No retraction, correction or concern found when last checked on Oct 21, 2026.

Contextualizing selection bias in Mendelian randomization: how bad is it likely to be?

Apostolos Gkatzionis, Stephen Burgess

International Journal of Epidemiology · 2018 · CC BY

Abstract

Background. Selection bias affects Mendelian randomization investigations when selection into the study sample depends on a collider between the genetic variant and confounders of the risk factor–outcome association. However, the relative importance of selection bias for Mendelian randomization compared with other potential biases is unclear.

Methods. We performed an extensive simulation study to assess the impact of selection bias on a typical Mendelian randomization investigation. We considered inverse probability weighting as a potential method for reducing selection bias. Finally, we investigated whether selection bias may explain a recently reported finding that lipoprotein(a) is not a causal risk factor for cardiovascular mortality in individuals with previous coronary heart disease.

Results. Selection bias had a severe impact on bias and Type 1 error rates in our simulation study, but only when selection effects were large. For moderate effects of the risk factor on selection, bias was generally small and Type 1 error rate inflation was not considerable. Inverse probability weighting ameliorated bias when the selection model was correctly specified, but increased bias when selection bias was moderate and the model was misspecified. In the example of lipoprotein(a), strong genetic associations and strong confounder effects on selection mean the reported null effect on cardiovascular mortality could plausibly be explained by selection bias.

Conclusions. Selection bias can adversely affect Mendelian randomization investigations, but its impact is likely to be less than other biases. Selection bias is substantial when the effects of the risk factor and confounders on selection are particularly large.

Full text

Its licence does not allow us to show the full text. Read it at PubMed Central:

References 38

1.
Davey Smith G, Ebrahim S. ‘ Mendelian randomization’: can genetic epidemiology contribute to understanding environmental determinants of disease?Int J Epidemiol2003;32:1–22.
Not in our library10.1093/ije/dyg070
2.
Burgess S, Thompson SG. Mendelian Randomization: Methods for Using Genetic Variants in Causal Estimation. Boca Raton, FL: Chapman & Hall, 2015.
Not in our library
3.
Angrist JD, Imbens GW, Rubin DB. Identification of causal effects using instrumental variables. J Am Stat Assoc1996;91:444–55.
Not in our library
4.
Didelez V, Sheehan N. Mendelian randomization as an instrumental variable approach to causal inference. Stat Methods Med Res2007;16:309–30.
Not in our library10.1177/0962280206077743
5.
Gail MH Selection bias In: Armitage P, Colton T (eds). Encyclopedia of Biostatistics. Hoboken, NJ: Wiley, 2005, pp. 4869–70.
Not in our library
6.
Greenland S. Quantifying biases in causal models: classical confounding vs collider-stratification bias. Epidemiology2003;14:300–06.
Not in our library
7.
Hernán MA, Hernández-Díaz S, Robins JM. A structural approach to selection bias. Epidemiology2004;15:615–25.
Not in our library10.1097/01.ede.0000135174.63482.43
8.
Cole SR, Platt RW, Schisterman EF et al Illustrating bias due to conditioning on a collider. Int J Epidemiol2010;39:417–20.
Not in our library10.1093/ije/dyp334
9.
Collider scope: when selection bias can substantially influence observed associations · 2017
No notice
10.
Swanson SA, Robins JM, Miller M, Hernán MA. Selecting on treatment: a pervasive form of bias in instrumental variable analyses. Am J Epidemiol2015;181:191–97.
Not in our library10.1093/aje/kwu284
11.
Canan C, Lesko C, Lau B. Instrumental variable analyses and selection bias. Epidemiology2017;28:396–98.
Not in our library10.1097/ede.0000000000000639
12.
Hughes RA, Davies NM, Davey Smith G, Tilling K. Selection bias in instrumental variable analyses. bioRxiv2017:192237.
Not in our library10.1097/ede.0000000000000972
13.
Nitsch D, Molokhia M, Smeeth L, DeStavola BL, Whittaker JC, Leon DA. Limits to causal inference based on Mendelian randomization: a comparison with randomized controlled trials. Am J Epidemiol2006;163:397–403.
Not in our library10.1093/aje/kwj062
14.
VanderWeele TJ, Arah OA. Bias formulas for sensitivity analysis of unmeasured confounding for general outcomes, treatments, and confounders. Epidemiology2011;22:42–52.
Not in our library10.1097/ede.0b013e3181f74493
15.
Watts G. UK Biobank gets 10% response rate as it starts recruiting volunteers. BMJ2007;334:659.
Not in our library10.1136/bmj.39167.407616.db
16.
Gaziano JM, Concato J, Brophy M et al Million Veteran Program: a mega-biobank to study genetic influences on health and disease. J Clin Epidemiol2016;70:214–23.
Not in our library10.1016/j.jclinepi.2015.09.016
17.
Guo Q, Burgess S, Turman C et al Body mass index and breast cancer survival: a Mendelian randomisation analysis. Int J Epidemiol2017;46:1814–22.
Not in our library10.1093/ije/dyx131
18.
Zewinger S, Kleber ME, Tragante V et al Relations between lipoprotein(a) concentrations, LPA genetic variants, and the risk of mortality in patients with established coronary heart disease: a molecular and genetic association study. Lancet Diabetes Endocrinol2017;5:534–43.
Not in our library10.1016/s2213-8587(17)30096-7
19.
Vansteelandt S, Dukes O, Martinussen T. Survivor bias in Mendelian randomization analysis. Biostatistics2017. doi: 10.1093/biostatistics/kxx050.
Not in our library10.1093/biostatistics/kxx050
20.
Estimating the causal influence of body mass index on risk of Parkinson disease: A Mendelian randomisation study · 2017
No notice
21.
Lewis SJ, Davey Smith G. Alcohol, ALDH2, and esophageal cancer: a meta-analysis which illustrates the potentials and limitations of a Mendelian randomization approach. Cancer Epidemiol Biomarkers Prev2005;14:1967–71.
Not in our library10.1158/1055-9965.epi-05-0196
22.
Cho Y, Shin SY, Won S, Relton CL, Smith GD, Shin MJ. Alcohol intake and cardiovascular risk factors: a Mendelian randomisation study. Sci Rep2016;5:18422.
Not in our library10.1038/srep18422
23.
Alcohol Intake and Blood Pressure: A Systematic Review Implementing a Mendelian Randomization Approach · 2008
No notice
24.
Burgess S, Thompson SG; CRP CHD Genetics Collaboration. Avoiding bias from weak instruments in Mendelian randomization studies. Int J Epidemiol2011;40:755–64.
Not in our library10.1093/ije/dyr036
25.
Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression · 2015
No notice
26.
Consistent Estimation in Mendelian Randomization with Some Invalid Instruments Using a Weighted Median Estimator · 2016
No notice
27.
Seaman SR, White IR. Review of inverse probability weighting for dealing with missing data. Stat Methods Med Res2013;22:278–95.
Not in our library10.1177/0962280210395740
28.
Hernán MA, Robins JM. Estimating causal effects from epidemiological data. J Epidemiol Community Health2006;60:578–86.
Not in our library10.1136/jech.2004.029496
29.
Weight Trimming and Propensity Score Weighting · 2011
No notice
30.
Boerwinkle E, Leffert CC, Lin J, Lackner C, Chiesa G, Hobbs H. Apolipoprotein(a) gene accounts for greater than 90% of the variation in plasma lipoprotein(a) concentrations. J Clin Invest1992;90:52–60.
Not in our library10.1172/jci115855
31.
Clarke R, Peden JF, Hopewell JC et al Genetic variants associated with Lp(a) lipoprotein level and coronary disease. N Engl J Med2009;361:2518–28.
Not in our library10.1056/nejmoa0902604
32.
Kamstrup PR, Tybjaerg-Hansen A, Steffensen R, Nordestgaard BG. Genetically elevated lipoprotein(a) and increased risk of myocardial infarction. J Am Med Assoc2009;301:2331–39.
Not in our library10.1001/jama.2009.801
33.
Greenland S, Robins J, Pearl J. Confounding and collapsibility in causal inference. Stat Sci1999;14:29–46.
Not in our library
34.
Burgess S; CHD CRP Genetics Collaboration. Identifying the odds ratio estimated by a two-stage instrumental variable analysis with a logistic regression model. Stat Med2013;32:4726–47.
Not in our library10.1002/sim.5871
35.
Mendelian Randomization Analysis With Multiple Genetic Variants Using Summarized Data · 2013
No notice
36.
Hu YJ, Schmidt AF, Dudbridge F et al Impact of selection bias on estimation of subsequent event risk. Circ Cardiovasc Genet2017;10:e001616.
Not in our library10.1161/circgenetics.116.001616
37.
Hernán MA, Robins JM. Instruments for causal inference: an epidemiologist’s dream? Epidemiology 2006;17:360–72.
Not in our library10.1097/01.ede.0000222409.00878.37
38.
Burgess S, Butterworth A, Malarstig A, Thompson S. Use of Mendelian randomisation to assess potential benefit of clinical intervention. BMJ2012;345:e7325.
Not in our library10.1136/bmj.e7325

Cited by papers we hold 8

Reads

Read by people on this page

None yet.

Checked or read by AI through naulon

None yet.

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.

Are you an author?

Is this your paper?

Claim it to add your note and be paid when agents cite it. Free.