We reanalysed the summary statistics using our regression-based adjustment with 29,715 LD-pruned SNPs with imputation \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$R^2 \ge 0.99$$\end{document}R2≥0.99 (Methods). The regression of prognosis log odds ratios on incidence log odds ratios gave a coefficient of −0.042, which decreased to −0.264 with SIMEX adjustment for regression dilution (95% CI: −0.299 to −0.236). Here, the Hedges–Olkin based adjustment gave a similar result of −0.272. Again the negative coefficient implies that there are common causes of incidence and prognosis with concordant directions of effect.
