Interpretation of Mendelian randomization using a single measure of an exposure that varies over time.
Mendelian randomization
causal inference
longitudinal
simulation
Journal
International journal of epidemiology
ISSN: 1464-3685
Titre abrégé: Int J Epidemiol
Pays: England
ID NLM: 7802871
Informations de publication
Date de publication:
13 12 2022
13 12 2022
Historique:
received:
30
11
2021
accepted:
15
06
2022
pubmed:
19
7
2022
medline:
17
12
2022
entrez:
18
7
2022
Statut:
ppublish
Résumé
Mendelian randomization (MR) is a powerful tool through which the causal effects of modifiable exposures on outcomes can be estimated from observational data. Most exposures vary throughout the life course, but MR is commonly applied to one measurement of an exposure (e.g. weight measured once between ages 40 and 60 years). It has been argued that MR provides biased causal effect estimates when applied to one measure of an exposure that varies over time. We propose an approach that emphasizes the liability that causes the entire exposure trajectory. We demonstrate this approach using simulations and an applied example. We show that rather than estimating the direct or total causal effect of changing the exposure value at a given time, MR estimates the causal effect of changing the underlying liability for the exposure, scaled to the effect of the liability on the exposure at that time. As such, results from MR conducted at different time points are expected to differ (unless the effect of the liability on exposure is constant over time), as we illustrate by estimating the effect of body mass index measured at different ages on systolic blood pressure. Univariable MR results should not be interpreted as time-point-specific direct or total causal effects, but as the effect of changing the liability for the exposure. Estimates of how the effects of a genetic variant on an exposure vary over time, together with biological knowledge that provides evidence regarding likely effective exposure periods, are required to interpret time-point-specific causal effects.
Sections du résumé
BACKGROUND
Mendelian randomization (MR) is a powerful tool through which the causal effects of modifiable exposures on outcomes can be estimated from observational data. Most exposures vary throughout the life course, but MR is commonly applied to one measurement of an exposure (e.g. weight measured once between ages 40 and 60 years). It has been argued that MR provides biased causal effect estimates when applied to one measure of an exposure that varies over time.
METHODS
We propose an approach that emphasizes the liability that causes the entire exposure trajectory. We demonstrate this approach using simulations and an applied example.
RESULTS
We show that rather than estimating the direct or total causal effect of changing the exposure value at a given time, MR estimates the causal effect of changing the underlying liability for the exposure, scaled to the effect of the liability on the exposure at that time. As such, results from MR conducted at different time points are expected to differ (unless the effect of the liability on exposure is constant over time), as we illustrate by estimating the effect of body mass index measured at different ages on systolic blood pressure.
CONCLUSION
Univariable MR results should not be interpreted as time-point-specific direct or total causal effects, but as the effect of changing the liability for the exposure. Estimates of how the effects of a genetic variant on an exposure vary over time, together with biological knowledge that provides evidence regarding likely effective exposure periods, are required to interpret time-point-specific causal effects.
Identifiants
pubmed: 35848950
pii: 6645008
doi: 10.1093/ije/dyac136
pmc: PMC9749705
doi:
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
1899-1909Subventions
Organisme : Medical Research Council
ID : MC_UU_00011/1
Pays : United Kingdom
Organisme : Medical Research Council
ID : MC_UU_00011/3
Pays : United Kingdom
Informations de copyright
© The Author(s) 2022. Published by Oxford University Press on behalf of the International Epidemiological Association.
Références
BMJ. 2018 Jul 12;362:k601
pubmed: 30002074
Eur J Epidemiol. 2018 Aug;33(8):723-728
pubmed: 29721747
Hum Mol Genet. 2010 Feb 1;19(3):545-52
pubmed: 19880856
BMJ. 2018 Jun 6;361:k2022
pubmed: 29875094
Am J Epidemiol. 2007 Oct 1;166(7):760-4
pubmed: 17641153
PLoS Genet. 2022 Jul 18;18(7):e1010290
pubmed: 35849575
Int J Epidemiol. 2020 Aug 1;49(4):1158-1162
pubmed: 31800042
Stat Med. 2019 May 20;38(11):2074-2102
pubmed: 30652356
J Child Psychol Psychiatry. 2021 Dec;62(12):1462-1474
pubmed: 33778956
Int J Epidemiol. 2004 Feb;33(1):30-42
pubmed: 15075143
Am J Epidemiol. 2019 Jan 1;188(1):231-238
pubmed: 30239571
Am J Hum Genet. 2014 Jul 3;95(1):24-38
pubmed: 24954895
Stat Methods Med Res. 2007 Aug;16(4):309-30
pubmed: 17715159
Hum Genet. 2020 Jan;139(1):121-136
pubmed: 31134333
Nat Commun. 2021 Jan 7;12(1):20211
pubmed: 33436567
Annu Rev Genomics Hum Genet. 2015;16:327-50
pubmed: 25939054
Int J Epidemiol. 2003 Feb;32(1):1-22
pubmed: 12689998
Cold Spring Harb Perspect Med. 2022 Jan 4;12(1):
pubmed: 34426474
Int J Epidemiol. 2015 Apr;44(2):700-12
pubmed: 25953783
JAMA Netw Open. 2019 Jun 5;2(6):e196587
pubmed: 31251383
Int J Epidemiol. 2016 Dec 1;45(6):1866-1886
pubmed: 28108528
PLoS Genet. 2021 Aug 26;17(8):e1009723
pubmed: 34437535
Cold Spring Harb Perspect Med. 2021 Feb 1;11(2):
pubmed: 32341063
J Epidemiol Community Health. 1995 Feb;49(1):5-9
pubmed: 7707006
Int J Epidemiol. 2019 Jun 1;48(3):713-727
pubmed: 30535378
BMJ. 2020 May 06;369:m1203
pubmed: 32376654
Hypertension. 2009 Jul;54(1):84-90
pubmed: 19470880
Eur J Epidemiol. 2021 Mar;36(3):253-257
pubmed: 33611685
Stat Methods Med Res. 2017 Oct;26(5):2333-2355
pubmed: 26282889
Curr Epidemiol Rep. 2018;5(3):214-220
pubmed: 30148040
Hum Mol Genet. 2014 Sep 15;23(R1):R89-98
pubmed: 25064373
Eur J Epidemiol. 2020 Feb;35(2):99-111
pubmed: 32207040
Stat Med. 2021 Nov 10;40(25):5434-5452
pubmed: 34338327
Elife. 2020 Jan 30;9:
pubmed: 31999256
Genet Epidemiol. 2016 Dec;40(8):744-755
pubmed: 27813215
Stat Med. 2013 Nov 30;32(27):4726-47
pubmed: 23733419
Int J Epidemiol. 2016 Aug;45(4):1271-1279
pubmed: 27371628
BMJ. 2016 Mar 08;352:i582
pubmed: 26956984
Nat Commun. 2019 Oct 1;10(1):4448
pubmed: 31575865