Download Obsessive-Compulsive Disorder and quality of life outcomes

Epidemiology Biostatistics and Public Health - 2015, Volume 12, Number 2
O RIGINAL ARTICLES
Obsessive-Compulsive Disorder and
quality of life outcomes: protocol for a
systematic review and meta-analysis of
cross-sectional case-control studies
Anna Coluccia(1), Andrea Fagiolini(2), Fabio Ferretti(1), Andrea Pozza(1), Arianna Goracci(2)
Background: Obsessive-compulsive disorder (OCD) is a chronic disabling psychological condition,
which can severely affect quality of life (QOL). Growing interest has been dedicated to assessing
which domains of QOL are more severely affected in patients with OCD but the findings have not
been unanimous. Research on QOL in OCD could suggest the domains in which intervention could
be improved.
Methods: Here we propose a systematic review of cross-sectional case-control studies according to
PRISMA guidelines, comparing QOL outcomes of patients with a primary OCD diagnosis with healthy
controls. Primary objectives will be to examine differences in QOL outcomes between patients with OCD
and healthy controls, and to assess which QOL domains are more severely impaired in patients than
controls, particularly subjective well-being, social and interpersonal functioning, work functioning
and family functioning. The study will then investigate potential moderators of QOL in OCD, including
participant characteristics (age, gender, presence of comorbid personality disorders, OCD symptom
severity, severity of concurrent depressive symptoms, duration of OCD symptoms and generational
cohort) and study characteristics (date of publication and methodological quality of studies).
Online databases will be searched (PsycINFO, PubMED, Science Direct, Cinahl, Biological Abstracts,
Psyclit, Embase, The Cochrane Central Register of Controlled Trials, and Google Scholar). To locate
unpublished records, conference abstracts, doctoral dissertations and theses will be handsearched,
and experts will be contacted. Statistical analysis will be performed by random effect model metaanalysis. Risk of bias assessments will be conducted using the instrument Methodological Index for
Non-Randomized Studies.
Key words: quality of life, compulsive disorder, systematic review, protocol
(1) Department of Medical Sciences, Surgery and
Neurosciences, Santa Maria alle Scotte University Hospital,
Siena, Italy.
(2) Department of Molecular Medicine, Santa Maria alle
Scotte University Hospital, Siena, Italy.
Corresponding author: Fabio Ferretti - Santa Maria
alle Scotte University Hospital - Viale Bracci 16, 53100 Siena,
Italy - Email: [email protected] - Phone: +39 0577 586409
Fax: +39 0577 233222
The study was not supported by any institutional grant. The
authors declare that they have no conflict of interest. The
authors have no competing interests to declare.
DOI: 10.2427/10037
Accepted on January 19, 2015
Obsessive-Compulsive Disorder and quality of life outcomes
e11037-1
Epidemiology Biostatistics and Public Health - 2015, Volume 12, Number 2
O RIGINAL ARTICLES
Background
Quality of life in Obsessive-Compulsive Disorder
Obsessive-Compulsive Disorder (OCD) is a
chronic psychological condition with a lifetime
prevalence of 2% in the general population [1-3].
OCD is characterized by intrusive thoughts,
impulses or mental images and repetitive
behaviours or mental compulsions, which can
strongly affect quality of life (QOL) of the
individual [4]. The World Health Organization
(WHO) has ranked OCD as the 10th leading
cause of disability of all health conditions in the
industrialized world [5].
Increasing evidence suggested that patients
with OCD have worse QOL than healthy individuals
[6]. For example, in a cross-sectional case-control
study, Huppert et al. [7] found that a group of
66 patients with primary OCD had significantly
lower QOL outcomes than a group of 36 controls
without any lifetime psychiatric diagnosis.
Growing interest has been dedicated by
researchers and clinicians to assessing which
domains of QOL are more severely affected in
patients with OCD. However, research to date
has yielded conflicting findings. Some studies
report that OCD is associated with impairment in
all domains of QOL [8]. For example, Fontenelle
et al. [9] reported that a group of patients with a
primary OCD diagnosis had significantly lower
scores than controls on all scales of the Medical
Outcomes Survey Short Form-36 [10], but not
on the scale measuring physical pain. In other
studies [11], OCD has been associated with
impairment in the specific domain of social
functioning. Indeed, lengthy pervasive rituals,
and the chronic course of the disorder itself, can
break down the social networks of the patient
and his/her family [12]. In other studies [13], OCD
was significantly associated with impairment only
in the family functioning domain but not in other
QOL domains. OCD can negatively affect the
family functioning of patients, since rituals may
often involve informal caregivers [14-15]. Impaired
family functioning caused by the expression of
compulsions in the home environment, tends to
be associated with stronger expressed emotions
and hypercriticism by caregivers [16].
Rationale for the current study
Given the inconsistency of findings from
available studies, a quantitative summary
e11037-2
of current evidence on the QOL domains
more severely impaired in OCD would be
useful for research and clinical practice. No
systematic review and meta-analysis specifically
addressing this topic has yet been conducted.
In the literature, Olatunji, Cisler and Tolin [17]
conducted a meta-analysis of 23 cross-sectional
case-control studies examining QOL in patients
with anxiety disorders, including OCD, and
healthy controls. The findings suggested that,
although patients with anxiety disorders had
lower QOL than healthy controls, the difference
was not statistically significant [17]. It could,
however, be argued that impairment was
stronger in specific domains of QOL for certain
anxiety disorders [17], and, importantly, the
meta-analysis did not specifically assess QOL in
patients with primary OCD, due to the limited
number of studies that could be included.
Assessment of QOL in OCD could have
major implications for practice by suggesting
integrated treatment strategies, pharmacological
and psychotherapeutic, aimed at improving
prognosis [18]. Although some evidence implies
that pharmacological [19] and psychological
treatments [20] can enhance sustained change on
OCD symptom outcomes, scores on QOL outcomes
after trials tend to remain lower than those of the
general population [21]. Low QOL can also be a
predictor of negative response to pharmacological
treatment and psychotherapy [22]. In addition,
poor QOL can increase the risk of early drop out
from treatment [23] or the probability of relapse
after a successful treatment course. Research on
QOL in OCD could suggest the domains in which
intervention could be improved.
The current report presents the protocol of
a systematic review and meta-analysis of crosssectional case-control studies on QOL in OCD.
Aims of the study are:
To examine differences in total QOL
outcomes scores between patients with
primary OCD and healthy controls (screened
without any current or lifetime psychiatric
disorder, individuals recruited from the general
population, and undergraduates);
To determine which QOL domains are
more severely impaired in patients with
primary OCD compared to healthy controls;
differences between patients with OCD and
healthy controls will be examined for the
following QOL domains: subjective well-being,
social and interpersonal functioning, work
functioning and family functioning;
Obsessive-Compulsive Disorder and quality of life outcomes
Epidemiology Biostatistics and Public Health - 2015, Volume 12, Number 2
O RIGINAL ARTICLES
To investigate potential moderators of QOL
in OCD, Specifically: participants characteristics
(age, gender, comorbid personality disorders,
OCD symptom severity, severity of concurrent
depressive symptoms, duration of OCD
symptom, and generational cohort, coded as
a categorical variable: child/adolescent versus
adult samples, and study characteristics (date of
publication and methodological quality)
Methods
Eligibility criteria
Following a modified version of the PICOS
approach for observational cross-sectional casecontrol studies, defined in the PRISMA guidelines
[24], the criteria considered for inclusion of
studies will be: (a) participants; (b) outcomes;
(c) comparators and; (d) design characteristics:
Participant Characteristics. Studies will
be included if they were conducted on patients
with a primary diagnosis of OCD, provided
the diagnosis was made through a semistructured interview based on standardized
diagnostic criteria, such as the Structured
Clinical Interview for DSM-IV [25]. Studies will
only be included if they involved patients with
a current primary diagnosis of OCD. Thus,
studies on patients with a lifetime diagnosis
of OCD and studies using participants with
subclinical OCD symptoms will be excluded.
Studies where patients were assessed only
through psychometric self-report measures will
be excluded. Studies will be included if they
used either adult or adolescent/chiln samples.
Studies on primary compulsive hoarding will
be excluded, as the treatment for hoarding
differs from CBT for OCD, and hoarding
is a separate diagnosis in DSM-5 [4], albeit
included in the OCD-related disorders chapter.
Studies where all patients had OCD and a
specific comorbid psychological or medical
disorder will be excluded. However, studies
where some of the patients had comorbid
disorders will be included. Studies conducted
on mixed samples, consisting of patients with
primary OCD and patients with other primary
conditions (e.g, mood or anxiety disorders)
will be excluded. Concurrent psychological
or pharmacological treatments, will not be
considered as an exclusion criterion.
Outcome Characteristics. Studies will be
included if they used outcome measures of
QOL with known psychometric properties,
either self-report questionnaires or interviews,
such as the SF-36.
Comparator characteristics. Studies will be
included if they used healthy control groups
consisting of screened participants who did not
report any psychiatric disorder during a clinical
interview. Such participants may be matched
with patients with OCD on socio-demographic
variables. Undergraduates or individuals
recruited from the general population (i.e.
community participants) are also acceptable.
Design characteristics. Studies will be included
if they used observational cross-sectional casecontrolled design, comparing QOL outcomes
between groups of patients with primary OCD
diagnosis and healthy control groups.
Information sources and search procedure
The following search strategies will be
used to identify studies eligible for inclusion.
Electronic search. Studies will be retrieved
by online systematic literature searches, in
which key words related to OCD (“obsessive
compulsive
disorder”,
“obsessions”,
“compulsions”, “rituals”) will be combined by
the boolean operator “AND” with key words
and text words indicative of the QOL construct
(“quality of life”, “psychological well-being”,
“satisfaction with life”, “global functioning”,
“social functioning”, “family functioning”,
“subjective well-being”) and of QOL measures
(“Medical Outcome Survey Short Form-36”,
“Quality of Life Inventory”, “Satisfaction with
Life Scale”, “WHOQOL-100”, “WHOQOLBREF”). The following online databases will be
searched: PsycINFO, PubMED, Science Direct,
Cinahl, Biological Abstracts, Psyclit, Embase,
The Cochrane Central Register of Controlled
Trial, and Google Scholar. No date restriction
will be applied.
Corresponding authors. To request any
further papers, published or unpublished, all
corresponding authors of studies to be included
in the meta-analysis will be contacted.
Handsearch Conference proceedings and
doctoral dissertations will be handsearched for
abstract books of the following international
associations relevant to the issue of QOL in
OCD: European Health Psychology Society,
International Society for Quality of Life Research,
American Psychiatry Association, American
Obsessive-Compulsive Disorder and quality of life outcomes
e11037-3
Epidemiology Biostatistics and Public Health - 2015, Volume 12, Number 2
O RIGINAL ARTICLES
Psychology Association, International Society
for Quality of Life Studies, World Psychiatric
Association, The Royal College of Psychiatrists,
British Psychology Society.
Reference lists. Reference lists of the studies
included in the meta-analysis will be examined.
In addition, references will be examined for
the following peer-reviewed journals relevant
to this topic: Anxiety, Stress and Coping,
Applied Research in Quality of Life, Behaviour
Research and Therapy, British Journal of
Health Psychology, Comprehensive Psychiatry,
Depression and Anxiety, European Psychiatry,
Health Psychology, Health Psychology Review,
Journal of Clinical Psychiatry, Journal of
Psychiatric Research, Psychotherapy and
Psychosomatics, Quality of Life Research.
Study selection
Studies will be assessed on eligibility
criteria by two independent reviewers (FF, AC)
in three stages.
During the first and second stages, studies
will be examined with regard to inclusion criteria
after reading the title and abstract, respectively.
Studies for which reviewers are not unanimous
about inclusion will be retained and assessed for
eligibility after reading the full-text. After each
stage, the reviewers will discuss their reasons
for inclusion and discrepancy will be addressed
during meetings with two other independent
reviewers (AF, AG) in order to obtain a shared
pool of included studies for the meta-analysis.
Between-reviewer agreement on inclusion
will be calculated by Cohen’s Kappa index [26].
Assessment of methodological quality and
risk of bias
Methodological quality and risk of bias
for the studies included in the meta-analysis
will be assessed by the two independent
reviewers (FF, AC) using the Methodological
Index
for
Non-Randomized
Studies
[MINORS; 27]. Between-reviewer agreement
on methodological quality scores will be
calculated by Cohen’s Kappa index [26].
Quality of studies and data reporting
will be assessed according to Strengthening
the Reporting of Observational Studies in
Epidemiology guidelines [STROBE; 28].
e11037-4
Moderatos coding
If inconsistency analyses indicates
significant heterogeneity of effect sizes, the
role of moderators will be investigated. Two
independent reviewers (FF and AC) will code
the moderators, extract the data from primary
studies and enter it in an excel worksheet.
Subsequently, the two reviewers will meet
to check worksheet data entry for accuracy,
and any discrepancy will be discussed and
resolved. The following variables will be coded
as moderators: Participans characteristics: (1)
mean age of sample; (2) gender of sample
(coded as percentage of female participants);
(3) co-occurence of comorbid personality
disorders (percentage of participants with
comorbid personality disorders in sample); (4)
OCD symptom severity (coded as a continuous
variable based on Y-BOCS scores); (5) severity
of concurrent depressive symptoms (coded as a
continuous variable based on BDI-II scores); (6)
duration of OCD symptoms (coded as number
of years from first diagnosis of OCD made by
a mental health professional or as patient selfreported age of onset; (7) generational cohort
(categorical variable: child/adolescent versus
adult samples);
Study characteristics: (1) date of
publication; (2) methodological quality (as
continuous variable based on MINORS scores).
Statistical analysis
Power calculations
An a priori power analysis will be
performed to investigate the number of studies
requested to achieve a statistically powerful
analysis for detecting a medium effect size, as
defined by Lipsey [29]. Power analysis will be
conducted according to the guidelines provided
by Borenstein and colleagues [30] using Power
and Precision software version 4.00.
Data extraction and summary measures
Before calculating the effect sizes, outlier
studies will be identified by the sample-adjusted
meta-analytic deviance method [SAMD; 31].
As we expect heterogeneity across the
studies included, effect sizes will be calculated
using a random effects model. Random effects
models assume that the studies included are
Obsessive-Compulsive Disorder and quality of life outcomes
Epidemiology Biostatistics and Public Health - 2015, Volume 12, Number 2
O RIGINAL ARTICLES
drawn from populations of studies that differ
systematically from each other. According to
these models, the effect sizes derived from
included studies differ not only because of
random error within studies (as in the fixed
effects model) but also because of true variation
in effect sizes from one study to another [30].
Data requited to calculate effect sizes
will be extracted independently by two metaanalysts (FF and AP) using formula [32]:
(a) ES= MOCD – MCONTROL/ SDPOOLED
where MOCD is the mean of the groups of
patients with OCD, MCONTROL is the mean of the
healthy controls and SDPOOLED is the pooled
standard deviation.
The effect size for each study will be weighted
by application of correction formula (b):
Inconsistency analysis
In order to assess between-study
heterogeneity, two complementary indices
will be used, the I 2 index [35] and the
Q statistic [36]. The I2 index determines
percentage of heterogeneity of effect sizes of
included studies [35]. A value approaching
zero suggests homogeneity, whereas values
of 25-50%, 50-75% and 75-100% represent
low, medium and high heterogeneity,
respectively [35].
Analysis of moderators
If inconsistency analysis suggests
high heterogeneity, the above-mentioned
moderators will be analyzed by mixed model
ANOVA and weighted least squares metaregression.
(b) Wzr= 1/SE2zr
where SE2zr is the standard error of the
effect size computed for each study. This
formula will be further corrected with formula
(c) for sampling error:
(c) W’= Wzr (αQOL) (αDOC)
where αQOL is the Cronbach reliability
coefficient [33] as internal consistency for QOL
measures and e αDOC is the coefficient for
OCD symptom measures.
Effect sizes will be estimated using a 95%
confidence interval and interpreted according
to criteria suggested by Cohen [32]. Thus, effect
sizes of 0.80 or more will be assumed to be
large, 0.50 moderate, and 0.20 small [32]. The
Hedges' correction for small sample bias will be
applied to all effect sizes [34]. A global effect
size will be calculated as mean effect size by
combining effect sizes related to different QOL
measures. For studies using QOL measures
related to different QOL domains (e.g. social
functioning and subjective well-being), a mean
effect size will be calculated for each study.
Subsequently, effect sizes will be calculated
separately for specific domains related to
QOL. Mean effect sizes will be computed for
the following domains: subjective well-being,
social and interpersonal functioning, work
functioning and family functioning.
Publication bias
To investigate the likelihood that effect
sizes are subject to publication bias, Orwin’s
fail-safe N method [37] and visual inspection
of the funnel plot will be used.
The meta-analysis will be performed
using Comprehensive Meta-analysis software,
version 2.00.
Discussion
The systematic review will examine
QOL outcomes in OCD and particularly
which QOL domains are more severely
affected in this psychological condition.
Although no meta-analysis has ever been
conducted on this topic, some limitations
should be considered. The inclusion of
different observational designs prevents
any conclusions about causal connection
between OCD and QOL. Another source of
heterogeneity could come from aggregating
different types of QOL measures to calculate
a global QOL outcome. A further potential
source of bias could arise from using different
types of control sample as comparators (e.g.
screened participants, individuals from the
general population, undergraduates).
In conclusion, by highlighting which
Obsessive-Compulsive Disorder and quality of life outcomes
e11037-5
Epidemiology Biostatistics and Public Health - 2015, Volume 12, Number 2
O RIGINAL ARTICLES
domains of QOL to target for treatment
and which patient characteristics could be
associated with poorer QOL, the results will
be useful for clinical practice. The study
could also be a starting point for future
research on predictors of poor QOL in OCD.
References
[1] Crino R, Slade T, Andrews G. The changing prevalence
and severity of obsessive-compulsive disorder criteria from
DSM-III to DSM-IV. Am J Psychiatry 2005; 162: 876–882.
[2] Kessler RC, Berglund P, Demler O, Jin R, Merikangas KR,
Walters EE. Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity
Survey Replication. Arch Gen Psychiatry 2005; 62: 593-602.
[3] Masellis M, Rector NA, Richter MA. Quality of life in
OCD: differential impact of obsessions, compulsions, and
depression comorbidity. Can J Psychiatry 2003; 48: 72-77.
[4] American Psychiatric Association: Diagnostic and statistical manual of mental disorders (5th ed.). Washington,
DC: American Psychiatric Association; 2013.
[5] Murray C, Lopez A. The global burden of disease: a comprehensive assessment of mortality and disability from diseases,
injury, and risk factors in 1990 and projected to 2020.
Cambridge, MA: Harvard School of Public Health; 1996.
[6] Eisen JL, Mancebo MA, Pinto A, et al. Impact of
obsessive-compulsive disorder on quality of life. Compr
Psychiatry 2006; 47: 270-75.
[7] Huppert JD, Simpson HB, Nissenson KJ, Liebowitz MR,
Foa EB. Quality of life and functional impairment in
obsessive–compulsive disorder: A comparison of patients
with and without comorbidity, patients in remission, and
healthy controls. Depress Anxiety 2009; 26: 39-45.
[8] Lack CW, Storch EA, Keeley ML, et al. Quality of life in children and adolescents with obsessive-compulsive disorder:
base rates, parent–child agreement, and clinical correlates.
Soc Psychiatry Psychiatr Epidemiol 2009, 44: 935-42.
[9] Fontenelle IS, Fontenelle LF, Borges MC, Prazeres AM,
Rangé BP, Mendlowicz MV, Versiani M. Quality of life
and symptom dimensions of patients with obsessive–
compulsive disorder. Psychiatry Res 2010; 179: 198-203.
[10]Ware JE, Sherbourne CD. The MOS 36-item short-form
health survey (SF-36). Med Care 1992; 30: 473–83.
[11] Rosa AC, Diniz JB, Fossaluza V, et al. Clinical correlates of
social adjustment in patients with obsessive-compulsive disorder. Journal of Psychiatric Research 2012; 46: 1286-92.
[12]Bobes J, Gonzalez MP, Bascaran MT, Arango C, Saiz PA,
Bousono M. Quality of life and disability in patients with
obsessive-compulsive disorder. European Psychiatry
2001; 16: 239-45.
[13] Lochner C, Mogotsi M, du Toit PL, Kaminer D, Niehaus DJ,
Stein DJ. Quality of life in anxiety disorders: a comparison
e11037-6
of obsessive-compulsive disorder, social anxiety disorder,
and panic disorder. Psychopathology 2003; 36: 255-62.
[14]Leckman JF, Mayes LC, Feldman R, Evans DW, King RA,
Cohen DJ. Early parental preoccupations and behaviors and their possible relationship to the symptoms of
obsessive compulsive disorder. Acta Psychiatr Scand
1999; 100: 1-26.
[15]Storch EA, Geffken GR, Merlo LJ, et al. Family accommodation in pediatric obsessive–compulsive disorder.
Journal of Clinical Child and Adolescent Psychology
2007; 36: 207-16.
[16] Pace SM, Thwaites R, Freeston MH. Exploring the role of
external criticism in obsessive-compulsive disorder: A narrative review. Clinical Psychology Review 2011; 31: 361-70.
[17]Olatunji BO, Cisler JM, Tolin DF. Quality of life in the
anxiety disorders: a meta-analytic review. Clin Psychol
Rev 2007; 27: 572-81.
[18]Jacoby RJ, Leonard RC, Riemann BC, Abramowitz JS.
Predictors of quality of life and functional impairment in
obsessive-compulsive disorder. Compr Psychiatry 2014;
55: 1195-202.
[19]Eddy KT, Dutra L, Bradley R, Westen D. A multidimensional meta-analysis of psychotherapy and pharmacotherapy for obsessive-compulsive disorder. Clin Psychol
Rev 2004; 24: 1011-30.
[20]Moritz S, Rufer M, Fricke S, et al. Quality of life in obsessive-compulsive disorder before and after treatment.
Compr Psychiatry 2005; 46: 453-9.
[21]Safren SA, Heimberg RG, Brown EG, Holle C. Quality
of life in patients with social phobia. Depression and
Anxiety 1997; 4: 126−33.
[22]Maher MJ, Huppert JD, Chen H, et al. Moderators and
predictors of response to cognitive-behavioral therapy
augmentation of pharmacotherapy in obsessive–compulsive disorder. Psychol Med 2010; 40: 2013-23.
[23]Hollander E, Stein DJ, Fineberg NA, Marteau F, Legault
M. Quality of life outcomes in patients with obsessivecompulsive disorder: relationship to treatment response
and symptom relapse. J Clin Psychiatry 2010; 71: 784.
[24]Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred
reporting items for systematic reviews and meta-analyses: the PRISMA statement. Annals of Internal Medicine
2009; 151: 264-69.
[25]First MB, Gibbon M, Spitzer RL, Williams JBW, Benjamin
Obsessive-Compulsive Disorder and quality of life outcomes
Epidemiology Biostatistics and Public Health - 2015, Volume 12, Number 2
O RIGINAL ARTICLES
LS. Structured Clinical Interview for DSM-IV Axis II personality disorders (SCID-II). Washington, DC: American
Psychiatric Press; 1997.
[26] Cohen J. A coefficient of agreement for nominal scales.
Educational and psychological measurement 1960; 20: 37-46.
[27]Slim K, Nini E, Forestier D, Kwiatkowski F, Panis Y,
Chipponi J. Methodological index for non-randomized
studies (MINORS): Development and validation of a
new instrument. Australian and New Zealand Journal of
Surgery 2003; 73: 712-6.
[28]von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche
PC, Vandenbroucke JP. The Strengthening the Reporting
of Observational Studies in Epidemiology (STROBE)
statement: guidelines for reporting observational studies.
Preventive medicine 2007; 45: 247-51.
[29]Lipsey MW. Design sensitivity: statistical power for
experimental research. Sage: New York Park; 1990.
[30]Borenstein M, Hedges LV, Higgins JPT, Rothstein, HR.
Introduction to meta-analysis. Chichester: John Wiley &
Sons; 2009.
[31] Huffcutt AI, Arthur W. Development of a new outlier statistic for meta-analytic data. J Appl Psychol 1995; 80: 327.
[32]Cohen J. Statistical power analysis for the behavioral sciences. New York. Routledge; 1988.
[33]Cronbach LJ. Coefficient alpha and the internal structure
of tests. Psychometrika 1951; 16: 297-334.
[34]Hedges L. Distribution theory for Glass’ estimator of
effect size and related estimators. Journal of Educational
Statistics 1981; 6: 107-28.
[35]Higgins JPT, Thompson SG, Deeks JJ, Altman DG.
Measuring inconsistency in meta-analyses. BMJ 2003,
327: 557-60.
[36]Hedges LV, Olkin I. Statistical methods for meta-analysis.
San Diego, CA: Academic Press; 1985.
[37]Orwin RG. A fail-safe N for effect size in meta-analysis.
Journal of Educational Statistics 1983; 8: 157-9.
Obsessive-Compulsive Disorder and quality of life outcomes
e11037-7