Download Antenatal syphilis in sub-Saharan Africa: missed

HEALTH POLICY AND PLANNING; 16(1): 29–34
© Oxford University Press 2001
Antenatal syphilis in sub-Saharan Africa: missed opportunities
for mortality reduction
STEPHEN GLOYD,1 SANDERS CHAI2 AND MARY ANNE MERCER1
1Department of Health Services and 2Department of Environmental Health, University of Washington School of
Public Health and Community Medicine, Seattle, Washington, USA
Purpose: Between 4–15% of pregnant women are believed to be infected with syphilis in sub-Saharan Africa.
Active infection with syphilis in pregnant women results in foetal or infant death or disability for 50–80% of
affected pregnancies, and is a major cause of adult morbidity as well. Antenatal syphilis screening is cheap
and effective; however, it is often poorly implemented in countries with high syphilis risk. This study sought
to estimate the missed opportunities for antenatal syphilis screening in sub-Saharan Africa.
Methods: Survey data were collected from 22 ministries of health in sub-Saharan Africa, complemented by
data from published sources and key informants. Informants described their country’s policies and experience with antenatal syphilis screening and estimated their national syphilis screening rates.
Findings: Seventy-three percent of women are reported by WHO to receive antenatal care in the study countries. Of women in antenatal care, 38% were estimated by survey respondents to be screened for syphilis.
Costs and the organization of services were the principal reported obstacles to screening. With syphilis seroprevalence estimated at 8.3%, approximately 1 640 000 pregnant women with syphilis are undetected annually, including 1 030 000 women who attend antenatal care.
Discussion: Syphilis testing and treatment is a cost-effective intervention that deserves much greater attention, particularly in sub-Saharan Africa and other countries where syphilis infection is high.
Introduction
Active infection with syphilis in pregnant women has long
been recognized as a major cause of death or disability in the
infants born to infected women. It is estimated that active
syphilis infection in pregnancy causes adverse outcomes in
50–80% of pregnancies surviving past 12 weeks gestation, primarily as spontaneous abortions in the second and early third
trimester, stillbirths, and congenital syphilis.1 Syphilis infection in pregnancy is highly prevalent in many areas of the
world. Among women attending antenatal clinics in Africa,
estimates of syphilis sero-reactivity range from 4–15%.1 Data
from Zambia and Malawi suggest that between 26–42% of
stillbirths and 8% of infant deaths in those countries may be
attributable to syphilis alone.1,2 Syphilis is also a significant
cause of morbidity in adults, and has been identified as a principal cofactor facilitating the transmission of HIV.3,4
The technology to address maternal syphilis has been widely
available for decades. Syphilis screening and treatment in
antenatal care has been cited as one of the most cost-effective ways to reduce foetal and infant mortality and morbidity in the developing world, comparing very favourably with
immunizations against childhood infectious diseases.5 The
treatment of syphilis-infected mothers represents an opportunity to also treat their partners, and thus interrupt further
transmission of syphilis as well as HIV infection. In Tanzania, a reduction in rates of sexually transmitted infections
(STIs) in a community was associated with a reduction in the
risk of transmitting HIV by 40%.6
Although syphilis is an important cause of morbidity and
mortality, and antenatal screening is relatively feasible and
effective, this intervention has not been promoted to the
extent of other, less efficacious interventions. Even where
antenatal syphilis screening is a national policy and where
antenatal care is widely utilized, syphilis screening appears to
be sporadically implemented at best. Few estimates exist as to
how well this problem is being addressed in areas of the world
where screening and treatment could have an enormous
impact.
This study was designed to determine the extent to which
health systems in sub-Saharan Africa have addressed the
issue of antenatal syphilis screening. The research objectives
were to (1) identify the national policies in sub-Saharan
Africa regarding antenatal syphilis screening, and (2) estimate the proportion of pregnant women already attending
antenatal care who are not being screened for syphilis.
Recognizing the paucity of precise data regarding both infection levels and programme efforts, the emphasis of the study
was to provide orders of magnitude estimates of the current
30
Stephen Gloyd et al.
state of antenatal syphilis screening for policy-makers, providing guidance for allocation of scarce health sector
resources. Researchers thus sought to estimate the magnitude
of the missed opportunities to reduce maternal, foetal and
infant mortality and morbidity due to syphilis among women
already receiving antenatal care. The study focused on the
countries of sub-Saharan Africa where syphilis prevalence is
known to be high and where the potential for reducing mortality and morbidity due to maternal syphilis infection is correspondingly great.
Methods
The data for this study were collected from a survey of
national ministries of health conducted by the authors, complemented by data from published sources and key informants of participating countries. Published sources include The
state of the world’s children,7 the Demographic and Health
Survey series,8 World Health Organization (WHO) publications9,10 and epidemiological studies of maternal syphilis
infection rates.11–13 Whenever possible, the authors corroborated information from more than one source. Estimates of
the effectiveness of syphilis screening on neonatal outcomes
were derived from Schulz et al.5
The survey instrument consisted of a one-page, 14-item questionnaire. Questions covered national syphilis screening
policy and targets, prenatal care rates, rates of antenatal
syphilis screening and reported results of that screening, location of screening, source of information for the survey, including whether they had methods of routinely assessing syphilis
screening, and reported obstacles to screening.
The survey was directed at the 110 countries in the world with
populations over 1 million and with the highest reported
under-5 mortality rates. This analysis focuses on a sub-sample
of 38 countries in mainland sub-Saharan Africa and Madagascar. Surveys were sent between January and July of 1996
to the 33 countries from this group for which a fax number or
postal address was available to the authors; follow-up continued through July of 1997. In five countries, personal visits
were made with Ministry of Health officials responsible for
maternal health services to obtain questionnaire information.
All responses were confidential.
Antenatal care coverage figures for each country were based
on DHS surveys and/or WHO estimates,10 using the most
recent national figure when two or more differing estimates
were available. The proportion of women screened for
syphilis was based on rates provided by the respondents from
each country. Syphilis seroprevalence was calculated based
on published literature and the survey estimates. Pregnancies
per year were calculated by multiplying the crude birth rate
times the country population by 1.15 in order to adjust for
pregnancy losses.10 ‘Missed opportunities’ were estimated as
the number of pregnant women with active syphilis already
attending antenatal care but who were not screened, and thus,
whose infections were not detected. These estimates were
based on several assumptions discussed below.
The literature used in estimating syphilis infection rates
among antenatal care attendees sometimes failed to specify
whether screening (RPR or VDRL) or confirmatory treponemal (FTA or TPHA) tests were used in the studies
described.i In such cases, we cross-referenced the cited figures
with other comparable studies in the country. If other study
data could not be located, we assumed that ‘screening’ meant
unconfirmed (RPR or VDRL only) seropositive rates and
projected that 80% of those screened positive by nontreponemal tests would have been confirmed cases of syphilis
by treponemal tests.14 In countries where multiple values for
antenatal syphilis seroprevalence were substantially different, data from the most recent and most representative
sample were used. Aggregate means of antenatal care attendance, syphilis screening, and syphilis prevalence levels were
weighted by country population for the calculations. ‘Nonrespondent countries’ were defined as sample countries not
responding to mailed or faxed questionnaires and countries
to which questionnaires were not sent.
Results
Of the 33 target countries in sub-Saharan Africa to which the
survey was sent, respondents from 17 countries returned the
questionnaire. We added questionnaires filled out during the
personal visits to five countries which had not returned questionnaires; thus, the total sample was 22 countries. Although
the questionnaire was directed to the maternal–child health
(MCH) division of each ministry of health, only three of the
17 returned questionnaires were signed by MCH division
staff, nine were signed by STD/AIDS programme staff, four
were signed by statistics or epidemiology staff, and one was
signed by someone whose division was unknown to the
authors.
Seventeen (77%) of the 22 respondents reported universal
syphilis screening to be a national norm of pregnancy care in
their country. VDRL was reported to be used as the principal
treponemal screening test by 10 (45%) of the responding
countries, RPR by seven (32%), and both VDRL and RPR
by five (23%) countries. Verification of positive screening
tests by TPHA or FTA was reported to be the norm in six
(27%) of the responding countries. No country reported
national targets or indicators for monitoring progress of antenatal syphilis screening. Routine contact tracing was reported
to be an integral part of antenatal syphilis follow-up in eight
(36%) of the respondent countries. No country reported
routine repeat screening for women near the end of pregnancy.
Nineteen respondents provided estimates of syphilis screening rates among women in antenatal care. The data were estimated by seven respondents, six used specific investigations
from sample populations (not necessarily representative),
and nine reported data from STI registries or routine statistics (some indicated more than one source). Eight of the
above respondents were able to estimate screening levels by
capital city, other urban, or rural areas.
Table 1 provides overall summary data on population size,
pregnancies, antenatal care and syphilis screening among the
22 respondent countries. The population of those 22 countries
Antenatal syphilis screening in Africa
Table 1.
31
Estimated antenatal syphilis screening in 22 reporting countries of sub-Saharan Africa (in millions)
Characteristic
n (millions)
(%)
Total population represented by sample (% of total population of sub-Saharan Africa)
Estimated pregnancies/year in reporting countries (% of total population in reporting countries)
Estimated pregnant women who received antenatal care (% of pregnancies/year)
Estimated pregnant women screened for syphilis (% of those receiving antenatal care)
397.5
19.9
14.5
5.5
(73)
(5)
(73)
(38)
Obstacle reported to be important
Number reporting
each obstacle
antenatal syphilis prevalence from 14 countries of subSaharan Africa, 11 of which were from the 22 sample countries.11–13 The weighted mean for syphilis seroprevalence
from these studies was 8.3%, the median was 8.75% and the
range was 0.9–20.8%.
Cost (to patient) of testing
Organization of services
Cost (to patient) of treatment
Transport costs to testing facility
Inadequate priority of MOH
Social/cultural resistance
Vacations, absence of health workers
Provider compliance/awareness
10
7
6
4
3
3
2
2
Table 3 provides estimates of syphilis screening and missed
opportunities for all of sub-Saharan Africa, based on the data
from the study sample and using the weighted syphilis prevalence estimate of 8.3%. Over 2 million pregnant women in
sub-Saharan Africa would be actively infected with syphilis
each year if it were possible to generalize this prevalence
through the region.
Table 2. Reported obstacles to universal antenatal syphilis screening in sub-Saharan Africa (n = 21)
represents approximately 73% of the total population of subSaharan Africa. The weighted mean of antenatal care coverage reported by WHO for respondent countries was 73%
compared to 50% for the non-respondent countries. The
weighted mean of the estimated proportions of antenatal
patients screened for syphilis in the 19 countries that provided
that information was 38%, with a range of 1–92%.
Table 2 summarizes survey data on reported obstacles to
screening for the 21 countries that responded to this question.
Cost of testing or treatment and the organization of services
were reported to be the principal obstacles.
The weighted mean reported syphilis screening positivity rate
among pregnant women in antenatal care estimated by the 14
countries responding to this question was 10.9%. Assuming
that 80% of those screened as positive would be women with
active syphilis infection, the weighted mean proportion of
antenatal active syphilis in responding countries was estimated to be 8.7%. The authors’ reviewed 23 recent studies on
We estimated the antenatal syphilis screening rate for all of
sub-Saharan Africa by applying the 38% reported screening
rate of responding countries to all countries times their antenatal care attendance rates. If survey data on antenatal screening are representative of the rest of sub-Saharan Africa,
approximately 630 000 pregnant women with syphilis are
currently detected and treated (assuming 100% accurate
detection and timely treatment). Another 1 640 000 pregnant
women with syphilis are undetected, including about 1 030 000
women who attend antenatal care.
Discussion
The survey data demonstrate that although antenatal syphilis
screening is a national policy in over three-quarters of respondent countries, most pregnant women who attend antenatal
care are not screened. These findings suggest that inadequate
syphilis screening and treatment in pregnancy represents an
enormous missed opportunity to reduce maternal, perinatal
and infant mortality and morbidity in sub-Saharan Africa. In
addition, inadequate syphilis screening probably represents
Table 3. Estimated annual missed opportunities for prevention of stillbirths, late abortion, and congenital syphilis for sub-Saharan Africa
(in millions)
Characteristic
Number
(millions)
Method
Total pregnant woman per year
Pregnant women with active syphilis infection
Pregnant women with active syphilis infection – detected
27.4
2.3
0.63
@5% of total population
@8.3% infection rate
@73% ANC coverage
@38% screening
Total women with syphilis minus those detected
@73% ANC coverage
@62% not screened
Pregnant women with active syphilis – undetected
Missed opportunities: pregnant women with active syphilis who are
already in antenatal care – undetected
1.64
1.03
32
Stephen Gloyd et al.
an important missed opportunity for reduction of HIV infection among adults.
The study data are imprecise and subject to error. The
authors believe that the antenatal syphilis screening rate from
our survey of 38% may be an overestimate of the real syphilis
screening rate in all of sub-Saharan Africa for the following
reasons.
• The screening rates for respondents to the survey are likely
to be higher than non-respondents because of response
bias. Respondent countries had much higher levels of antenatal care utilization (using WHO figures) than nonrespondent countries. Among respondents, countries with
higher antenatal care coverage had significantly higher
antenatal syphilis screening than those with lower antenatal coverage, suggesting that non-respondent countries
may also have lower antenatal syphilis screening rates.
• The rates of antenatal syphilis screening within responding
countries may be overestimated. Three respondents
appeared to have overestimated their syphilis screening
rates according to key informants, possibly because they
might have wanted to put a favourable face on their antenatal syphilis activities.
• Where screening is done, treatment may be given too late
in gestation to effectively prevent adverse perinatal outcomes. Studies in Mozambique suggest that even when
screening is operational, only 20% of those receiving prenatal care received treatment prior to 28 weeks of gestation
due to late first antenatal visits and delayed (or no) treatment.15
Although validation of the accuracy of the data reported in
the survey is difficult, it appeared that nearly all respondents
filled out the questionnaires with considerable care, and most
indicated where data were not available. All of the officials
from the five countries for whom personal visits were made
said that no routine data were kept in their countries regarding syphilis screening rates; thus, their estimates were guesses
based on their experience. The fact that most of the respondents were from epidemiology or STD sectors of their Ministries of Health casts some doubt on the accuracy of the
antenatal care data. However, it would have been difficult for
these officials to obtain information on antenatal syphilis
screening policy and prenatal care attendance without consulting the MCH officials in the ministries. Moreover, the
unweighted mean antenatal screening rate reported by study
respondents (79.5%) was nearly identical to that available
from WHO reports (78.5%).12 In addition, the weighted
mean reported syphilis seroprevalence from the sample countries (8.7%) was very similar to the weighted mean from the
literature (8.3%). These findings support the likelihood that
information gained from the survey is a reasonably accurate
picture of syphilis screening practices in Africa.
A series of assumptions were needed to generate estimates of
the missed opportunities, any of which could lead to errors.
We assume that having active syphilis in pregnancy is pathogenic, although only observational data have been presented
for this assumption.2 The precise level of risk to infants of
active syphilis infection is difficult to measure, since the risk
depends on the length of time that the mother has been
infected as well as the stage of infection and the gestation of
the pregnancy when treatment is provided. We base our
assumptions on 100% effectiveness of testing; however, false
negatives do occur, especially among early infections. A high
proportion of false negatives would reduce the number of
women among those screened who are detected. We did not
take into account false negative laboratory results caused by
early syphilis infection or poor quality testing; a high proportion of false negative results could reduce the potential to
prevent adverse outcomes. Nevertheless, our estimates are
consistent with what is currently known about gestational
syphilis, its risks, its diagnosis in common practice, and
efficacy of treatment.
Despite some uncertainties, it is clear that a substantial unmet
need for syphilis screening and treatment exists. According to
WHO/CDC estimates, 58% of pregnancies with active gestational syphilis will suffer adverse foetal outcomes (stillbirth,
third trimester abortion and congenital syphilis). By our
rough calculations, every year approximately 600 000 opportunities are missed to reduce adverse foetal and infant outcomes due to maternal syphilis infection in sub-Saharan
Africa. These estimates, however imprecise, suggest perinatal wastage on a substantial scale.
Although the benefits are substantial, the cost of antenatal
syphilis screening is low. The cost of the screening tests
(RPR) for each woman is approximately US$0.20.5 Treatment of RPR-positive women (and their partners) with benzathine penicillin is very effective and costs about US$1.00
per dose. The cost per death or adverse event prevented with
antenatal syphilis screening is comparable to the highest predicted mortality reduction from other interventions, including immunization programmes. A study in The Gambia
estimated the cost of immunization per childhood death prevented to be approximately US$41, US$100 and US$153 for
measles, pertussis and neonatal tetanus, respectively. By
comparison, in most of sub-Saharan Africa, US$12 would
avert an adverse outcome associated with syphilis during
pregnancy, taking into account the costs of screening and the
drug treatment.5
At a national level, the fact that most respondents were from
STD, statistics or epidemiogy units of the MOH rather than
maternal health units suggests that antenatal syphilis screening may be less a priority of the MCH section than of the STD
sector. Most antenatal care norms include a long list of interventions and risk assessment with little prioritization of these
activities. Few countries have clear strategies of how these
many interventions are to be carried out in low resource settings. Syphilis screening is usually not a high priority and, like
many other antenatal norms, is carried out sporadically. In
areas with limited resources, programmes which are not
viewed as important or which do not have explicit targets tend
to be placed at a low priority by peripheral health workers
who have many other competing demands.
In light of the potential importance of antenatal syphilis
screening, this intervention is clearly accorded inadequate
priority. International donor agencies infrequently make
Antenatal syphilis screening in Africa
explicit reference to syphilis screening in their lists of priority
interventions in antenatal care. Donor funding for antenatal
syphilis screening is minimal. The recent trend promoting
screening of STIs during pregnancy may refocus efforts on
syphilis screening. However, syphilis screening per se should
be emphasized in countries with high prevalence, since the
other common STIs usually represent less serious problems
and are much more difficult to screen and treat.
The authors suggest the following to re-prioritize antenatal
syphilis screening and treatment within current health programmes. The following measures need to be applied in any
country where syphilis prevalence is high.
• Establish antenatal syphilis screening as a clearly identified
priority. Syphilis screening should be identified as one of
the three or four principal interventions of antenatal care
with strategies developed to carry it out.
• Develop systems to assure that adequate stocks of syphilis
screening tests and treatment are available and distributed
to appropriate health facilities. This implies allocation of
adequate resources for supplies, logistics and training. If
syphilis screening were carried out in all pregnant women,
current stocks of VDRL/RPR tests and penicillin would be
inadequate in most countries.
• Promote early attendance at antenatal care and publicize
the existence of syphilis screening and treatment as an
incentive to come early.
• Reduce costs of testing and treatment for pregnant women.
• Strengthen methods to improve contact tracing and treatment for partners.
• Establish clear indicators and targets for adequacy of
screening and mechanisms for monitoring the achievement
of these indicators at the health facility level, using methods
proven successful in immunization programmes.
Antenatal syphilis is not just a problem of sub-Saharan
Africa. Recent reports suggest that syphilis prevalence may
be of a similar order of magnitude in some countries in Asia,
Latin America and the Caribbean.16 In our survey, respondents from outside of sub-Saharan Africa estimated syphilis
prevalence ranging from 0.16–14.5%. The highest figures
emanated from Asia and Eastern Europe, where country estimates were as high as 8 and 14.5%, respectively.
The cost of antenatal syphilis screening and treatment is well
within the means of most, if not all, developing countries.
Donor agencies can provide direction and start-up funds to
initiate activities, with ministries of health mustering the
political will to maintain them as priorities. This study suggests that the impact on the health of infants and their
mothers will be enormous.
33
Non-treponemal tests are widely available and relatively inexpensive; however, their effectiveness is limited by lack of sensitivity in
early and late syphilis and by false-positive reactions. Treponemal
tests are more expensive, but have higher sensitivity and specificity
than non-treponemal tests.
References
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Schulz KF, Cates W Jr, O’Mara PR. Pregnancy loss, infant death,
and suffering: legacy of syphilis and gonorrhea in Africa. Genitourinary Medicine 1987; 63: 320–5.
McDermott J, Steketee R, Larsen S, Wirima J. Syphilis-associated
perinatal and infant mortality in rural Malawi. Bulletin of the
World Health Organization 1993; 71: 773–80.
Gelmon LJ, Piot P. The interactions between HIV and other sexually transmitted infections. In: Mann J, Tarantola D (eds). AIDS
in the world II. New York: Oxford University Press, 1996: 99.
Hutchinson CM, Rompalo AM, Reichart CA, Hook EW. Characteristics of patients with syphilis attending Baltimore STD
clinics. Archives of Internal Medicine 1991; 151: 511–6.
Schulz KF, Schulte JM, Berman SM. Maternal health and child
survival: opportunities to protect both women and children
from the adverse consequences of reproductive tract infections.
In: Germain A et al. (eds). Reproductive tract infections. New
York: Plenum Press, 1992.
Grosskurth H, Mosha F, Todd J et al. Impact of improved treatment of sexually transmitted diseases on HIV infection in rural
Tanzania: randomised controlled trial. Lancet 1995; 346: 530–6.
UNICEF. The state of the world’s children, 1995. Oxford: Oxford
University Press, 1995.
Macro International, Demographic and Health Services Internet
Web Site, 1998: http://www.macroint.com/dhs/
WHO. Report of WHO consultation on maternal and perinatal
infections, 28 November–2 December 1988. WHO/MCH/91.10.
Geneva: World Health Organization, 1991.
WHO. Coverage of maternity care: A listing of available information. WHO/RHT/MSM/96.28. Geneva: World Health
Organization, 1997.
De Schryver A, Meheus A. Epidemiology of sexually transmitted
diseases: the global picture. Bulletin of the World Health
Organization 1990; 68: 639–54.
Over M, Piot P. HIV Infection and Sexually Transmitted Diseases.
In: Jamison DT, Mosley WH, Measham AR, Bobadilla JL (eds).
Disease control priorities in developing countries. Washington
DC: Oxford University Press for IBRD/WB, 1993: 458.
Tsui AO, Wasserheit JN, Haaga JG. Reproductive health in
developing countries: expanding dimensions, building solutions.
Washington DC: National Academy Press, 1997: 46.
Tramont EC. Treponema Pallidum. In: Mandell G, Bennett J,
Dolin R (eds). Mandell, Douglas and Bennett’s principles and
practices of infectious disease. New York: Church, Livingstone
Inc., 1995.
Lafort Y, Seunda M, Floriano F, Gloyd S. Evaluation of screening
for syphilis and high risk pregnancy in prenatal clinics of
Mozambique. Presentation at IX Jornadas de Saude, Maputo,
Mozambique, December 13–16, 1994.
UNAIDS. Global prevalence and incidence of selected curable
sexually transmitted diseases: overview and estimates. WHO/
GPA/STD/95.1. Geneva: UNAIDS, 1995.
Biographies
Endnotes
i
VDRL = Veneral Disease Research Laboratory test; RPR =
Rapid Plasma Reagin test; FTA = Fluorescent Treponemal Antibody test; TPHA = Treponema Pallidum Hemaglutination test.
These tests are all serological, either treponemal (FTA, TPHA),
which test for presence of antibody to the syphillis organism, or nontreponemal (VDRL, RPR), which test for presence of antibodies to
other substances present in the sera of patients infected with syphilis.
Stephen Gloyd, MD, MPH, is an Associate Professor in the Department of Health Services at the School of Public Health and Community Medicine at the University of Washington (UW), as well as
the Director of the UW International Health Program. He is also
Project Director of Health Alliance International (HAI), a Seattlebased non-governmental organization with health-related projects in
Africa and Asia. His research interests are in primary health care
systems; programme evaluation; rapid assessment methods; and the
34
Stephen Gloyd et al.
epidemiology and control of measles, tuberculosis, diarrhoea and
HIV. Steve has experience in Mozambique, Somalia, Honduras,
Mexico, Zimbabwe, Côte d’Ivoire and Swaziland.
Sanders Chai, MD, is an Acting Instructor in the Department of
Pediatrics at the University of Washington School of Medicine and a
Senior Fellow in the Department of Environmental Health in the
School of Public Health and Community Medicine. He is currently
conducting research in Vietnam and Nicaragua on environmental
risk factors for children’s respiratory health. He also sits on several
local and national committees on children’s environmental health
and childcare health. A board certified pediatrician and board eligible occupational and environmental medicine physician, Dr Chai
has had experience in India implementing a TB control programme
and in Vietnam conducting course work for the Ministry of Health.
Mary Anne Mercer, DrPH, MPH, is a Lecturer with the Department
of Health Services at the School of Public Health and Community
Medicine at the University of Washington. She is also Deputy Director of Health Alliance International (HAI), a Seattle-based nongovernmental organization with health-related projects in Africa
and Asia. Her research interests are in the design, management,
training and evaluation for community-based maternal–child health
programmes, including primary health care and programmes
responding to HIV/AIDS. Mary Anne has an additional focus on the
social and economic determinants of health status, and has experience in Nepal, Thailand and East and Southern Africa.
Correspondence: Stephen Gloyd, Department of Health Services,
Box 357660, University of Washington School of Public Health and
Community Medicine, Seattle, WA 98195, USA.