Skip to main content

Table 3 The effect estimates, test of heterogeneity and test of pleiotropy of exposure on cholecystitis

From: Mendelian randomization rules out the causal relationship between serum lipids and cholecystitis

  MR methodology Effect estimates on cholecystitis Test of heterogeneity Test of pleiotropy p value
OR Lower Upper p value Cochrane Q test p value MR-Egger intercept
Total cholesterol Inverse variance weighted (multiplicative random effects) 0.9994228 0.9994222 0.9994233 0.2960 135.009 0.053   
  MR Egger 0.9987183 0.9987161 0.9987205 0.1893 133.856 0.053 0.0000406 0.335
  Inverse variance weighted (fixed effects) 0.9994228 0.9994223 0.9994232 0.2347 135.009 0.053   
  Weighted median 0.9994726 0.9994719 0.9994733 0.5640     
HDL-C Inverse variance weighted (multiplicative random effects) 0.9997020 0.9997017 0.9997023 0.5650 188.737 0.000   
  MR Egger 0.9988892 0.9988871 0.9988913 0.2510 187.188 0.000 0.0000446 0.319
  Inverse variance weighted (fixed effects) 0.9997020 0.9997018 0.9997023 0.4740 188.737 0.000   
  Weighted median 0.9996111 0.9996106 0.9996117 0.5870     
LDL-C Inverse variance weighted (multiplicative random effects) 0.9984483 0.9984499 0.9984468 0.0080 116.489188 0.004851838   
  MR Egger 0.9987160 0.9987180 0.9987140 0.1893 116.200524 0.004109598 − 0.000021 0.658974602
  Inverse variance weighted (fixed effects) 0.9984483 0.9984496 0.9984471 0.0008 116.489188 0.004851838   
  Weighted median 0.9991214 0.9991225 0.9991203 0.3380     
Triglycerides Inverse variance weighted (multiplicative random effects) 0.9990893 0.9990882 0.9990903 0.2380 96.010 0.021   
  MR Egger 0.9980924 0.9980889 0.9980958 0.1680 93.378 0.027 0.0000655 0.168
  Inverse variance weighted (fixed effects) 0.9990893 0.9990884 0.9990902 0.1360 96.010 0.021   
  Weighted median 0.9983449 0.9983424 0.9983474 0.1360     
  1. TC total cholesterol, HDL-C HDL cholesterol, LDL-C LDL cholesterol, TG triglycerides, OR odds ratio, upper upper bounds of 95% confidence interval, lower lower bounds of 95% confidence interval. p value was adjusted by false discovery rate approach