Skip to main content
. 2019 Nov 6;50(14):2435–2443. doi: 10.1017/S0033291719002678

Table 2.

Two-sample MR analyses of the effect of schizophrenia and depression on smoking

Exposure Outcome Method N SNP β (95% CI) p value
Schizophrenia Lifetime smoking Inverse-variance weighted 102 0.022 (0.005–0.038) 0.009
MR Egger (SIMEX)
Weighted median 102 0.015 (0.004–0.026) 0.009
Weighted Mode 102 0.016 (−0.014 to 0.045) 0.31
MR RAPS 102 0.018 (0.003–0.032) 0.015
Depression Lifetime smoking Inverse-variance weighted 34 0.091 (0.027–0.155) 0.005
MR Egger (SIMEX)
Weighted median 34 0.100 (0.058–0.141) 2.77 × 10−06
Weighted mode 34 0.109 (0.037–0.182) 0.005
MR RAPS 34 0.078 (0.014–0.141) 0.016
Schizophrenia Smoking initiation Inverse-variance weighted 107 0.010 (0.000–0.021) 0.04
MR Egger (SIMEX) 107 −0.030 (−0.093 to 0.033) 0.35
Weighted median 107 0.003 (−0.006 to 0.012) 0.53
Weighted Mode 107 −0.008 (−0.033 to 0.017) 0.54
MR RAPS 107 0.008 (−0.002 to 0.017) 0.11
Depression Smoking initiation Inverse-variance weighted 34 0.083 (0.039–0.127) 2.32 × 10−04
MR Egger (SIMEX)
Weighted median 34 0.077 (0.042–0.112) 1.55 × 10−05
Weighted mode 34 0.062 (0.007–0.117) 0.03
MR RAPS 34 0.077 (0. 037–0.117) 1.64 × 10−04

SIMEX, simulation extrapolation, MR RAPS, robust adjusted profile score.

Note: Due to low regression dilution I2GX (see online Supplementary Table S3), MR Egger estimates could not be conducted apart from for the effect of schizophrenia on smoking initiation, where a weighted MR Egger SIMEX was conducted. Smoking initiation scores are given in βs by GSCAN calculated from the meta-analysed z-score statistic by assuming a prevalence for the binary trait (Liu et al., 2019).