent: Academia Press; 2006. 2. Levensverwachting Belg bedraagt net geen 80 jaar [http://www.belgium. be/nl/nieuws/2011/news_levensverwachting-belg.jsp] 3. Van Oyen H, Deboosere P, Lorant V, Charafeddine R, Demarest S, Portet F: Programma Samenleving en Toekomst. Eindverslag- deel “Synthese van het onderzoek”. In. Brussel: Federaal Wetenschapsbeleid; 2010. 4. Marmot M, Allen J, Goldblatt P, Boyce T, McNeish D, Grady M, Geddes I: Fair Society, Healthy lives. The Marmot review. Executive Sumary; 2010. 5. Mackenbach JP, Stirbu I, Roskam AJ, Schaap MM, Menvielle G, Leinsalu M, Kunst AE: Socioeconomic inequalities in health in 22 European countries. N Engl J Med 2008, 358(23):2468-2481. 6. Drieskens S, Van der Heyden J, Hesse E, Gisle L, Demarest S, Tafforeau J: Gezondheidsenquête België, 2008. Rapport V - Sociaal-economische ongelijkheden in gezondheid. In.; 2010. 7. Link BG, Phelan J: Social conditions as fundamental causes of disease. J Health Soc Behav 1995, Spec No:80-94. 8. Whitehead M, Dahlgren G: Concepts and principles for tackling social inequities in health: Levelling up Part 1. 2006. 9. Social determinants of health. Key Concepts. [http://www.who.int/social_ determinants/thecommission/finalreport/key_concepts/en/index.html] 42 10. Whitehead M, Dahlgren G: Levelling up (part 1): a discussion paper on concepts and principles for tackling social inequities in health. In.: WHO collaborating Centre for Policy Research on Social Determinants of Health, University of Liverpool; 2006. 11. Interim second report on social determinants of health and the health divide in the WHO European Region. In.: Regional Committee for Europe; 2011. 12. Chang WC: The meaning and goals of equity in health. J Epidemiol Community Health 2002, 56(7):488-491. 13. CSDH: Closing the gap in a generation: health equity through action on the social determinants of health. Final Report of the Commission on Social Determinants of Health. In. Geneva: World Health Organisation; 2008. Chapter 1: Introduction 14. Vandenbroucke F: Social Justice and Individual Ethics in an Open Society. Equity, Responsibility and Incentives.: Springer; 2001. 15. Wilkinson RG, Pickett KE: The Spirit level. Why Equality is better for everyone: Penguin Books; 2010. 16. Dahlgren G, Whitehead M: Levelling up (part 2): a discussion paper on European strategies for tackling social inequities in health; 2006. 17. Diderichsen F, Evans DB, Whitehead M: The social basis of disparities in health. In: Challenging inequities in health - from ethics to action. edn. New York: Oxford University Press; 2001. 18. Link BG, Phelan JC: Understanding sociodemographic differences in health--the role of fundamental social causes. Am J Public Health 1996, 86(4):471-473. 19. Wilkinson RG, Pickett KE: Income inequality and population health: a review and explanation of the evidence. Soc Sci Med 2006, 62(7):1768-1784. 20. Wilkinson RG: the impact of inequality: how to make sick societies healthier. London: Routledge; 2005. 21. Health inequalities: Europe in profile. An independent expert report commissioned by and published under the auspices of the United Kingom Presidency of the European Union [http://ec.europa.eu/health/ph_determinants/socio_economics/documents/ev_060302_rd06_en.pdf ] 22. McLaren D, Cottray O, Taylor M, Pipes S, Bullock S: The geographic relation between household income and polluting factories. In. Edited by Earth Fot; 1999. 23. Hemmingsson T, Lundberg I, Diderichsen F, Allebeck P: Explanations of social class differences in alcoholism among young men. Soc Sci Med 1998, 47(10):1399-1405. 24. Pope CA, 3rd, Burnett RT, Thun MJ, Calle EE, Krewski D, Ito K, Thurston GD: Lung cancer, cardiopulmonary mortality, and long-term exposure to fine particulate air pollution. JAMA 2002, 287(9):1132-1141. 25. Eriksson JG, Forsen T, Tuomilehto J, Winter PD, Osmond C, Barker DJ: Catch-up growth in childhood and death from coronary heart disease: longitudinal study. BMJ 1999, 318(7181):427-431. 26. Power C, Matthews S: Origins of health inequalities in a national population sample. Lancet 1997, 350(9091):1584-1589. 27. Lindholm C, Burstrom B, Diderichsen F: Class differences in the social consequences of illness? J Epidemiol Community Health 2002, 56(3):188192. 43 Chapter 1: Introduction 28. Baum F: Cracking the nut of health equity: top down and bottom up pressure for action on the social determinants of health. Promot Educ 2007, 14(2):90-95. 29. Marmot M: Status Syndrome. How your social standing directly affects your health and life expectancy: Bloomsbury; 2004. 30. Health equity - an election manifesto? The Lancet 2010, 375(9714):525. 31. Marmot M: Interim first report on social determinants of health and the health divide in the WHO European Region. In: European Social Determinants and Health Divide Review. World Health Organization Regional office for Europe; 2010. 32. SARWGG: Visienota Maatschappelijk Verantwoorde zorg. In. Brussels; 2011. 33. Vlaanderen in actie. PACT 2020. Levenskwaliteit [http://vlaandereninactie. be/actie/pact-2020/levenskwaliteit/] 34. Annual report 2010/11. In. Edited by Health SDoP; 2010. 44 35. Declaration of Alma-Ata. In: International Conference on Primary Health Care. USSR; 1978. 36. What are the advantages and disadvantages of restructuring a health care system to be more focused on primary care services? In. Edited by (HEN) WROfEsHEN: World Health Organization Europe; 2004. 37. Starfield B: Primary Care. Concept, evaluation and policy. New York: Oxford University Press; 1992. 38. Starfield B: Contribution of Primary Care to Health Systems and Health. The Milbank Quarterly 2005, 83(3):457-502. 39. Gerkens S, Farfan M, Desomer A, Stordeur S, De Waroux M, Van de Voorde C, Van de Sande S, Leonard C: Het Belgische Gezondheidssysteem in 2010. In. Edited by research HS. Brussel: Federaal Kenniscentrum voor de Gezondheidszorg (KCE); 2010. 40. Gerkens S, Merkur S: Belgium. Health System Review, vol. 12; 2010. 41. Van Der Heyden J: Patiënttevredenheid. Gezondheidsenquête, België 2008. In. Brussel: Wetenschappelijk Instituut Volksgezondheid; 2010. 42. Speech to the International Federation of Red Cross and Red Crescent Societies Global Health and Care Forum 2008: Primary care starts with people [http://www.who.int/dg/speeches/2008/20080514/en/index.html] Chapter 1: Introduction 43. Droomers M, Westert GP: Do lower socioeconomic groups use more health services, because they suffer from more illnesses? Eur J Public Health 2004, 14(3):311-313. 44. Regidor E, Martinez D, Calle ME, Astasio P, Ortega P, Dominguez V: Socioeconomic patterns in the use of public and private health services and equity in health care. BMC Health Serv Res 2008, 8:183. 45. Westert GP, Satariano WA, Schellevis FG, van den Bos GA: Patterns of comorbidity and the use of health services in the Dutch population. Eur J Public Health 2001, 11(4):365-372. 46. Hart JT: The inverse care law. Lancet 1971, 1(7696):405-412. 47. Carlisle R, Avery AJ, Marsh P: Primary care teams work harder in deprived areas. J Public Health Med 2002, 24(1):43-48. 48. Baz L, Muller N, Beluchin E, Kloos C, Lehmann T, Wolf G, Muller UA: Differences in the quality of diabetes care caused by social inequalities disappear after treatment and education in a tertiary care centre. Diabetic medicine : a journal of the British Diabetic Association 2012, 29(5):640-645. 49. van Doorslaer E, Masseria C, Koolman X: Inequalities in access to medical care by income in developed countries. CMAJ 2006, 174(2):177-183. 50. Van der Heyden JH, Demarest S, Tafforeau J, Van Oyen H: Socio-economic differences in the utilisation of health services in Belgium. Health Policy 2003, 65(2):153-165. 51. Weissman JS, Stern R, Fielding SL, Epstein AM: Delayed access to health care: risk factors, reasons, and consequences. Ann Intern Med 1991, 114(4):325-331. 52. WHO report 2008: Primary health care now more than ever. In.; 2008. 53. Culyer AJ: Equity - some theory and its policy implications. Journal of medical ethics 2001, 27(4):275-283. 54. Bayoumi AM: Equity and health services. Journal of public health policy 2009, 30(2):176-182. 55. Hanafin S, Houston AM, Cowley S: Vertical equity in service provision: a model for the Irish public health nursing service. J Adv Nurs 2002, 39(1):68-76. 56. Goddard M, Smith P: Equity of access to health care services: theory and evidence from the UK. Soc Sci Med 2001, 53(9):1149-1162. 57. Knox PL: The accessibility of primary care to urban patients: a geographical analysis. J R Coll Gen Pract 1979, 29(200):160-168. 45 Chapter 1: Introduction 58. Field KS, Briggs DJ: Socio-economic and locational determinants of accessibility and utilization of primary health-care. Health Soc Care Community 2001, 9(5):294-308. 59. Leira EC, Hess DC, Torner JC, Adams HP, Jr.: Rural-urban differences in acute stroke management practices: a modifiable disparity. Arch Neurol 2008, 65(7):887-891. 60. van Ryn M, Burgess D, Malat J, Griffin J: Physicians’ perceptions of patients’ social and behavioral characteristics and race disparities in treatment recommendations for men with coronary artery disease. Am J Public Health 2006, 96(2):351-357. 61. van Ryn M, Burke J: The effect of patient race and socio-economic status on physicians’ perceptions of patients. Soc Sci Med 2000, 50(6):813-828. 62. Burgess DJ, van Ryn M, Crowley-Matoka M, Malat J: Understanding the provider contribution to race/ethnicity disparities in pain treatment: insights from dual process models of stereotyping. Pain Med 2006, 7(2):119134. 46 63. De Maesschalck S: Linguistic and cultural diversity in the consultation room: a tango between physicians and their ethnic minority patients. Ghent: Ghent University; 2012. 64. Balsa AI, McGuire TG: Statistical discrimination in health care. J Health Econ 2001, 20(6):881-907. 65. Doescher MP, Saver BG, Franks P, Fiscella K: Racial and ethnic disparities in perceptions of physician style and trust. Arch Fam Med 2000, 9(10):1156-1163. 66. Burgess DJ, Fu SS, van Ryn M: Why do providers contribute to disparities and what can be done about it? J Gen Intern Med 2004, 19(11):1154-1159. 67. Campbell JL, Ramsay J, Green J: Age, gender, socioeconomic, and ethnic differences in patients’ assessments of primary health care. Qual Health Care 2001, 10(2):90-95. 68. Martins DC: Experiences of homeless people in the health care delivery system: a descriptive phenomenological study. Public Health Nurs 2008, 25(5):420-430. 69. Riley AJ, Harding G, Underwood MR, Carter YH: Homelessness: a problem for primary care? Br J Gen Pract 2003, 53(491):473-479. 70. Kushel MB, Vittinghoff E, Haas JS: Factors associated with the health care utilization of homeless persons. JAMA 2001, 285(2):200-206. 71. Hwang SW: Homelessness and health. CMAJ 2001, 164(2):229-233. Chapter 1: Introduction 72. Gelberg L, Gallagher TC, Andersen RM, Koegel P: Competing priorities as a barrier to medical care among homeless adults in Los Angeles. Am J Public Health 1997, 87(2):217-220. 73. Mercer SW, Cawston PG, Bikker AP: Quality in general practice consultations; a qualitative study of the views of patients living in an area of high socio-economic deprivation in Scotland. BMC Fam Pract 2007, 8:22. 74. Bernheim SM, Ross JS, Krumholz HM, Bradley EH: Influence of patients’ socioeconomic status on clinical management decisions: a qualitative study. Ann Fam Med 2008, 6(1):53-59. 75. Ziglio E, Barbosa R, Charpak Y, Turner S: Health Systems confront poverty. Public Health Case Studies, no. 1. In. Edited by Organization WH. 76. Dierckx D, Van Herck N, Vranken J: Armoede en sociale uitsluiting in Europa en België. In: Armoede in België. edn. Edited by OASeS. Leuven/ Den Haag: Acco; 2010: 37-60. 77. Drieskens S, Van der Heyden J, Hesse E, Gisle L, Demarest S, Tafforeau J (eds.): Gezondheidsenquête België, 2008. Rapport IV - Gezondheid en Samenleving; 2010. 78. Whitehead M, Dahlgren G, Evans T: Equity and health sector reforms: can low-income countries escape the medical poverty trap? Lancet 2001, 358(9284):833-836. 79. Physicians funding and health care systems - an international perspective. In: A summary of a conference hosted by the WHO, WONCA and RCPG at St John’s College, Cambridge. Edited by Brown PS. 80. De Maeseneer J, Willems S, De Sutter A, Van de Geuchte M, Billings M: Primary health care as a strategy for achieving equitable care: a literature review commissioned by the Health Systems Knowledge Network. In.; 2007. 81. Annemans L, Closon J, Heymans I, Lagasse R, Mendes Da Costa E, Moureaux C, Roch I: Vergelijking van kost en kwaliteit van twee financieringssystemen voor de eerstelijnsgezondheidszorg in België. KCE reports 85A. In.; 2008. 82. Aday LA, Begley CE, Lairson DR, Slater CH, Richard AJ, Montoya ID: A framework for assessing the effectiveness, efficiency, and equity of behavioral healthcare. The American journal of managed care 1999, 5 Spec No:SP25-44. 83. van der Meer JB, van den Bos J, Mackenbach JP: Socioeconomic differences in the utilization of health services in a Dutch population: the contribution of health status. Health Policy 1996, 37(1):1-18. 47 Chapter 1: Introduction 84. Fiscella K, Franks P, Gold MR, Clancy CM: Inequality in quality: addressing socioeconomic, racial, and ethnic disparities in health care. JAMA 2000, 283(19):2579-2584. 85. Calle EE, Flanders WD, Thun MJ, Martin LM: Demographic predictors of mammography and Pap smear screening in US women. Am J Public Health 1993, 83(1):53-60. 86. Asch SM, Kerr EA, Keesey J, Adams JL, Setodji CM, Malik S, McGlynn EA: Who is at greatest risk for receiving poor-quality health care? N Engl J Med 2006, 354(11):1147-1156. 87. Le H, Ziogas A, Lipkin SM, Zell JA: Effects of socioeconomic status and treatment disparities in colorectal cancer survival. Cancer Epidemiol Biomarkers Prev 2008, 17(8):1950-1962. 88. Willems S, De Maesschalck S, Deveugele M, Derese A, De Maeseneer J: Socio-economic status of the patient and doctor-patient communication: does it make a difference? Patient Educ Couns 2005, 56(2):139-146. 48 89. DeVoe JE, Wallace LS, Fryer GE, Jr.: Measuring patients’ perceptions of communication with healthcare providers: do differences in demographic and socioeconomic characteristics matter? Health Expect 2009, 12(1):7080. 90. Alter DA, Naylor CD, Austin P, Tu JV: Effects of socioeconomic status on access to invasive cardiac procedures and on mortality after acute myocardial infarction. N Engl J Med 1999, 341(18):1359-1367. 91. Braaten T, Weiderpass E, Lund E: Socioeconomic differences in cancer survival: the Norwegian Women and Cancer Study. BMC Public Health 2009, 9:178. 92. Bhayat F, Das-Gupta E, Smith C, McKeever T, Hubbard R: The incidence of and mortality from leukaemias in the UK: a general population-based study. BMC cancer 2009, 9:252. 93. Denvir MA, Lee AJ, Rysdale J, Walker A, Eteiba H, Starkey IR, Pell JP: Influence of socioeconomic status on clinical outcomes and quality of life after percutaneous coronary intervention. J Epidemiol Community Health 2006, 60(12):1085-1088. 94. Leslie SJ, Rysdale J, Lee AJ, Eteiba H, Starkey IR, Pell J, Denvir MA: Unemployment and deprivation are associated with a poorer outcome following percutaneous coronary angioplasty. International journal of cardiology 2007, 122(2):168-169. 95. Shimony A, Zahger D, Ilia R, Shalev A, Cafri C: Impact of the community’s socioeconomic status on characteristics and outcomes of patients Chapter 1: Introduction undergoing percutaneous coronary intervention. International journal of cardiology 2010, 144(3):379-382. 96. Begley C, Basu R, Lairson D, Reynolds T, Dubinsky S, Newmark M, Barnwell F, Hauser A, Hesdorffer D: Socioeconomic status, health care use, and outcomes: persistence of disparities over time. Epilepsia 2011, 52(5):957964. 97. Re-orienting health systems: towards modern, responsive and sustainable health promoting systems: draft. In. Edited by EuroHealthNet: EuroHealthNet; 2012. 98. Macinko J, Starfield B, Shi L: The contribution of primary care systems to health outcomes within Organization for Economic Cooperation and Development (OECD) countries, 1970-1998. Health Serv Res 2003, 38(3):831-865. 99. Franks P, Fiscella K: Primary care physicians and specialists as personal physicians. Health care expenditures and mortality experience. J Fam Pract 1998, 47(2):105-109. 100. Shi L, Macinko J, Starfield B, Politzer R, Wulu J, Xu J: Primary care, social inequalities and all-cause, heart disease and cancer mortality in US counties: a comparison between urban and non-urban areas. Public Health 2005, 119(8):699-710. 101. Shi L, Starfield B, Politzer R, Regan J: Primary care, self-rated health, and reductions in social disparities in health. Health Serv Res 2002, 37(3):529550. 102. Mark DH, Gottlieb MS, Zellner BB, Chetty VK, Midtling JE: Medicare costs in urban areas and the supply of primary care physicians. J Fam Pract 1996, 43(1):33-39. 103. Rivo ML, Satcher D: Improving access to health care through physician workforce reform. Directions for the 21st century. JAMA 1993, 270(9):1074-1078. 104. Moore GT: The case of the disappearing generalist: does it need to be solved? Milbank Q 1992, 70(2):361-379. 105. Franks P, Clancy CM, Nutting PA: Gatekeeping revisited--protecting patients from overtreatment. N Engl J Med 1992, 327(6):424-429. 106. Aaraas I, Sorasdekkan H, Kristiansen IS: Are general practitioner hospitals cost-saving? Evidence from a rural area of Norway. Fam Pract 1997, 14(5):397-402. 49 Chapter 1: Introduction 107. Dale J, Lang H, Roberts JA, Green J, Glucksman E: Cost effectiveness of treating primary care patients in accident and emergency: a comparison between general practitioners, senior house officers, and registrars. BMJ 1996, 312(7042):1340-1344. 108. Murphy AW, Bury G, Plunkett PK, Gibney D, Smith M, Mullan E, Johnson Z: Randomised controlled trial of general practitioner versus usual medical care in an urban accident and emergency department: process, outcome, and comparative cost. BMJ 1996, 312(7039):1135-1142. 109. Ward P, Huddy J, Hargreaves S, Touquet R, Hurley J, Fothergill J: Primary care in London: an evaluation of general practitioners working in an inner city accident and emergency department. J Accid Emerg Med 1996, 13(1):11-15. 110. Shear CL, Gipe BT, Mattheis JK, Levy MR: Provider continuity and quality of medical care. A retrospective analysis of prenatal and perinatal outcome. Med Care 1983, 21(12):1204-1210. 50 111. Leape LL, Park RE, Solomon DH, Chassin MR, Kosecoff J, Brook RH: Does inappropriate use explain small-area variations in the use of health care services? JAMA 1990, 263(5):669-672. 112. Vedsted P, Olesen F: Are the serious problems in cancer survival partly rooted in gatekeeper principles? An ecologic study. Br J Gen Pract 2011, 61(589):e508-512. 113. Richards MA: The size of the prize for earlier diagnosis of cancer in England. British journal of cancer 2009, 101 Suppl 2:S125-129. 114. Whitehead M, Hanratty B: Health care for rich and poor alike. In: Accessing health care Responding to diversity. edn. Edited by Healy J, McKee M. Oxford: Oxford University Press; 2004: 71-88. 115. Shi L: Primary care, specialty care, and life chances. Int J Health Serv 1994, 24(3):431-458. 116. Campbell RJ, Ramirez AM, Perez K, Roetzheim RG: Cervical cancer rates and the supply of primary care physicians in Florida. Fam Med 2003, 35(1):60-64. 117. Ferrante JM, Gonzalez EC, Pal N, Roetzheim RG: Effects of physician supply on early detection of breast cancer. The Journal of the American Board of Family Practice / American Board of Family Practice 2000, 13(6):408-414. 118. Roetzheim RG, Pal N, van Durme DJ, Wathington D, Ferrante JM, Gonzalez EC, Krischer JP: Increasing supplies of dermatologists and family Chapter 1: Introduction physicians are associated with earlier stage of melanoma detection. Journal of the American Academy of Dermatology 2000, 43(2 Pt 1):211-218. 119. Gulliford MC: Availability of primary care doctors and population health in England: is there an association? J Public Health Med 2002, 24(4):252254. 120. Ahmed SM, Lemkau JP, Nealeigh N, Mann B: Barriers to healthcare access in a non-elderly urban poor American population. Health Soc Care Community 2001, 9(6):445-453. 121. van Doorslaer E, Koolman X, Jones AM: Explaining income-related inequalities in doctor utilisation in Europe. Health Econ 2004, 13(7):629-647. 51 52 Chapter 2 Research aim 53 54 Chapter 2: Research aim Research aim In the previous chapter we introduced the concept of inequity in health and health care and its determinants. Next, we focused on the role primary health care can play in reducing inequities in health. One of the actions in tackling health inequities is providing an equitable primary health care system. The aim of this doctoral thesis is to focus on the equity in primary health care in Belgium in terms of equal access, equal treatment and equal outcomes for vulnerable population groups. Part I of the thesis will focus on the access to primary health care, where access is adressed by the definition of Goddard et al (2001) “the ability to secure a specified range of services, at a specified level of quality, subject to a specified maximum level of personal inconvenience and costs, whilst in the possession of a specified level of information” [1]. In part II of the thesis the emphasis is on the treatment and outcomes of primary care, more specifically on the attitude of the doctor, doctor-patient communication and decision making. In some of the studies, primary health care use is compared to other types of care (e.g. specialist care) or is studied in specific settings (e.g. care for the homeless people, low income people). Figure 8 provides an overview of the main focus of the thesis and the hypotheses. 55 Chapter 2: Research aim Equity in Health Care Equal treatment for patients in equal need Equal access to care for patients in equal need e.g. communication, attitude, decision making Equal outcomes for patients in equal need e.g. financial, cultural and geographical access Hypothesis 1 The specialist is the doctor of the high income people and the GP is the doctor of the low to middle income people (paper 1). 56 Hypothesis 2 Homeless people have a higher likelihood to consult the emergency department than the general population. A universal primary health care system with active guidance for homeless people towards this system might counter this (paper 2). Hypothesis 3 People who are not able to cope with their available income have a higher likelihood to postpone a visit to a GP (paper 3). Hypothesis 4 The social gradient in doctor-patient communication persists, where patients from lower socio-economic classes are less involved during the consultation, receive less control and a more paternalistic consultation style (paper 4). Hypothesis 5 Physicians have a negative attitude towards patients with a low SES and as a result make changes in their clinical decision which do not favor the outcomes for the patient (paper 5). Figure 8: an overview of the main focus of this thesis, the hypotheses and the research papers Part I: Access to primary care for people with low SES in Belgium In the first part of this thesis we focus on access to care. Three hypotheses are formulated: • Hypothesis 1: The specialist is the doctor of the higher-income people and the GP is the doctor of the lower-income people. • Hypothesis 2: Homeless people have a higher likelihood to consult the emergency department than the general population. A universal primary health care system with active guidance for homeless people towards this system might counter this. • Hypothesis 3: people who are not able to cope with their available income have a higher likelihood to postpone a visit to a GP. Chapter 2: Research aim Hypothesis 1: The specialist is the doctor of the higher-income people and the GP is the doctor of the lower-income people. International literature shows a social gradient in health care use, where people from lower social classes tend to visit a GP more frequently than people from middle or higher social classes, while people from higher social classes tend to use more frequently specialist care than the less well-of. However, when taking into account health care need, in some studies this gradient persists, in others this gradient disappears [2-6]. Most of the studies describing the social variations in health care utilization are based on data that are obtained from self-administered questionnaires or interviews. We believe that analyzing health care use by means of registered medical utilization data on the one hand and detailed self-reported information on respondent’s health on the other hand, we can overcome the methodological restraints caused by the use of self-reported utilization data or the lack of information about health status that many other studies are confronted with. Paper 1 aims to describe the social gradient in the use of specialist care and primary care in Belgium. Hereto data from the Belgian Health Interview are linked with the corresponding registered medical utilization data delivered by the sickness funds. The innovation of this study lays in the fact that this linked database allows us to research the social gradient on actual registered health care utilization data and not on reported data. Hypothesis 2: Homeless people have a higher likelihood to consult the emergency department than the general population. A universal primary health care system with active guidance for homeless people towards this system might counter this. Literature shows that both homeless people and low-income people are vulnerable population groups facing a higher disease burden leading to a higher health care need [7-9]. Despite their higher health care need, these people do not always find the way to the appropriate health care, or they might even postpone their needed medical care [10-14]. Paper 2 presents the findings of a survey conducted in a sample of 122 homeless people living in Ghent. Data collected by face-to-face interviews are compared with comparable data from the Belgian Health Survey which gives information on the general population. The health care use of the homeless people is compared to the use of the non-homeless people living in Ghent, taking into account several health indicators. The merits of this study lie in the fact that a relatively large sample of 57 Chapter 2: Research aim the homeless population residing in Ghent, an area with a nearly universal health care system and with active guidance towards that system, was questioned following the research protocol and instruments of the Belgian Health Interview Survey. Hypothesis 3: People who are not able to cope with their available income have a higher likelihood to postpone a visit to a GP. Paper 3 aims to identify which subgroups of low-income people are most at risk for postponing a visit to a GP. Data are collected among 606 low-income people living in Flanders. An important strength of this study is the success of questioning 606 low-income people about their health and healthcare use. In other studies this population group is often under-represented, making it very difficult to formulate strong conclusions. Part II: Treatment and outcomes of care for vulnerable social groups 58 Part two of this thesis focuses on equal treatment and outcomes in primary health care. Two hypotheses are formulated: • • Hypothesis 4: The social gradient in doctor-patient communication persists over time, where patients from lower socio-economic classes are less involved during the consultation, receive less control and a more paternalistic consultation style. Hypothesis 5: Physicians have a negative attitude towards patients with a low SES and as a result make changes in their clinical decision which do not favor the outcomes for the patient. Hypothesis 4: The social gradient in doctor-patient communication persists, where patients from lower socio-economic classes are less involved during the consultation, receive less control and a more paternalistic consultation style. Several studies on health care use have repeatedly shown the importance of a good doctor-patient interaction [15, 16]. Good doctor-patient communication is associated with better patient health and outcomes [17, 18]. By communicating with a patient, a physician gets to know the patient’s problems and creates a therapeutic relationship necessary for its management and, if possible, its solution [19]. Chapter 2: Research aim Nevertheless, qualitative studies show that patients from lower social classes are often less involved in the decision making process or do not always understand what the doctor says [20]. In Paper 4 we aimed, by means of a systematic review of the literature, to answer the question whether the doctor-patient communication varies according to the socio-economic status of the patient. And if so, which aspects of the consultation are affected. The social gradient in doctor-patient communication is an emerging topic in the research on health care use. The importance of this topic is made clear by the large number of citations of the review on doctor-patient communication by Willems et al. (2005) [21]. Therefore it is important to keep this information up to date and address possible new issues. Hypothesis 5: Physicians have a negative attitude towards patients with a low SES and as a result make changes in their clinical decision which do not favor the outcomes for the patient. Physicians are expected not to judge patients on their social class, and not to treat patients differently because of their social class. Qualitative studies have shown that patients from lower social classes have the feeling that physicians have a different attitude towards them. They have the feeling they are sometimes not respected by the doctor, or the doctor does not take his/her time to understand the complex situation of the patient [20]. In Paper 5 we aimed to explore the influence of the socio-economic status of the patient on the clinical decision making of the general practitioner by means of a qualitative study among 11 GP’s in Flanders. By means of face-to-face interviews, physicians were asked to which extent they think their clinical decisions are influenced when working with low SES patients. There is only little evidence on the influence of patient’s SES on the clinical decisions of the GP, and the evidence is nearly almost restricted to the US. These results may be context-dependent and cannot be generalized. 59 Chapter 2: Research aim References 60 1. Goddard M, Smith P. Equity of access to health care services: theory and evidence from the UK. Soc Sci Med. 2001;53(9):1149-62. Epub 2001/09/15. 2. Van der Heyden JH, Demarest S, Tafforeau J, Van Oyen H. Socio-economic differences in the utilisation of health services in Belgium. Health Policy. 2003;65(2):153-65. Epub 2003/07/10. 3. Bongers IM, van der Meer JB, van den Bos J, Mackenbach JP. Socio-economic differences in general practitioner and outpatient specialist care in The Netherlands: a matter of health insurance? Soc Sci Med. 1997;44(8):11618. Epub 1997/04/01. 4. Fernandez de la Hoz K, Leon DA. Self-perceived health status and inequalities in use of health services in Spain. Int J Epidemiol. 1996;25(3):593603. Epub 1996/06/01. 5. van Doorslaer E, Masseria C, Koolman X. Inequalities in access to medical care by income in developed countries. CMAJ. 2006;174(2):177-83. Epub 2006/01/18. 6. van Doorslaer E, Koolman X, Jones AM. Explaining income-related inequalities in doctor utilisation in Europe. Health Econ. 2004;13(7):629-47. Epub 2004/07/20. 7. Droomers M, Westert GP. Do lower socioeconomic groups use more health services, because they suffer from more illnesses? Eur J Public Health. 2004;14(3):311-3. Epub 2004/09/17. 8. Hwang SW. Homelessness and health. CMAJ. 2001;164(2):229-33. Epub 2001/05/03. 9. Morris DM, Gordon JA. The role of the emergency department in the care of homeless and disadvantaged populations. Emerg Med Clin North Am. 2006;24(4):839-48. Epub 2006/09/20. 10. Kushel MB, Vittinghoff E, Haas JS. Factors associated with the health care utilization of homeless persons. JAMA. 2001;285(2):200-6. Epub 2001/02/15. 11. Power R, French R, Connelly J, George S, Hawes D, Hinton T, et al. Health, health promotion, and homelessness. BMJ. 1999;318(7183):590-2. Epub 1999/02/26. Chapter 2: Research aim 12. Drieskens S VdHJ, Hesse E, Gisle L, Demarest S, Tafforeau J. Gezondheidsenquête België, 2008. Rapport V - Sociaal-economische ongelijkheden in gezondheid. 2010 2032-9172. 13. Gelberg L, Gallagher TC, Andersen RM, Koegel P. Competing priorities as a barrier to medical care among homeless adults in Los Angeles. Am J Public Health. 1997;87(2):217-20. Epub 1997/02/01. 14. Whitehead M, Hanratty B. Health care for rich and poor alike. In: Healy J, McKee M, editors. Accessing health care Responding to diversity. Oxford: Oxford University Press; 2004. p. 71-88. 15. Street RL, Jr., Millay B. Analyzing patient participation in medical encounters. Health Commun. 2001;13(1):61-73. Epub 2001/05/24. 16. Bensing J. Doctor-patient communication and the quality of care. Soc Sci Med. 1991;32(11):1301-10. Epub 1991/01/01. 17. Duberstein P, Meldrum S, Fiscella K, Shields CG, Epstein RM. Influences on patients’ ratings of physicians: Physicians demographics and personality. Patient Educ Couns. 2007;65(2):270-4. Epub 2006/11/28. 18. Cheraghi-Sohi S, Bower P, Mead N, McDonald R, Whalley D, Roland M. What are the key attributes of primary care for patients? Building a conceptual ‘map’ of patient preferences. Health Expect. 2006;9(3):275-84. Epub 2006/08/17. 19. Bensing J, Kerssens JJ, Vanderpasch M. Patient-directed gaze as a tool for discovering and handling psychosocial problems in general practice. Journal of Nonverbal Behavior. 1995;19:223-42. 20. Willems S. The socio-economic gradient in health: a never-ending story? A descriptive and explorative study in Belgium. Ghent: Ghent University; 2005. 21. Willems S, De Maesschalck S, Deveugele M, Derese A, De Maeseneer J. Socio-economic status of the patient and doctor-patient communication: does it make a difference? Patient Educ Couns. 2005;56(2):139-46. Epub 2005/01/18. 61 62 Chapter 3 Research Papers 63 64 Chapter 3: Research papers Paper 1: Verlinde E., Pasteels I., De Maeseneer J., Willems S. Socio-economic differences in the utilization of FP and specialist care: do patients form lower social class visit more often a doctor when taking into account their higher health care need? Health Policy. Submitted. IF: 1.383 SW contributed to the application for the National Privacy Commission to obtain permission to use and link the data. IP contributed to the data manipulation and linkage of the data. EV, IP, and SW contributed to the analysis. EV and SW contributed to the writing of the article. IP and JDM participated in the revision of the manuscript. Paper 2: Verlinde E., Verdee T., Van de Walle M., Art B., De Maeseneer J, Willems S. Unique health care utilization patterns in a homeless population in Ghent. BMC Health Services Research. 2010; 10(242). IF:1.66 MVDW, TV, BA , SW and JDM contributed to the design of the study. MVDW and TV contributed to the data retrieval. EV and SW contributed to the analysis and the writing of the paper. BA, SW and JDM participated in the revision of the manuscript. Paper 3: Verlinde E., Poppe A., DeSmet A., Hermans K., De Maeseneer J., Van Audenhove C., Willems S. Social differences in postponing a GP visit: Which low-income patients are most at risk? Health and Social Care in the Community. Under review. IF: 1.008 EV, AP, AD, JDM, CVA, and SW contributed to the design of the study; EV, AP, and SW contributed to the analysis and the writing of the article; and AD, KH, JDM, CVA, and SW participated in the revision of the manuscript. Paper 4: Verlinde E., De Laender N., De Maesschalk S., Deveugele M., Willems S. The social gradient in doctor-patient communication. Journal for equity in health. 2012, 11:12.IF:1.30 EV, NDL, SW and SDM equally contributed to the research and the report of the study. EV, NDL and SW contributed to the writing of the paper. DM and SDM contributed to the revision of the manuscript. 65 Chapter 3: Research papers Paper 5: Verlinde E., Bonte E., Willems S. Socio-economic status of the patient: does it influence the decisions of the GP? Patient education and Counseling. Submitted. IF: 2.237 EV and SW contributed to the design of the study. EB and EV contributed to the data retrieval and the analysis. EV, and SW contributed to the writing of the paper. 66 Chapter 3: Research papers Socio-economic differences in the utilization of FP and specialist care. Evelyn Verlinde, MA§ Phd Candidate Department of Family Medicine and Primary Healthcare Ghent University De Pintelaan 185, 9000 Ghent Belgium Evelyn.Verlinde@UGent Inge Pasteels, MA Coordinator “Divorce in Flanders” Research Center for Longitudinal and Life Course Studies (CELLO) Antwerp University Sint-Jacobstraat 2, 2000 Antwerp Belgium [email protected] Jan De Maeseneer, MD PhD Full Professor Department of Family Medicine and Primary Healthcare Ghent University De Pintelaan 185, 9000 Ghent Belgium [email protected] Sara Willems MD PhD Professor Department of Family Medicine and Primary Healthcare Ghent University De Pintelaan 185, 9000 Ghent Belgium [email protected] §Corresponding author Evelyn Verlinde De Pintelaan 185 – 6K3 9000 Ghent Phone: 0032 9 332 6082 Fax: 0032 9 332 4967 [email protected] 67 Chapter 3: Research papers Abstract International studies report higher primary care utilization rates for people with a lower socio-economic status (SES) and higher specialist care utilization rates for people high on the socio-economic ladder. These studies are often limited due to the use of self-reported utilization data, or the lack of detailed information on health status. In this paper, we describe the social variation in primary care and specialist care use in Belgium based on registered medical utilization data, taking into account health status. Data from the Belgian Health Interview Survey are linked to registered medical utilization data. SES was measured in terms of education and income. Lower-educated and middle-educated people make greater use of FP than do higher-educated, whereas higher-educated make greater use of specialist care, even after controlling for health status. 68 Although the Belgian health care system covers almost 100% of the population, this does not correspond with equity in health care use. This could be due to the lack of a gatekeeper in primary care and free access to secondary care in the Belgian health care system. Keywords: Primary care, social class, registered utilization data Chapter 3: Research papers Background Equity in health can be defined as “the absence of systematic disparities in health (or in major social determinants of health) between social groups who have different levels on underlying social advantage”[1]. In order to obtain equity in health there needs to be equity in health care. Equity in health care means that patients who are alike should be treated in the same way (horizontal equity) and that patients who are unlike, should be treated in the same unlike way (vertical equity) [2-5]. An important aspect of delivering equitable health care is providing equal care for those in equal need [6]. In the light of providing equal care to people in equal need, it would be assumed that in the absence of barriers to health care, the demand for health services is similar across socioeconomic groups after controlling for their health status [6]. Nevertheless, across countries, higher utilization rates of primary care are reported for people with a lower socio-economic status (SES) [7-10]. On the contrary, the higher the SES, the higher the use of specialist care, physiotherapy, and dentistry [7]. However, as low SES is known to be associated with poor health, adjustment for health status is necessary for interpretation of social gradients in health care use [11]. For example, the Belgian Health Interview Survey 2004 shows that the social gradient found in the use of general practitioner care, nursing care at home, and admissions to hospitals (all higher for patients with a lower educational attainment) disappears when self-rated health status is taken into account [7]. Based on the data from 21 OECD (Organization for Economic Cooperation and Development) countries, the social gradient in specialist care seems to remain after controlling for need differences: Higher-income people are still more likely to see a specialist than lower-income people [10]. Prudence in the interpretation of these results is of utmost importance. Most of the studies describing the social variations in health care utilization are based on data that are obtained from self-administered questionnaires or interviews. This has some implications. First of all, reliability studies show that self-reported use of health services, in general, leads to an over-reporting of the number of visits to a health care provider. In case of an actual high number of visits to a health care provider or when the recall period lengthens (for example, in the question “how many times did you visit your FP in the past 12 months” compared to “how many times did you visit your FP in the past month” ), patients tend to under-report [12-14]. Secondly, there could be a recall bias according to social class [12]. 69 Chapter 3: Research papers In order to overcome these problems, studies on the use of health care services can make use of registered medical data. However, there are only a small number of studies that analyse the social gradient in health care use based on registered data [11, 15-19]. All of them confirm the significance of primary care in the health care for disadvantaged populations. For example, the Universal Health Insurance in Canada successfully achieves equity in its access to primary care but not to specialist care [11, 15]. The results of these studies are highly context-dependent, and so, the social gradient in health care use might differ according to different health care systems. 70 Our objective is to examine equity in the use of primary care and specialist care in Belgium, a country in which the health care system achieves almost universal public coverage of the population for the majority of health care services with direct access to any general practitioner and specialist at any time for any problem [10]. In the Belgian fee-for-service system, patients pay about 30% co-payment for primary health care and 40% for specialist care, with reduced co-payment rates (8%) for lower socio-economic groups [10]. We hypothesize that, in Belgium, the specialist is the doctor of the higher income people and the FP the doctor of the lower income people. By analyzing individuals’ registered medical utilization data on the one hand and detailed self-reported information on their health on the other hand, this study aims to overcome the methodological restraints caused by the use of self-reported utilization data or the lack of information about health status that many other studies in this research domain are confronted with. Methods In this study, data from the Belgian Health Interview Survey (BHIS) 1997 are linked with the corresponding registered medical utilization data provided by the Sickness Funds. The linkage between these databases is unique for Belgium and has not been repeated in the case of more recent BHIS data. The Belgian Health Interview Survey The Belgian Health Interview Survey is a cross-sectional survey based on a representative sample of the non-institutionalized population residing in Belgium who are aged 15 years or older. Sampling has been based on a combination of stratification, multistage sampling, and clustering [7]. Chapter 3: Research papers The first stage was a stratification by region (Flemish Region, Walloon Region, and the Brussels Region) with a fixed sample size. The second stratification was at the level of the provinces with the sample size defined proportional to the population size of the province. In the third step, the primary sampling unit (PSU) was selected within each stratum by a weighted systematic sampling procedure. In the fourth stage, a number of households were selected, called the secondary sampling units (SSUs), by a clustered systematic sampling procedure after ranking the household hierarchically by the statistical sector, household size, and age of the reference person. This method allowed to minimise selection bias. In the last step, the individual or the third sampling unit was selected within the household. Up to a maximum of four persons could be interviewed within each household. The first two subjects were selected by definition: the reference person and his/her partner. The two other individuals were selected based on a birthday rule: the persons who had their birthday coming up first after the interview [20]. Overall, 10.221 individuals, living in 4.664 households, were questioned, that is 0.1% of the Belgian population [7, 21]. The variables used in this study were collected by means of a face-to-face interview carried out at home [7]. More detailed information on the methodology of the Belgian Health Interview Survey has been published elsewhere [20]. Medical care utilization data registered by the Sickness Funds In the Belgian fee-for-service system, patients advance the fee for ambulatory medical care and get a refund from the Sickness Funds afterward. In this context, the Sickness Funds registers all medical utilization data. The database from the Sickness Funds analyzed in this study contains the medical care use of almost all individuals included in the Belgian Health Interview Survey, for the reference period of 1996 – 1997. It contains information on all medical care subject to a refund by the Sickness Funds. This database contains no information on medical care that is not refunded (such as supplements). However, these are scarce in the Belgian primary care. In addition, the self-employed respondents were excluded, as they are not covered by the public health insurance the same way as employees. This implies that the data concerning the health care use of the self-employed is registered in another way, making comparison of the datasets not possible. 71 Chapter 3: Research papers Linking both databases To protect the privacy of the participants, the linking of the two databases was carried out by the Crossroads Bank for Social Security, a governmental organization in charge of promoting information security and privacy protection and delivering integrated statistical information to politicians and researchers in order to support the social policy [22]. The linked database contains information on 7.364 individuals who are aged 15 years or older. The reduction compared with the 10.221 individuals in the HIS was due to small loss of data at the different levels of the linking process, persons who died during the reference period and the exclusion of the self-employed respondents. In order to match the Belgian population, the weighing procedure was recalculated according to the original weighing procedure applied but with taking into account the drop-out of observations after linking both datasets, taking into account the age, sex, and province. 72 Measuring socio-economic status To measure SES, both education and income are commonly used [23, 24]. In this study, educational level is based on the highest educational attainment in the family. To determine educational level, the educational attainment of the reference person was compared with his/her partner, if present. Educational level is categorized in three groups: higher (higher education), middle (higher secondary school or Alevel), and lower (no education, primary school, or lower secondary school). Income is defined using the equivalent household income based on the total household income and the number of household members. Equivalent income was categorized as lower (<500–750 euro), middle (750–1500 euro), and higher (> 1.500 euro). Measuring health status The measures of health status used in this study are self-rated health, reporting two or more chronic diseases, psychological distress, and self-reported disability. Self-rated health is measured by a validated WHO instrument, recoded in three categories: very bad–bad, fair, and good–very good [25]. Chronic diseases are questioned by means of a checklist of chronic conditions and the question whether the respondent experienced any of these conditions in the past 12 months. The cut-off point for analysis was set on having two or more chronic conditions. Disability due to chronic conditions or diseases is categorized in three categories (severe to Chapter 3: Research papers moderate disability, light disability, trouble of chronic disease without disability to neither trouble nor disability due to chronic disease). Psychological distress is measured using the mental health scale of the GHQ12 questionnaire (cut-off point at 90% probability for having psychological distress). Health care utilization Registered data on health care utilization were obtained from the Sickness Funds. The utilization of the health care delivered by the family physician and specialist is studied in terms of the number of consultations during the two years’ reference period (1996–1997). These number of consultations were recalculated from two years to one year for a better understanding and interpretation of the analysis. Data analysis Considering that we are working with an over-dispersed count date (when the conditional variance exceeds the conditional mean), we used a negative binomial regression. We modelled the number of visits to a FP and the number of visits to a specialist. The dependent variables were the number of visits to a FP (model 1) or a specialist (model 2) and the independent variables were income and education. Furthermore, the models controlled for age, sex, self-rated health, chronic conditions, disability, and mental distress. The level of statistical significance was set at p < 0.05. All analyses were performed using SPSS 19. Ethics approvement This study was approved by the Ethics Committee of the Ghent University Hospital (Dossier No. 2001/64). Results Description of the research sample Table 1 shows the distribution of the study sample according to sex, age, educational level, income, and health status. There are slightly more women (51.4%) than men (48.6%) in the sample. More than half (65.1%) are aged between 25 and 65. With regard to educational level, 34.9% of the respondents have a lower educational level; 33.2% have a middle educational 73 Chapter 3: Research papers level; and 32.0% have a higher educational level. The proportion of the respondents with a higher, middle, and lower equivalent income is, respectively, 25.8%, 61.5%, and 12.7%. For self-rated health, 3.5% reported bad to very bad health compared with 20.5% and 75.9% with fair and good to very good health, respectively. Considering chronic diseases and disability, 16.7% suffered from two or more chronic diseases, and 11.5% experienced moderate to severe disability. With regard to mental health, 9.8% had a bad mental health status. The number of consultations a year with a FP varied between 0 and 208. 25% had seven or more consultations with a FP. The number of consultations with a specialist varied between 0 and 32. 25% had three or more consultations with a specialist. Table 1: Distribution of the study sample 74 Variable Sex Men Women Age 15–24 25–65 65+ Educational level Lower Middle Higher Equivalent income Lower Middle Higher Self-rated health Good–very good Fair Bad–very bad Chronic diseases < 2 > 2 N (%) 7364 3578 (48.6) 3786 (51.4) 7364 11106 (15.0) 4792 (65.1) 1466 (19.9) 7331 2556 (34.9) 2431 (33.2) 2344 (32.0) 7019 1810 (25.8) 4316 (61.5) 894 (12.7) 6950 5278 (75.9) 1427 (20.5) 245 (3.5) 7362 6131 (83.3) 1231 (16.7) Chapter 3: Research papers Disability due to chronic disease Severe to moderate disability Light disability Troubles of chronic disease without disability to neither troubles nor disability Mental health < 90% > 90% Number of consultations with FP Q 25 Q 50 Q 75 Number of consultations with specialist Q 25 Q 50 Q 75 7318 838 (11.5) 644 (8.8) 5836 (79.3) 5836 (79.3) 7156 6453 (90.2) 703 (9.8) 1 3 7 0 1 3 The association between SES and the number of consultations with a FP In table 2, the associations between the number of consultations and education and income after controlling for age, sex, education, income, and health status are shown. As established by literature, we find significant effects for age, sex, self-rated health, chronic diseases and disability on the number of consultations with a FP. Women have a higher ratio of visiting a FP than men (odds ratio 1.318, p < 0.001). People with bad health have a higher ratio of visiting a FP than respondents with good health (odds ratio 1.607 p < 0.001). Also those with a fair health have a higher ratio of visiting a FP compared to people with a good health (odds ratio 1.416, p<0.001). People aged 65 years or older have a higher ratio of visiting a FP than those aged between 15 and 24 years (odds ratio 1.847, p<0.001). Those aged between 25 and 65 years also have a higher ratio of visiting a FP compared to people older than 65 years (odds ratio 1.099, p=0.007). Furthermore, people with more than two chronic conditions have a higher ratio contacting a FP than people with no or less than two chronic conditions (odds ratio: 1.286, p < 0.001). Respondents who experience no disability due to their chronic conditions have a lower ratio to visit a 75 Chapter 3: Research papers FP compared to those perceiving moderate to severe disabilities (odds ratio 0.726, p<0.001). No significant results were found for those perceiving a light disability compared to those with a moderate to severe disability (p=0.622). Furthermore, no significant results were found for mental distress (p=0.841). After controlling for these health measures, we find significant results for educational level and income on the number of consultations with a FP. For educational level, we find a social gradient in visiting the FP: people with a higher educational level have a lower ratio of visiting a FP, compared to people with a lower educational level (odds ratio 0.751, p<0.001). But also people with a middle educational level have a lower ratio of visiting a FP than those with a lower educational level (odds ratio 0.867, p<0.001). For income, we find a significant association for middle versus lower income: people with a middle income have a higher ratio in contacting a FP compared to those with a low income (odds ratio 1.079, p=0.008). No significant association was found between higher and lower income people. 76 Table 2: Negative binomial distribution for visiting a FP during the reference period and socio-economic status adjusted for age, sex, and health care status Sex women vs men B SE 0.276 0.0240 Odds ratio 1.318 95% CI p* 1.257-1.381 < 0.001 Age 65+ versus 15–24 0.628 0.0425 1.874 1.724-2.037 < 0.001 25–65 versus 15–24 0.095 0.0349 1.099 1.026-1.177 0.007 Educational level Higher versus lower -0.286 0.331 0.751 0.704-0.802 < 0.001 Middle versus lower -0.142 0.0296 0.867 0.818-0.919 < 0.001 Equivalent income Higher versus lower -0.16 0.0464 0.985 0.899-1.078 0.737 Middle versus lower 0.076 0.0289 1.079 1.020-1.142 0.008 Self-rated health Bad versus good 0.474 0.0722 1.607 1.395-1.851 < 0.001 Fair versus good 0.348 0.0337 1.416 1.325-1.512 < 0.001 Chronic disease 0.252 0.0343 1.286 1.203-1.376 < 0.001 Chapter 3: Research papers Disability Troubles of chronic disease without disability to neither troubles nor Severe to moderate 0.0444 0.726 0.666-0.792 < 0.001 0.026 0.0529 1.026 0.925-1.139 0.622 -0.008 0.0406 0.992 0.916-1.074 0.841 disability versus Light disability vs severe -0.320 to moderate Mental health *All significant results are indicated in bold. The association between SES and the number of consultations with a specialist The results of the analysis on the use of specialist care are depicted in table 3. As with the number of consultations with a FP, we also find significant results for age, sex, self-rated health, chronic diseases and disability on the number of consultations with a specialist. Women have a higher ratio of contacting a specialist than men (odds ratio 1.615, p<0.001). People aged older than 65 years have a higher ratio of contacting a specialist compared with people aged between 15 and 24 (odds ratio 1.489, p<0.001) and people aged between 25 and 65 years also have a higher ratio of visiting a specialist compared to people aged between 15 and 24 years (odds ratio 1.220, p<0.001). People suffering from more than two chronic conditions have a higher ratio of visiting a specialist compared to people with less than two or no chronic conditions (odds ratio 1.342, p<0.001). Respondents who experience no disability due to their chronic conditions have a lower ratio to visit a specialist compared to those perceiving moderate to severe disabilities (odds ratio 0.693, p<0.001). No significant results were found for those perceiving a light disability compared to those with a moderate to severe disability (p=0.616). We find no significant results with regard to mental distress (p=0.691). After controlling for health status, only a significant association between educational level and the number of consultations with a specialist is found: people with a higher educational level have a higher ratio of contacting a specialist compared to those with a lower educational level (odds ratio 1.196, p<0.001). People with a middle educational level have a higher ratio of contacting a specialist compared to 77 Chapter 3: Research papers those with a lower educational level (odds ratio 1.137, p=0.001). For income, we find no significant associations between those with a higher versus lower income and those with a middle versus lower income (odds ratio resp. 1.085, 1.073). Table 3: Negative binomial distribution for visiting a specialist during the reference period and socio-economic status adjusted for age, sex, and health care status. Sex women vs men B 0.479 SE Odds ratio 0.0310 1.615 95% CI P* 1.520-1.716 < 0.001 Age 65+ versus 15–24 0.398 0.0561 1.489 1.334-1.662 < 0.001 25–65 versus 15–24 0.199 0.0460 1.220 1.115-1.335 < 0.001 Educational level Higher versus lower 0.179 0.427 1.196 1.100-1.301 < 0.001 Middle versus lower 0.128 0.0388 1.137 1.054-1.227 0.001 Equivalent income 78 Higher versus lower 0.081 0.0583 1.085 0.967-1.216 0.164 Middle versus lower 0.070 0.0377 1.073 0.997-1.155 0.062 0.001 Self-rated health Bad versus good 0.307 0.0919 1.359 1.135-1.627 Fair versus good 0.338 0.0439 1.402 1.286-1.528 < 0.001 Chronic disease 0.294 0.0437 1.342 1.232-1.462 < 0.001 Disability Troubles of chronic to neither troubles nor Severe to moderate disease without disability 0.0566 0.693 0.620-0.774 < 0.001 -0.034 0.0670 0.967 0.848-1.103 0.616 -0.021 0.0516 0.980 0.885-1.084 0.691 disability versus Light disability vs severe -0.367 to moderate Mental health *All significant results are indicated in bold. Chapter 3: Research papers Discussion and conclusion Studies exploring the social gradient in health care utilization are often impeded by important methodological limitations. Many are based on self-reported utilization data that are subject to under- or overreporting. When registered data are used, detailed information on health and health care need is often lacking [9, 12]. This study aims to examine equity in the use of primary care and specialist care in Belgium by linking the data on health and social status from the respondents from the Belgian Health Interview Survey with their corresponding registered medical utilization data delivered by the Sickness Funds. This study shows that the specialist is not the doctor of the rich, but the doctor of the higher educated, and the FP is not the doctor of the poor, but the doctor of the lower educated: higher-educated people have more contact with a specialist and less contact with a FP than the lower-educated people. While for income, only a significant result was found on the number of contacts with a FP for middle income people compared to lower income people. No significant results were found in relation to specialist care. Attention is needed when interpreting the results on income as literature shows that the measurement of income is far less stable and valid than the measurement of education [26, 27]. These results contrast with findings reported in studies using self-reported utilization data. A study using the Belgian Health Interview Survey indicates that the social gradient in the utilization of FP care disappears when health status is taken into account [7]. A study on the health care use in 21 OECD countries reports that contacting a FP is distributed according to need and not much according to income [10]. With regard to the use of specialist care, our findings are consistent with the evidence based on self-reported data: Taking into account the disease burden, higher SES people seek specialist care more often than do lower SES people [8, 10]. A limited number of other studies have linked population survey data with registered utilization. The study from Nova Scotia using the 1990 Nova Scotia Nutrition Survey reported a higher use of FP and underuse of specialist care by people in lower income groups. However, this study did not adjust for health status [16]. A study on the Ontario portion of the 1994–1995 National Population Health Survey found an increased use of specialist care among those with a higher educational level but not among those with a high income. This study controlled for health status as being self-reported health as a single measure [17]. A more recent study on the 2000–2001 Canadian Community Health Survey examined equity in the use of FP and specialist care in Ontario, after controlling for health status using both 79 Chapter 3: Research papers self-reported and diagnosis-based measures. They found no differences according to income or education with regard to primary care use. With regard to specialist care, they found inequity favouring the well-educated people [11]. Important to notice on these studies is that they were all conducted in Canada which provides a universal health care coverage for all its residents. 80 This study, in the context of a health care system with free choice of FP and specialist care (no gate-keeping), seems to indicate that the inequity in health care use in Belgium is an education issue where higher educated visit more frequently a specialist and lower educated visit more frequently a FP, independent of the patient’s health status. No significant associations were found for specialist care and primary care for low income people compared to high income people. These findings confirm the accessibility of primary care in Belgium for vulnerable population groups such as low-income or low-educated people. The social differences in utilization patterns could partly be explained by the complex social context that this patient group is confronted with and by the greater health-damaging environment they live in: lower-educated patients are often simultaneously exposed to multiple health-threatening risk factors (e.g. low income, unhealthy working and living conditions) and experience a greater impact when exposed to risk factors than higher-educated patients [6]. Next to the social context, there could also be the issue of the cultural access to health care such as language barriers, miscommunication between health care provider and patient due to different priorities on the patient side, not explaining things in a way patients can understand, the attitude of the health care provider,… [6]. It might be possible that FP’s have a better doctor-patient relationship with their patients than specialists do and have better understanding of the needs and priorities of patients from lower social classes, resulting in a preference of consulting a FP instead of a specialist when needing medical care among lower educated people. The results indicate that education has an important role in the health care use in Belgium. First of all, there is the aspect of health literacy. Higher educated people suffering from a disease or condition wherefore specialist care is needed will probably go immediately to a specialist because they believe it’s the best care for that disease. While people with a low educational level might visit the FP because they do not know where else to go with this disease. Furthermore, there is also the organization of the health care system. It is possible that lower educated people find the health care system too complicated (e.g. waiting lists, appointments,…). This stresses the importance of the central function of the FP. The FP can navigate people through the different Chapter 3: Research papers health services and is able to mobilize the support of other facilities by referring patients or calling on the support of specialized services [28]. This study finds no significant results for specialist care and FP care for low income people. This might indicate that the health care system in Belgium has little financial barriers. There is a reduced co-payment for lower socio-economic groups in the fee-for-service system, there is the third-party payment (where the patient only has to pay the co-payment), the maximum bill (where the total amount of co-payment each year is limited to a certain amount), and the capitation system in the Community Health Centres and some larger group practices. Another important finding of the study is that middle income people have a higher use of the FP than lower income people. It is possible that the measures in order to reduce the financial barriers to health care use for vulnerable populations, have made it more easily for middle income people to visit the FP. However, we cannot make statements on the excessive or underuse of certain populations groups. This study only describes that there are social differences in the utilization patterns of FP and specialist care after controlling for health status. This study has strengths and limitations. Among its strengths, we used both education and income to determine socio-economic status, took into account health status using several measures for physical and mental health, and created a linkage with registered health care data. By doing so, we overcome the bias of under- or over-reporting health care use when using self-reported health care utilization data. These strengths also touch the limitations of this study. We used self-reported educational level and income as a measure for socio-economic status. Including recorded information on income might add to the validity of the results seen the sensitivity of people providing information on their personal income [26]. Further, it would be interesting for further studies to take into account other determinants of care, such as the strength of the patient’s social network or his/her coping abilities. Furthermore, we did not have information on the “reasons for encounter”, but instead we used self-rated health, chronic conditions and disability due to chronic conditions to measure the respondent’s health status. Taking into account reasons for encounter might add to the validity of the results. Another limitation consists of the drop-out of people by creating a linkage between the two databases. First of all, for some respondents, no identification number was found, which made linking the data of these individuals impossible. In addition, the self-employed were excluded from the research group, because the Sickness Funds do not have complete data on their consumption patterns. Including the self-employed might give a more 81 Chapter 3: Research papers complete picture on health care use in Belgium. Another limitation of this study is the time frame: We analyzed data from the 1997 National Health Interview Survey. Designing the technical procedure to link the databases, obtaining permission from the National Privacy Commission and from the administrators of the databases to link the data, and collecting the data took several years. This could largely be explained by the fact that it was the first and only time in Belgium that the link between administrative data and NHIS data was made. Since a similar linkage with more recent data did not appear possible, this study was done on the 1997 database. It is possible that since then, the social gradient in health care use changed. Furthermore, as Van der Heyden et al. already indicated, a weakness of assessing socio-economic differences based on data from a health survey is that specific subgroups of people such as homeless and illegal immigrants are not included in the sample. Even though they might be smaller groups in the Belgian society, these people especially may have major problems with access to health care [7]. 82 Health care utilization data are important in health care research, and the accuracy of this information is of utmost importance. This study shows that analyzing registered utilization data leads to different findings than when self-reported utilization data are analyzed, especially for the use of FP care. There is an inequity in the use of primary and specialist care, after taking health status into account. Lower-educated people have a higher tendency of contacting a FP than do higher-educated people, whereas higher-educated people have a higher tendency in using specialist care than do lower-educated people. The results of this study also stress the importance of taking into account the “need for health care” when assessing the social gradient in medical care use. By doing so, a more precise picture of the social difference in the utilization of medical care appears. Although the Belgian health care system has an almost 100% coverage of the population, this does not lead to an equitable utilization pattern. An explanatory factor could be the lack of a gate-keeping system in the context of a fee-for service with co-payments, which might lead to the higher utilization (taking into account health needs) of specialist care by higher-educated people. Competing interests. The authors declare that they have no competing interests. Chapter 3: Research papers Authors’ contribution SW contributed to the application for the National Privacy Commission to obtain permission to use and link the data. IP contributed to the data manipulation and linkage of the data. EV, IP, and SW contributed to the analysis. EV and SW contributed to the writing of the article. IP and JDM participated in the revision of the manuscript. All the authors have read and approved the final manuscript. Acknowledgements The authors would like to thank all Sickness Funds, the National Institute for Health and Disability Insurance (NIHDI), the Scientific Institute for Public Health, the Crossroad Bank for Social Security, and the Federal Agency for Social Security for their cooperative collaboration. This article outlines the results of a pilot study conducted as part of a research program concerning the accessibility of the Belgian healthcare system for disadvantaged people and was funded by the Federal Agency for Social Security. 83 Chapter 3: Research papers References 1. Braveman P, Gruskin S. Defining equity in health. J Epidemiol Community Health 2003; 57:254-8. 2. Goddard M, Smith P. Equity of access to health care services: theory and evidence from the UK. Soc Sci Med 2001; 53:1149-62. 3. Culyer AJ. Equity - some theory and its policy implications. Journal of medical ethics 2001; 27:275-83. 4. Bayoumi AM. Equity and health services. Journal of public health policy 2009; 30:176-82. 5. Hanafin S, Houston AM, Cowley S. Vertical equity in service provision: a model for the Irish public health nursing service. J Adv Nurs 2002; 39:6876. 6. Whitehead M, Dahlgren G. Levelling up (part 1): a discussion papar on concepts and principles for tackling social inequities in health. WHO collaborating Centre for Policy Research on Social Determinants of Health, University of Liverpool, 2006. 7. Van der Heyden JH, Demarest S, Tafforeau J, Van Oyen H. Socio-economic differences in the utilisation of health services in Belgium. Health Policy 2003; 65:153-65. 8. Bongers IM, van der Meer JB, van den Bos J, Mackenbach JP. Socio-economic differences in general practitioner and outpatient specialist care in The Netherlands: a matter of health insurance? Soc Sci Med 1997; 44:1161-8. 9. Fernandez de la Hoz K, Leon DA. Self-perceived health status and inequalities in use of health services in Spain. Int J Epidemiol 1996; 25:593-603. 84 10. van Doorslaer E, Masseria C, Koolman X. Inequalities in access to medical care by income in developed countries. CMAJ 2006; 174:177-83. 11. Glazier RH, Agha MM, Moineddin R, Sibley LM. Universal health insurance and equity in primary care and specialist office visits: a populationbased study. Ann Fam Med 2009; 7:396-405. 12. Ritter PL, Stewart AL, Kaymaz H, Sobel DS, Block DA, Lorig KR. Selfreports of health care utilization compared to provider records. J Clin Epidemiol 2001; 54:136-41. 13. Bellon JA, Lardelli P, Luna JD, Delgado A. Validity of self reported utilisation of primary health care services in an urban population in Spain. J Epidemiol Community Health 2000; 54:544-51. Chapter 3: Research papers 14. Roberts RO, Bergstralh EJ, Schmidt L, Jacobsen SJ. Comparison of selfreported and medical record health care utilization measures. J Clin Epidemiol 1996; 49:989-95. 15. Ferrer RL. Pursuing equity: contact with primary care and specialist clinicians by demographics, insurance, and health status. Ann Fam Med 2007; 5:492-502. 16. Veugelers PJ, Yip AM. Socioeconomic disparities in health care use: Does universal coverage reduce inequalities in health? J Epidemiol Community Health 2003; 57:424-8. 17. Finkelstein MM. Do factors other than need determine utilization of physicians’ services in Ontario? CMAJ 2001; 165:565-70. 18. Danchin N, Neumann A, Tuppin P, De Peretti C, Weill A, Ricordeau P, Allemand H. Impact of free universal medical coverage on medical care and outcomes in low-income patients hospitalized for acute myocardial infarction: an analysis from the FrenchNational Health Insurance system. Circulation Cardiovascular quality and outcomes 2011; 4:619-25. 19. Tuppin P, Blotiere PO, Weill A, Ricordeau P, Allemand H. [Mortality and hospital admissions rates and diagnosis among individuals with low income and full health insurance coverage in France, 2009]. Presse medicale (Paris, France : 1983) 2011; 40:e304-14. 20. Van oyen H, Tafforeau J, Hermans H, Quataert P, Schiettecatte E, Lebrun L, Belamer L. The Belgian Health Interview Survey. Arch Public Health 1997; 55:1-13. 21. Demarest S LP, Tafforeu J, Tellier V, Van der Heyden J, Van Oyen H. De gezondheid van de bevoling in België. Gezondheidsenquête door middel van interview, België 1997. Brussel: Wetenschappelijk Instituut Volksgezondheid, Afdeling Epidemiologie, 1998. 22. Crossroads Bank for Social Security. Mission. Brussels: Crossroads Bank for Social Security. 23. Isaacs SL, Schroeder SA. Class - the ignored determinant of the nation’s health. N Engl J Med 2004; 351:1137-42. 24. Smith GD, Hart C, Watt G, Hole D, Hawthorne V. Individual social class, area-based deprivation, cardiovascular disease risk factors, and mortality: the Renfrew and Paisley Study. J Epidemiol Community Health 1998; 52:399-405. 25. de bruin A PH, Nossikov A. Health interview surveys. Toward international harmonization of methods and instruments. Copenhagen:WHO-Europe, CBD Netherlands. 1996. 85 Chapter 3: Research papers 26. Galobardes B, Shaw M, Lawlor DA, Lynch JW, Smith DG. Indicators of socioeconomic position (part1). J Epidemiol Community Health 2006; 60:7-12. 27. Liberatos P, Link BG, Kelsey JL. The measurment of social class in epidemiology. Epidemiologic reviews 1988; 10:87-121. 28. WHO report 2008: Primary health care now more than ever. 2008. 86 Chapter 3: Research papers Verlinde et al. BMC Health Services Research 2010, 10:242 http://www.biomedcentral.com/1472-6963/10/242 RESEARCH ARTICLE Open Access Unique health care utilization patterns in a homeless population in Ghent Evelyn Verlinde*, Tine Verdée, Mieke Van de Walle, Bruno Art, Jan De Maeseneer, Sara Willems Abstract Background: Existing studies concerning the health care use of homeless people describe higher utilisation rates for hospital-based care and emergency care, and lower rates for primary care by homeless people compared to the general population. Homeless people are importantly hindered and/or steered in their health care use by barriers directly related to the organisation of care. Our goal is to describe the accessibility of primary health care services, secondary care and emergency care for homeless people living in an area with a universal primary health care system and active guidance towards this unique system. Methods: Observational, cross-sectional study design. Data from the Belgian National health survey were merged with comparable data collected by means of a face-to-face interview from homeless people in Ghent. 122 homeless people who made use of homeless centres and shelters in Ghent were interviewed using a reduced version of the Belgian National Health survey over a period of 5 months. 2-dimensional crosstabs were built in order to study the bivariate relationship between health care use (primary health care, secondary and emergency care) and being homeless. To determine the independent association, a logistic model was constructed adjusting for age and sex. Results and Discussion: Homeless people have a higher likelihood to consult a GP than the non-homeless people in Ghent, even after adjusting for age and sex. The same trend is demonstrated for secondary and emergency care. Conclusions: Homeless people in Ghent do find the way to primary health care and make use of it. It seems that the universal primary health care system in Ghent with an active guidance by social workers contributes to easier GP access. Background Homelessness is worldwide an important societal issue as it can be considered as unacceptable and unfair. In the USA, each year over three million people experience homelessness of which 1,3 million children [1]. In England 99.5000 households were officially recognised as newly homeless in 2007 [2]. In the Flemish region each year 12.000 people get support in residential care for homeless persons [3]. Compared to a decade ago, the homeless population today is younger (between 30 and 50 years) with a growing number of women and children. This trend continues both in the USA and in Europe [4-7]. The way people become homeless can be attributed to a number of factors such as a disruptive family environment * Correspondence: [email protected] Department of General Practice and Primary Healthcare, Ghent University, Belgium characterized by extreme poverty, marital discord, addiction, financial problems (no job, no money) and mental health problems [7-10]. Homelessness and health Homeless people face many challenges. They lack the access to basic human needs such as shelter, clothing and healthy food, and have a disproportionally higher burden of disease than the average population [11]. They suffer from a wide range of medical problems, acute as well as chronic, e.g. tuberculosis, [8,12-14] hepatitis [7,14], HIV [4,7,14-16], influenza [13] and skin and soft tissue infections [8,13,15,16]. On top of the physical diseases, homeless people are more frequently caught in the downward spiral of having mental health problems or substance abuse: suffering from disorders such as depression, bipolar disorder or schizophrenia makes it even more difficult to find a job, stable housing © 2010 Verlinde et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. 87 Chapter 3: Research papers Verlinde et al. BMC Health Services Research 2010, 10:242 http://www.biomedcentral.com/1472-6963/10/242 or medical care [4,10,12,15,17]. These higher morbidity rates within the homeless population are also reflected in raised mortality rates [4,7,9,15]: studies on the European and US situation report life expectancies between 41 and 47 years [4-6,18,19]. Homeless and health care utilisation 88 The literature concerning the health care utilisation of homeless people is scarce and almost exclusively reporting on the situation in the UK and the US. These studies show low utilisation rates of primary care among homeless people, with emergency care often being the only form of health care use [6,7,14,16,20-23]. As a result homeless people often consult with severe diseases and complications that could be prevented through earlier contact with a health care provider [22]. This particular pattern in health care use by homeless people is largely determined by the way the health care system is organised [7,20-23]. In the USA, the health care system relies almost entirely on the private sector for both the financing [24] and the delivery of the health care. Most health care facilities are privately owned and operated. Some vulnerable groups are covered by social insurance programs such as Medicare, Medicaid and the State Children’s Health Insurance Program. Even though these programs contribute to a more accessible health care for those groups, they are not able to provide access to all types of medical care nor do they cover all groups of vulnerable patients [25]. Unsurprisingly, US based studies reporting on barriers in healthcare by homeless people describe several financial barriers to health care because of the lack of insurance (high insurance fees, refusing homeless patients because of missing identification cards,...), high costs of medication, competing priorities (finding housing or employment is more important than addressing health problems) [6,16,20]. Literature from the UK shows a slightly different picture, reflecting the particularity of the UK health care system. The UK has a universal health care system free at the point of service if people are registered with a GP [26]. It is in fact this registration procedure which forms an important barrier to primary health care for homeless people in the UK: they often don’t know they have to register or are scared off by the complexity of the registration procedure, GPs are reluctant to accept them on their list,... [7,14,21,23,24]. Although the UK provides a health service for all, marginal groups, such as homeless people, are poorly served and sometimes excluded. Since 1997 the NHS has focused on improved primary health care services for these special needs groups [27]. Specialised general practices that register only homeless people have become more common [28]. But despite the installation of this selective health care system, parallel to the Page 2 of 9 universal system, the majority of the homeless is not aware of its existence and use emergency care as a substitute for primary care [29]. Information on the health care use of homeless people in other countries and the influence of health system characteristics seems to lack. Furthermore, literature also reports on a wide range of factors and mechanisms not related to the organisation of the health care system contributing to the particular health care use patterns of homeless patients. The majority of these studies emphasize the importance of the attitude of health care providers towards homeless patients. Care is often postponed because homeless people feel labelled, stigmatized or considered as “problem patients”[11,21]. In the region of Ghent (Belgium) the care for homeless people is organised in a unique way. In general, Belgium has a universal health care system with direct access to any general practitioner or specialist, without gate-keeping nor a patient-list. There is a fee-for-service system with about 30% co-payment for primary health care and 40% for specialist care. The lower socio-economic groups have a reduced co-payment of 8%. This co-payment is limited each year by an income-based threshold: the maximum bill [30]. Since 1982, the legal possibility to work in a capitation system in primary health care has been put in practice. In the area of Ghent (a midsized city in Belgium with 225 000 inhabitants) the 19 th century belt of deprived areas is to a large extent covered by Community Health Centres that provide interdisciplinary comprehensive primary health care using a capitation payment system. They adopt a universal approach in which all people residing in the neighbourhood can access the services of the centre. Less privileged residents such as the homeless staying in the area (e.g. frequenting the night shelters, living in squats, ...) are actively guided towards the community health centres by social welfare services. More specifically, there are formal agreements between the shelters for homeless people and the Community Health Centres in Ghent; in case people staying in the shelter need medical care, the nearby Community Health Centre is warned and (in most cases) a doctor comes to the shelter to see the patient. In rare occasions, the patient consults the doctor in the Community Health Centre (e.g. when he/she has been there before and he/she is able to walk to the Centre). Also for the payment of the costs, arrangements are made: all costs are directly refunded by the shelter or by the patient’s Sickness Fund. When the patient is not insured, the Community Health Centres tend to ask no fee. Most homeless people living in the streets are usually known by the social workers working in the neighbourhood such as the outreachers and street corner workers. In case a homeless person Chapter 3: Research papers Verlinde et al. BMC Health Services Research 2010, 10:242 http://www.biomedcentral.com/1472-6963/10/242 needs medical care, the social workers refer him/her (or go with them) to one of the shelters (and usually do a follow-up whether the person did go to the shelter) or takes him/her directly to a doctor. There is no formal agreement between the street corner workers and the Community Health Workers, however the strong informal networks between them and the low financial and other thresholds of the Community Health Centres, makes that also the street corner workers usually lead the homeless people towards the Community Health Centres. In one part of the area where there is no community health centre, people are guided towards a large group practice with a comparable way of working to the community health centres, except for the payment system which is fee-for-service based without co-payment for the patient. Seen the particular organisation of the health care use in Ghent, we want to describe the accessibility of primary health care services, secondary care and emergency care for homeless people living in an area with a universal primary health care system and active guidance towards the system. Methods The National Health Survey is a rich source of information when studying health and health care use. This cross-sectional survey is used to collect data from a representative sample of the non-institutionalised population residing in Belgium. For the 2004 survey, 12.945 Belgians were interviewed using both a face-to-face survey and a self-registering survey. Sampling has been based on a combination of stratification, multistage sampling and clustering. Stratification was done at the regional level and the level of the provinces. The number of selected individuals per province was determined by the proportions of the Belgians living in that province. In order to keep the fieldwork manageable, the number of interviews in each municipality should at least be 50. Per stratum, households were selected by randomly selecting so called ‘reference persons’ from the State Register. In every selected household, maximum four persons were questioned: the ‘reference person’, his/her partner and one or two other household members. Two hundred and thirty five inhabitants of Ghent were questioned. Finally, the data were anonymized by a trusted third party (the National Institute for Statistics) and made available for external researchers. For this study a variable was add to the dataset, indicating whether the participant is an inhabitant of Ghent. However, homeless people are not included in the study. Therefore additional data from homeless people in Ghent was collected using a face-to-face survey, according to an adapted form of the National Health Page 3 of 9 Survey protocol and using a reduced version of the NSH questionnaires. The final database, used for this paper, contains data from 235 inhabitants with a fixed address in Ghent and data from 122 homeless people staying in a shelter in Ghent. Participants and sample For this study the study population was restricted to all the homeless people making use of the homeless centres and shelters in Ghent. This choice was partly taken based on pragmatic reasons (homeless centres and shelters are the places where one can reach the homeless easily and with a good level of safety for the interviewers). The exact total number of homeless people visiting shelters and homeless centres is not available however social workers estimated this total number at the time of the data collection on 250. All the homeless people who made use of homeless centres and shelters in Ghent during the inclusion period of five months were invited to participate in the study by the researchers or the organisation’s personnel (in most cases the social worker). Hereby no distinction was made between people occasionally sleeping in the shelter or people semi-permanently sleeping in the shelter. Inclusion criteria for the study were: at the moment of the survey staying in an emergency or temporary accommodation in Ghent (shelter, homeless centre,...) (1), being homeless according to the definition of the Health Council (2), being aged between 18 and 65 years (3), and giving oral and written consent (4). People who were noticeably intoxicated at the moment of the interview or were aggressive towards the interviewers were not included in the study as were those visitors who did not speak Dutch, French or English, or refugees. All participants gave written consent. This study was approved by the Ethics Committee of Ghent University Hospital (project EC UZG 2006/060). Data collection Data were collected by means of a face-to-face survey being a reduced version of the Belgian National Health Survey (questions on the gender, age and nationality of the respondent, his/her socio-economic status, health status and smoking, and health care use). The following variables were used to study the possible confounding influence of other variables. Socio-economic status, based on the highest qualification the responded acquired and was categorised in 4 categories: no education/primary school, first 3 years of vocational school, vocational school completed (6 or 7 years), higher education/university. Health Status was measured 89 Chapter 3: Research papers Verlinde et al. BMC Health Services Research 2010, 10:242 http://www.biomedcentral.com/1472-6963/10/242 by means of subjective health status (How is your health in general?). Smoking status was measured by ‘Are you a current smoker’. The utilisation of health care was studied in terms of the likelihood to consult a GP, secondary care and emergency care and defined in terms of whether or not the individual consulted a general practitioner (GP), secondary care or an emergency department (ED) during the last two months. Administrating the survey took place in a separate room of the organisation to respect the participant’s privacy. Interviews lasted between 30 and 60 mins. Data analyses 90 The data were collected in the context of a larger study on the social gradient in health care use. For this paper, a selection of the database was used. More specifically the data from the homeless people were compared to the data from the 18 to 65 year old inhabitants of Ghent available in the most recent National Health Survey data base (2004). None of the latter were homeless. Data were analysed using SPSS version 16. Initially, 2dimensional crosstabs with c ² -statistic were built in order to study the bivariate relationship between the homeless and non-homeless respondents: controlling for sex, age, education, nationality, subjective health, smoking behaviour and health care use. Secondly, 2-dimensional crosstabs with c²-statistic were built in order to study the bivariate relationship between the health care use of the homeless and possible influencing variables such as sex, age, education, nationality, subjective health, smoking behaviour and type of practice. To determine the independent association between health care use and being homeless, a logistic regression model was constructed adjusting for age and sex. An odds ratio and 95% CI for contact with GP and contact with the ED in past two months was determined for each variable used in the model. The enterwise method was used to select the final model. The level of significance was tested by Wald tests, and P < 0,05 was set as being the level of statistical significance. Results Table 1 describes the homeless respondents compared to the non-homeless respondents. It demonstrates that the homeless respondents are significantly younger (p < 0,001) and lower educated (p < 0,001) than the general population. Also the proportion of males (p < 0,001) and of people with a nationality other than Belgian (p = 0,002) is significantly higher in the homeless population. Forty percent of the homeless respondents report to be in fair, bad or very bad health, compared to 34 percent in the general population (p < 0,001) and the smoking rate is almost 3 times as high than in the general population (p < 0,001). Page 4 of 9 Concerning health care utilisation, bivariate analysis demonstrate a significant association between being homeless and the likelihood to consult a GP (p < 0,001), secondary care (p = 0,005) and emergency care (p < 0,001), with higher likelihoods for all three forms of care in the homeless people. Bivariate analysis also show an additional significant association between likelihood to consult a GP and perceived health status (p = 0,019). For the likelihood to consult an ED and likelihood to consult secondary care we found no significant associations (see table 2). When homeless people consult a GP, they do so at a group practice (56 % versus 5,4%), more often than when non-homeless patients consult a GP (see table 3). To determine the independent impact of being homeless on the likelihood to consult a GP, secondary care and an ED, three multiple logistic regression models were built adjusting for sex and age (see table 4). The model focusing on likelihood to consult a GP explained 13,8% of the variance and demonstrated that even when adjusting for age and gender, homeless people have a significant higher likelihood to consult a GP (see table 4). The odds ratio for the homeless is 3,641 representing the homeless persons’ increase in likelihood to consult a GP (95% CI = 2,160 - 6,139). The model focusing on likelihood to consult secondary care explained 8,7% of the variance and demonstrated after adjusting for age and gender, being homeless increases the likelihood to consult secondary care. The odds ratio for the homeless is 3,271 representing the homeless persons’ increase in likelihood to consult secondary care (95% CI = 1,800 - 5,943). The comparable model for likelihood to consult an ED explained 27,7% of the variance and demonstrates that being homeless also increases the likelihood to consult an ED, after adjusting for age and sex (see table 3). Homeless patients have a 13,351 times higher likelihood to consult an ED compared to the general population (OR: 13,351 95% CI = 5,114 - 34,857). Regression models additionally adjusting for perceived health status follow the same trend. The explained variance increases to 21,5% for the model focussing on likelihood to consult a GP, to 11,9% for the model focussing on likelihood to consult secondary care and to 33,3% for the model focussing on likelihood to consult an ED. In these models, the OR for a homeless person to be likely to consult a GP is 2.592, to consult secondary care 2,683 and to consult an ED 10,752. Discussion and conclusion Homeless people have a disproportionately higher burden of health problems than the general population. Existing studies (mainly from the UK and the US) describe higher utilisation rates for hospital-based care and emergency Chapter 3: Research papers Verlinde et al. BMC Health Services Research 2010, 10:242 http://www.biomedcentral.com/1472-6963/10/242 Page 5 of 9 Table 1 Homeless people versus non-homeless people: descriptive and univariable results Homeless Not homeless n (%) n (%) P* 121 253 < 0,001 Male 101 (83,5) 124 (49,0) Female 20 (16,5) 129 (51,0) Sex Age 121 253 18-29 42 (34,7) 71 (28,1) 30-39 31 (25,6) 51 (20,2) 40-49 50-59 31 (25,6) 15 (12,4) 56 (22,1) 47(18,6) 60-65 2 (1,7) 28 (11,1) Educational level 121 223 No education/primary school lower education 17 (14,0) 18 (8,1) Lower vocational school 23 (19,0) 27 (12,1) Higher vocational school 74 (61,2) 87 (39,0) 7 (5,8) 91 (40,8) Higher education/University Nationality Belgian Non-Belgian 122 251 105 (86,8) 16 (13,2) 240 (95,6) 11 (4,4) Subjective health perception 120 237 Good or very good 73 (60,8) 203 (85,7) Fair, bad or very bad 47 (39,2) 34 (14,3) Current smoker Yes Contact with GP last 2 months Yes Contact with secondary care last 2 months Yes Contact with ED in last 2 months Yes Type of GP practice Single handed Two handed Group practice < 0,001 < 0,001 0,002 < 0,001 < 0,001 112 253 100 (89,3) 72 (30,5) 120 238 77 (64,2) 121 104 (43,7) 240 38 (31,4) 44 (18,3) 0,005 120 243 < 0,001 33 (27,5) 7 (2,9) 75 224 27 (36,0) 151 (67,4) 6 (8,0) 61 (27,2) 42 (56,0) 12 (5,4) < 0,001 < 0,001 * All significant results are indicated in bold. care, and lower rates for primary care by homeless people compared to the general population. Some studies indicate that homeless people do not access primary care at all and use emergency care as the only form of health care [6,16,20]. Homeless people are importantly hindered and/ or steered in their health care use by barriers directly related to the organisation of care [6,7,21,24]. In this study we describe the likelihood to consult a GP, secondary care and emergency care of homeless people living in an area with a primary health care system in which homeless people with health problems are actively guided to low-threshold but universal accessible (i.e. not selective) community health centres. Hereto the Belgian National Health Survey Database was extended with data of 122 homeless persons residing in one of the night shelters or temporarily accommodations for homeless people in Ghent. The homeless respondents in this study are significantly younger, lower educated, more likely to smoke, have more often a non-Belgian nationality and report to be in worse health than the general population. These findings are concordant with other studies reporting that homeless people have worse health, suffer more from multimorbidity, have higher smoking prevalence (up to 80%), and are mostly less educated, decreasing the chance to find a stable job [4,6,14,18,31,32]. Literature shows a growing number of women in the homeless population [33]. This could not be confirmed in this study in which over 80% of the respondents are male. This could partly be due to a selection bias; 91 Chapter 3: Research papers Verlinde et al. BMC Health Services Research 2010, 10:242 http://www.biomedcentral.com/1472-6963/10/242 Page 6 of 9 Table 2 health care use of the homeless people: descriptive and univariable results (total n homeless: 121) Had contact with GP n (%) Sex Total n Had contact with ED P* Total n n (%) 0,151 Had contact with secondary care P* 121 0,507 63 (80,8) 29 (87,9) 30 (78,9) Female 15 (19.2) 4 (152,1) 8 (21,1) 18-29 24 (31,2) 12 (36,4) 12 (31,6)15 (39,5) 30-39 40-49 20 (26,0) 20 (26,0) 11 (33,3) 5 (15,2) 7 (18,4) 4 (10,5) 50-59 11 (14,3) 4 (12,1) 0 (0,0) 60-65 2 (2,6) 1 (30,0) 0,693 Educational level 120 0,621 121 0,466 0,740 No education/primary school 92 n (%) Male Age lower education 11 (14,1) 6 (18,2) 15 (19,2) 7 (21,2) 9 (23,7) Higher vocational school 46 (59,0) 19 (57,6) 25 (65,8) Higher education/ University 6 (7,7) 1 (3,0) 1 (2,6) Nationality 0,344 Belgian 66 (84,6) Non-Belgian 12 (15,4) Subjective health perception Good or very good Yes 121 0,827 29 (87,9) 33 (86,8) 4 (12,1) 5 (13,2) 0,019 41 (52,6) 19 (50,0) 37 (47,4) 19 (50,0) 0,426 64 (87,7) Type of GP practice Single handed 113 0,683 28 (87,5) 121 0,274 121 0,171 121 0,274 121 0,989 121 0,116 113 0,257 0,396 19 (31,7) 76 0,629 33 (94,3) 76 8 (44,4) 0,647 10 (40,0) Two handed 5 (8,3) 1 (5,6) 1 (4,0) Group practice 36 (60) 9 (50,0) 14 (56,0) women (often with children) who become homeless usually stay in crisis centres for women and children which were not included in this study. However, possibly there are also less homeless women in Belgium than in other countries because of the relatively good safety net for women with children; they have priority in getting social housing or -in case of the unavailability of social housing-emergency housing (temporarily housing in a premises owned or rented by the council). Table 3 Type of practice Type of practice P* 3 (7,9) Lower vocational school Fair, bad or very bad Current smoker Total n Homeless Ghent sample Solo 32,2% 64,6% Duo 8,5% 28,3% Group practice 59,3% 7,1% Multivariate analyses show that homeless persons in Ghent have a higher likelihood to consult a GP (i.e. have consulted the GP at least once in the 2 months prior to the interview) than the general population after adjusting for sex, age and perceived health status. This finding indicates that in Ghent the homeless people find the way to primary health care and use it. This contrasts with studies from other countries. Several US based studies show lower utilisation rates of ambulatory care and up to 5 times higher utilisation rates of acute hospital-based care and emergency care in the homeless population compared to the US average [6,16,20]. Studies from the United Kingdom describe the same trend: homeless people contact secondary services more frequently than the housed population. Some studies describe that a considerable proportion of the Chapter 3: Research papers Verlinde et al. BMC Health Services Research 2010, 10:242 http://www.biomedcentral.com/1472-6963/10/242 Page 7 of 9 Table 4 Logistic regression model for going to the GP, the ED and secondary care in the past two months Variable Estimate SE Wald P* Odds ratio 95% CI 5,926 0,015 1,835 1,126-2,993 Contact with GP Sex 0,607 0,249 19,895 0,001 30-39 -0,146 0,313 0,218 0,641 0,864 0,468-1,595 40-49 0,070 0,305 0,052 0,819 1,071 0,589-1,951 50-59 1,173 0,359 10,672 0,001 3,233 1,599-6,537 60-65 Homeless in Ghent 1,242 -1,292 0,457 0,266 7,381 23,523 0,007 0,000 3,462 3,641 1,413-8,480 2,160 -6,139 Sex -0,032 0,475 0,005 0,969 0,382-2,456 5,929 0,205 30-39 -0,107 0,450 0,057 0,811 0,898 0,372-2,168 40-49 -1,214 0,563 4,655 0,031 0,297 0,099-0,895 50-59 60-65 -0,743 0,149 0,621 0,866 1,434 0,030 0,231 0,863 0,476 1,161 0,141-1,605 0,213-6,339 -2,592 0,490 28,016 0,000 13,351 5,114-34,857 0,002 2,470 1,387-4,398 Age 18-29 Contact with emergency department 0,946 Age 18-29 Homeless in Ghent Contact with secondary care Sex 0,904 0,294 9,441 4,362 0,359 30-39 40-49 0,391 -0,089 0,359 0,376 1,187 0,056 0,276 0,813 1,478 0,915 0,732-2,986 0,438-1,911 50-59 0,355 0,405 0,768 0,381 1,426 0,645-3,151 0,819 0,509 2,587 0,108 2,268 0,836-6,150 -2,592 0,305 15,130 0,000 3,271 1,800-5,943 Age 18-29 60-65 Homeless in Ghent * All significant results are indicated in bold. Nagelkerke R² contact with GP: 0,138. Nagelkerke R² contact with ED: 0,277, Nagelkerke R² contact with secondary care: 0,087 homeless population does not find the way to primary health care at all: a survey conducted by Crisis (the national charity for solitary homeless people in the UK) in 2002 reported that homeless people were three times more likely not to have had contact with a GP in the last year and were nearly five times more likely than the general population to attend the emergency departments [34]. A 2002 survey conducted in the UK also looked at the likelihood to consult a GP. They found that homeless persons were 3 times more likely not to contact their GP in the last year than the general population[34]. This higher likelihood in Ghent to consult a GP in the homeless population compared to the general population could be explained by the fact that primary health care for homeless people in Ghent is organised in a way that contributes to the reduction of some of the barriers demonstrated in others types of primary health care organisation such as financial barriers (insurance status), registration, accessibility, stigmatisation,... [7,11,14,21,23,24]. Tracing homeless people with health problems and active guidance towards primary care centres by social workers, might contribute to an easier access to the GP. The Community Health Centres are situated close the shelters and social restaurants, so transportation is often not an issue. There are also agreements on the payment procedure so the homeless people do not need to worry about money, reducing the financial barrier. Additionally, Community Health Centres are known not to stigmatise or refuse patients because of being homeless or having no money. This hypothesis is supported by the finding that the homeless people who visit a GP, do this considerably more often at a group practice (including the health centres) (56% versus 5,6% in the general population). Also, the universal approach of the primary care centres (providing care to everyone residing in the geographical area) leads to less stigmatization and labelling than in a selective setting focussing solely on the homeless population. In this study, the higher likelihood to consult a GP does not go hand in hand with a lower 93 Chapter 3: Research papers Verlinde et al. BMC Health Services Research 2010, 10:242 http://www.biomedcentral.com/1472-6963/10/242 94 likelihood to consult the emergency department or secondary care. This indicates that homeless people, although they have a higher likelihood to make use of the primary health care system, have not totally abandoned the way to emergency care. Probably this can partly be explained by the particularity of the health problems homeless people experience (e.g. postponing care until it’s really urgent or care during the evening or night). Homeless people are also more often the victim of drug related accidents and road accidents and suffer from health problems that need more emergency care than the general population (e.g. hypothermia)[35]. Additionally, this study has a cross-sectional design which makes that we do not know whether the homeless people would use more frequently the emergency services if there was the universal primary care system would have been organised differently. Moreover, since this study is the first to study the health care use of homeless people in Ghent, it is not possible to describe an increase or decrease in the emergency care use of the homeless population over time. This study has some limitations. First of all there is no reliable information available on the response rate of the homeless people nor on the respondents excluded from the study. The study was presented to all homeless persons present at different moments in one of the shelters in Ghent during the five months inclusion period and everyone was invited when fulfilling the inclusion criteria. However, there was no registration of any information concerning the people who did not want to participate. This was mainly due to anonymity reasons thrown up by the managers of the shelters. Also the exact total number of homeless people visiting shelters and homeless centres is not available. Most of the crisis centres have no registration of their visitors. A second and important limitation of this study is that only likelihood to consult is studied. The National Health Survey contains self reported frequencies of health care use, so analogously these frequencies were also included in the survey for the homeless people. However, most of the respondents found it very hard to remember the exact number of their visits to a care provider. For a more nuanced view on health care use, information on the number of consultations is necessary. The concept of likelihood to consult is in any case usable to evaluate whether people know and find their way to a certain care provider but not to evaluate the intensity of their health care use. Although this study cannot prove the impact of a better accessible primary health care system on emergency care utilisation rates or on secondary care utilisation rates, a recent study commissioned by the Belgian government gives an important indicator of the possible impact. In this report the authors compared the traditional, fee-for-service Page 8 of 9 based primary health care services in Belgium with capitation based primary health care services such as the community health centres. Their findings complement the findings of our study as they state that the community health centres indeed succeed in reaching the most vulnerable of society. Additionally this study provides evidence that people who found their way to these centres tend to contact less frequently the emergency department or secondary care[36]. The merits of this study lie in the fact that a relatively large sample of homeless people was questioned (an estimated third of the homeless population in Ghent visiting a homeless centre in the research period), following the research protocol and instruments of the National Health Survey as close as possible so that a comparison with National Health Survey data was made possible. However, prudence is recommended in the generalisation of the results. Data were collected from homeless people seeking help or assistance, mainly for food or a place to sleep, in the homeless centres or night shelters in Ghent. Without doubt their help seeking behaviour towards health care is different from that of homeless people not seeking help in the homeless centres. Also no information is available on the response rate and the characteristics of the people who refused to participate in the study, which makes it very difficult to evaluate the selection bias. Further research addressing the health care use of the homeless people living in Belgium is needed to provide policymakers with the evidence needed to take action in order to make the health care system accessible and suitable for every person disregarding their different socioeconomic backgrounds. The performance of universal primary health care services, as demonstrated in this study, may stimulate the debate between a universal or a selective approach. Acknowledgements We would like to thank the personnel of the homeless centres and shelters in Ghent for their cooperative collaboration. Authors’ contributions MVDW, TV, BA, SW and JDM contributed to the design of the study. MVDW and TV contributed to the data retrieval. EV and SW contributed to the analysis and the writing of the paper. BA, SW and JDM participated in the revision of the manuscript. All authors have read and approved the final manuscript. Competing interests The authors declare that they have no competing interests. Received: 8 April 2010 Accepted: 19 August 2010 Published: 19 August 2010 References 1. Homelessness and Poverty in America. [http://www.nlchp.org/hapia.cfm]. 2. Official Homelessness Statistics Overview. [http://www.crisis.org.uk/ policywatch/pages/homelessness_statistics.html]. Chapter 3: Research papers Verlinde et al. BMC Health Services Research 2010, 10:242 http://www.biomedcentral.com/1472-6963/10/242 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. Van Menxel G, Lescrauwaet D, Parijs I: Verbinding verbroken. De zorg voor thuislozen in het almeneen welzijnswerk. Berchem: Steunpunt Algemeen Welzijnswerk 2003. Hwang SW, Lebow JM, Bierer MF, O’Connell JJ, Orav EJ, Brennan TA: Risk factors for death in homeless adults in Boston. Arch Intern Med 1998, 158(13):1454-1460. Kushel MB, Perry S, Bangsberg D, Clark R, Moss AR: Emergency department use among the homeless and marginally housed: results from a community-based study. Am J Public Health 2002, 92(5):778-784. Kushel MB, Vittinghoff E, Haas JS: Factors associated with the health care utilization of homeless persons. JAMA 2001, 285(2):200-206. O’Carroll A, O’Reilly F: Health of the homeless in Dublin: has anything changed in the context of Ireland’s economic boom? Eur J Public Health 2008, 18(5):448-453. Hill RP: Health care and the homeless: a marketing-oriented approach. J Health Care Mark 1991, 11(2):14-23. Johnson TP, Fendrich M: Homelessness and drug use: evidence from a community sample. Am J Prev Med 2007, 32(6 Suppl):S211-218. O’Toole TP, Conde-Martel A, Gibbon JL, Hanusa BH, Freyder PJ, Fine MJ: Where do people go when they first become homeless? A survey of homeless adults in the USA. Health Soc Care Community 2007, 15(5):446-453. Martins DC: Experiences of homeless people in the health care delivery system: a descriptive phenomenological study. Public Health Nurs 2008, 25(5):420-430. Daiski I: Perspectives of homeless people on their health and health needs priorities. J Adv Nurs 2007, 58(3):273-281. O’Connell JJ: Dying in the shadows: the challenge of providing health care for homeless people. CMAJ 2004, 170(8):1251-1252. Wright NM, Tompkins CN: How can health services effectively meet the health needs of homeless people? Br J Gen Pract 2006, 56(525):286-293. Morris DM, Gordon JA: The role of the emergency department in the care of homeless and disadvantaged populations. Emerg Med Clin North Am 2006, 24(4):839-848. Hwang SW: Homelessness and health. CMAJ 2001, 164(2):229-233. Woollcott M: Access to primary care services for homeless mentally ill people. Nurs Stand 2008, 22(35):40-44. Cuddy R: Anonymous demise: mortality in the homeless. J Emerg Med 1997, 15(3):373-374. Hwang SW, Orav EJ, O’Connell JJ, Lebow JM, Brennan TA: Causes of death in homeless adults in Boston. Ann Intern Med 1997, 126(8):625-628. Gelberg L, Gallagher TC, Andersen RM, Koegel P: Competing priorities as a barrier to medical care among homeless adults in Los Angeles. Am J Public Health 1997, 87(2):217-220. Riley AJ, Harding G, Underwood MR, Carter YH: Homelessness: a problem for primary care? Br J Gen Pract 2003, 53(491):473-479. Power R, French R, Connelly J, George S, Hawes D, Hinton T, Klee H, Robinson D, Senior J, Timms P, et al: Health, health promotion, and homelessness. BMJ 1999, 318(7183):590-592. Victor CR, Connelly J, Roderick P, Cohen C: Use of hospital services by homeless families in an inner London health district. BMJ 1989, 299(6701):725-727. Crane M, Warnes AM: Primary health care services for single homeless people: defects and opportunities. Fam Pract 2001, 18(3):272-276. Docteur E, Suppanz H, Woo J: The US Health System: An Assessment and prospective directions for reform economics department working papers No. 350. 2003. Health Care Systems in Transition. United Kingdom. European Observatory on Health Care Systems 1999. Lewis R, Jenkins C, Gillam S: Personal Medical Services Pilots in London: Rewriting the Red Book. London: The King’s Fund 1999. Wright N: Homelessness: a primary care responde: London. Royal College of General practitioners 2002. Little GF, Watson DP: The homeless in the emergency department: a patient profile. J Accid Emerg Med 1996, 13(6):415-417. Willems S: The socio-economic gradient in health: a never-ending story? A desriptive and explorative study in Belgium. Ghent 2005. Gelberg L, Leake BD: Substance use among impoverished medical patients: the effect of housing status and other factors. Med Care 1993, 31(9):757-766. Page 9 of 9 32. van Laere IR, de Wit MA, Klazinga NS: Pathways into homelessness: recently homeless adults problems and service use before and after becoming homeless in Amsterdam. BMC Public Health 2009, 9:3. 33. Lewis JH, Andersen RM, Gelberg L: Health care for homeless women. J Gen Intern Med 2003, 18(11):921-928. 34. Critical condition: Vulnerable Single Homeless People and Access to GP’s. [http://www.crisis.org.uk/downloads/php/101/GP_policybrief.pdf]. 35. Wright NM, Tompkins CN, Oldham NS, Kay DJ: Homelessness and health: what can be done in general practice? J R Soc Med 2004, 97(4):170-173. 36. Annemans L, Closon J, Heymans I, Lagasse R, Mendes Da Costa E, Moureaux C, Roch I: Vergelijking van kost en kwaliteit van twee financiëringssystemen voor de eerstelijnsgezondheidszorg in België. KCE reprots 85A. 2008. Pre-publication history The pre-publication history for this paper can be accessed here: http://www.biomedcentral.com/1472-6963/10/242/prepub doi:10.1186/1472-6963-10-242 Cite this article as: Verlinde et al.: Unique health care utilization patterns in a homeless population in Ghent. BMC Health Services Research 2010 10:242. 95 Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color figure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Chapter 3: Research papers 96 Chapter 3: Research papers Social differences in postponing a GP visit in Flanders, Belgium: Which low-income patients are most at risk? 1 1 2 3 1 Verlinde E. *, Poppe A. , DeSmet A. , Hermans K. , De Maeseneer J. , Van Au3 1 denhove C. , Willems S. 1Department of General Practice and Primary Healthcare, Ghent University, Belgium 2 Department of Movement and Sports Sciences, Ghent University, Belgium 3 Lucas, Center for Care Research and Consultancy at K.U. Leuven, Leuven, Belgium *Corresponding author Abstract One of the main goals of primary care is providing equitable healthcare, meaning equal access, equal treatment, and equal outcomes of healthcare for patients in equal need. Some studies show that patients from lower socioeconomic groups visit a GP more often, while other studies show that they are more likely to postpone a visit to a GP. In this study we want to explore within the social group of low-income patients living in Flanders, Belgium, which patients have a higher risk of postponing a visit to a GP. A face-to-face questionnaire was administered among 606 low-income users of Public Social Services. The questionnaire consisted of questions on socioeconomic and demographic characteristics, social networks, health and healthcare use. A multivariate logistic regression model was built in order to study the relationship between postponing or cancelling a GP visit respondents thought they needed and variables on health and socio-demographic background. The multivariate regression indicates that depression, self-rated health, and trust in the GP independently predict postponing a visit to a GP. Low-income people with a low trust in the GP, people with ill health and people suffering from a severe depression are more likely to postpone or cancel a GP visit they thought they needed compared to other low-income patients. This might indicate that the access to healthcare for low-income people might be hindered by barriers which are not directly linked to the cost of the consultation. Keywords: Primary Care, Health Disparities, Consultation, Doctor-Patient relationship, Social class, low-income people 97 Chapter 3: Research papers What is known about this topic: • Studies show relatively high rates of postponing a visit to the GP by patients from lower socioeconomic groups due to factors related to the health care system. What this paper adds: • • 98 Within a low-income population in Belgium, people with low trust in the GP, with ill health, and people suffering from a severe depression have a higher chance of postponing a visit to the GP than other lowincome patients. The inability to make ends meet does not seem to lead to further postponing of healthcare. The universal approach for low income people in Flanders seems to be able to reduce the financial barriers towards health care use, but the access to health care for low income people is still hindered by barriers that are not directly linked to the price of the consultation Introduction The socioeconomic conditions in which people live powerfully influence their chances to be healthy. Poverty, social exclusion, discrimination, poor housing, unhealthy early childhood conditions, and low occupational status are important determinants of the systematic variations in health and illness between social groups, both between and within countries (Whitehead and Dahlgren, 2006). Despite the marked improvements in the health of the general population, these health inequities seem to persist and even increase for some diseases and/or population groups (Hart, 1971, Whitehead and Dahlgren, 2006). Massive investment in further stimulating the rise of (healthy) life expectancy will not automatically lead to less inequity in health. However, actions that strengthen primary health care might. Evidence shows that a stronger primary care system—in contrast with specialist care—is associated with a more equitable distribution of health among populations (Marmot et al., 2010). The beneficial effects of primary care can be attributed to the cumulative effect of its main delivery characteristics: the focus on the person rather than on the management of particular diseases, coordinating and integrating care provided elsewhere or by others in the healthcare system, the provision of a low-threshold entry for relatively deprived population groups, delivering high-quality clinical care, providing Chapter 3: Research papers better access to preventive activities and early management of health problems and facilitating entry to the rest of the health system (Starfield et al., 2005, WHO). Thus, one of the main goals of primary care is providing equitable healthcare, meaning equal access, treatment, and outcomes of healthcare for patients in equal need. Studies show higher crude utilization rates of primary health care (PHC) (i.e. utilization rates without taking into account the disease burden) by people from lower socioeconomic groups, in contrast to people from higher socioeconomic groups, who tend to visit specialists more frequently. When taking into account the higher morbidity of people from the lower social classes, the difference in PHC utilization between the higher and lower social classes seems to melt away partially, whereas specialist care still seems to favor the better off (van der Meer et al., 1996, Van der Heyden et al., 2003, Regidor et al., 2008, Campbell, 1999, Droomers and Westert, 2004, Veugelers and Yip, 2003). Although these results tell a positive story about the accessibility of primary healthcare, other studies show relatively high rates of postponing a visit to the GP by patients from lower socioeconomic groups (Drieskens et al., 2010, Vilhjalmsson, 2005, Burstrom, 2002). In Belgium, each five years a Health Interview Survey (HIS) is conducted. This is a cross-sectional survey on health, life style and health care use based on a representative sample of the noninstitutionalized population residing in Belgium who are aged 15 years or older (Van der Heyden et al., 2003). In the 2008 HIS, 18% of the households with a low educational level reported having postponed a visit to a GP in the last 12 months while they actually needed one, whereas only 9% of the households with a high educational level reported having done the same thing (Drieskens et al., 2010). The results from the Health Survey in Iceland show that younger individuals, the nonwidowed, the economically troubled, the chronically ill and those who have to pay high out-of-pocket payments relative to their income were more likely than others to postpone or cancel a doctor visit they thought they needed (Vilhjalmsson, 2005). There are several reasons why people might postpone healthcare: Some people rely more on self-care than others or find informal help for their medical problems in their social network (eg. from a friend or a relative working in the healthcare sector). However, sometimes people postpone healthcare because of reasons related to the organization of the healthcare system. Thus, a distinction can be made between problems in the physical access of care (eg. problems with transportation), the financial access, psychosocial (eg. feelings of shame or embarrassment) and cultural access (eg. language problems), administrative access (eg. problems of finding the way in a complex healthcare system), or doctor-patient communication, which does not 99 Chapter 3: Research papers succeed in meeting the patient’s need and expectations (Willems, 2005, Willems et al., 2005, Ahmed et al., 2001). Postponing or cancelling a needed visit to a doctor might have an important impact on the patient’s health status. People from a lower social class suffer more often from a higher disease burden (Droomers and Westert, 2004). Postponing medical care can lead to more serious health problems that could be prevented in an earlier stage, but now require hospitalization and specialist care. In this study, we want to explore within a vulnerable population group, namely of low-income people, which patients have a higher risk of postponing a needed GP visit due to factors related to the organization of the healthcare system. We believe that people who are not able to cope with their monthly income have a higher risk of postponing a visit to a GP. Furthermore, special attention will be given to the vulnerable subgroups within a low income population: people without a paid job, people with young children living with them, people with a poor social network and people with mental health problems. 100 Methods Participants, sample, and procedure This study is a part of the ‘KANS’ project, which aims at describing the use of healthcare and welfare services among low-income people in eight Flemish regions: Ypres, Ostend, Genk, Ghent, Oudenaarde, Tielt-Winge, Geel, and Antwerp. Region selection is based on maximal variations in geographical spread, level of urbanization, socioeconomic profile of the region, and the availability of health and welfare services (Hermans, 2008). Between September 2008 and December 2009, social workers of the Public Social Services in the selected regions invited all their clients who met the inclusion criteria to participate in the study. The inclusion criteria were as follows: master the Dutch language, living in one of the eight selected regions, being aged between 18 and 65 years and having a low income (being a gross taxable family income lower than 13.543, 71 Euro, increased with 2.507,30 Euro per additional family member). People were excluded from the study when the social worker felt that participating in the study would unnecessarily confuse the client, he/she would feel worse by participating, or completing the questionnaire would create a larger burden than Chapter 3: Research papers the responder could handle. Once clients gave their consent to participate in the study, a trained interviewer contacted them by phone to set a date for completing the questionnaire. Data collection took place in the homes of the respondents or at their regular Public Social Service center by means of a combination of a written and an oral questionnaire. Filling in both questionnaires took between 45 minutes and 1h30 min. A total of 1148 low-income people were included in the study. Overall, 606 people completed the questionnaire. The interviewers could not contact 339 persons due to absence of the included person or due to a wrong telephone number. 203 included respondents refused to participate after all at the moment the interviewer contacted them. Additional information on these respondents was not available. Furthermore, no information is available on how many clients were invited to participate in the study by the social workers. Instruments The questionnaire consisted of questions on socio-economic and demographic characteristics, social networks, health, and healthcare use. Socio-demographic variables Socio-demographic variables included gender, age, educational level, subjective feeling of poverty, urbanization, having young children living with them and having a paid job at the moment of the study. Age was categorized into five groups (18-24, 25-34, 35-44, 45-54, and 55+). Education was categorized into lower (no education, primary school or lower secondary school), middle (higher secondary school or A-level), and higher (higher education) groups. Subjective feeling of managing with the available income (“To what extent can you manage with your current available household income?”) was categorized as a dichotomous variable (“very difficult,” “difficult,” or “rather difficult” versus “rather easy,” “easy,” or “very easy”). Level of urbanization was categorized into urban or rural. Having young children was categorized as having children younger than 12 years. Health care-related variables The health care-related variables included are as follows: level of trust in the GP (“complete distrust”, “distrust,” and “trust” versus “complete trust”) and satisfaction 101 Chapter 3: Research papers with the care delivered by the GP. The latter was measured by questions from the Consumer Quality Index (CQI) Family Medicine. The CQI contains 11 questions on what people expect from primary care and 11 questions on to what extent these expectations are met. Statements regarding the importance of several aspects of primary care were rated using four categories from “totally unimportant” to “very important” (0-3). Next, the occurrence of these aspects in the general practice they usually visit was measured in four categories from “never” to “always” (0–3). The scores for importance were deducted from the scores for the actual situation. This way, a score could be interpreted: 0 indicates that the actual situation exactly meets the desired needs; a negative score indicates a situation where the actual situation exceeds the needs; and a positive score indicates that the actual situation does not meet the desired level. Finally, a mean score for all questions regarding what people find important in primary care and the occurrence of it was calculated. If information on four or more questions was missing, then the mean was not calculated for this respondent. 102 Health status Health status is measured by self-rated health, disability due to a chronic condition and mental distress. Self-rated health was measured using a validated WHO instrument: “poor,” or “very poor” versus “fair” versus “good” or “very good” (de bruin A, 1996). Disability due to a chronic condition is categorized as: “no chronic condition”, “chronic condition without or with mild disability”, “chronic condition with moderate disability”, “chronic condition with severe disability”. The severity of mental distress is measured by the Beck Depression Inventory (BDI). People experiencing a form of mental distress based on the validated screening instrument developed by Whooley (Whooley et al., 1997) were asked to fill out the BDI. The Whooley instrument contains two questions: “During the past month, have you often been bothered by feeling down, depressed, or hopeless?” and “During the past month, have you often been bothered by little interest of pleasure in doing things.” When people answered positive on one of these questions, they scored positive on mental distress. Next, the Beck Depression Inventory was administered. The BDI is a 21-question multiple-choice self-report inventory and one of the most widely used instruments for measuring the severity of depression. When the test is scored, a value of 0 to 3 is assigned for each answer and then the total score is compared to a key to determine the depression’s severity. The standard cut-offs are as follows: 0–9: indicates minimal depression; 10–18: indicates mild depression; 19–29: indicates Chapter 3: Research papers moderate depression; 30–63: indicates severe depression. Social networks The aspects of social networks included in the analyses are network diversity and how often people feel lonely. Network diversity was measured based on the Social Network Index developed by Cohen et al (Cohen et al., 1997). Network diversity represents the number of social networks in which the respondent has regular contact with at least one other person. Lower is defined as zero to three social networks; middle is defined as four to five social networks; and higher is defined as six or more social networks. How often people feel lonely is categorized as “never” to “seldom,” “occasionally,” and “often” to “all the time.” Postponing a visit to a GP Respondents were asked whether they, during the past 12 months, had postponed or cancelled a GP visit they thought they needed. This postponement or cancellation can be due to reasons related to the organization of the healthcare system: financial reasons, no transportation, on a waiting list, uninsured, afraid to go to the doctor, practical reasons (no child care), did not know where to go, did not know how to express themselves, afraid the GP could not help, ashamed to go, or physically not able to go. Respondents answering positive on one of these reasons were included in the variable “postponing a visit with a GP in the past 12 months.” Data analyses In order to build the logistic model, two-dimensional crosstabulations with c² test statistics were used to study the bivariate relationship between having postponed a visit to a GP during the past 12 months and socio-demographics, health status, social networks, and healthcare-related variables. The independent variables that had a p value <0.25 in the crosstabs were used to build the multivariate logistic regression model. Before the logistic model was build, the variables were checked with regard to multi-collinearity. The cut-off point used was at a Pearson correlation coefficient of 0.6. The logistic models were fit using enterwise regression. The level of significance was tested by Wald tests, and a p < 0.05 was set as the level of statistical significance. An odds ratio and 95% CI were determined for each variable in the model. 103 Chapter 3: Research papers Ethics approvement The study was approved by the Ethics Committee of the Ghent University Hospital (Dossier No. 2008/343). Results Description of the participants 104 In this study, 606 low-income people were interviewed using an oral and written questionnaire. Table 1 provides a description of the study group. 32.1% postponed or cancelled a GP visit they thought they needed during the past 12 months. Slightly more women (327 – 54.0%) than men (279 – 46.0%) participated in the study. Overall, 75.4% was 35 years or older; 47.1% had a lower educational level; 39.7% had a middle educational level; and 13.2% had a higher educational level. In fact, 70.6% of the participants perceived difficulties in coping with their current available income. This could be related to the finding that 75.9% does not have a paid job at the moment of the survey. 16.2% of the respondents had children younger than 12 years living with them and 63.7% lived in a rural area. Self-rated health was rated from “very poor to poor” by 20.3% of the participants, 34.3% rated their health as fair and 45.4% reported experiencing good to very good health. At the moment of the survey, 39.8% did not suffer from any chronic condition, 8.1% had a chronic condition but did not perceive any kind of disability of very mild disability, 22.6% had a chronic condition from which they perceived moderate disability and 29.5% had a chronic condition from which they perceived severe disability. 38.1% did not cope with mental distress, 27.3 % had feelings of minimal depression, 11.5% suffered from mild depression, 11.2% suffered from moderate depression and 11.9% suffered from severe depression. 41.8% of the respondents had a lower network diversity; and 26% felt lonely often or all the time. Overall, 66.3% of the respondents reported high satisfaction with the care delivered by the GP; 45.7% had complete trust in the GP. Chapter 3: Research papers Table 1: distribution of the study sample Variable Sex Female Male Age n(%) 606 327 (54.0) 279 (46.0) 595 18-24 35-44 137 (23.0) 55+ 109 (18.5) 25-34 45-54 Education 49 (8.2) 98 (16.4) 202 (33.9) 554 Lower 106 (19.1) Higher 73 (13.2) Middle Perception of coping with income Rather difficult / very difficult Rather easy / very easy Self-rated health 375 (67.7) 591 417 (70.6) 174 (29.4) 606 Poor / very poor 123 (20.3) Good /very good 275(45.4) Fair Chronic condition with disability 208 (34.3) 606 No chronic condition 241 (39.8) Chronic condition with moderate disability 137 (22.6) Chronic condition without or with mild disability Chronic condition with severe disability Mental distress 49 (8.1) 179 (29.5) 598 None 228 (38.1) Mild 69 (11.5) Minimal Moderate Severe 163 (27.3) 67 (11.2) 71 (11.9) 105 Chapter 3: Research papers Satisfaction with care delivered by GP Low satisfaction (needs > actual situation) 118 (25.8) High satisfaction (needs < actual situation) 303 (66.3) Perfect match (needs = actual situation) Trust in the GP Quite trust / distrust / complete distrust Complete trust Postpone a visit to a GP during the past 12 months Yes Network diversity 36 (7.9) 579 302 (52.2) 277 (45.7) 474 152 (32.1) 594 Lower 248 (41.8) Higher 89 (15.0) Middle How often do you feel lonely 106 457 257 (43.3) 600 Seldom / never 233 (38.8) Often / all the time 160 (26.7) Occasionally Urbanization Rural Urban Having young children Yes Having a paid job Yes 207 (34.5) 606 386 (63.7) 220 (36.3) 606 98 (16.2) 585 141 (24.1) Postponing a visit to a GP during the past 12 months The results from the chi-square tests indicated that the variables sex, age, education, perception of coping with income, self-rated health, trust in the GP, having a paid job, having young children and mental distress had a p value <0.25 (results not shown). These variables were included in the multivariate logistic regression model. Table 2 shows the results of the multivariate model, including the variables sex, age, education, perception of coping with income, self-rated health, trust in the GP, having a paid job, having young children and mental distress The model explains Chapter 3: Research papers 16.5% of the total variance of postponing or cancelling a GP visit respondent thought they needed during the past 12 months. The results show that people with a severe depression are 2,209 times more likely to postpone a visit to a GP than people with no signs of depression or mental distress (p=0.036, CI: 1.052-4.641). People with a good to very good health also tend to postpone less frequently a visit to a GP compared to people with a poor to very poor health (odds: 0.482, p=0.040, CI: 0.240-0.967). People with a complete trust in their GP tend to postpone less frequently a visit to a GP than those without complete trust. These results have a trend to significance (odds: 0.649, p=0.064, CI: 0.410-1.026). No significant associations were found for sex, age, educational level, coping with available monthly income, having young children and having a paid job. Table 2: Regression model for postponing health care with a GP in the past 12 months* 95% CI p 0.911-2.305 0.118 Age 0.016 25-34 vs 18-24 0.478 1.613 0.603-4.315 0.341 35-44 vs 18-24 0.379 1.461 0.558-3.829 0.440 45-54 vs 18-24 -0.298 0.742 0.287-1.918 0.538 55+ vs 18-24 -0.820 0.440 0.152-1.281 0.132 Education 0.164 Middle versus low 0.125 1.133 0.700-1.835 0.610 High versus low -0.611 0.543 0.254-1.161 0.115 -0.370 0.691 0.400-1.194 0.185 Self-rated health 0.080 Sex Perception of coping with income vs B Odds ratio 0.371 1.449 not coping with income Fair versus Poor/very poor -0.213 0.808 0.435-1.500 0.499 Good/very good versus poor/very poor -0.729 0.482 0.240-0.967 0.040 Complete trust in the GP vs -0.433 0.649 0.410-1.026 0.064 -0.215 0.807 0.454-1.433 0.464 Having young children vs 0.202 1.224 0.640-2.343 0.541 no complete trust Having a paid job vs no paid job no young children 107 Chapter 3: Research papers Mental distress 0.108 Minimal vs none -0.290 0.748 0.411-1.364 0.344 Mild vs none 0.039 1.039 0.477-1.364 0.923 Moderate vs none 0.198 1.219 0.587-2.530 0.596 Severe vs none 0.793 2.209 1.052-4.641 0.036 Nagelkerke R²: 0.165 *All significant results are indicated in bold Discussion 108 Literature indicates that, on the one hand, people from the lower socioeconomic groups visit a GP more frequently, whereas people from the higher socioeconomic groups tend to use more specialist care (Van der Heyden et al., 2003, van der Meer et al., 1996). These higher utilization rates of PHC can partly be explained by a higher disease burden (Droomers and Westert, 2004). However, on the other hand, low-income patients are also at risk of postponing a visit to the GP more often than do patients from the higher socioeconomic groups (Whitehead and Hanratty, 2004, Vilhjalmsson, 2005). This study examines which subgroup of patients with a low income are particularly at risk for postponing or cancelling a visit to a GP they thought they needed. Knowing this might contribute to more fine-tuned actions and interventions that will further improve the accessibility of the primary healthcare system in Flanders. The participants of this study are 606 low-income people, aged between 18 and 65 years, visiting a Public Social Service in one of the eight selected areas in Flanders (Belgium). The respondents of the study are in worse health compared with the Flemish population aged between 18 and 65 years: 54.6% of the respondents report having fair to poor health compared with 17.6% of the Flemish population (Van Der Heyden et al., 2010). More than half of the respondents suffer from a chronic condition (60.6%) compared with 26.2% of the Flemish population (Van Der Heyden et al., 2010). Additionally, 61.2% have feelings of mental distress, of which 11.9% cope with a severe depression. Our study reached the low income people: 75% had no paid job and 70% had difficulties of coping with their available monthly income. Earlier studies describe that people living in poverty feel lonely more often and have a higher risk of social isolation than the general population (Böhnke, 2008). Results from this study confirm these findings. With regard to the Chapter 3: Research papers delivery of healthcare, participants are generally satisfied with the care delivered by their GP, and half of the participants have complete trust in their GP. Despite the high disease burden and the complex social situation of our respondents, 32% of the respondents postponed or cancelled a GP visit they thought they needed during the past 12 months. The multivariate analysis shows that not the feeling of not being able to cope with the daily costs of life with the available family income but feelings of depression and self-rated health are independent predictors of postponing or cancelling a GP visit respondents thought they needed. Furthermore having no complete trust in the GP has a trend to independently predict postponing or cancelling a visit to a GP. First, the finding that people who experience a severe depression are more likely to postpone a visit to a GP is quite alarming. Literature indicates that people with a lower social status are more vulnerable for depression than their better of counterparts (Van Der Heyden et al., 2010). Furthermore, studies show that people from lower social classes and young people are more reluctant and have a negative attitude towards seeking help for their mental illnesses ( Jagdeo et al., 2009). These findings suggest that people in the greatest need for services have the poorest attitudes. Therefore there is a need for an outreach to help people who suffer from a mental illness. Second, patients with poor self-rated health tend to postpone or cancel a visit to their GP more often than those with good self-rated health. Nevertheless, patients with a poor health have a higher likelihood to have multiple conditions and a higher disease burden leading to a higher health care need. But often, low income patients postpone the needed medical care because they do not know where to go, have other priorities (getting through the day),…(Willems, 2005). These are exactly the people who providers need to reach, before their ill health gets even worse. Effective interventions to improve health care access should not only focus on financial barriers to care e.g. by increasing the number of insured patients, introducing reduced fees or even free care for low-income patients, … It is as important to tackle non-financial barriers to health care e.g. by providing walk-in hours, the availability of public transportation, training healthcare workers in communication skills, … Also, adapting an outreaching approach could help to identify and help people who postpone care before they get too ill and hospitalization is needed. A third finding of this study is that people with complete trust in their GP are less likely to postpone a visit to their GP compared with those with no complete trust. 109 Chapter 3: Research papers This strengthens the importance of a good doctor-patient relationship. Several studies show that building an environment which creates trust is crucial in providing longitudinal care (Ahmed et al., 2001, Ridd et al., 2009) and that people who have a regular GP receive more continuous care, have better health outcomes (e.g. better compliance, pregnancy outcomes…), lower service utilization, and lower healthcare costs (Nutting et al., 2003, Shear et al., 1983, Wasson et al., 1984, Christakis et al., 2001). In a good doctor-patient relationship, compassion and shared respect are important factors, leading to a higher level of trust between a patient and a GP (Teutsch, 2003, Mainous et al., 2001). A Scottish study shows that people from deprived areas want holistic GPs who understand the realities of life and whom they can trust as being both competent and genuinely caring. A GP without these is judged as being socially distant and emotionally detached. Relational continuity, empathy, and taking sufficient time in consultations are key factors in building a good doctor-patient relationship (Mercer et al., 2007). 110 Having young children, a paid job, age, sex and perception of coping with income did not give significant results in the multivariate logistic model. But this does not mean that these measures aren’t relevant. Removing one of these variables from the model, reduced the explained variance with minimum two percent. Contrary to our expectations, this study shows that low-income patients who find it difficult to manage with their available income do not postpone a visit to their GP more often than do those low-income patients who are able to make ends meet despite their low family income. Possible explanations are the success of the interventions that try to lower the financial burden for low-income patients. In Belgium, we have a nearly universal healthcare system with direct access to a general practitioner or a specialist and no gate-keeping or patient list. There is a fee-for service payment according to which patients pay about 30% co-payment for primary healthcare and 40% for specialist care. For people with a low income (being a gross taxable family income lower than 13.543,71 Euro, increased with 2.507,30 Euro per additional family member), there is a reduced co-payment of 8%. The total amount of co-payment per year is limited by an income-related threshold (Björnberg et al., 2009). Furthermore, in the fee-for-service system, low-income patients can visit their GP for only one euro. Patients visiting Community Health Centres and group practices using the capitation system can visit the doctor for free when they are registered with Chapter 3: Research papers that practice. This kind of universal approach for low-income patients in Flanders reduces the financial barriers of the Health care system. An important strength of this study is the success of questioning 606 low-income patients about their health and healthcare use. When reviewing literature on healthcare use, this population group is often under-represented, making it very difficult to formulate strong conclusions. In this study, the design was developed based on actively inviting a large number of potential participants and minimizing the barriers for participation in the study for this specific study group as much as possible. With the aim of reaching the study population and inviting them to participate in the study, we collaborated with Public Social Services in the selected areas. Their massive time investment led to a large number of included participants (n = 1148). However, using this specific method of inclusion might have led to a selection bias at the moment of inclusion and to a selection bias because of differential refusal rates among subgroups. First, although the social workers were asked to invite all their clients who met the inclusion criteria, they might have made a pre-selection (e.g. not inviting clients during busy consultation hours, not inviting clients whom they had a difficult working relationship with). Second, the information on the number of clients who refused to participate and their profiles is limited. In the study protocol, the social workers were asked to complete a minimal information form for each client whom they had invited to participate in the study. This form included information on the sex and age of the person, whether they responded to the inclusion criteria, and whether they wanted to participate in the study or not. In case the person refused to participate, his/her motivation to refuse was asked. However, filling in this form was considered as being too time consuming, and it was not rigorously used. This also implies that there was no additional information available on the respondents who were included in the study but eventually did not participated. Several actions were taken to minimize the barriers for participation in the study for this specific study group. The interviewer made up to 10 attempts to contact the respondents by phone to make an appointment for the interview. In case this did not result in contact, for example, because the telephone number was no longer active, the interviewer contacted the social worker who had included the respondent to trace the respondent’s new contact details. In addition, at the moment of the interview, the respondents were sometimes not at home, no longer lived at the provided address, or perceived the timing as inconvenient. In this case, the interviewer made a new appointment. Respondents were also given the option of being 111 Chapter 3: Research papers interviewed at a place other than their home (e.g. in the premises of the Local Social Service Center). Finally, all study material was developed using a clear language and was pre-tested in the study population, and the procedures were developed to minimize the burden for the respondent. The interviewers were extensively trained to clearly communicate about the study, to answer the respondent’s questions, and were expected to be flexible in finding solutions for problems when contacting respondents or finding a time and location to interview them. This resulted in the participation of 606 respondents completing the entire survey (i.e. a response rate of 54%), which is, seen the population and the encountered problems, a relatively large sample. 112 When using this specific inclusion procedure, there are also some limitations. First of all, immigrants or undocumented people who do not speak the Dutch language cannot participate in the study, although they are often financially vulnerable and experience important problems in accessing healthcare. Second, all participants in this study are clients in a Public Social Service. Although these services reach the majority of low-income Belgians since they administratively arrange the minimum wages and remittances, we might have missed (the small group of ) people who do not find access to these services or who lead completely isolated lives. Third, social workers did not always manage to fill in the selection form or did a pre-selection without the form. Therefore we have no (complete) information on the number of people excluded in the study nor the number of people visiting the Public Social Services during the inclusion period of the study. Fourth, this study was conducted in eight Flemish regions. Although these regions were selected with care, the results may not apply to the entire Flemish region. Furthermore, prudence is called when interpreting the results. Nearly half of the participants stated to have complete trust in their GP. Due to the skewed distribution, we could not make a comparison between people who trust their GP and those who did not. So the cut-off point was taken between “complete trust” and all the other answer categories. Next, the model we used explained 16.5% of the total variance in postponing a visit to a GP while they actually needed one. This result stresses the complexity of the problem. Other possible factors explaining the variance could be: having a regular GP, the availability of public transportation, the opening hours of the practice,… Therefore a more detailed questionnaire on the organization of the GP practices is needed. Chapter 3: Research papers Finally we have no information on how long people postponed a visit to a GP (a few days or several months). Furthermore, in Belgium, people are not obligated to make an appointment with a GP. In that way we can not compare the reported postponement of a GP visit to registered postponement. This makes it very difficult for policymakers to identify the magnitude of the problem. Conclusion This study shows that especially people suffering form a severe depression, people with no complete trust in the GP and people with ill health, are more likely to postpone or cancel a GP visit they thought they needed compared to other lowincome patients. The inability to make ends meet does not seem to lead to further postponing of healthcare. This might indicate that the access to healthcare for low-income people might also be hindered by barriers that are not directly linked to the price of the consultation. Effective interventions should focus on outreaching activities, a family-friendly organization of care, physical accessibility of care, and building a trustful doctorpatient relationship. Further research concerning the accessibility and quality of healthcare for low-income people is needed to provide policymakers with the evidence needed to take action in order to make the healthcare system easily accessible for every person. Competing interests All authors declare that no competing interests exist. Acknowledgements EV, AP, AD, JDM, CVA, and SW contributed to the design of the study; EV, AP, and SW contributed to the analysis and the writing of the article; and AD, KH, JDM, CVA, and SW participated in the revision of the manuscript. All the authors have read and approved the final manuscript. 113 Chapter 3: Research papers References 114 - AHMED, S. M., LEMKAU, J. P., NEALEIGH, N. & MANN, B. 2001. Barriers to healthcare access in a non-elderly urban poor American population. Health Soc Care Community, 9, 445-53. - BJÖRNBERG, A., GARROFÉ, B. C. & LINDBLAD, S. 2009. Euro Health Consumer Index 2010. In: POWERHOUSE, H. C. (ed.). - BÖHNKE, P. 2008. Art the poor socially integrated? The lkink between poverty and social support in different welfare regimes. Journal of European Social Policy, 18, 133-150. - BURSTROM, B. 2002. Increasing inequalities in health care utilisation across income groups in Sweden during the 1990s? Health Policy, 62, 11729. - CAMPBELL, J. L. 1999. Patients’ perceptions of medical urgency: does deprivation matter? Fam Pract, 16, 28-32. - CHRISTAKIS, D. A., MELL, L., KOEPSELL, T. D., ZIMMERMAN, F. J. & CONNELL, F. A. 2001. Association of lower continuity of care with greater risk of emergency department use and hospitalization in children. Pediatrics, 107, 524-9. - COHEN, S., DOYLE, W. J., SKONER, D. P., RABIN, B. S. & GWALTNEY, J. M., JR. 1997. Social ties and susceptibility to the common cold. JAMA, 277, 1940-4. - DE BRUIN A, P. H., NOSSIKOV A 1996. Health interview surveys. Toward international harmonization of methods and instruments. Copenhagen:WHO-Europe, CBD Netherlands. - DRIESKENS, S., VAN DER HEYDEN, J., HESSE, E., GISLE, L., DEMAREST, S. & TAFFOREAU, J. 2010. Gezondheidsenquête België, 2008. Rapport V - Sociaal-economische ongelijkheden in gezondheid. - DROOMERS, M. & WESTERT, G. P. 2004. Do lower socioeconomic groups use more health services, because they suffer from more illnesses? Eur J Public Health, 14, 311-3. - EUROPE, T. 2008. European Practice Assessment-EPA. Easy to use and scientifically developed quality management for general practice. In: EUROPE, T. (ed.). - HART, J. T. 1971. The inverse care law. Lancet, 1, 405-12. Chapter 3: Research papers - HERMANS KOEN, D. M. D. A. V. J. M. L. D. M. J. V. A. C. 2008. Werknota “De slectie van de SWVG-onderzoeksregio’s” Programma 1-Kencijfers en monitoring. Steunpunt Welzijn Volksgezondheid en Gezin. - JAGDEO, A., COX, B. J., STEIN, M. B. & SAREEN, J. 2009. Negative attitudes toward help seeking for mental illness in 2 population-based surveys from the United States and Canada. Can J Psychiatry, 54, 757-66. - MAINOUS, A. G., 3RD, BAKER, R., LOVE, M. M., GRAY, D. P. & GILL, J. M. 2001. Continuity of care and trust in one’s physician: evidence from primary care in the United States and the United Kingdom. Fam Med, 33, 22-7. - MARMOT, M., ALLEN, J., GOLDBLATT, P., BOYCE, T., MCNEISH, D., GRADY, M. & GEDDES, I. 2010. Fair Society, Healthy lives. The Marmot review. Executive Sumary. - MERCER, S. W., CAWSTON, P. G. & BIKKER, A. P. 2007. Quality in general practice consultations; a qualitative study of the views of patients living in an area of high socio-economic deprivation in Scotland. BMC Fam Pract, 8, 22. - NUTTING, P. A., GOODWIN, M. A., FLOCKE, S. A., ZYZANSKI, S. J. & STANGE, K. C. 2003. Continuity of primary care: to whom does it matter and when? Ann Fam Med, 1, 149-55. - REGIDOR, E., MARTINEZ, D., CALLE, M. E., ASTASIO, P., ORTEGA, P. & DOMINGUEZ, V. 2008. Socioeconomic patterns in the use of public and private health services and equity in health care. BMC Health Serv Res, 8, 183. - RIDD, M., SHAW, A., LEWIS, G. & SALISBURY, C. 2009. The patientdoctor relationship: a synthesis of the qualitative literature on patients’ perspectives. Br J Gen Pract, 59, e116-33. - SHEAR, C. L., GIPE, B. T., MATTHEIS, J. K. & LEVY, M. R. 1983. Provider continuity and quality of medical care. A retrospective analysis of prenatal and perinatal outcome. Med Care, 21, 1204-10. - STARFIELD, B., SHI, L. & MACINKO, J. 2005. Contribution of primary care to health systems and health. Milbank Q, 83, 457-502. - TEUTSCH, C. 2003. Patient-doctor communication. Med Clin North Am, 87, 1115-45. - VAN DER HEYDEN, J., GISLE, L., DEMAREST, S., HESSE, E. & TAFFOREAU, J. 2010. Gezondheidsenquête België 2008. Rapport I: gezondheidstoestand. Brussel: Wetenschappelijk instituut Volksgezondheid Operationele Directie Volksgezondheid en surveillance. 115 Chapter 3: Research papers - VAN DER HEYDEN, J. H., DEMAREST, S., TAFFOREAU, J. & VAN OYEN, H. 2003. Socio-economic differences in the utilisation of health services in Belgium. Health Policy, 65, 153-65. - VAN DER MEER, J. B., VAN DEN BOS, J. & MACKENBACH, J. P. 1996. Socioeconomic differences in the utilization of health services in a Dutch population: the contribution of health status. Health Policy, 37, 1-18. - VEUGELERS, P. J. & YIP, A. M. 2003. Socioeconomic disparities in health care use: Does universal coverage reduce inequalities in health? J Epidemiol Community Health, 57, 424-8. - VILHJALMSSON, R. 2005. Failure to seek needed medical care: results from a national health survey of Icelanders. Soc Sci Med, 61, 1320-30. - WASSON, J. H., SAUVIGNE, A. E., MOGIELNICKI, R. P., FREY, W. G., SOX, C. H., GAUDETTE, C. & ROCKWELL, A. 1984. Continuity of outpatient medical care in elderly men. A randomized trial. JAMA, 252, 2413-7. - WHITEHEAD, M. & DAHLGREN, G. 2006. Levelling up (part 1): a discussion paper on concepts and principles for tackling social inequities in health. WHO collaborating Centre for Policy Research on Social Determinants of Health, University of Liverpool. - WHITEHEAD, M. & HANRATTY, B. 2004. Health care for rich and poor alike. In: HEALY, J. & MCKEE, M. (eds.) Accessing health care. Responding to diversity. Oxford: Oxford University Press. - WHO. Available: http://www.euro.who.int/en/what-we-do/health-topics/ Health-systems/primary-health-care/country-work/primary-care-evaluation-tool. - WHOOLEY, M. A., AVINS, A. L., MIRANDE, J. & BROWNER, W. S. 1997. Case-finding instruments for Depression. Two Questions Are as Good as Many. J Gen Intern Med., 12, 439-445. - WILLEMS, S. 2005. The socio-economic gradient in health: a never-ending story? A descriptive and explorative study in Belgium., Ghent University. - WILLEMS, S., DE MAESSCHALCK, S., DEVEUGELE, M., DERESE, A. & DE MAESENEER, J. 2005. Socio-economic status of the patient and doctor-patient communication: does it make a difference? Patient Educ Couns, 56, 139-46. 116 Chapter 3: Research papers Verlinde et al. International Journal for Equity in Health 2012, 11:12 http://www.equityhealthj.com/content/11/1/12 ANNOTATED BIBLIOGRAPHY Open Access The social gradient in doctor-patient communication Evelyn Verlinde1,2*, Nele De Laender1, Stéphanie De Maesschalck1, Myriam Deveugele1 and Sara Willems1 Abstract Objective: In recent years, the importance of social differences in the physician-patient relationship has frequently been the subject of research. A 2002 review synthesised the evidence on this topic. Considering the increasing importance of social inequalities in health care, an actualization of this review seemed appropriate. Methods: A systematic search of literature published between 1965 and 2011 on the social gradient in doctorpatient communication. In this review social class was determined by patient’s income, education or occupation. Results: Twenty original research papers and meta-analyses were included. Social differences in doctor-patient communication were described according to the following classification: verbal behaviour including instrumental and affective behaviour, non-verbal behaviour and patient-centred behaviour. Conclusion: This review indicates that the literature on the social gradient in doctor-patient communication that was published in the last decade, addresses new issues and themes. Firstly, most of the found studies emphasize the importance of the reciprocity of communication. Secondly, there seems to be a growing interest in patient’s perception of doctor-patient communication. Practice implications: By increasing the doctors’ awareness of the communicative differences and by empowering patients to express concerns and preferences, a more effective communication could be established. Keywords: Communication, Physician-patient relations, Social class Introduction In 1977 a commission under the lead of Sir Francis Black published the famous Black report, illustrating the existence of a social gradient in health in the UK. The publication of this report was the start of a new wave of research on social inequity [1]. Since then many studies have confirmed the gradient in health between social classes [2-4]. Health differences between social groups due to underlying social mechanisms such as differential access to care, social exclusion or poverty are a matter of major concern in today’s public health research but in spite of marked health improvements of the overall population and efforts to overcome health inequalities, higher morbidity and mortality rates for the socio-economically disadvantaged are still found [5-8]. The causes for these inequalities in health are multiple and * Correspondence: [email protected] 1 Department of Family Medicine and Primary Health Care, Ghent University, Ghent, Belgium Full list of author information is available at the end of the article complex: a different distribution of power and resources among social classes, different levels of exposure to health hazards, same level of exposure leading to differential impacts, life-course effects and different social and economic effects of being sick [2,9-16]. A prerequisite for equity in health is equity in health care, defined as equal care for people in equal need of care. As Dahlgren and Whitehead quoted: “Equity in health care includes fair arrangements that allow equal geographic, economic and cultural access to available services for all in equal need of care” [17]. An essential component of the delivery of health care is the relationship between the patient and the health care provider [18]. Several studies on communication in health care have repeatedly shown the importance of the doctor’s communication skills [19]. By communicating with a patient, a physician gets to know the patient’s problem and creates a therapeutic relationship necessary for its management and, if possible, its solution [20]. The quality of the relationship between a doctor and a © 2012 Verlinde et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. 117 Chapter 3: Research papers Verlinde et al. International Journal for Equity in Health 2012, 11:12 http://www.equityhealthj.com/content/11/1/12 patient is a key factor in the effectiveness of care. Good doctor-patient communication is associated with a higher level of patient satisfaction and better compliance [19,21]. Furthermore, optimizing doctor-patient communication can lead to better patient health and outcomes [22,23]. Available evidence suggests that low-income populations and people without health insurance report lower communication satisfaction and a reduced access to care [24]. In recent years the importance of social inequalities in the physician-patient relationship has frequently been the subject of research [19,21,24]. A previous review showed that doctor-patient communication indeed varies according to the social class of the patient [25]. Considering the increasing importance of social inequities in health care, an actualization of the review seemed appropriate. If differences in the physicians’ communicative behaviour vary according to the socio-economic status of the patient, this could be a new focus in the battle against socio-economic inequities in health. In this paper we want to answer the following questions based on a systematically review of the literature: 118 Does the doctor-patient communication varies according to the socio-economic status of the patient? If so, which aspects of the consultation are affected? Are the findings of studies published after 2002 different than those of the publication of the first (and the latest) review on this topic [25]. Methods Before starting the review, a protocol was developed, including the following steps. Search strategy In step one, a systematic search in MEDLINE, PsycINFO and Web Of Science was conducted to identify publications on doctor-patient communication and social class of the patient. The following search strings were used: MeSH: communication AND (physician-patient relations OR provider-patient relations OR physicianfamily relations) AND (social class OR socio-economic factors) Text-words: (doctor-patient communication OR physician-patient communication OR providerpatient communication) AND (social class OR socioeconomic status). The search was limited to publications from 1965 on. No specific search software was used. Page 2 of 14 Articles that were not original research articles, opinion articles and reviews were excluded. Furthermore, the search strategy was narrowed to studies performed within industrialized countries. Outcome measures To make the comparison of results possible, articles were included when they mentioned the interaction between the socioeconomic status (SES) of the patient or one of its indicators (educational level, income or occupation) as well as determinants of doctor-patient communication. At the initial stage, doctor-patient communication was not yet defined into specific categories, in order to obtain a wide range of studies. This means that studies mentioning any form of doctor-patient communication were selected. We included as well studies from primary care as specialist care to gain insight in the overall social gradient in doctor-patient communication. This resulted in a list of 129 articles. Study selection Figure 1 provides an overview of the study selection process. Of the 129 studies under review, 51 were excluded based on title and abstract review since they were after all not related to doctor-patient communication and social class. The abstracts of the remaining 78 publications were screened for explicit references to social class related concepts (education, income or occupation) and doctor-patient communication. Fourty-six articles determining SES by other variables than education, income or occupation (e.g. race, gender, health literacy) and articles focussing on disease-specific communication were excluded. In the last step of the selection process, an independent full text analysis of the remaining 32 publications was performed independently by two of the researchers to confirm the relationship between social class and doctor-patient communication in the publications. Publications labelled as “doubtful relevance concerning social class and doctor-patient communication” by one of the reviewers, were discussed until consensus was reached. Twelve publications were rejected in this phase. Eventually, 20 publications were labelled as relevant to asses doctor-patient communication and social class of the patient (Table 1). Analysis and synthesis of the study findings A narrative review was conducted. Meta-analysis was not attempted due to heterogeneity of populations and outcome measures among included studies. Findings were compared according to investigated communication measure and according to the social class. To cluster the communication variables, several communication assessment approaches were considered [26]. Communicative behaviour can be categorised in terms of verbal and non- Chapter 3: Research papers Verlinde et al. International Journal for Equity in Health 2012, 11:12 http://www.equityhealthj.com/content/11/1/12 Page 3 of 14 119 Figure 1 Selection procedure. verbal behaviour. Verbal behaviour can be defined as ‘the spoken communication’. The verbal elements of communication can be divided into instrumental or task-focused verbal behaviour (e.g. question asking, information giving, etc.) and affective or socio-emotional behaviour (counselling, positive and negative talk, etc.) reflecting the distinction between cure and care [27,28]. However, some of the determinants of communication do not fit into the above categories but are related to the concept patient-centeredness. Patient-centeredness is about seeing the patient as a person with a unique personal history and individual needs. We can identify five dimensions: (1) using the bio-psycho-social perspective, (2) approach the patient as a whole person, (3) Setting professional health care providers primary care multipurpose clinic, pediatric consultation primary care primary care Hall Street Street Martin Fiscella direct observation, chart audits, patiënt reports 2538 1972 115 audiotapes questionnaires 41 157 (mean) education occupation educational level education social class indices, education or income Nb of Variable patients SES audiovisual analysis meta-analysis Method time use, preventive tasks, satisfaction, attributes of primary care listening, explaining, advice giving, examination partnership building information giving (diagnostic, treatment, procedural) information giving, question asking, task and interpersonal competence, partnership building and socio-emotional behaviour Variable communication doctor Interpersonal communication, patient satifaction listening, explaining, advice giving, examination m parent’s question asking and opinion giving communicative style: affective expressiveness More question asking by patient leads to more information giving. No relation between educational level and question asking Physician information giving: 0.82 Partnership building: 0.87 Patient’s opinion giving:0.82 Patient affective expressiveness: 0.75 Patient’s question asking: 0.96 Time use: Davis Observation Code Attributes of Primary care: Components of Primary Care Instrument Patient satisfaction: items from the Medical Outcomes Survey Lower education more physical examination and nutritional counseling, less time on questions, assessing health knowledge, negotiation and counseling, chatting and screening tests. Patients perceive no difference Higher social class: more examination, listening and explaining. Physician’s perception of consultation: emphasis on active listening, supporting and giving advice. No information available Patient perception of consulation: emphasis on prescribing, reassuring and referring Reliability: Higher education: more expressive, - physician response higher level of opinionated and asking more questions. (0.72-0.95) - patient response (0.680.91) Personal characteristics less influence on physician response than own communication behavior? Higher educated patients: more diagnostic and health information. Higher social class: more overall communication and more information. Conclusion of the study Unitizing reliability for utterances: Cohen’s kappa = 0.84 Reliability categorizing: Correlation and standard norm deviate was extracted for each study if possible Variable communication Information on patient validity and reliability 120 First author (Ref. nb.) Table 1 Overview of the selected articles Chapter 3: Research papers Verlinde et al. International Journal for Equity in Health 2012, 11:12 http://www.equityhealthj.com/content/11/1/12 Page 4 of 14 Employees primary care primary care solo & multispecialty practices primary care primary care primary care Taira Pendleton Street Kaplan McKinstry Roter Stewart 8316 41 79 6549 audiotapes audiotape RIAS; questionnaires 140 537 structured 410 interview, video vignettes questionnaires videotaped consultations videotaped consultations questionnaires educational level income social class indices education education social class income Table 1 Overview of the selected articles (Continued) information giving; patiënt centredness narrowly biomedical, expanded biomedical, biopsychosocial, psychosocial, consumerist pattern shared decision making style PDM (Participatory DecisionMaking) style: involve them in treatment decisions, give them a sense of control over medical care and ask them to take some responsibility for care nonverbal behaviour consistency and adaptations amount of information given to the patiënt discussion of health risk idem health risk behaviours Same satisfaction as higher educated. Statement made by Higer education: more explanation on doctor: Bales Interaction drug prescription Process Analysis Communication on drugs: scheme developed by Svarstad and refined by Scherwitz and Evans Reliability physician: 0.76 Lower SES patients prefer narrowly Reliability patient: 0.81 biomedical pattern. No information available Lower educated patients: lower preference for shared decision making. Data from the Medical Lower educated patients: less mutual Outcomes Survey (MOS) decision making, less sense of control Reliability Participatory and given less responsibility. decision making style: 0.74 Cohen’s kappa: 0.82 for Physicians talking with higher educated speaking turns and patients used more body orientated talk response latencies then they did with lower educated. 0.71 for interruptive speakovers 0.92 for physicians’ taks touch 0.85 for illustrators 0.71 for adaptors 0.79 for procimity 0.93 for body orientation 0.90 for turn duration 0.81 for response latency 0.83 for pausing within speaking 0.75 for patient’s anxiety No information available High SES: more explanations Lower income: more smoking discussion No information available High income: more diet and exercise discussion. Reliability: doctor satisfaction = 0.90, nurse satisfaction 0.72 Chapter 3: Research papers Verlinde et al. International Journal for Equity in Health 2012, 11:12 http://www.equityhealthj.com/content/11/1/12 Page 5 of 14 121 primary care primary care audiotapes patients, lung cancer patients, patients with systemic lupus erythematosus oncology practices Jensen Street Siminoff audiotapes, RIAS survey, interview department of telephone Veterans interview Affairs (VA), universitybased and county health care system Piette & Schillinger [39] survey, PEPPI breast cancer treatment program 405 279 131 752 327 education, income education education, income education education, income patient-perceived selfefficacy educating and counseling the patient concerning biomedical and psychosocial issues-ask patients for information to indicate understanding, opinion or permissionattempting to built a relationship with the patientengagement in conversation about the patients emotional status-gathering relevant data and information partnership-building; encourage patient involvement; supportive talk patient communicates biomedical and psychosocial information; asking questions; building relationship with the physician, engaging in discussion; expression of feelings; asking questions; assertive responses, expressions of concern or other negative emotions explain things; listen carefully idem to what the patient has to say; show respect; spend enough time with the patient Interpersonal Processes of Care (IPC) questionnaire: general clarity; explanations; elicitation of patient’s preferences, emotional support interactive information-giving 122 Maly Table 1 Overview of the selected articles (Continued) Low education: more emotional support Doctor-patient communication: Roter Interaction Analysis System (RIAS) Coding system developed by Street and colleagues with reliability ranging from 0.61 to 0.97 depending on the behavior and the study Questions coming from the Medical Expenditures Panel Survey Revised scale of the Interpersonal Processes of Care (IPC) questionnaire with reliability of 0.91 High income patients: receive more biomedical talk, emotional talk, psychosocial counseling and education, ask more question and receive less questions about their disease than low income patients. Patients are more active communicators when physicians use partnershipbuilding. Higher educated patients are more active communicators, ask more questions and are more assertive, but they do not express more concerns Low income perceive some areas of tension in communication with their health care provider. Patients with high literacy skills are more critical on their physician. Low SES: better general and diabetesspecific communication than high SES Self-efficacy: validated Higher education: more interactive Perceived information giving by physician and Efficacy in Patientgreater perceived self-efficacy. Physician Interactions (PEPPI) questionnaire Language: Marin Acculturation Scale with reliability of 0.99 Chapter 3: Research papers Verlinde et al. International Journal for Equity in Health 2012, 11:12 http://www.equityhealthj.com/content/11/1/12 Page 6 of 14 primary care Bao Devoe [24] primary care american households Murray 5978 3209 secondary 16 700 analysis of data from Medical Expenditure Panel Survey (MEPS) (faceto-face interview) patient and physician surveys computerassisted telephone interview educational attainment, family income Income, education education, income Table 1 Overview of the selected articles (Continued) listen carefully; explain things; show respect; spend enough time with the patient self-assessment of communication; performance of communication behaviours when discussing cancer screening giving information; decisionmaking style preferred style of decision-making; experienced style of decision-making; Questionnaires coming from the Medical Expenditure Panel Survey (MEPS) Questionnaire from the Communication in Medical Care (CMC) Research Program series. Poor patients: receiving less explanations in a way they understand. Within-physicians differences by education. Between-physicians differences by income Low SES patients are more likely to discuss cancer screening then high SES patients. High SES patients are more likely to experience the preferred style Lowe SES patients prefer consumerism and paternalism No information available High SES patients prefer shared decision making Chapter 3: Research papers Verlinde et al. International Journal for Equity in Health 2012, 11:12 http://www.equityhealthj.com/content/11/1/12 Page 7 of 14 123 Chapter 3: Research papers Verlinde et al. International Journal for Equity in Health 2012, 11:12 http://www.equityhealthj.com/content/11/1/12 sharing power and responsibility, (4)building a therapeutic relationship and (5) considering the physician as a person and acknowledging the influence of its personal qualities [29,30]. The communication variables in the selected articles are classified to the following categories: verbal behaviour including instrumental and affective behaviour; non verbal behaviour and patient-centred behaviour. The selected studies were grouped and analysed by the researches according the three communication variables. Results Verbal/non-verbal behaviour Verbal behaviour: instrumental behaviour 124 Instrumental behaviour is considered as all interactions that serve the “cure” part of the consultation. It can be defined as technically based skills that are used in problem solving e.g. giving directions, giving information, asking clarification, asking questions, counselling, etc. [26,31]. Eleven studies explored the interaction between the instrumental behaviour of the physician and/or the patients, and the SES of the patient. A meta-analysis conducted by Hall et al. explored the correlation between physicians’ communicative behaviour and the patient’s outcome variables. Social class was measured by income, education or other non-specified social class indices. The study revealed a positive relationship between patient’s social class and information giving. Patients of a higher social class received not only more overall communication but also more information [32]. Not only patients’ social class but also his/ her communication style influences the doctor-patient communication. In a study by Street et al., social class was measured by educational level. Physicians’ information giving was positively influenced by the patient’s communicative style such as question-asking, affective expressiveness and opinion-giving. More affective expressiveness and being assertive on the patient’s sidewhich is strongly related to his/her educational levelleads to more information giving on the doctor’s side. More educated patients receive more diagnostic and health information than their lower educated counterparts. However this study did not find a relation between the frequency of the patients’ question asking and his/her educational background [33]. The fact that adaptations in the physicians’ responses may, besides a function of patients’ personal or social characteristics per se, also are the result of the patients’ communicative actions, was confirmed by a second study by Street et al. In this study they compared the degree to which parents’ personal and interactive characteristics accounted for variation in doctor-parent interactions during paediatric consultations. Social class was measured as educational level. More educated Page 8 of 14 parents are not only more expressive and assertive but they also ask more questions. All three of these communication aspects lead to more information and direction giving by the physician. Additionally, this study shows that the parent’s personal characteristics have less influence on the physicians’ responses than their own communication behaviour [34]. Besides patient’s communication style, doctors’ and patients’ perceptions are an important aspect of the consultation and for the outcome of the consultation. In an observational study, Martin et al. looked at how both physicians and patients perceive what happens during the consultation. Social class was measured by occupation. From the patients point of view most emphasis of the consultation is put on prescribing, reassuring and referring. Whereas physicians report that emphasis is put on active listening, supporting and giving advice. Furthermore, physicians perceived they explained and listened more to patients from higher social classes and also examined them more than patients from lower social classes, but gave the latter more “other help” which was not specified. They also said to examine more and to give less advice to patients from lower social classes. However, patients did not report having experienced any of these differences [35]. The study of DeVoe et al. where social class was determined by family income and educational level, shows different results. This study suggests that patients’ perceptions of communication in healthcare settings vary widely by demographics and other individual patient characteristics. The poorest patients were less likely to report that providers always explained things so that they understood. Surprisingly, different levels of education were not independently associated with any of the investigated communication measures [36]. These results are in line with the results from the study of Fiscella et al., exploring whether educational level affected the visits of family physicians. Patients with a low educational level had a slightly larger proportion of the consultation time spent on physical examination and nutritional counselling. Less time was spent on patients’ questions, assessing their health knowledge, negotiating and counselling, chatting, and less screening tests were provided to them. One could say that less educated people are approached in a more directive way during the consultation. Less educated patients also saw their expectations less met during the consultation, although they were as satisfied as the more educated patients [37]. When looking at the outcome of the consultation, Maly et al. studied the impact of physician-patient communication on women’s receipt of, or planning for, breast reconstructive surgery. Social class was determined by educational level and was included as a potentially confounding factor. Women who had graduated from high school were Chapter 3: Research papers Verlinde et al. International Journal for Equity in Health 2012, 11:12 http://www.equityhealthj.com/content/11/1/12 more likely to report planning of breast reconstructive surgery. This is positively associated with interactive information-giving by the physician and greater patient perceived self-efficacy. These two communication factors weakened the negative influence of education barriers. Empowering aspects of patient-physician communication and self-efficacy may overcome the negative effects of a lower education on receipt or planned breast reconstructive surgery [38]. In line with the different consultation style, Taira et al. investigated whether the patients’ income level had an influence on the physicians’ discussion of health risk behaviours. Concerning patients at risk, physicians tended to discuss diet and exercise more with high income patients and smoking more with low income patients [39]. Discussing health risk behaviour is very important in consultation, especially for chronic conditions like diabetes. In a cross-sectional survey by Piette et al., general communication processes and diabetesspecific communication was examined. Socio-economic status was measured by means of educational achievement. Patients with lower education levels reported better general and better diabetes-specific communication than their less-vulnerable counterparts. This could be due to the fact that these patients have lower expectations of their patient-provider relationship or greater discomfort with criticizing them. Another introduced explanation could be that health care providers spend more time counseling patients which they perceive as in need for extra attention or explanation [40]. Pendleton et al. considered four types of information giving. SES was measured by social class. There was a significant difference in voluntary explanations given to patients from different social classes, independent of the different types of problems; higher SES patients receive significantly more explanations even when the explanation was not explicitly requested by the patient [41]. Also the study from Siminoff et al. where social class was measured by income and educational level showed that more biomedical talk was provided to higher income patients compared to medium and low income patients and to patients with higher educational achievement. In general, physicians provided little psychosocial counseling and education, however, they provided more to their high and medium income patients as compared to low-income patients. Patients that had more than a high school education and patients that reported a medium or high income asked more questions and showed more proactive behavior such as volunteering information to the physician unasked. Physicians on their side asked less educated patients and low income patients more questions about their disease and medical history [42]. Page 9 of 14 Verbal behaviour: affective behaviour The affective behaviour in doctor-patient communication is part of the emotional domain [19] and consists of all forms of social behaviour and social talk. Possible affective expressions are: showing concern, reassurance, reflection, signs of agreement or disagreement and paraphrasing [27,28]. Only three studies investigated the effects and outcomes of affective behaviour. The meta-analysis by Hall et al. (supra) explores the socio-emotional behaviours such as social talk and positive and negative talk. Although a link between the aspects of affective behaviour and the patients’ satisfaction and compliance can be identified, none of these determinants were found to be related to any determinant of the patients’ social class [32]. On the other hand, the studies of Street et al. (supra) concluded that doctors provided more comments of reassurance, support and empathy to the parents of children with cancer which were more affectively expressive (more specifically who expressed more negative affect). As patients with a higher educational level are more affectively expressive than their counterparts, it can be assumed that physicians show more affective behaviour towards these patients [33,34]. In the observational study from Siminoff et al. (supra), the emotional expressions by physicians varied by patient’s demographic variables, with more emotional utterances from their physician [42]. Non-verbal behaviour Non-verbal behaviour is one of the least investigated topics of doctor-patient communication, especially when looking at its interaction with determinants of social class. The effect of non-verbal behaviour is only mentioned in two of the selected articles [41,43]. Non-verbal behaviour can be operationalised in different ways such as eye contact, tone of voice, laughter, facial expression, physical distance, nodding, etc. [26]. The meta-analysis by Hall et al. (supra) could not find any research that was done on the association between the physicians’ non-verbal behaviour and the patients’ social class [32]. The same year of the Hall review, Street and Buller examined the non-verbal behaviour in doctor-patient interactions and the relationship with patient’s age, sex and social class measured as educational level. No differences were found in the level of non-verbal communication towards patients with different educational level. However, when talking to higher educated patients the physicians reciprocated their body orientations more than they did with lower educated patients. Finally, this article refers to specific difficulties in coding non-verbal behaviour, which is much more complex than categorising the verbal interactions [44]. 125 Chapter 3: Research papers Verlinde et al. International Journal for Equity in Health 2012, 11:12 http://www.equityhealthj.com/content/11/1/12 Patient-centeredness 126 Patient-centeredness can be classified into several aspects such as supportive talk, being attentive to patients’ psychosocial as well as physical needs, enabling the disclosure of patients’ concerns, conveying a sense of partnership and actively facilitating patient involvement in the decision-making [45]. In 10 of the 20 selected articles, patient centeredness in relationship with the Social class of the patient is described. First of all there is the relationship between patients’ social class and the decision making style of the doctor, described in three studies. In a study by Kaplan et al. social class was measured as educational achievement. Patients with a high school education or less were less involved in treatment decisions, less given a sense of control over treatment decisions and less asked to take responsibility for care than patients with post-graduate college education [46]. Also McKinstry observed the patients’ preference for shared decision making. Social class was determined by education. Patients’ preference for shared decision style or directive approach was associated with their social class, age, the scenario and their perception of the consultation style of their own physician being shared or directive. A lower social class predicted a lower preference for shared decision making style [47]. Murray et al. attempted to determine the congruence between patients’ preferred style of clinical decision-making and the style they usually experienced. Social class includes household income and educational achievement. People of high SES were more likely to prefer shared decision-making, and people of low SES were more likely to prefer consumerism and paternalism. Wealthier patients also were more likely to experience their preferred style of shared decision making. The results also point out that SES was strongly associated with reporting having enough information. Respondents who had not completed high school were less likely than those with an advanced degree to report having enough information to make the right decision [48]. Roter et al. Studied the preferred communication style of the patient. They described five communication patterns and their relationship with several patient characteristics, among social class measured as income. Patients approached in the narrowly biomedical pattern were more likely to be poorer than patients approached in other patterns [49]. Jensen et al. surveyed whether literacy, numeracy and optimism are related to satisfaction with health care providers’ communication skills. Participants’ social class was measured as educational level and admitted as a predictor variable. Almost half of the low-income patients were displeased with the amount of time health providers spent with them during interactions. As displayed in earlier studies, communication Page 10 of 14 dissatisfaction appears to be more common in lowincome adults than in higher income adults [50]. Certain aspects of communication can vary widely among different doctors or among patients. Bao et al. aimed to determine the extent to which socio-economic differences (income and education) in cancer screening discussion between a patient and his or her primary care physician are due to inter-physician versus intraphysician variation. Patients with low SES were less likely than their high-SES counterparts to have discussed cancer screening with their physicians. Differences by income are mainly ‘between-physicians’ While the ‘within-physician’ differences by income were minimal. The education gradient in cancer screening discussion mainly existed in ‘within physicians’. Except for mammogram the rate of discussion more than doubled among college graduates compared with those with a less than high school education. This may indicate that education plays an important role in determining what happens during clinical encounters [51]. Not only the communication style of the physician, but also patient participation in an essential topic in patient-centeredness. Street et al. examined the extent to which patient participation in medical interactions is influenced by the patient’s personal characteristics (among social class measured by education), the physician’s communication style and the clinical setting. Patients with at least some college education tended to be more active communicators than were less educated patients. Although the more educated persons asked more questions and are more assertive than less educated patients, they do not more often express concerns. The degree to which patients actively participate in medical encounters is a function of multiple patient, physician and contextual factors. It seems that patients are more active participants when interacting with physicians who more frequently engaged in partnershipbuilding and supportive talk [52]. A final aspect described in three studies is building a relationship between patient and physician. The study of Siminoff (supra) et al. shows educational level as significant independent factor on relationship building. It seems that both patients and their physicians spent more time trying to establish an interpersonal relationship with each other. Nevertheless, patients did more effort in relationship building than did their physicians. These results confirm previous evidence that providers communicate differently with patients by education and income [53]. In the study of Maly et al. (supra) on patients with breast cancer both the physician information-giving and patient empowerment in interacting with physicians were found to be significant determinants of breast reconstructive surgery, controlling for possible confounders. These two communication factors Chapter 3: Research papers Verlinde et al. International Journal for Equity in Health 2012, 11:12 http://www.equityhealthj.com/content/11/1/12 diminished the negative influence of education barriers and acculturation [38]. The study of Stewart, where educational level was a measure for social class, showed that physicians were more likely to appeal to the intellect of a patient with a university degree by justifying the drug prescription while on the other hand they offer more emotional support and solidarity to patients with a lower educational level [54]. As presented above, the amount of information given to patients is related to patients’ characteristics and to the patients’ communicative style. Hereby, the patients’ communicative style is not only influenced by his/her educational level but also by the level of partnership building of the physician [33,34]. In the observational study mentioned above by Street et al. it appeared that higher educated patients received more partnership building utterances [33]. Discussion and conclusion The aim of the current review is to give the state of the art on the social gradient in doctor-patient communication, to describe which aspects of the consultation are affected by this social gradient, and whether an evolution over time can be noticed comparing the results of older studies with those of newer studies. In this review we found that patients from lower social classes (measured by income, education or occupation) receive less socioemotional talk, a more directive and a less participatory consulting style characterised by for example less involvement in treatment decisions; a higher percentage of biomedical talk and physicians’ question asking; lower patient control over communication; less diagnostic and treatment information and more physical examination. Doctors give more information, more explanations, more (emotional) support and adapt more often a shared decision making style with higher SES participants. This review also indicates that the literature on the social gradient in doctor-patient communication that was published after 2002, at least addresses new issues and themes. Firstly, in the period 1965-2002, 42 articles were selected for this review, while for the period 20022011, 87 articles were selected. These numbers indicate that doctor-patient communication becomes a more emerging topic in the research on delivering qualitative care. Secondly, most of the more recent studies emphasize the importance of the reciprocity of communication: the doctor might communicate differently according to the social status of the patient, and patients may adapt a different communication style according their social class. Patients with a high SES tend to ask more questions, ask for explanations, are more expressive and have a higher level of being opinionated than their lower SES counterparts [33,34,42,55]. Furthermore, there seems to be a growing interest in patient’s perception of doctor-patient communication. Page 11 of 14 While in the past, patient’s perception was not taking into account or no differences in perception were found, more recent studies show that low SES patients have the feeling doctors fail to explain things in a way they can understand and spend less time with them [21,24,41]. These findings emphasise that doctor-patient communication is a complex interactional system. To depict this complexity, Street et al. (2007) applied an ecological model that takes into account the interplay of multiple physician, patient and contextual factors that collectively influence doctor-patient interactions [56]. The influence of any variable (e.g. ethnicity) may vary depending on the presence of other factors (e.g., the patients’ level of education, income, doctors’ communication style) [53]. The ecological approach recognizes that within the context of any medical encounter, a number of processes affect the way physicians and patients communicate and perceive one another. There are four important sources of potential influence: the physician’s communication style, patients’ characteristics, physician-patient demographic concordance and the patients’ communication. First, how a physician communicates with a patient may depend on his or her style. Some physicians provide more information, ask more questions, are more supportive and use more partnership-building than other physicians [33,34,49]. Second, variability in physicians’ communication and perceptions may be related to the patients’ demographic characteristics (education, income, occupation) [57]. Finally, the patients’ communication style can have a strong effect on physician behaviour and beliefs [18]. Important in this model is that patient interaction not only depends on the physician’s behaviour but also on patients’ characteristics and preferences. Patients from lower social classes more often suffer from (multiple) chronic conditions and more severe acute conditions [58]. But also they often have lower levels of health literacy-the degree to which persons have the capacity to obtain, process and understand basic health information and services needed to make appropriate health decisions [59,60]. Furthermore lower social class is associated with a lower sense of personal control also known as external locus of control. This means that the person perceives that certain events such as health and sickness are beyond his/her control [61]. This might explain why low SES people show lower levels of participation. Also, because they are less used to or feel less capable to interact during consultation, they might prefer a more directive consultation style. Recently, an international consortium of research teams in the UK, the Netherlands, Italy and Belgium set up the Gulliver study which focuses on the patient’s preferences in doctor-patient communication. Analysing a possible social 127 Chapter 3: Research papers Verlinde et al. International Journal for Equity in Health 2012, 11:12 http://www.equityhealthj.com/content/11/1/12 gradient in these preferences is hereby one of the points of attention of the researchers. Limitations of the study 128 Many of the limitations the review of 2005 encountered are still applicable today. Affective and non-verbal behaviour are important aspects in physician-patient communication e.g. through their influence on patient satisfaction [44]. Still a limited number of studies described the interaction between social class and non-verbal physicianpatient communication. All studies indicates the difficulties measuring and coding non-verbal behaviour. Therefore, these limited number of studies entails important methodological difficulties and does not allow us to draw conclusions concerning non-verbal and affective behaviour. Further research on this topic is still needed. Secondly, it is very difficult to compare the results of the studies due to the great diversity of measurements and frameworks organising these measurements in the different studies. Socio-economic status of the patient was measured by means of educational level, income or occupation [3,62]. An alternative to determine SES is to use “proxy” measures e.g. the insurance status, house tenure, car ownership, socio-demographic measures (race, etc.). Articles using proxy-variables as the only measure for SES were excluded. However, some of the selected articles used these variables in combination with educational level, income or occupational class. Next to the SES of the patient, also communication variables can be classified in many different ways. The variables used in these classifications are not always comparable, making if very difficult to compare the studies using different classification systems. We chose to categorise most of the communication variables according to the axis verbal/ non-verbal behaviour. The determinants of communication that did not fit into the categories of this axis were related to patient centeredness. In order to improve the comparability of future research, the use of a uniform definition and classification of communication variables is indispensable. Practice implications This review of the literature has revealed the complex relationship of doctor-patient communication and reinforces the practice implications of the former review. Physicians behave differently with patients from different SES and patients communicate differently with their doctor depending on their SES. The finding that the physician’s communicative behaviour is related to the communicative style of the patient and to his/her personal or social characteristics, may have important implications for the daily practice of the physician. Physicians need to be aware of the differences in giving information to and involving patients from lower Page 12 of 14 social classes in the consultation, as well as of the underlying causes [63]. It is important that physicians pay attention to the attitudes that they have toward patients, and have to remain aware of how their feelings might impact their behaviour and thus be perceived by patients [64]. They should consider the possibility that conscious or unconscious stereotyping may influence their behaviours, including their interpersonal style [65]. Physicians have to encourage patients to discuss their concerns and to ask questions, and they should listen actively. Communication skills and attitudes training can be an important tool to improve these defaults: the effects of such training have been proven and can persist over time [66]. Patients have a certain power to control communication during the consultation and to influence the physicians’ communicative behaviour. However, patients from lower social classes seem to exercise this control less than patients from higher educated groups. It seems that not only patient’s personal characteristics but more importantly their communicative behaviour has an influence on the doctor-patient communication. Therefore it is important to empower the patients towards more self-efficacy and towards learning how to express their concerns and preferences [33,34,38,67]. It has been shown that interventions to increase the participation of patients with low education obtain a good response and lead to measurable and clinically important improvements in health outcomes [68]. By understanding processes that facilitate or hinder patient involvement, physicians should be better able to adapt their own communication and office practices to help patients more effectively participate in medical encounters [52]. On the other hand, it is important not to take patientcentered care as the obvious and only choice. When working patient-centered, physicians should not only focus on normative thinking regarding participatory decision making but they also have to pay greater attention to a broader set of considerations relating to respect for patients as individuals. It is important to enable and empower patients but it is perhaps even more important to enable and empower them to the degree that they desire. As McKinstry states: “Doctors need both communication skills and time in consultations, along with knowledge of their patients, to determine at which times, with which illnesses, and at which level their patients wish to be involved in decision making” [47]. Future research should further investigate low-SES patients’ perceptions and expectations of their health care providers’ communication skills and of being involved in the decision making process. Special attention should be paid to the relationship between patient skills, patient activism and communication satisfaction [50]. Chapter 3: Research papers Verlinde et al. International Journal for Equity in Health 2012, 11:12 http://www.equityhealthj.com/content/11/1/12 Authors’ contributions and acknowledgements VE, NdeL, WS and MdeS equally contributed to the research and the report of the study, DM helped formulating the core idea and was involved in the final editing of the manuscript. All authors read and approved the final manuscript Author details 1 Department of Family Medicine and Primary Health Care, Ghent University, Ghent, Belgium. 2Verlinde Evelyn, Department of Family Medicine and Primary Health Care, Ghent University, UZ-1 K3, De Pintelaan 185, B-9000 Ghent, Belgium. Competing interests The authors declare that they have no competing interests. Received: 17 November 2011 Accepted: 12 March 2012 Published: 12 March 2012 References 1. Inequalities in health: Report of a research Working Group London: DHSS; 1980. 2. Field KS, Briggs DJ: Socio-economic and locational determinants of accessibility and utilization of primary health-care. Health Soc Care Community 2001, 9:294-308. 3. Mackenbach JP, Kunst AE: Measuring the magnitude of socio-economic inequalities in health: an overview of available measures illustrated with two examples from Europe. Soc Sci Med 1997, 44:757-771. 4. Mackenbach JP, Stirbu I, Roskam AJ, Schaap MM, Menvielle G, Leinsalu M, Kunst AE: Socioeconomic inequalities in health in 22 European countries. N Engl J Med 2008, 358:2468-2481. 5. Health for all in the 21st century. Proceedings of the WHO 50th anniversary congress. World Health Organisation; 1998. 6. Lynch J, Smith GD, Hillemeier M, Shaw M, Raghunathan T, Kaplan G: Income inequality, the psychosocial environment, and health: comparisons of wealthy nations. Lancet 2001, 358:194-200. 7. Turrell G, Mathers C: Socioeconomic inequalities in all-cause and specificcause mortality in Australia: 1985-1987 and 1995-1997. Int J Epidemiol 2001, 30:231-239. 8. Mackenbach JP, van de Mheen H, Stronks K: A prospective cohort study investigating the explanation of socio-economic inequalities in health in The Netherlands. Soc Sci Med 1994, 38:299-308. 9. Dahlgren G, Whitehead M: European strategies for tackling social inequities in health: Levelling up Part 2. 2006. 10. Power C, Matthews S: Origins of health inequalities in a national population sample. Lancet 1997, 350:1584-1589. 11. Benzeval M, J K, Whitehead M: Tackling inequalities in health. 1995. 12. van de Mheen H, Stronks K, Looman CW, Mackenbach JP: Does childhood socioeconomic status influence adult health through behavioural factors? Int J Epidemiol 1998, 27:431-437. 13. Lundberg O: The impact of childhood living conditions on illness and mortality in adulthood. Soc Sci Med 1993, 36:1047-1052. 14. Sturm R, Gresenz CR: Relations of income inequality and family income to chronic medical conditions and mental health disorders: national survey. BMJ 2002, 324:20-23. 15. Alter DA, Naylor CD, Austin P, Tu JV: Effects of socioeconomic status on access to invasive cardiac procedures and on mortality after acute myocardial infarction. N Engl J Med 1999, 341:1359-1367. 16. Wyke S, Hunt K, Walker J, Wilson P: Frequent attendance, socioeconomic status and burden of ill health. An investigation in the west of Scotland. Eur J Gen Pract 2003, 9:48-55. 17. Whitehead M, Dahlgren G: Concepts and principles for tackling social inequities in health: Levelling up Part 1. 2006. 18. Street RL Jr: Millay B: Analyzing patient participation in medical encounters. Health Commun 2001, 13:61-73. 19. Bensing J: Doctor-patient communication and the quality of care. Soc Sci Med 1991, 32:1301-1310. Page 13 of 14 20. Bensing JM, Kerssens JJ, Vanderpasch M: PATIENT-DIRECTED GAZE AS A TOOL FOR DISCOVERING AND HANDLING PSYCHOSOCIAL PROBLEMS IN GENERAL-PRACTICE. Journal of Nonverbal Behavior 1995, 19:223-242. 21. Jensen JD, King AJ, Guntzviller LM, Davis LA: Patient-provider communication and low-income adults: age, race, literacy, and optimism predict communication satisfaction. Patient Educ Couns 2010, 79:30-35. 22. Cheraghi-Sohi S, Bower P, Mead N, McDonald R, Whalley D, Roland M: What are the key attributes of primary care for patients? Building a conceptual ‘map’ of patient preferences. Health Expect 2006, 9:275-284. 23. Duberstein P, Meldrum S, Fiscella K, Shields CG, Epstein RM: Influences on patients’ ratings of physicians: Physicians demographics and personality. Patient Educ Couns 2007, 65:270-274. 24. DeVoe JE, Wallace LS, Fryer GE Jr: Measuring patients’ perceptions of communication with healthcare providers: do differences in demographic and socioeconomic characteristics matter? Health Expect 2009, 12:70-80. 25. Willems S, De Maesschalck S, Deveugele M, Derese A, De Maeseneer J: Socio-economic status of the patient and doctor-patient communication: does it make a difference? Patient Educ Couns 2005, 56:139-146. 26. Ong LM, de Haes JC, Hoos AM, Lammes FB: Doctor-patient communication: a review of the literature. Soc Sci Med 1995, 40:903-918. 27. van den Brink-Muinen A, Verhaak PFM, Bensing JM, Bahrs O, Deveugele M, Gask L, Mead N, Leiva-Fernandez F, Perez A, Messerli V, et al: Communication in general practice: differences between European countries. Family Practice 2003, 20:478-485. 28. Schouten BC, Meeuwesen L, Tromp F, Harmsen HAM: Cultural diversity in patient participation: The influence of patients’ characteristics and doctors’ communicative behaviour. Patient Education and Counseling 2007, 67:214-223. 29. Zandbelt LC, Smets EMA, Oort FJ, Godfried MH, de Haes H: Patient participation in the medical specialist encounter: Does physicians’ patient-centred communication matter? Patient Education and Counseling 2007, 65:396-406. 30. Mead N, Bower P: Patient-centredness: a conceptual framework and review of the empirical literature. Soc Sci Med 2000, 51:1087-1110. 31. Roter DL, Hall JA, Katz NR: Relations between physicians’ behaviors and analogue patients’ satisfaction, recall, and impressions. Med Care 1987, 25:437-451. 32. Hall JA, Roter DL, Katz NR: Meta-analysis of correlates of provider behavior in medical encounters. Med Care 1988, 26:657-675. 33. Street RL Jr: Information-giving in medical consultations: the influence of patients’ communicative styles and personal characteristics. Soc Sci Med 1991, 32:541-548. 34. Street RL: Communicative styles and adaptations in physician-parent consultations. Soc Sci Med 1992, 34:1155-1163. 35. Martin E, Russell D, Goodwin S, Chapman R, North M, Sheridan P: Why patients consult and what happens when they do. BMJ 1991, 303:289-292. 36. DeVoe JE, Wallace LS, Fryer GE: Measuring patients’ perceptions of communication with healthcare providers: Do differences in demographic and socioeconomic characteristics matter? Health Expectations 2009, 12:70-80. 37. Fiscella K, Goodwin MA, Stange KC: Does patient educational level affect office visits to family physicians? J Natl Med Assoc 2002, 94:157-165. 38. Maly RC, Liu YH, Kwong E, Thind A, Diamant AL: Breast Reconstructive Surgery in Medically Underserved Women With Breast Cancer The Role of Patient-Physician Communication. Cancer 2009, 115:4819-4827. 39. Buetow SA: What do general practitioners and their patients want from general practice and are they receiving it? A framework. Soc Sci Med 1995, 40:213-221. 40. Piette JD, Schillinger D, Potter MB, Heisler M: Dimensions of patientprovider communication and diabetes self-care in an ethnically diverse population. J Gen Intern Med 2003, 18:624-633. 41. Pendleton DA, Bochner S: The communication of medical information in general practice consultations as a function of patients’ social class. Soc Sci Med Med Psychol Med Sociol 1980, 14A:669-673. 42. Siminoff LA, Graham GC, Gordon NH: Cancer communication patterns and the influence of patient characteristics: disparities in information-giving and affective behaviors. Patient Educ Couns 2006, 62:355-360. 129 Chapter 3: Research papers Verlinde et al. International Journal for Equity in Health 2012, 11:12 http://www.equityhealthj.com/content/11/1/12 130 43. Stewart MRD: Which facets of communication have strong effects on outcome-a meta-analysis. Communicating with medical patients. 1989. 44. Street RL, Buller DB: PATIENTS CHARACTERISTICS AFFECTING PHYSICIANPATIENT NONVERBAL-COMMUNICATION. Human Communication Research 1988, 15:60-90. 45. Mead N, Bower P, Hann M: The impact of general practitioners’ patientcentredness on patients’ post-consultation satisfaction and enablement. Soc Sci Med 2002, 55:283-299. 46. Kaplan SGB, Greenfield S, Rogers W, Ware J: Patient and visit characteristics related to physician’s participatory decision making style. Med Care 1995, 1176-1187. 47. McKinstry B: Do patients wish to be involved in decision making in the consultation? A cross sectional survey with video vignettes. BMJ 2000, 321:867-871. 48. Murray E, Pollack L, White M, Lo B: Clinical decision-making: Patients’ preferences and experiences. Patient Education and Counseling 2007, 65:189-196. 49. Roter DL, Stewart M, Putnam SM, Lipkin M Jr: Stiles W, Inui TS: Communication patterns of primary care physicians. JAMA 1997, 277:350-356. 50. Jensen JD, King AJ, Guntzviller LM, Davis LA: Patient-provider communication and low-income adults: Age, race, literacy, and optimism predict communication satisfaction. Patient Education and Counseling 2010, 79:30-35. 51. Bao YH, Fox SA, Escarce JJ: Socioeconomic and racial/ethnic differences in the discussion of cancer screening: “between-” versus “within-” physician differences. Health Services Research 2007, 42:950-970. 52. Street RL, Gordon HS, Ward MM, Krupat E, Kravitz RL: Patient participation in medical consultations-Why some patients are more involved than others. Medical Care 2005, 43:960-969. 53. Siminoff LA, Graham GC, Gordon NH: Cancer communication patterns and the influence of patient characteristics: Disparities in information-giving and affective behaviors. Patient Education and Counseling 2006, 62:355-360. 54. Stewart M: Patient characteristics which are related to the doctor-patient interaction. Fam Pract 1984, 1:30-36. 55. Street RL Jr, Gordon HS, Ward MM, Krupat E, Kravitz RL: Patient participation in medical consultations: why some patients are more involved than others. Med Care 2005, 43:960-969. 56. Street RL Jr: Gordon H, Haidet P: Physicians’ communication and perceptions of patients: is it how they look, how they talk, or is it just the doctor? Soc Sci Med 2007, 65:586-598. 57. Burgess DJ, Fu SS, van Ryn M: Why do providers contribute to disparities and what can be done about it? J Gen Intern Med 2004, 19:1154-1159. 58. Droomers M, Westert GP: Do lower socioeconomic groups use more health services, because they suffer from more illnesses? Eur J Public Health 2004, 14:311-313. 59. Parker RM, Gazmararian JA: Health literacy: essential for health communication. J Health Commun 2003, 8(Suppl 1):116-118. 60. Beck RS, Daughtridge R, Sloane PD: Physician-patient communication in the primary care office: a systematic review. J Am Board Fam Pract 2002, 15:25-38. 61. Kraus MW, Piff PK, Keltner D: Social class, sense of control, and social explanation. J Pers Soc Psychol 2009, 97:992-1004. 62. Van Berkel-van Schaik AB TB: Naar een standardoperationalisatie van sociaal-economische status voor epidemiologisch en sociaal-medisch onderzoek. 1990. 63. Epstein AM, Taylor WC, Seage GR: Effects of patients’ socioeconomic status and physicians’ training and practice on patient-doctor communication. Am J Med 1985, 78:101-106. 64. Beach MC, Roter DL, Wang NY, Duggan PS, Cooper LA: Are physicians’ attitudes of respect accurately perceived by patients and associated with more positive communication behaviors? Patient Educ Couns 2006, 62:347-354. 65. Piette JD, Bibbins-Domingo K, Schillinger D: Health care discrimination, processes of care, and diabetes patients’ health status. Patient Education and Counseling 2006, 60:41-48. 66. Simpson M, Buckman R, Stewart M, Maguire P, Lipkin M, Novack D, Till J: Doctor-patient communication: the Toronto consensus statement. BMJ 1991, 303:1385-1387. 67. Waitzkin H: Information giving in medical care. J Health Soc Behav 1985, 26:81-101. Page 14 of 14 68. Kaplan SH, Greenfield S, Ware JE Jr: Assessing the effects of physicianpatient interactions on the outcomes of chronic disease. Med Care 1989, 27:S110-S127. doi:10.1186/1475-9276-11-12 Cite this article as: Verlinde et al.: The social gradient in doctor-patient communication. International Journal for Equity in Health 2012 11:12. Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color figure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Chapter 3: Research papers Socio-economic status of the patient: does it influence the decisions of the GP? Verlinde E., Bonte E., Willems S. Corresponding author Verlinde Evelyn Department of Family Medicine and Primary Health Care, Ghent University, Belgium De Pintelaan 185 – 6K3 9000 Ghent [email protected] Telephone: 0032 9 332 6082 Fax: 0032 9 332 49 67 Bonte Eleonora Department of Family Medicine and Primary Health Care, Ghent University, Belgium Willems Sara Department of Family Medicine and Primary Health Care, Ghent University, Belgium Abstract Objectives: Our aim is to explore the influence of the SES of the patient on the clinical management decision of GPs in a country with a universal health care system. Methods: Eleven semi-structured interviews with GPs from varied practice settings in Flanders. Results: The attitude of GPs towards low SES patients differs according to the number of patients visiting the practice. Physicians with frequent contacts have the feeling to cope with prejudices and a fatalistic view. All physicians indicated that they changed 131 Chapter 3: Research papers their clinical decisions according to the SES of the patient. Those with occasional contacts with low SES patients said to made changes in order to enhance the outcomes for the patients while GPs with frequent contacts said the changes were not always in favor of the patients. Conclusion: Physicians working frequently with low SES patients feel caught between delivering a good standard of care and tailoring the care according to the needs and possibilities of the patients. Practice implications: Additional training, supervision, (psychological) coaching and other forms of remuneration could support physicians working with low SES patients. Keywords: social class, General Practice, attitude. Introduction 132 Since many years, equity in healthcare is an important topic in global health policy [1]. People with a low socio-economic status (SES) suffer disproportionate more from ill health and from multiple chronic conditions than their higher class counterparts [2, 3]. Similarly, the use of available health care is also not distributed across socio-economic groups according to the principle of equity. An equitable distribution of health care implies equal care for people in equal need, yet several studies show the opposite: patients with a lower socio-economic status less frequently visit a specialist (despite their higher disease burden) or a dentist, receive less chronic and preventive care (e.g. fewer mammograms, diabetic eye examination) and the received care is often from a lower quality compared to higher socio-economic groups [4-7]. An equitable health care system does not only succeed in equitable outcomes of care (e.g. equal vaccination rates or success rates after surgery in patient groups with the same health care need, independent from their socio-economic or cultural background or their sex) but also in equity in access to care and equity in treatment. Nor does it only reach equity in the structural components of care (e.g. an equal distribution of GPs in deprived areas as in affluent areas) but also in the process components of care. Concerning the latter, the doctor patient relationship is of crucial importance and can be a source of inequity but also a major contributor to Chapter 3: Research papers more equality in care for people in equal need for care. Variations in doctor-patient relations depend on the knowledge, skills, preferences, perceptions, attitudes and prejudices of both patient and health care provider [8]. Physicians are expected and expect themselves not to judge patients on their social class and above all, not to treat patients differently because of their sociodemographic characteristics [9]. Literature on the influence of socio-economic status on the clinical decisions physicians make is scarce and almost exclusively reporting on the situation in the US. Bernheim et al. interviewed 18 primary care physicians in different practice settings on the influence of patients’ socio-economic status on their clinical management decisions. Physicians indicated that patient’s SES did affect their decisions. Some doctors reported that the changes were made to enhance patient outcomes, others had the feeling the outcomes were not always in favor of the patient. An important finding of Bernheim and colleagues is that physicians said to no longer accept Medicaid patients, leaving the most vulnerable patients out of the loop [10]. We believe that, based on qualitative studies with patients, physicians working with low SES patients make changes in their clinical decision which not always favor the outcomes for the patient[11]. The aim of this study is to explore the influence of the socio-economic status of the patient on the clinical management decision of general practitioners in a country with a universal health care system for the patients. The universal Belgian health care system covers over 98% of the population and provides direct access to a general practitioner or specialist without gate-keeping or patient-list. There is a fee-for service payment system with reduced co-payment for the less privileged (about 1 euro per consultation with a GP). In 1982 the legal possibility to work in a capitation system in primary care was introduced in order to reduce the financial barrier towards primary health care. This payment system is mostly but not solely used in the Community Health Centres. These centres provide interdisciplinary comprehensive primary health care which all people residing in the neighbourhood can access without cost. 133 Chapter 3: Research papers Methods Study Design and Participants To explore general practitioners’ perceptions on working with patients from lower social classes and their vision on how this might influence their clinical decisions and the outcomes for the patients, we choose for a qualitative research design. In the selection of the participants we strived for maximum variation regarding the GPs age, sex, ethnicity (country of birth of both parents), years of experience, type of practice (solo, duo, group practice), location of the practice (rural versus urban) and frequency of contacts with patients with a low SES. The number of interviews was not pre-defined and interviews were conducted until thematic saturation was reached. Eleven semi-structured interviews with physicians were conducted. 134 The participants in this study were selected in three phases. In the first phase, 25 primary care physicians were identified based on the above characteristics. They were selected from the Flemish Community Health Centers and from lists of primary care physicians who previously participated in research conducted by the department of General Practice and Primary Health Care, practices that were known to the Department as outplacement practice for students or were informally known by Department staff. They were sent an invitation to participate in the study and received an information leaflet. Two weeks later, they were all contacted by phone. Six primary care physicians agreed to participate in the study. Based on the characteristics of the included doctors, and with the aim of maximum variation in mind, two additional physicians were added to the sample. They were identified by the physicians in the first phase by means of the snowball technique. After the second phase of inclusion, interviews were conducted. However, thematic saturation was not yet reached. Therefore, in the third phase of the inclusion the characteristics of the physicians were checked a last time. Ethnicity of the physicians and contact with patients with low SES did not yet vary enough. By means of the snowball technique another three physicians were identified and all were willing to participate. The reasons physicians gave for not participating in the study were: ‘already participating at other studies’ or ‘no time to participate’. Chapter 3: Research papers Data collection All interviews were conducted by a medical student – researcher (E.B.). Prior to the study, the interviewer received an intensive training in interviewing and during this phase of the data collection she was permanently supervised by a senior researcher in qualitative research methods (E.V.). The interviews took place in the physician’s office and lasted between 40 and 75 minutes. An interview guide was developed including open-ended questions and additional prompts and was tested in two interviews with physicians working at the Department of Family Medicine and Primary Health Care of Ghent University. The interviews started with the question whether the doctors frequently saw patients with a low SES in their practice and how they experienced these consultations. Next, physicians were asked to reflect on how the patient’s SES influences the care they give, including their clinical decision making. Physicians were prompted to give examples of changes in their clinical decision process. The concept of “SES” was not pre-defined by the researchers because one of the aims of this study is to gain insight into how physicians perceive and define this patient characteristic. Additionally, all physicians completed a brief written questionnaire on their demographic background, type of practice, payment system and years of experience. All interviews were recorded on audiotape and transcribed verbatim by the researchers. Data Analysis To code and analyze the interviews we used the software program NVIVO 9. Line-by-line open coding was used to develop the coding structure. Analysis was undertaken by the first two authors, EB and EV. Interview transcripts were coded by the researchers independently from each other. Consensus on the meaning of codes and the code structure was reached through numerous discussions. Coding and interviewing occurred simultaneously and as new concepts were identified during the coding, the interview guide was modified to reflect these emerging themes. To guard the quality of the coding, the researchers reviewed data within codes several times to ensure consistency in their use and full understanding of their scope and meaning. Ethics approval Participants were provided with written information about the study and all gave written consent. Any potentially identifying information that was given on the 135 Chapter 3: Research papers consent form or during the interviews was removed from the file before analysis. The interview tapes were deleted six months after the end of the study. The study was approved by the Ethics Committee of the Ghent University Hospital (Project number 2009/448). Results 136 A total of 11 primary care physicians participated the study (table 1); five men and six women. Six GPs were younger than 40, three were aged between 40 and 50 and two were older than 50. Five had less than 10 years of experience as a GP, four practiced between 10 and 20 years and two had more than 20 years of experience. One physician was not born in Belgium, the others were born in Belgium and had Belgian (grand) parents. They practiced in different settings: one physician had a solo practice, one had a duo practice, four physicians were part of a group practice and five physicians worked in a Community Health Centre. Five physicians predominantly treated patients with a low SES, five physicians predominantly treated patients with a middle or higher socio-economic status and one physician had almost no patients of lower social class in his patient population. Variable n (%) Sex Male 5 (45.0) Female 6 (55.0) <40 6 (55.0) 40-50 3 (27.0) >50 2 (18.0) Age Years of experience <10 years 5 (45.0) 10-20 4 (36.0) >20 2 (18.0) Country of Birth Born in Belgium Other country 10 (91.0) 1 (9.0) Chapter 3: Research papers Type of practice Solo 1 (9.0) Duo 1 (9.0) Group 4 (36.0) Community Health Centre 5 (45.0) Location practice Antwerp 2 (18.0) Vlaams-Brabant 1 (9.0) Oost-Vlaanderen 4 (36.0) West-Vlaanderen 4 (36.0) Number of patients with low SES Small number 1 (9.0) Average 5 (45.0) Large number 5 (45.0) Table 1: characteristics of the physician sample Perception of socio-economic status of the patient When asking the physicians how they defined the socio-economic background of their patients this resulted in a very diverse pallet of answers. They firstly referred to economic aspects such as being unemployed, being entitled with a social welfare benefit, being uninsured, homeless, tenants, or having a lot of debts. Secondly, they also referred to socio-cultural aspects (being immigrant, having a low educational level, being illiterate, having a restricted social network or living in social isolation or having a large family), living conditions (low quality housing, living in a deprived neighborhood, lack of personal hygiene, poor diet), health challenges (addiction, multi-morbidity, mental health problems, more unexpected pregnancies), childhood education, attitude and knowledge about health (poor health literacy, competing priorities, postpone health care, not interested in prevention, very susceptible to advertisements). Physicians used a combination of both positive (e.g. loyal) and negative (e.g. disrespectful) connotations to describe their perceptions of low SES people. 137 Chapter 3: Research papers Attitude towards low SES patients All GPs showed a higher commitment and were more concerned about these patients. Physicians said it is very important to be empathic with these patients. However, they indicated that it is not always easy to work with low SES patients. Annotations about working with patients from lower social class varied from “challenging” to “very frustrating” and “out of control, no matter how hard they tried”. How physicians cope with these feelings seems to correlate with the number of patients of low social class visiting the practice. Physicians with occasional contacts with low SES patients seemed to perceive them as more grateful and very loyal patients. They described working with these patients as a positive challenge and very enriching, nevertheless the difficulties that also come across. “They are very dedicated to you, because they feel supported by you (physician 8).” 138 Physicians with frequent contacts with low SES patients and the one physician barely seeing low SES patients reported a less positive feelings towards low SES patients. “Sometimes I have the feeling they don’t appreciate you. You do a lot of effort for these people, but these are the kind of patients who take it for granted (physician 1).” They report to lower their expectations towards low SES patients e.g. concerning drug adherence or progression in healthy eating habits. Often this is induced by earlier experiences of failure when setting targets, leading to frustration and a kind of “tiredness” in staying motivated. “ You always have some kind of expectations, but you never achieve them (physician 6). “ As a result of this frustration or “tiredness” the physicians in this study report to sometimes acquiesce in the patient’s situation or behavior, not trying to motivate the patient to change their behavior (eg. smoking habits, dietary habits) or to take action (eg. to find a healthier house) anymore. Earlier failures with patients might result in less attempts to motivate other patients with a similar socio-demographic profile. “After a while, you make peace with the situation. For non-urgent matters, it’s often not possible to make a change (physician 10).” Chapter 3: Research papers The participating physicians were aware of their so called fatalistic view and emphasize the danger of prejudices. “ You are confronted with your own prejudices, I have to be aware not being too fatalistic (physician 10).” “ You have to be careful with prejudices. Sometimes I think: “Should I really need to explain this, they won’t understand it anyway”. But it’s not because they have a lack of money, that they don’t have brains (physician 5).” Influence of patients’ SES on treatment All physicians stated they try to work evidence based, although they had the feeling this was not always possible. Sometimes they had to adapt the treatment or medication in order to achieve the best possible outcome. “Sometimes you have to do it yourself. The patients need specialist care, but he/she cannot afford it. You have to try with trial and error until you succeed (physician 1).” The one physician barely seeing low SES patients said always to work evidence based and made no exceptions. Also on drug prescribing, referring, additional tests and payment all physicians aligned. However, those with fewer contacts often did more effort. All physicians prescribed generic medication or tailored the medication to an easier scheme. GPs with fewer contacts with low SES patients made additional agreements with pharmacists e.g. on payment or phased delivery. Sometimes these GPs also gave free samples of medication when they felt that their patients could not afford it. “When prescribing medication, I look for the cheapest medication. Euhm, while with other patients you prescribe the medication you are used to prescribe. You put more effort in searching for cheaper medication (physician 1).” All physicians thought twice before referring a patient to a specialist or performing additional tests. They only referred a person or made an extra test if this could determine or change the treatment. “…, another aspect is that you try to avoid additional cost, by thinking twice. Will the referral or the extra test add something to my decision (Physician 8).” 139 Chapter 3: Research papers All physicians tried to make agreements with certain specialists to drop the copayment in order to make specialist care financially more accessible for their low SES patients. Physicians not working under the capitation system, adapted the third-party payment (where the patient only needs to pay the co-payment and the physician gets paid by the sickness fund) to lower the financial burden of the consultation. GPs with fewer contacts with patients of lower SES sometimes adapted their own payment rule (e.g. drop the co-payment, pay later,…) “… If we use the third-party payment it is no problem to let them come back for follow up. This lowers the financial burden. So I think we removed a huge obstacle in this way (physician 4).” 140 Physicians with less contact with patients from lower social classes often performed more home visits. They stated that working with appointments creates a higher burden for the organization of the consultations and patients often forget or don’t show up. These physicians found house visits especially functional for the follow up of chronic conditions. They go see the patient once or twice a month to make sure he/she is fine. This way they also have more control on the therapy compliance, which was stated by all physicians to be a large problem among patients from lower social classes. “We do more house visits for these patients. Making an appointment or transportation is often a problem. By doing more house visits, we try to tackle these problems (physician 4)”. Physicians who frequently have contact with low SES people, work almost solely on appointment. All physicians agreed that working with low SES patients takes more time and has an influence on their consultation length. They have to explain things better, ask more questions to see if the patient actually understands what they have said,… “ You make sure you have enough time for the consultation… You put them on them last on the day, so you don’t have to stress when the consultation takes longer (physician 8).” All physicians agreed on the basics of delivering patient-centered care and shared decision making. They found it important to consider the patient as a person and to listen to their individual needs even if these needs are not health care related (e.g. housing problems). Chapter 3: Research papers “For example a patient whose refrigerator is broken at the moment, so he cannot store his insuline… Euhm so this patient, it’s something we see now and then, he lives in a hard socio-economic situation. He doesn’t know anybody in the neighborhood, therefore there is nobody he can ask to store his insulin. So now his insulin is getting warm, he keeps on using it, but you can see that it has lost its effect. So we switch temporary to oral medication until he has a new refrigerator. These are things that often happens within this group of patients (physician 10).” Influence of patients’ SES on the decision making and the outcomes for the patient All physicians stated that they do make changes in their decisions according to the SES of the patient. However, the outcomes of these decisions varied according to the number of patients of low social class visiting the practice. Physicians with less contact with patients from lower social classes said to do the best they can to give each patient the same treatment and strive for equal health outcomes. They do make changes in their clinical decisions but these changes are made to enhance the patient’s outcomes. “I think in most cases, the outcomes are in favor of the patient (physician 3).” Physicians with frequent contact with patients from lower social classes stated always to start with equal treatment, but they feel it is not always possible to achieve this. Some said it is even not always possible to give good qualitative care to these patients. They must make the best decision within a limited number of options. Evidence based medicine is not always an option. Some physicians have the feeling that the changes they make do not always favor the patients, some even say the decisions may disfavor the patient. “First, we do not change our behavior according to the socio-economic status of the patient. We will try, wherever possible, to give the same treatment we give to anybody (physician 10).” “I think that the quality of care is less. Because it is difficult to communicate with these people, which is the biggest obstacle. Also for some matters you need a financial reserve (physician 2).” “Interviewer: do you have the feeling that these changes are in favor of the patient?” Physician 2: “No, not at all.” 141 Chapter 3: Research papers “There was a patient who needed an MRI scan, however, this patient had no health insurance, so I decided not to perform the test (physician 5)”. “Sometimes it is difficult to prescribe the right medication. For example, the patient has a respiratory infection. Paracetamol would be the best way to relieve his symptoms, however, paracetamol does not get reimbursed. So we prescribe NSAID’s (non-steroidal anti-inflammatory drug) because they get reimbursed. (physician 10)”. Discussion The aim of this study was to explore how patients’ SES influences the perceptions and the decision making of the GP, and how this could affect the health care outcomes. 142 First of all, SES was not defined by the researchers but by the perception of the physicians. Physicians did not define SES solely in terms of occupation, educational level or income, the indicators most commonly used for SES in research [12-14]. Physicians gave a much broader explanation including economic aspects, socio-cultural, childhood conditions, health literacy,… These results confirm the complexity of measuring SES in daily practice. All physicians said that working with low SES patients takes more time and can sometimes be frustrating. However, handling these kinds of feelings seemed to vary according to the number of contacts with patients from lower social classes. Physicians with occasional contacts with low SES patients perceived working with these patients as more enriching and a positive experience. They put more effort in these patients (e.g. more house visits) and perceived them rather as loyal and grateful. While physicians with frequent contacts with low SES patients said to sometimes lower their expectations towards their patients and accept situations as they are. Some said to have a fatalistic view and caught themselves on prejudices. Regarding to the changes made in the clinical decisions, the results also vary according to the number of contacts with low SES patients physicians have. On the one hand, the physician with nearly no contact with low SES patients said not to change his/her decisions or treatment according to the social class of the patient. All other physicians on the other hand, said to change their treatment or decisions according to patients’ SES. Those with less contact with low SES patients, put a Chapter 3: Research papers lot of effort in treating all patients equal. The changes they made (medication, house visit, payment system,…) were made in order to enhance the outcomes for the patient. For example: thinking twice before referring a patient to a specialist, avoiding unneeded additional costs could enhance the outcomes for the patient in terms of good qualitative care. Those with frequent contacts with low SES patients felt that they were not always able to give qualitative care due to a limited number of options and resources. They strive for equal treatment, but often have to change their treatment or decisions. They are concerned that the outcomes not always favor the patients. These results point out some important issues when caring for low SES patients in a primary care setting. First of all, all physicians said to make changes in their care and decision making according to the SES of the patient. These changes, they say, are not always in favor of the patient’s outcomes. This may contribute to the socio-economic disparities in health care. Physicians make changes on what they believe is best for the patient. However, this does not necessarily reflect the needs and the values of the patients. Secondly, physicians with occasional contacts with patients from lower social classes often try harder: out-of-office visits, samples of medications, making appointments with a specialist,… As stated before, they make these changes in order to enhance the outcomes for the patients. However, next to the positive aspects of these changes (good qualitative care) these changes could lead to more passive patients. Patients are getting used that their GP does everything for them and as a consequence they show less interaction during the consultation. As a physician, it is important to empower patients towards higher levels of self-efficacy and interaction during the consultation. It has been proved that interventions on more patient-interaction and patient control lead to better health outcomes [15]. Thirdly, there seems to be a difference in the perceptions and health care outcomes for the patients, between GPs with frequent contacts with low SES patients and those with occasional contacts. GPs with frequent contacts seem to feel more often frustrated, fatalistic and sometimes cope with prejudices. They feel caught between delivering a good standard of care and tailoring the care according to the needs and possibilities of the patients. Sometimes they feel that delivering good qualitative care is not always possible due to a lack of (financial) resources. A Belgian study has indicated that these feelings and the additional load when caring for low SES people, raises the chances of a burn-out for the GP [16]. Next 143 Chapter 3: Research papers to these personal strains, another aspect when caring for low SES patients could be the financial strain. The physicians with frequent contact with low SES patients mostly worked in Community Health Centers. These centers apply the capitation payment system. Within this system the health centres are paid a specified sum of money for the ongoing care of a patient for a particular period over time no matter how much or little this patient uses this service [17]. Physicians stated that care for low SES patients often takes more time and more visits. Within the capitation system, based on the average cost of a patient, there is no financial compensation for this more intense care. A possibility of tackling this problem is the principle of weighted capitation, where GPs working with a deprived population group, get a higher share. This system is already successfully applied in the UK with the Jarman underprivileged area-score [18]. 144 There are several strengths and limitations to our study that must be taken into account. First of all, we limited our sample to physicians working in Flanders. Therefore, the results cannot be generalized for Belgium. Secondly, when sampling the physicians, special attention was given to the diversity of the physicians (age, gender, ethnicity, years of experience), type of practice and their contact with patients from low SES. Nevertheless, the largest part of the physicians worked in an urban area. The perceptions of physicians working in a rural area could differ, due to a different sample of patients. Thirdly, there could be a selection bias in the recruiting of the physicians: physicians with a positive attitude or caring for low SES patients could be more likely to participate the study than those not working with low SES patients. In order to reduce this selection bias, we aimed at maximal variation according to the number of low SES patients in a GP practice. Fourth, we only interviewed the physicians, the patient’s side nor the progress of the consultation was taken into account. Physicians could make subconscious changes in their behavior which can only be defined when videotaping the consultation. Fourth, the interviewer was a medical student inexperienced in qualitative research. This could lead to possible biases in reporting the results. For this reason, all interviews were additionally coded by a second investigator, a health services researcher with substantial qualitative research experience and experience in quality of care. Finally, SES was not predefined, so we were able to gain insight into physicians’ own conception of SES and of which factors associated with SES they believed influenced their management decisions. Chapter 3: Research papers Conclusion Physicians make changes in their clinical management decision according to the social class of the patient. The impact on the quality and outcomes of care seems to differ according to the number of contacts physicians have with low SES patients. Physicians caring for many low SES people say to have difficulties finding the balance between providing good quality of standard care and tailoring the care to the capacities and needs of low SES patients. When caring for too many low SES patients, GPs seem not to be able to maintain their high standard of qualitative care (figure 1). Quality Additional resources Critical number 145 N° of patients with low SES Figure 1: Additional resources supporting physicians working with low SES patients. Practice Implications Additional support should be given to GPs caring for low SES people in order to maintain this standard of care. First of all, medical training focuses on adapting communication behavior (such as explaining things in a non-medical way) to patients with low SES. However it does not focus on the challenges and the personal and financial constraints physicians perceive when working with patients from low SES. Through early and repeated exposure of students in undergraduate curriculum to people living in poverty, these future GPs learn to know and how to handle this population group. Secondly it is Chapter 3: Research papers important to give additional training on the available financial measures (existing and new) in order to lower the financial burden for the patient. A Belgian study showed that the Belgian GPs have a limited knowledge of the available financial measures [19]. Thirdly, additional resources e.g. supervision, (psychological) coaching , administrative and financial (other forms of remuneration) support, could help physicians maintaining delivering good quality of care for a large number of low SES patients. Authors’ contribution EV and SW contributed to the design of the study. EB and EV contributed to the data retrieval and the analysis. EV, EB and SW contributed to the writing of the paper. Acknowledgements 146 We would like to thank the participating physicians for their motivation to participate in their interest in this research theme. Competing Interests All authors declare that no competing interests exist. Chapter 3: Research papers References 1. Marmot M, Allen J, Goldblatt P, Boyce T, McNeish D, Grady M, Geddes I: Fair Society, Healthy lives. The Marmot review. Executive Sumary; 2010. 2. Droomers M, Westert GP: Do lower socioeconomic groups use more health services, because they suffer from more illnesses? Eur J Public Health 2004, 14(3):311-313. 3. Westert GP, Satariano WA, Schellevis FG, van den Bos GA: Patterns of comorbidity and the use of health services in the Dutch population. Eur J Public Health 2001, 11(4):365-372. 4. Van der Heyden JH, Demarest S, Tafforeau J, Van Oyen H: Socio-economic differences in the utilisation of health services in Belgium. Health Policy 2003, 65(2):153-165. 5. Veugelers PJ, Yip AM: Socioeconomic disparities in health care use: Does universal coverage reduce inequalities in health? J Epidemiol Community Health 2003, 57(6):424-428. 6. van Doorslaer E, Masseria C, Koolman X: Inequalities in access to medical care by income in developed countries. CMAJ 2006, 174(2):177-183. 7. Fiscella K, Franks P, Gold MR, Clancy CM: Inequality in quality: addressing socioeconomic, racial, and ethnic disparities in health care. JAMA 2000, 283(19):2579-2584. 8. Goddard M, Smith P: Equity of access to health care services: theory and evidence from the UK. Soc Sci Med 2001, 53(9):1149-1162. 9. van Ryn M, Burke J: The effect of patient race and socio-economic status on physicians’ perceptions of patients. Soc Sci Med 2000, 50(6):813-828. 10. Bernheim SM, Ross JS, Krumholz HM, Bradley EH: Influence of patients’ socioeconomic status on clinical management decisions: a qualitative study. Ann Fam Med 2008, 6(1):53-59. 11. Willems S: The socio-economic gradient in health: a never-ending story? A descriptive and explorative study in Belgium. Ghent: Ghent University; 2005. 12. Oakes JM, Rossi PH: The measurement of SES in health research: current practice and steps toward a new approach. Soc Sci Med 2003, 56(4):769784. 13. Shavers VL: Measurement of socioeconomic status in health disparities research. J Natl Med Assoc 2007, 99(9):1013-1023. 147 Chapter 3: Research papers 14. Adler NE, Newman K: Socioeconomic disparities in health: pathways and policies. Health Aff (Millwood) 2002, 21(2):60-76. 15. Kaplan SH, Greenfield S, Ware JE, Jr.: Assessing the effects of physicianpatient interactions on the outcomes of chronic disease. Med Care 1989, 27(3 Suppl):S110-127. 16. Jonckheer P, Stordeur S, Lebeer G, Roland M, De Schampheleire J, De Troyer M, Kacenelenbogen N, Offermans A, Pierart J, Kohn L: Burnout bij Huisartsen: preventie en aanpak. Health Services Research (HSR). In. Edited by (KCE) FKvG, vol. KCE Reprots 165A; 2011. 17. Physicians funding and health care systems - an international perspective. In: A summary of a conference hosted by the WHO, WONCA and RCPG at St John’s College, Cambridge. Edited by Brown PS. 18. Art B, De Maeseneer J: Forfaitaire betaling: een beter systeem voor patiënt en huisarts. [Capitation system: a better system for the patient and doctor]. 148 19. Broeckaert O, De Troeyer K: Sociale dienstverlening en financiële tegemoetkomingen voor de patiënt: hoe ver reikt de kennis van de huisarts? [Social services and financial measures for the patient: what is the knowledge of the GP?] ; 2011. Chapter 4 Discussion 149 150 Chapter 4: Discussion 4. Discussion 4.1 What this thesis adds: summary of the results This doctoral thesis focuses on the equity in primary health care in Belgium in terms of equal access, equal treatment and equal outcomes for vulnerable population groups in equal need compared to non vulnerable population groups. Figure eight provides an overview of the hypotheses of this thesis. Five studies were set up: one qualitative study focusing on the influence of the patient’s socio-economic status on the clinical management decisions of the GP, one systematic literature review on the impact of patient’s social class on the doctor-patient communication and three quantitative studies focusing on the social inequities in health care use in Belgium and on the health care use of two vulnerable population groups namely homeless people and low-income people. Equity in Health Care 151 Equal treatment for patients in equal need Equal access to care for patients in equal need e.g. communication, attitude, decision making Equal outcomes for patients in equal need e.g. financial, cultural and geographical access Hypothesis 1 The specialist is the doctor of the high income people and the GP is the doctor of the low to middle income people (paper 1). Hypothesis 2 Homeless people have a higher likelihood to consult the emergency department than the general population. A universal primary health care system with active guidance for homeless people towards this system might counter this (paper 2). Hypothesis 3 People who are not able to cope with their available income have a higher likelihood to postpone a visit to a GP (paper 3). Hypothesis 4 The social gradient in doctor-patient communication persists, where patients from lower socio-economic classes are less involved during the consultation, receive less control and a more paternalistic consultation style (paper 4). Hypothesis 5 Physicians have a negative attitude towards patients with a low SES and as a result make changes in their clinical decision which do not favor the outcomes for the patient (paper 5). Figure 8: an overview of the main focus of this thesis, the hypotheses and the research papers Chapter 4: Discussion Exploring the social gradient in health care utilization in Belgium using registered data on health care use. In paper one, we linked data from the Belgian Health Interview Survey with corresponding registered medical utilization data provided by the Sickness Funds. Despite the nearly universal health care system, we found inequity in utilization patterns according to educational level. The findings of this study indicate that the specialist is not the doctor of the ‘rich people’, but rather is the doctor of the higher educated. While the GP is not the doctor of the ‘poor’, but the doctor of the lower educated. Patients with a lower or middle educational level have a higher ratio for contacting their GP compared to higher educated patients, even after controlling for health care need (including subjective health, chronic diseases and disability). When looking at specialist care, this picture changes: higher educated patients have a higher tendency in contacting a specialist compared to lower educated people. These results remain significant even after controlling for health care need. However, attention is needed when interpreting results on income as the measurement of income is far less stable and valid than the measurement of education. 152 These results contrast with findings reported in studies using self-reported utilization data. Data from the Belgian Health Interview Survey indicate that the social gradient in the utilization of the GP disappears when health status is taken into account [1]. With regard to the use of specialist care, our findings are consistent with the evidence based on self-reported data: taking into account the disease burden, higher SES people seek specialist care more often than do lower SES people [2, 3]. The conclusions of this study are fourfold. First of all, this study stresses the importance of using accurate information on health care utilization. Secondly, these findings indicate that primary care succeeds in being accessible for vulnerable population groups such as low-income people or low-educated people. Third, primary care teams are in a position to take on a coordinating role by helping people navigate through the health system and mobilizing support by referring the patients or calling on the support of specialized services. Fourth, in a health care system without gate-keeping, higher educated tend to utilize relatively more specialist care compared to lower-educated people (paper 1). Chapter 4: Discussion Exploring the health care use of two vulnerable population groups: lowincome people and homeless people. Both homeless people and low income people are vulnerable population groups facing a higher disease burden leading to a higher health care need [4-6]. However, despite their higher health care need, these people do not always find the way to the appropriate health care service, or they postpone medical care [7-11]. In paper 2 and 3 we aim to explore the healthcare use of these two vulnerable population groups. The homeless people residing in Ghent have a higher likelihood to consult a GP than the non-homeless people in Ghent, even after adjusting for age and sex. The same trend is seen for secondary care and the use of emergency care. This contrasts with findings from existing studies (mainly from the UK and the US) describing higher utilization rates for hospital-based care and emergency care and lower rates for primary care by homeless people compared to the general population [5, 7, 12]. The higher likelihood in Ghent to consult a GP in the homeless population compared to the general population could be explained by the fact that primary health care for homeless people in Ghent is organized in a way that contributes to the reduction of some of the barriers such as financial barriers, accessibility, registration,… The area of Ghent is to a large extent covered by Community Health Centers that provide interdisciplinary comprehensive primary health care using a capitation system without co-payments. Additionally, homeless people with health problems are actively guided towards Community Health Centres and large group practices by social workers. These practices are located close to homeless shelters and social restaurants so transportation is not an issue. There are agreements on the payment procedure – even for those with administrative problems with health insurance- so the financial barriers are minimalized. Furthermore, Community Health Centers, are known not to stigmatize or refuse patients because of being homeless or having no financial resources. This hypothesis is supported by the finding that the homeless people who visit a GP, visit more often a group practice compared to the general population (56% versus 5.6% in general population in Ghent) (paper 2). In the third paper, 606 low-income people were interviewed on their health care use. Nearly one in three respondents postponed or cancelled a visit to the GP while they actually needed one. The reasons for postponing these visits were reasons that were linked to the health system such as accessibility, stigmatization, not knowing where to go,… The sub-groups within the low-income people that have a higher tendency to postpone a visit to a GP were people suffering from a severe depres- 153 Chapter 4: Discussion sion, people with a poor self-rated health, and people with moderate or low trust in their GP. Against our expectations, having difficulties to cope with the available monthly income had no effect on postponing a visit to a GP. These findings could indicate that the health care system in Flanders does not create huge financial barriers. Possible explanations are the success of the interventions that try to lower the financial burden for low-income patients such as the third-party system and the reduced co-payment in the fee-for-service payment system and the usage of the capitation system in the Community Health Centers and some larger group practices. However, these results also indicate that the access to primary care for low-income patients might be hindered by barriers which do not directly link to the costs of consultation. Such as: finding appropriate child care, not knowing where to go with a particular health problem, being afraid or ashamed, thinking the doctor will not be able to solve the problem or having the feeling the doctor does not understand their struggle to survive and its implications for their health and compliance (paper 3). 154 Doctor-patient interaction: the role of patient’s socio-economic status on the doctor’s communication style In the studies on the health care use of homeless and low-income people described above, the doctor-patient interaction and relationship is mentioned as a possible barrier or facilitator in the utilization of medical care in Belgium (paper 2 and 3). Also other studies conclude that the interaction between the patient and the health care provider is an essential component of the delivery of health care. A good doctor-patient relationship adds to a better quality of care, better compliance and a higher level of patient satisfaction. When building a doctor-patient relationship, one of the components is doctor-patient communication [13, 14]. Paper four presents the findings of a systematic literature review. Since the review by Willems et al. in 2005, the social gradient in doctor-patient communication persists. Patients from lower social classes receive less socio-emotional talk, a more directive and less participatory consultation style characterized by for example less involvement in treatment decisions, a higher percentage of biomedical talk and physician’s question asking; lower patient control over communication; less diagnostic and treatment information and more physical examination. While with patients from higher social classes, doctors gave more information, more explanations, more (emotional) support and adapted more often a shared decision making Chapter 4: Discussion style. Since 2005, emphasis is on the importance of the perception of the patient and the reciprocity of the communication: doctors may communicate differently according to the social status of the patient, but also patients may (subconsciously) adapt a different communication style according to their social class (paper 4). The role of the socio-economic status of the patient on the clinical decisions made by their GP Variations in doctor-patient interactions depend on the knowledge, skills, preferences, perceptions, attitudes and prejudices of both patient and health care provider [15]. Physicians are expected and expect themselves not to judge patients on their social class and above all, not to treat patients differently because of their sociodemographic characteristics [16]. Paper five presents the results of a qualitative study on the influence of the socioeconomic status of the patient on the perception, attitude and clinical decision making of the GP and the health outcomes for the patient. There seems to be a difference in the perception and attitude of GPs towards patients from lower social classes according to the number of patients with a low SES visiting the practice. All physicians agree that working with low SES patients takes more time, is often beyond the control of the GP and can lead to frustration. How GPs handle these feelings seems to depend on the number of low SES patients visiting the practice. GPs with occasional contacts with patients from low SES perceive working with low SES patients more often as a positive challenge. They take more time for these patients and invest more energy (for example: more house visits, distribution of free samples of medication, helping with paperwork,…). They perceive their low SES patients as loyal and thankful. However, among GPs who have very often contact with patients from low SES, frustration can lead to a fatalistic view and prejudices. Next to the differences in attitude and perception, there are also differences in the decision making process and the health outcomes according to the number of contacts with low SES patients. All physicians acknowledge that they change their clinical decisions according to the social class of the patient. Those with occasional contacts with low SES people reported that the changes were made in order to enhance the health outcomes for the patient, while those with frequent contacts were not always sure that their changes in decision led to better health outcomes for the patient. These results stress the challenges for a GP when caring for patients with low SES. GPs feel caught between delivering a good standard of 155 Chapter 4: Discussion care and tailoring the care according to the needs and possibilities of the patients. Sometimes delivering high quality care seems not possible due to a lack of (financial) resources (paper 5). Box 4:What this thesis adds • The specialist is not the doctor of the higher-income people but the doctor of the higher educated people. While the GP is not the doctor of the lowerincome people, but the doctor of the lower-educated people. These findings indicate that the Belgian health care system is financially accessible, but still has other access issues related to the educational level of the patient. • Low-income people suffering from bad health, those suffering from a severe depression and those with moderate or low trust in their GP have a higher risk of postponing a visit to a GP. People not able to cope with their monthly income do not have a higher risk of postponing a visit to a GP. • A universal health care approach for homeless people, with active guidance by social workers, seems to be far more effective than a selective setting focusing solely on that specific population. This universal approach leads to less stigmatization and labeling than a selective categorical approach, leading to a higher use of these services. • The social gradient in doctor-patient communication persists over time, where patients from lower social classes receive less socio-emotional talk, a more directive and less participatory consultation style characterized by less involvement in treatment decisions, lower patient control over communication, less diagnostic and treatment information and more physical examinations. • The attitude and decisions of GPs toward patients from lower social classes depend on the number of contacts physicians have with this population group. GPs with occasional contact with patients from low SES take more time for these patients, invest more energy and see it as a positive challenge. They make changes in their clinical decisions, but these changes are made in order to enhance the patients’ outcomes. However, when the number of low SES patients is too large, GPs say they are not always capable to maintain providing good quality care. They have the feeling that the changes they make in their decisions are not always in favor of the outcomes for the patient. 156 Chapter 4: Discussion 4.2 Conclusion In order to provide equitable primary health care, equal access, equal treatment and equal outcomes for patients in equal needs is essential [17]. 4.2.1 Equal access for equal need Since 2007, a set of policy measures are taken in Belgium to strengthen the accessibility of the health system, to assure health care quality and to maintain the financial sustainability of the system [18]. The studies in this doctoral thesis confirm the strengths of the Belgian health care system in its accessibility. First of all, the health care system in Belgium does not create huge financial barriers: (1) low income people in Flanders who have the feeling not being able to manage with their available income do not postpone more often primary health care than those having the feeling being able to cope with their income; (2) homeless people in Ghent do find their way to the primary care centers, which contrasts the findings from other countries (3) low-educated people find their way to primary care. Belgium has already come a long way in reducing the financial barriers towards health care. Looking at primary care, there is the reduced co-payment for the less privileged, the maximum bill limiting the health care expenses, the third-party system, the introduction of the capitation system; the preferential reimbursement,…. However, the success of the primary health care for homeless people is not solely the result of a low financial threshold but also of the active guidance by social workers towards the primary health care system. In contrast to these positive results concerning the financial access, there still seems to be inequity on the other dimensions of access to health care: (1) lower educated people have a higher likelihood of consulting a GP compared to higher educated people, while higher educated people have a higher likelihood to consult a specialist compared to lower educated people. These results persist even after taking into account patients’ health need; (2) low-income people with no complete trust in their GP tend to postpone more often a needed medical care visit compared to people who completely trust their GP. These inequities could be explained by cultural barriers such as language barriers and miscommunication between the patient and the health care provider. Lower educated persons might not always understand the things a specialist explains or 157 Chapter 4: Discussion they are afraid of being labeled as not smart or being stigmatized. But there is also often a gap between the priorities of a low SES patient and those of a health care provider. Often people from lower social classes focus on getting through the day, providing food for the children, finding shelter… instead of caring for their own health problems [19]. Next to these communication barriers, there are also practical barriers (for example finding child care or transportation), the organization of the health care system, the attitude of the health care provider and the doctor-patient relationship. Having trust in the GP leads to a more efficient and longitudinal health care [13, 14]. In order to develop a trust-creating environment, a good doctor-patient communication and a regular source of care are essential components. Having the opportunity to visit the same GP, who knows your medical history, but also your personal strains and worries, who you feel comfortable to talk to and who you can trust adds to the quality of the delivered health care. 158 The results indicate that primary care favors access to care for the population in general and vulnerable populations in particular. By taking on a coordinating role, primary care can counteract the current fragmentation of health care by guiding patients through the system where necessary [20]. Furthermore, a gate-keeping system could even strengthen these effects. Finally, one of our studies indicates that a universal approach for specific population groups, such as the homeless people, seems to be far more effective than a selective setting focusing solely on that specific population. This universal approach leads to less stigmatization and labeling than a selective approach, leading to a higher use of these services. Moreover, it may contribute to more social cohesion in the local community. 4.2.2 Equal treatment and equal outcomes for equal need Next to some inequities in the access to health care, this thesis illustrates inequities in treatment and outcomes of health care: (1) doctors communicate differently according to the social class of the patient (lower SES patients receive less socioemotional talk, a more directive consultation style, receive lower patient control, less diagnostic and treatment information,…); (2) the perception and attitude of GPs working with low SES patients varies according to the number of contacts with these patients: GPs with frequent contacts feel more fatalistic than GPs with occasional contacts; (3) GPs change their clinical decisions according to the social Chapter 4: Discussion class of the patient; (4) the outcomes of these changes in decision-making always seem to favor the low-SES patient. It is important to draw special attention on these findings. First of all, there is the aspect of the reciprocity of the communication: the doctor might communicate different according to the social status of the patient, but also the patient may communicate different according to their social class. Patients from lower social classes seem to exercise less control on the consultation than patients from higher social classes. Patients from lower classes often have lower levels of health-literacy (the degree to which persons have the capacity to obtain, process and understand basic health information and services needed to make appropriate health decisions), a lower sense of personal control and often feel not capable to interact during the consultation [21, 22]. Besides teaching communication skills to GPs and providing attitude-training (e.g. through early and repeated exposure of students in undergraduate curriculum to people living in poverty [23]), it is important to empower the patients towards more self-efficacy and towards learning how to express their concerns and preferences [24]. However, this empowerment should include the capabilities and perspectives of the patient: it is important that a GP empowers a patient, but it is even more important to empower a patient to a degree he/she desires. Secondly, GPs working occasionally with low SES patients said to try harder when working with these patients and to make changes in their decisions in order to improve the health outcomes for these patients. This can enhance the quality of the delivered care (medication, referring, tests,…). However, trying too hard for these patients can lead to positive (health) outcomes on short term, but can lead to negative (health) outcomes on long term. For example: if the GP always makes the appointments with a specialist or makes arrangements for medications or always does home visits, the patient never “learns” to do it himself. This can lead to passive patients in the long run with even lower levels of health-literacy and interaction. Guiding patients in the health care process, is more intense in the beginning, but might lead to better self-efficacy and better health outcomes for the patient (health) and lowers the long term work burden for the GP. Thirdly, physicians working frequently with low income people report more often frustrations which can lead to prejudices on the one hand, and a fatalistic outlook on the other. This might result in a less adequate treatment and less quality of care, leading to worse health outcomes for the patient. But these feelings also form a higher risk of burn-out for the doctors. In figure 9 we illustrate a hypothesis on 159 Chapter 4: Discussion the quality of care and on the need for additional resources when working with low SES patients. A GP needs a certain number of patients with low SES in their practice in order to give good qualitative care. However, when caring for too many low SES patients (critical number), it seems not possible to maintain this high standard of qualitative care. Providing physicians with additional resources (financial, administrative,…) could help them to maintain good quality of the delivered care for a larger number of low SES patients. In Belgium, this critical number might be highly context dependent. For example: physicians working in an area with a shortage of GPs, an area with a high population density or in a deprived area can have the advantages of the Impulseo I fund (grant interest-free loans and subsidies). With these additional resources they can improve access and maintain the quality of care for social vulnerable population groups in the population. In order to increase the critical number even further, other, probably non-financial, resources will be needed (e.g. administrative support, organizational, multidisciplinary team…). Quality 160 Additional resources Critical number N° of patients with low SES Figure 9: Additional resources supporting physicians working with low SES patients. Chapter 4: Discussion 4.3. What this thesis could not add: suggestions for future research It is important when doing research on the inequity in health care use, that accurate and valid data are used. We have showed that it is possible to link registered medical utilization data with data coming from Health Interview Surveys. We suggest that this is implemented as a routine in survey-analysis. However, we noticed that people are reluctant to give information on their income and therefore income is not valid as a measure for socioeconomic status. Including recorded information on income might add to the validity of the results. These data for Belgium are available in the “Datawarehouse Arbeidsmarkt” (DWA). Furthermore data on health care need in the Health Interview Survey is also very subjective. A possibility for overcoming this issue could be by linking to data coming from Health Examination Survey (HES). The HES contains high quality data on the most important health indicators (eg. diabetes, cholesterol, high blood pressure,…). However, HES data for the Belgian population are not yet available. Another possibility is using the “Permanente Steekproef ” from the “Inter Mutualistisch Agentschap” (IMA) containing registered medical utilization data. Linking registered medical utilization data with data from the DWA and HES could lead to very accurate data and possible new findings on the equity in health care use in Belgium. When striving for equity in health care use, it is important not to lose sight of the quality of care. It has no use striving for the same treatment for all in equal need, when this treatment does not fulfill the quality issues. Receiving high quality of care is a fundamental right of every individual. This includes good access to health care facilities, effective care according to the latest evidence, efficient, well-organized and safe care processes, and, most of all, care directed to the needs of patients whatever their demographic background [25]. Nevertheless, there still are disparities in the quality of health care delivered to different socioeconomic groups in several studies and countries. A study on socio-economic disparities in health care quality in four countries, each with a different health system (Canada, England, New Zealand and the United Sates), found that disparities in a variety of quality indicators persist. This underscores the importance of further exploring the quality issues outside the health care system [26]. When studying quality of care, one of the challenges is to develop a valid instruments which captures the perception of quality of the delivered care on the patients’ side as well as on the physicians’ side. For example, in this thesis the study on the 161 Chapter 4: Discussion influence of patient’s SES on the attitude and clinical decision making of the doctor shows some interesting results. In order to verify these results, a quantitative study on the influence of patient’s SES on the attitude, decision making process and the impact on the quality of care is needed. The challenge is to develop a comprehensive instrument that measures attitude as well as decision making and quality of care. Quality of the services in primary care covers a wide range of dimensions: the prescribing behavior, the diagnosis and treatment, the managing of chronic disease, mental health care, maternal and child care, health promotion and preventive care [27]. Furthermore, it is also important to take into account the perception of the patients in terms of satisfaction with the delivered care, perception of quality and satisfaction with the outcomes of the delivered care. 162 One of the upcoming challenges in providing equitable health care is inequity by disease. There is an increasing fragmentation of healthcare quality programs, which often focus on specific groups of patients or diseases. Guidelines may unintentionally exacerbate this vertical ‘disease management’ approach, leading to fragmentation of care and a reductionist approach (treat-the-target instead of treat-the-patients). In that way a new inequity can be created: if the patients does not suffer from a ‘targeted’ disease, they are effectively denied care. Therefore, a paradigm-shift from “problem-oriented” towards “goal-oriented” care is needed, reorienting the care towards the goals formulated by the patient [28-30]. When studying the effect of the primary health care system on the health care use of specific population groups, it is important to use the correct reference population. For example, in this thesis the study on the health care use of homeless people is restricted to Ghent. In order to evaluate the effects of a universal health care system with active guidance, we need to compare the homeless populations living in areas with a universal health care system to other homeless populations living in areas with a different, more traditional, health care approach for homeless populations. Furthermore, the results of this paper only apply to homeless people residing in Ghent who make use of the homeless centers and shelters in Ghent. This study does not contain information on the homeless population living on the street who do not find their way to the shelters. Including this population could lead do different results according to their (primary) health care use. Finally, this thesis aims to study the equity in primary health care use in Belgium in terms of equal access, equal treatment and equal outcomes for patients in equal need. This work should be considered as a starting point for a more profound study Chapter 4: Discussion on equity in the health system in Belgium. A future step could be to evaluate the equity in primary health care by studying the Belgian primary health care system in an international context. Some studies already focus on establishing a sustainable health information and knowledge system on the state and development of primary care systems in Europe. There is the PHAMEU (Primary Health Care Activity Monitor for Europe) that monitors the degree of development of primary care systems by means of a measurement instrument that is applicable to all national situations in Europe and that is able to capture the essential elements of primary care. There is also QUALICOPC (Quality and Costs of Primary Care study) which focuses on equity, avoidable hospitalizations, healthcare costs and identification of good practices. Nevertheless, in order to counteract the large amount of studies coming from the US, more European studies on the effect of primary care on tackling health inequities are needed. 4.4. Policy and Practice implications In order to achieve equity in health care utilization, action is needed on different levels. Policy level Policy plays a crucial role in optimizing the (access to) health care. First of all, the Belgian health care system appears to have a relatively low financial burden in primary care for patients of lower social class thanks to the several measures taken. However, the adaption of some financial measures such as third payer system, preferential reimbursement, maximum bill ... depends on the one hand on the knowledge of the patient but on the other hand also on the will and knowledge of the GP. Additionally, some of these financial measures bring a lot of paperwork for the patient and the GP. In order to tackle the inequity in primary health care, it is important to strive for a primary health care system without any financial barriers. This is possible by means of two parallel systems. On the one hand by adapting the capitation system, the financial burden is nearly non existing. In the capitation system, the patient has to be registered with a GP or a group of primary care providers consisting of e.g. GPs, nurses, physiotherapists,... and does not have to pay for the consultation. On the other hand, for practices not working with the capitation system but on a fee-for-service basis, adapting the third-party payment should be generalized and there needs to be an automatic attribution of the financial 163 Chapter 4: Discussion measures such as the OMNIO and the third-party system for vulnerable people. This automatic attribution could help lowering the feelings of stigmatization for the patients and help reaching the vulnerable population groups. 164 GPs working in deprived areas and with a vulnerable population should get additional support financial as well as practical. In Belgium there are the Impulseo funds. Impulseo I was created to grant interest-free loans (up to 15.000 €) and subsidies (of 20.000 €) to doctors starting a GP practice in “deprived urban areas” or areas with a shortage of GPs, or in areas with a population density of less than 125 per km² and with less than 120 GPs per 100.000 population. Also an additional loan was provided to self-employed GPs and free administrative assistance was offered during the first 18 months following the start of the practice. Already 5 % of active GPs has used this procedure [18]. As the Impulseo funds focus on area characteristics, but not on individual practices, GPs working with a lot of low social class patients in a non-Impulseo area are left out of the loop. A possibility of tackling this problem is the principle of weighted financial mechanisms, where GPs working with a deprived population get a higher share. This system is already successfully applied in the Netherlands (achterstandsindex) and in the UK ( Jarman underprivileged area-score, Townsend poverty-index) [31]. It is important to have a universal primary health care system, that is easily accessible at any time for every person. In this universal approach, primary care could function as the gate keeping system towards secondary care for new health problems in order to counter the inappropriate use of specialist services and to avoid cost of high technological tests. Furthermore, no health care delivery system, even if it scores high on accessibility and quality, can tackle the diseases related to poverty on his own [32]. Especially when there are clear social inequities by disease and health care use, intersectoral action on national and on community level, is essential to redress these inequities [32-34]. This includes working together with e.g. nurses, dietician, physiotherapists, but also agriculture, education, housing,… Community Health Centres already provide a multidisciplinary team, including a nurse, a dentist, physiotherapist,…. But this is also important and possible for GPs not working in Community Health Centres. Solo-, duo- or group practices can work locally together with nurses, dietician, social workers etc. Collaborating with specific programs and investing in team building across several sectors can enable the primary care team in a community to Chapter 4: Discussion play an important role in the multi-level intersectoral initiative [30, 35]. Providing additional resources to establish such a cooperation could especially be helpful for the smaller GP practices such as the solo- and duo-practices. This coordinating role is a key feature of good quality primary care ensuring that patients see the most appropriate health care professional for their particular problem. General practice can fill this pivotal role if the structural conditions allow it. Developing a team around the patients with all kinds of health professionals ( nurses, dentist, physiotherapist, dieticians,…) will benefit the quality of care. By managing the interface with other specialties the discipline ensures that those requiring high technology services based on secondary care can access them appropriately [36]. Furthermore there is a need to harmonize the healthcare and welfare sectors in first line in Belgium (horizontal integration) in order to provide a better interaction between health care and welfare providers. An integration of healthcare and welfare can lead to a more targeted approach towards the patient, keeping in mind his/her financial and personal situation. However, to establish such a cooperation, action is needed on several levels: education of health care and welfare providers, creating a uniform registration and classification system, … To fulfill this achievement should be a medium-term priority. Practice level The findings that the doctor’s communicative behavior, treatment and decision making is related to the social class of the patients has important implications for the daily practice of the doctor. Doctors need to be aware of the differences in attitude, information giving and involving patients from lower social classes in the consultation, as well as of the upstream causes of these differences [37]. But above all, doctors cannot lose sight of the perception and preferences of the patient. It is important to empower a patient and involve them into the consultation to a degree they wish. Therefore schools and universities training doctors, should provide communication skills and attitudes training on low SES patients, so that doctors can learn how to handle this difficult balance. This stresses the importance of medical schools to become socially accountable. Medical schools cope with a different set of challenges: improving quality, equity, relevance and effectiveness in health care delivery, reducing the mismatch with societal priorities, redefining roles of health professionals and providing evidence of impact on people’s health status. To address these challenges, a Global Consensus 165 Chapter 4: Discussion for Social Accountability of Medical Schools was developed. The Consensus consists of ten strategic directions for medical schools to become socially accountable, highlighting required improvements to (1) respond to current and future health needs and challenges in society, (2) reorient their education, research and service priorities accordingly, (3) strengthen governance and partnerships with other stakeholders, (4) use evaluation and accreditation to assess performance and impact [38]. Physicians caring for low SES people seem to have difficulties finding the balance between providing good quality of standard care and tailoring the care to the capacities and needs of low SES patients. Additional support should be given to GPs caring for low SES people. Medical training focuses on adapting communication behavior (such as explaining things in a non-medical way) to patients with low SES. However it does not focus on the challenges and the personal and financial constraints physicians perceive when working with patients from low SES. Additional training, supervision, and (psychological) coaching could support physicians working with low SES patients. 166 This stresses the importance of translational research, where the results of a research or a study are adapted in practice by local and global communities. In order to improve health and reduce health-care disparities locally and globally, there needs to be a continuum that spans discovery and translational science to provide integrated care delivery and improved global health. New organizational structures, external partnerships, research priorities, models of education, care delivery and investments in institutional information systems can make the achievement of this goal possible [39]. A new generation of educational reform is now needed. It should be a system based on improving the performance of health systems by adapting core professional competencies to specific contexts, while drawing on global knowledge [40]. A key role for the discipline of primary care is to provide advocacy, protecting patients from the harm which may ensue through unnecessary screening, testing and treatment, and also guiding them through the complexities of the health care system [41]. The GP continues to serve as the advocate for the patient regardless of the level of care within the system which the patient requires. Advocacy by the general practitioner also includes working with government and private authorities to maximize equitable services to all members of society [42]. Having in mind the goals of Flanders 2020 on establishing more group practices and Chapter 4: Discussion the importance of providing longitudinal and a regular source of care, it is important for group practices that they could work out a system that helps patients always to see their ‘personal GP’ if possible. Some Community Health Centers in Ghent already apply such a system. When a patient wants to see a doctor, the receptionist can see which GP is their regular GP and make an appointment with that GP. This system is quite efficient, because patients often do not know the consultation hours of their GP, or they do not dare to ask it, or do not remember the name of their ‘personal doctor’. Finally, it will be of utmost importance that every provider integrates continuously a professional ethical approach, witnessing that health is a fundamental human right that should be put in practice in a socially just and universal health system. Patient level Patients should be informed, taking into account the possible cultural barriers, how the health care system is organized and how they can make use of the system in the most adequate way. Patients need to know where to go for a particular health care problem, but they also need to know where they can find this kind of information. Furthermore it is important to empower patients through education and enhance their self-efficacy. This includes empowering people to develop their own solutions, helping them to disseminate the lessons they have learned and joining forces with them in showing that the competences and skills necessary to bring about change are within reach [43]. It has been showed that interventions to increase the participation of patients with low education, obtain a good response and lead to measurable and clinically important improvements in health outcomes [44]. However, we should be careful not to “blame the victims”. On the one hand, capitalizing the relationship of trust is very important. Yet very often GPs have difficulties to contextualize the condition of the patients. So integrating health and social welfare is important in order to cope with this challenge. Furthermore, the context (area deprivation, pollution, job,…) should be integrated in the electronic patient file so that the physician and primary health care workers are able to relate possible conditions to the context patients live in. 167 Chapter 4: Discussion References 168 1. Van der Heyden JH, Demarest S, Tafforeau J, Van Oyen H: Socio-economic differences in the utilisation of health services in Belgium. Health Policy 2003, 65(2):153-165. 2. Bongers IM, van der Meer JB, van den Bos J, Mackenbach JP: Socioeconomic differences in general practitioner and outpatient specialist care in The Netherlands: a matter of health insurance? Soc Sci Med 1997, 44(8):1161-1168. 3. van Doorslaer E, Masseria C, Koolman X: Inequalities in access to medical care by income in developed countries. CMAJ 2006, 174(2):177-183. 4. Droomers M, Westert GP: Do lower socioeconomic groups use more health services, because they suffer from more illnesses? Eur J Public Health 2004, 14(3):311-313. 5. Hwang SW: Homelessness and health. CMAJ 2001, 164(2):229-233. 6. Morris DM, Gordon JA: The role of the emergency department in the care of homeless and disadvantaged populations. Emerg Med Clin North Am 2006, 24(4):839-848. 7. Kushel MB, Vittinghoff E, Haas JS: Factors associated with the health care utilization of homeless persons. JAMA 2001, 285(2):200-206. 8. Power R, French R, Connelly J, George S, Hawes D, Hinton T, Klee H, Robinson D, Senior J, Timms P et al: Health, health promotion, and homelessness. BMJ 1999, 318(7183):590-592. 9. Drieskens S, Van der Heyden J, Hesse E, Gisle L, Demarest S, Tafforeau J: Gezondheidsenquête België, 2008. Rapport V - Sociaal-economische ongelijkheden in gezondheid. In.; 2010. 10. Gelberg L, Gallagher TC, Andersen RM, Koegel P: Competing priorities as a barrier to medical care among homeless adults in Los Angeles. Am J Public Health 1997, 87(2):217-220. 11. Whitehead M, Hanratty B: Health care for rich and poor alike. In: Accessing health care Responding to diversity. edn. Edited by Healy J, McKee M. Oxford: Oxford University Press; 2004: 71-88. 12. Kushel MB, Perry S, Bangsberg D, Clark R, Moss AR: Emergency department use among the homeless and marginally housed: results from a community-based study. Am J Public Health 2002, 92(5):778-784. 13. Bensing J: Doctor-patient communication and the quality of care. Soc Sci Chapter 4: Discussion Med 1991, 32(11):1301-1310. 14. Jensen JD, King AJ, Guntzviller LM, Davis LA: Patient-provider communication and low-income adults: age, race, literacy, and optimism predict communication satisfaction. Patient Educ Couns 2010, 79(1):30-35. 15. Goddard M, Smith P: Equity of access to health care services: theory and evidence from the UK. Soc Sci Med 2001, 53(9):1149-1162. 16. van Ryn M, Burke J: The effect of patient race and socio-economic status on physicians’ perceptions of patients. Soc Sci Med 2000, 50(6):813-828. 17. Whitehead M, Dahlgren G: Levelling up (part 1): a discussion paper on concepts and principles for tackling social inequities in health. In.: WHO collaborating Centre for Policy Research on Social Determinants of Health, University of Liverpool; 2006. 18. Gerkens S, Farfan M, Desomer A, Stordeur S, De Waroux M, Van de Voorde C, Van de Sande S, Leonard C: Het Belgische Gezondheidssysteem in 2010. In. Edited by research HS. Brussel: Federaal Kenniscentrum voor de Gezondheidszorg (KCE); 2010. 19. Willems S: The socio-economic gradient in health: a never-ending story? A descriptive and explorative study in Belgium. Ghent: Ghent University; 2005. 20. Schafer W, Groenewegen PP, Hansen J, Black N: Priorities for health services research in primary care. Quality in primary care 2011, 19(2):77-83. 21. Parker RM, Gazmararian JA: Health literacy: essential for health communication. Journal of health communication 2003, 8 Suppl 1:116-118. 22. Beck RS, Daughtridge R, Sloane PD: Physician-patient communication in the primary care office: a systematic review. The Journal of the American Board of Family Practice / American Board of Family Practice 2002, 15(1):25-38. 23. Art B, De Roo L, Willems S, De Maeseneer AJ: An Interdisciplinary Community Diagnosis Experience in an Undergraduate Medical Curriculum: Development at Ghent University. Academic Medicine 2008, 83(7):675683. 24. Willems S, De Maesschalck S, Deveugele M, Derese A, De Maeseneer J: Socio-economic status of the patient and doctor-patient communication: does it make a difference? Patient Educ Couns 2005, 56(2):139-146. 25. Grol R, Dautzenberg M, Brinkman H: Quality Management in Primary Care. European Practice Assessment. In. Gütersloh: Verlag Bertelsmann Stiftung; 2004. 169 Chapter 4: Discussion 26. Hussey P, Anderson G, Berthelot JM, Feek C, Kelley E, Osborn R, Raleigh V, Epstein A: Trends in socioeconomic disparities in health care quality in four countries. Int J Qual Health Care 2008, 20(1):53-61. 27. Kringos DS, Boerma WG, Hutchinson A, van der Zee J, Groenewegen PP: The breadth of primary care: a systematic literature review of its core dimensions. BMC Health Serv Res 2010, 10:65. 28. De Maeseneer J, Boeckxstaens P: Care for noncommunicable diseases (NCDs): Time for a paradigm-shift. World Hospital and Health Services 2011, 47(4):30-33. 29. De Maeseneer J, Roberts G, R., Demarzo M, Heath I, Sewankambo N, Kidd R, M., van Weel C, Egilman D, Boelen C, Willems S: Tackling NCDs: a different approach is needed. The Lancet 2011. 30. Willems S, Christiaens T, Narayan T, B. M: Reducing Social Inequities in Health - The role of primary health care. 31. Art B, De Maeseneer J: Forfaitaire betaling: een beter systeem voor patiënt en huisarts. In. 170 32. Kotwani N, Danis M: Tackling the health-poverty nexus: primary care medicine and intersectoral health action. J Gen Intern Med 2007, 22(11):1632-1633. 33. Starfield B: Pathways of influence on equity in health. Soc Sci Med 2007, 64(7):1355-1362. 34. De Maeseneer J, Willems S, De Sutter A, Van de Geuchte M, Billings M: Primary health care as a strategy for achieving equitable care: a literature review commissioned by the Health Systems Knowledge Network. In.; 2007. 35. WHO report 2008: Primary health care now more than ever. In.; 2008. 36. Physicians funding and health care systems - an international perspective. In: A summary of a conference hosted by the WHO, WONCA and RCPG at St John’s College, Cambridge. Edited by Brown PS. 37. Epstein AM, Taylor WC, Seage GR, 3rd: Effects of patients’ socioeconomic status and physicians’ training and practice on patient-doctor communication. The American journal of medicine 1985, 78(1):101-106. 38. Global Consensus for Social Accountability of Medical Schools [www. healthsocialaccountability.org] 39. Dzau J, V., Ackerly D, C., Sutton-Wallace P, Merson H, M., Williams R, S., Krishnan K, R., Taber C, R., Califf M, R.: The role of academic health science systems in the transformation of medicine. The lancet 2009. Chapter 4: Discussion 40. Frenk J, Chen L, Bhutta ZA, Cohen J, Crisp N, Evans T, Fineberg H, Garcia P, Ke Y, Kelley P et al: Health professionals for a new century: transforming education to strengthen health systems in an interdependent world. Lancet 2010, 376(9756):1923-1958. 41. Allen J, Gay B, Crebolder H, Heyrman J, Svab I, Ram P: The European Defenition of Genera Practice/Family Medicine. In. Edited by Europe W; 2005. 42. Bentzen G, B., Bridges-Webb C, Carmichael L, Ceitlin J, Feinbloom R, Metcalf D, MCwhinney I, Rajakumar K: The role of the general practitione/family physician in health care systems: a statement from WONCA. 1991. 43. Ziglio E, Barbosa R, Charpak Y, Turner S: Health Systems confront poverty. Public Health Case Studies, no. 1. In. Edited by Organization WH. 44. Kaplan SH, Greenfield S, Ware JE, Jr.: Assessing the effects of physicianpatient interactions on the outcomes of chronic disease. Med Care 1989, 27(3 Suppl):S110-127. 171 172 Chapter 5 Dankwoord 173 174 Chapter 5: Dankwoord Dankwoord “Words do major things, they provide food for the mind and create light for understanding and awareness” Er is geen moeilijker taak dan goed te bedanken. De voorbije 4 jaar zijn een aaneenschakeling geweest van verschillende processen, uitdagingen, gebeurtenissen, waarin telkens iemand je heeft bijgestaan, ondersteund, een duw in de juiste richting heeft gegeven,… Dit maakt het onmogelijk om iedereen persoonlijk te bedanken. Daarom wil ik op voorhand zeggen in het geval ik iemand vergeet: “jullie zijn allemaal ongelooflijk bedankt!”. Enkele personen hebben de evolutie van “pas afgestudeerd” tot “het indienen van een proefschrift” van dichtbij meegemaakt. Prof. Dr. Willems, Sara, mijn eerste woord van dank gaat uit naar jou. Ik herinner mij nog mijn sollicitatiegesprek. Een mooie, zonnige, zomerse dag. Ik was super zenuwachtig! De week ervoor waren we op reis geweest naar Frankrijk. Daar had ik de vacature waarschijnlijk 20 keer van voor naar achter gelezen, mijn Frans geoefend en uitgezocht wat die DWA in godsnaam was. Beginnen als doctoraatsstudent was dan ook een stap in het onbekende. De personen die ik kende omschreven het als vier jaren van ware stress en uitputting. Nu ik terugkijk op deze 4 jaren, dien ik hen ongelijk te geven. Het waren vier leerrijke jaren met, zoals het hoort, de nodige uitdagingen. Je hebt me gedurende deze jaren geïntroduceerd in de wereld van wetenschappelijk onderzoek. Bij vragen of moeilijkheden kon ik altijd bij je terecht. Je stimuleerde het zelfstandig werken en gaf mij ook het vertrouwen dit te doen. Sara, ik wil je bedanken voor je geduld, je nuchterheid in moeilijke omstandigheden, je opbouwende feedback, de steun, en nog zoveel meer, maar vooral wil ik je bedanken voor de persoon wie je bent! Jan, als co-promotor van mijn proefschrift bracht je altijd een nieuwe, frisse wind. Wanneer ik niet goed wist hoe ik verder moest of moeite had met het interpreteren van resultaten, dan bracht een discussie met jouw steeds licht in de duisternis. Ik heb enorm veel van je geleerd tijdens de treinreizen naar Brussel, zowel op menselijk als op wetenschappelijk vlak. Bedankt! Daarnaast wil ik ook mijn bureaugenoten bedanken. Sarah, Maaike en Sofie, jullie hebben het begin meegemaakt en hebben mij vaak geholpen in het organiseren van 175 Chapter 5: Dankwoord het veldwerk en het zoeken naar creatieve oplossingen. Veerle, Annelien, Evelyne en Wim, bedankt om het met mij uit te houden in de laatste maanden. Jullie raad en aanmoediging heeft mij super geholpen! Annelien en Evelyne: bedankt om alles nogmaals te lezen. Ook alle collega’s van de vakgroep Huisartsgeneeskunde en Eerstelijnsgezondheidszorg wil ik bedanken. Claudine voor haar kennis Frans, Michèle voor de vele praktische zaken, Lieselot voor haar organisatietalent, Karine en Emilienne voor het helpen met de “cijfers”. Aan de “lotgenoten”: het valt echt mee! Anja, bedankt om de lay-out van dit boekje en de uitnodigingen te verzorgen. Je nauwkeurigheid en creativiteit heeft zeker zijn vruchten afgeworpen! 176 Daarnaast wil ik ook de co-auteurs bedanken voor de vruchtbare samenwerking: Eleonora Bonte en Nele De Laender voor jullie ongelooflijke inzet, Tine Verdee, Mieke Van de Walle, Bruno Art voor de dataverzameling en het concept van het ‘daklozen-artikel’, Inge Pasteels voor haar statistische kennis en input, Prof. Chantal Van Audenhove, Prof. Koen Hermans, Ann DeSmet en Annelien Poppe als medeauteurs en collega’s bij het KANS onderzoek. Dr. Stéphanie De Maesschalck en Prof. Myriam Deveugele voor hun communicatieve inbreng. Een speciaal woord van dank gaat uit naar alle geïnterviewde artsen, patiënten, de spoeddiensten van de ziekenhuizen, de OCMW’s, CAW’s en de Diensten Maatschappelijk werk van het Ziekenfonds. Daarnaast wil ik ook alle participerende Hogescholen en studenten bedankten voor hun bijdrage aan de dataverzameling. Dankzij jullie is dit proefschrift een feit. Dit proefschrift zou eveneens niet mogelijk geweest zijn zonder de steun van het Steunpunt Welzijn, Volksgezondheid en Gezin. Een bijzonder woord van dank gaat uit naar mijn ouders. Ze hebben me steeds alle kansen gegeven en hebben me gesteund in de keuzes die ik heb gemaakt. Wanneer ik dringend wat hulp nodig had, stonden ze altijd voor me klaar. Daarnaast wil ik ook Wouter bedanken. Je hebt deze vier jaar beleefd aan mijn zijde. Het feit dat je eigenlijk niet zo goed snapte waar ik mee bezig was heeft ervoor gezorgd dat ik mijn hoofd telkens kon leegmaken toen ik thuis kwam. Die ontspanning en afleiding deden me goed. “Alone we can do so little, together we can do so much” Gent, 03 oktober 2012 Chapter 6 Curriculum Vitae 177 178 Chapter 6: Curriculum vitae CURRICULUM VITAE Evelyn Verlinde Geboren te Roeselare op 8 december 1986 Opleiding - 2004-2007 Bachelor Geneeskunde - 2007-2008 Master in de gezondheidsvoorlichting en -bevordering - 2008-2012 Doctoraatsopleiding Medische Wetenschappen Getuigschriften - 19/01/2009 Introductie voor tutoren in probleemgestuurd onderwijs Docent: Prof. Dr. Anselm Derese, Voorzitter van het Centrum voor Onderwijsontwikkeling Organisatie: Universiteit Gent - 8/07/2009 -10/07/2009 Summer-course: three days of qualitative research in health care Docent: Prof. dr. Paul van Royen, Department of General Practice, University of Antwerp, Belgium Organisatie: Universiteit Antwerpen - 08/2009 Opleiding soft-warepakket Nvivo Docent: Prof. dr. Paul van Royen, Department of General Practice, University of Antwerp, Belgium Organisatie: Universiteit Antwerpen - 03/2010-05/2010 Mixed and Multilevel models Master of Statistics Organisatie: KULeuven 179 Chapter 6: Curriculum vitae - 01/2011-02/2011 Statistische analyse met behulp van SPSS (2011)- gevorderden, Docent: Prof. G. Van Maele Organisatie: Cel biostatistiek Universiteit Gent - 10/2010-12/2010 Effective Scientific communication Docent. JL Dumont, Principae Organisatie: Doctoral Schools - 02/2011-04/2011 Presentation Techniques Docent. JL Dumont, Principae Organisatie: Doctoral Schools 180 - 05/2011 Multilevel Analysis for Grouped and Longitudinal Data Docent: Prof. Dr. E.Goetghebeur Organisatie: Centrum voor Statistiek - 04/2011 Populair wetenschappelijk schrijven. Docent: Ann de Ron Organisatie: Doctoral Schools Professionele loopbaan - 01/09/2008-30/09/2010 Doctoraatsbeurs KANS onderzoek i.o.v. Steunpunt Welzijn Volksgezondheid en Gezin Vakgroep Huisartsgeneeskunde en Eerstelijnsgezondheidszorg, Universiteit Gent - 01/10/2010- heden Assistent Vakgroep Huisartsgeneeskunde en Eerstelijnsgezondheidszorg Chapter 6: Curriculum vitae PUBLICATIES A. Artikels A1. Artikels opgenomen in Science Citation Index, Social Science Citation Index, Arts and Humanities Citation Index 1. Verlinde E, Verdee T, Van de Walle M, Art B, De Maeseneer J, Willems S. Unique health care utilization patterns in a homeless population in Ghent. BMC Health Services Research. 2010; 10(242). IF:1.66 2. Verlinde E., De Laender N, De Maesschalk S, Deveugele M., Willems S. The social gradient in doctor-patient communication. Journal for equity in health.2012,11:12. IF:1.30 3. Verlinde E., Pasteels I., De Maeseneer J., Willems S. Socio-economic differences in the utilization of FP and specialist care: do patients form lower social class visit more often a doctor when taking into account their higher health care need? Annals of Family Medicine. Submitted. IF: 4.45 4. Verlinde E., Poppe A., DeSmet A., Hermans K., De Maeseneer J., Van Audenhove C., Willems S. Social differences in postponing a GP visit: Which low-income patients are most at risk? Health and Social Care in the Community. Submitted. IF: 1.008. 5. Verlinde E., Bonte, E. Willems, S. Socio-economic status of the patient: does it influence the decisions of the GP? International Journal for Quality in Health Care. Submitted. IF: 2.064 6. DeSmet A, Hermans K, Verlinde E, Willems S, De Maeseneer J, Van Audenhove C. Integrated care in mental health: need for collaboration, communication and customisation. Psychiatr Prax. 2011;38. A3. Artikels in nationale tijdschriften met leescomité, niet begrepen in (a1) of (a2), chronologisch geordend 1. B. 1. Willems S, Verlinde E. Sociale ongelijkheid in gezondheid: een maatschappelijk onaanvaardbaar fenomeen. Terzake. November 2011;6; 5-12. Andere Verlinde E. Toegankelijkheid, betaalbaarheid en kwaliteit van de gezondheidszorg. In: De Sociale Staat van Vlaanderen. 2011. Noppe J., Vanderleyden L., Callens M. 203-206 181 Chapter 6: Curriculum vitae C. Rapporten bestemd voor de verspreiding van wetenschappelijk werk naar buiten 182 - DeSmet, A., Hermans, K., Verlinde, E., Willems, S., De Maeseneer, J. & Van Audenhove, C. (Onderzoeksgroep KANS). Depressieve klachten en suïcidaliteit in de (I)CAW en OCMW: onderzoek naar de ernst en de relatie tot armoede. Steunpunt Welzijn, Volksgezondheid en Gezin, 2010. - DeSmet, A., Hermans, K., Verlinde, E., Willems, S., De Maeseneer, J. & Van Audenhove, C. (Onderzoeksgroep KANS). Het hulpaanbod voor mensen met depressieve klachten in (I)CAW en OCMW: beantwoordt het aanbod aan de vraag? Steunpunt Welzijn, Volksgezondheid en Gezin, 2010. - Poppe, A., Verlinde, E., Willems, S., DeSmet, A., Hermans, K., Van Audenhove, C. & De Maeseneer, J. (Onderzoeksgroep KANS). Beschrijving van de KANS steekproef: representativiteit, demografische en socio-economische kenmerken, gezondheidstoestand en zorggebruik. Steunpunt Welzijn, Volksgezondheid en Gezin, 2011. - Willems, S., DeSmet, A., Poppe, A., Hermans, K., Verlinde, E., Van Audenhove C. & De Maeseneer, J. KANS, theoretische achtergronden en onderzoeksopzet. Steunpunt Welzijn, Volksgezondheid en Gezin, 2011. Wetenschappelijke activiteiten 1. Lezingen op nationale congressen, symposia en colloquia “De kwaliteit van de zorg vanuit het perspectief van de gebruiker (KANS)” (invited session lecture) Studiedag “In-Zicht”, Leuven Steunpunt Welzijn, Volksgezondheid en Gezin, 02/12/2010 “Ongelijkheid in gezondheid: feiten, cijfers en aanbevelingen” Ontmoetingsdag ‘Ongelijk gezond, een lokale aanpak.’ Provinciehuis Leuven, 14/06/2010 “Gebruik van huisartsenzorg bij financieel kwetsbare zorggebruikers (KANS)” Studiedag “Wel-en Wee” Leuven Steunpunt Welzijn, Volksgezondheid en Gezin, 28/10/2011 Chapter 6: Curriculum vitae Posterpresentatie op internationale congressen - Verlinde, E., Willems, S., Hermans, K., Van Audenhove, C., & De Maeseneer, J. (2010). The health care and welfare use of people living in poverty in Flanders: the contribution of the neighborhood level. Poster presentation at the ESHMS congress. Ghent 2010. - Verlinde, E., Willems, S., Hermans, K., Van Audenhove, C., & De Maeseneer, J. (2010). Tales of the (un)expected: people living in financial difficulties and their challenges, Poster presentation Regions for Health Network Conference. Genk, 8-9 november 2010. - DeSmet, A., Hermans, K., Verlinde, E., Willems, S., De Maeseneer, J., & Van Audenhove, C. (2010). The potential role of social services in providing integrated care on mental health and welfare. Paper presentation Regions for Health Network Conference. Genk, 8-9 November 2010. Posterpresentatie op nationale congressen - Verlinde, E., Poppe, A., Willems, S. en De Maeseneer, J. Gebruik van huisartsenzorg bij financieel kwetsbare welzijnszorggebruikers. Steunpunt WVG studiedag “Wel en Wee in Vlaanderen”. Leuven, Oktober 2011. - Verlinde, E., Poppe, A., Willems, S. en De Maeseneer, J. Sociaal kapitaal en gezondheid. Steunpunt WVG studiedag “Wel en Wee in Vlaanderen”. Leuven, Oktober 2011. - DeSmet, A., Hermans, K., Van Audenhove, Ch. Steunbronnen bij psychisch onwelbevinden: veerkracht, sociale steun en ‘coping’. Steunpunt WVG studiedag “Wel en Wee in Vlaanderen”. Leuven, Oktober 2011. - Poppe, A., Verlinde, E., DeSmet, A., Willems, S., Hermans, K., Van Audenhove, C., & De Maeseneer, J. De psychosociale en financiële impact van chronische aandoeningen bij financieel kwetsbare personen. Posterpresentatie tijdens de Steunpunt WVG studiedag “Wel en Wee in Vlaanderen”. Leuven, Oktober 2011. - Poppe, A., Verlinde, E., DeSmet, A., Willems, S., Hermans, K., Van Audenhove, C., & De Maeseneer, J. Sociaal kapitaal als buffer voor de negatieve gezondheidseffecten bij personen in een financieel kwetsbare situatie. Posterpresentatie tijdens de Steunpunt WVG studiedag “Wel en Wee in Vlaanderen”. Leuven, Oktober 2011. 183