Twenty-nine twin papers were included in the metanalysis [7, 30–56, 70] (Table 2), comprising 26 different samples. These studies involved self-report of symptoms using a range of validated questionnaires. Table 2 and Figs. 5 and 6 depict the MZ and DZ twin correlations from 29 twin studies (eight with child/adolescent samples) and the metanalytic summary estimates. The meta-analysis revealed similar correlations across the age subgroups and the total sample and that MZ twins had significantly higher correlations for OCD when compared to DZ twins (children/adolescents: rMZ = 0.52, 95% CI 0.37–0.65, p < 0.01; rDZ = 0.27; 95% CI, 0.18–0.36, p < 0.01; adults: rMZ = 0.43, 95% CI 0.40–0.45, p < 0.01; rDZ = 0.20; 95% CI, 0.17–0.22, p < 0.01; total: rMZ = 0.47, 95% CI 0.40–0.54, p < 0.01; rDZ = 0.23; 95% CI, 0.19–0.27, p < 0.01). The use of the standard error of the tetrachoric correlation did not generate significant changes in the above results.Table 2OCD twin studies.Author/yearCountryInclusion age criteriaTwin pairsAssessment toolsTwin correlations by zigozityVariance component estimates based on ACE model coefficient, 95% CIMZ nDZ
