Article Text

Download PDFPDF
Concerns about instrumental variable selection for biological effect versus uptake of proton pump inhibitors in Mendelian randomisation analysis
  1. Shuai Yuan1,
  2. Susanna C Larsson1,2,
  3. Dipender Gill3,
  4. Stephen Burgess4,5
  1. 1Institute of Environmental Medicine, Karolinska Institute, Stockholm, Sweden
  2. 2Department of Surgical Sciences, Uppsala Universitet, Uppsala, Sweden
  3. 3Department of Epidemiology and Biostatistics, School of Public Health, Imperial College London, London, UK
  4. 4MRC Biostatistics Unit, University of Cambridge, Cambridge, UK
  5. 5Department of Public Health and Primary Care, University of Cambridge, Cambridge, UK
  1. Correspondence to Dr Shuai Yuan, Institute of Environmental Medicine, Karolinska Institute, Stockholm, Sweden; shuai.yuan{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

A recent population-based cohort study found an association between proton pump inhibitor (PPI) prescription and risk of inflammatory bowel disease which they attributed to protopathic bias (ie, the drug was prescribed in response to initial disease symptoms) rather than a causal effect of PPI usage.1 In a following Mendelian randomisation (MR) study, An et al selected genetic variants associated with PPI usage to proxy the effect of exposure to PPI treatment.2 However, interpretation of these results remains unclear due to concerns about the strategy for instrumental variable selection identifies genetic variants that mimic the treatment under investigation. The fundamental concern is the difference between genetic predictors of the drug target effect versus genetic predictors of taking the drug.

The major concern is that the association between genetic variants and drug use likely stems from their potential to exacerbate the underlying condition targeted by the drug or to intensify associated symptoms. For example, the variant rs11591147 in the PCSK9 gene region is associated with an increased use of cholesterol-lowering medication but also higher levels of cholesterol.3 We also found contradictory results between genetically predicted medication usage behaviour and genetically predicted drug target effect …

View Full Text


  • X @@Yuan_AS, @@LarssonSC, @@dpsg108, @@stevesphd

  • Contributors SY, SCL, DG and SB planned the study. SY performed data analysis and created the figure. All authors wrote the manuscript, made revisions, read and approved the final version of the paper.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.