Among the previously reported associations, the strongest association in the current results was observed for rs1799971 (p = 1.94 × 10–6, see Supplementary Table 20 for previously GWAS associated variants and genes)27. The only other previously reported variant tested in our gSEM was rs62103177 in the KCNG2 gene23, which was nominally associated with OA in our EA cohort (p = 0.0024). To capture variants reported to be association with OA that were not included in the gSEM results, we extended our lookup to standard logistic regression meta-analysis of the subsets of EA GWAS cohorts that were not included in the originally reported findings. The CNIH3 variant, rs10799590, reported by Nelson et al.24 and the PGC-SUD reported48 variant, rs201123820, were not statistically significant in our standard meta-analysis (p = 0.49 and p = 0.63, respectively). Of the two PH reported variants28, only rs10014685 was present in independent cohorts, but it was not significant in either (deCODE p = 0.89; UHS p = 0.36). Examining a recently reported GWAS of prescription opioid misuse (POU)49, we see a genetic correlation between OA and
