Nineteen eligible family studies were included in the meta-analysis (Table 1). Almost all exclusions of family studies were due to the absence of a control group or the lack of an OCD diagnosis confirmation using standardized instruments and/or direct interviews of probands or relatives (see Table S1). Four studies (comprising three samples) assessed families of child/adolescent probands [17, 20–22]. Another 14 studies, comprising nine samples, reported data on family risk for adult OCD probands [4, 5, 16, 18, 23–29, 65–67]. One study did not compare OCD familial risk between case and control probands (probands without OCD) but were included because they compared OCD familial risk between early and late-onset OCD probands [68]. Finally, from the nineteen included publications, 18 studies [4, 5, 16–18, 20–29, 65–67] (comprising twelve different samples) were considered for the main analyses.Table 1OCD family studies.OCD and comorbid disorders assessment toolsDiagnosis criteria and controlSample sizeDiagnosis in first-degree relativesAuthor/yearCountryProband age inclusion criteriaProbandRelativesBlind to proband status (y/n)Best estimate diagnosis procedure (y/n)Diagnostic criteriaControls matched for gender and age (y/n)ProbandFirst-degree RelativesDefinite OCDProbable OCDDefinite/Probable OCDCaseControlCaseControlCaseControlCaseControlCaseControlnMean age (SD)nMean age (SD)nnnnnnnnChild probandReddy et al. [20]IndiaJuvenile
