\begin{document}$${N}_{k}$$\end{document}Nk is the total number of cases and controls. For GWAS on continuous traits, EffN can be replaced by the total sample size (N). The remaining four key indicators are standard estimates of LD Score regression (version 1.0.1; Bulik-Sullivan et al. 2015).Table 1Summary of GWAS summary statistics with and without 23andMe data for seven externalizing-related disorders and behaviorsPhenotypeEXT (Karlsson Linnér et al. 2021)Down-sampled EXT (EXT-minus-23andMe)Max N (EffN)h2(SE)λGCMean χ2InterceptRatioMax N (EffN)h2(SE)λGCMean χ2InterceptRatioADHD53,293 (49,017)0.235 (0.015)1.251.2971.0340.11353,293 (49,017)0.260 (0.017)1.251.2971.0340.113ALCP164,684 (150,640)0.055 (0.004)1.151.1741.0130.073164,684 (150,640)0.055 (0.004)1.151.1741.0130.073CANN186,875 (179,534)0.066 (0.004)1.231.2671.0260.098164,192 (157,230)0.068 (0.004)1.221.2451.0280.113FSEX*357,1870.115 (0.004)1.621.8691.0360.041357,1870.115 (0.004)1.631.8681.0360.041NSEX336,1210.097 (0.004)1.491.6821.0270.041336,1210.099 (0.004)1.491.6741.0270.041RISK426,3790.053 (0.002)1.371.4611.0190.041426,3790.053 (0.002)1.371.4611.0190.041SMOK1,251,809 (1,232,397)0.078 (0.002)2.333.1521.1260.058652,520 (652,518)0.079 (0.003)1.732.0621.0370.035EXT: Externalizing. Bolded rows indicate down-sampled summary statistics. EffN = sum of cohort-level effective sample sizes. The statistics reported in this table were all estimated with LD Score regression (v1.0.1) (Bulik-Sullivan et al. 2015): Heritability (h2) is on the observed scale. The genomic inflation factor, λGC, is the median χ2 statistic divided by the expected median of the χ2 distribution with 1 degree of freedom. Mean χ2 is the average χ2 statistic. Intercept is the estimated LD Score regression intercept. The ratio measures stratification bias, defined as (intercept − 1)/(mean χ2 − 1)ADHD attention-deficit/hyperactivity disorder; ALCP problematic alcohol use; CANN lifetime cannabis use; FSEX age at first sexual
