An introduction to instrumental variable assumptions, validation and estimation
Open accessAbstract
The instrumental variable method has been employed within economics to infer causality in the presence of unmeasured confounding. Emphasising the parallels to randomisation may increase understanding of the underlying assumptions within epidemiology. An instrument is a variable that predicts exposure, but conditional on exposure shows no independent association with the outcome. The random assignment in trials is an example of what would be expected to be an ideal instrument, but instruments can also be found in observational settings with a naturally varying phenomenon e.g. geographical variation, physical distance to facility or physician’s preference. The fourth identifying assumption has received less attention, but is essential for the generalisability of estimated effects. The instrument identifies the group of compliers in which exposure is pseudo-randomly assigned leading to exchangeability with regard to unmeasured confounders. Underlying assumptions can only partially be tested empirically and require subject-matter knowledge. Future studies employing instruments should carefully seek to validate all four assumptions, possibly drawing on parallels to randomisation.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
IV analyses are an alternative estimation method that can be used to derive an estimate of the causal effect of the treatment on the outcome by accounting for the size of the association between randomization and the exposure.
“IV analyses are an alternative estimation method that can be used to derive an estimate of the causal effect of the treatment (here, CRP) on the outcome (SBP) by accounting for the size of the association between randomization and CRP.3,4,12–15”
From Mendelian randomization (2022) · cites this paper as [14]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Identification of Causal Effects Using Instrumental Variables
Angrist, J. D., Imbens, G. W., & Rubin, D. B. (1996) · Journal of the American Statistical Association · doi:10.1080/01621459.1996.10476902 · Cited by 4,230 · see citing works
A meta-analysis of the association between adherence to drug therapy and mortality
Simpson, S. H., Eurich, D. T., Majumdar, S. R., Padwal, R. S., Tsuyuki, R. T., Varney, J., & Johnson, J. A. (2006) · BMJ · doi:10.1136/bmj.38875.675486.55 · Cited by 1,353 · cited 1× · see citing works
“This type of confounding has been investigated within the fields of oral contraceptives, postmenopausal hormone therapy, statins and influenza vaccines and termed “compliance bias” [6], “prevention bias” [7], “healthy adherer effect” [8] and “healthy user effect/bias” [9].”
BackgroundAn introduction to instrumental variables for epidemiologists
Greenland, S. (2000) · International Journal of Epidemiology · doi:10.1093/ije/29.4.722 · Cited by 1,156 · see citing works
Instruments for Causal Inference: An Epidemiologist's Dream?
Hernán, M. A., & Robins, J. M. (2006) · Epidemiology · doi:10.1097/01.ede.0000222409.00878.37 · Cited by 961 · see citing works
Instrumental variable methods for causal inference
Baiocchi, M., Cheng, J., & Small, D. S. (2014) · Statistics in Medicine · doi:10.1002/sim.6128 · Cited by 687 · see citing works
Identifiability, Exchangeability, and Epidemiological Confounding
GREENLAND, S., & ROBINS, J. M. (1986) · International Journal of Epidemiology · doi:10.1093/ije/15.3.413 · Cited by 686 · see citing works
Instrumental Variables: Application and Limitations
Martens, E. P., Pestman, W. R., de Boer, A., Belitser, S. V., & Klungel, O. H. (2006) · Epidemiology · doi:10.1097/01.ede.0000215160.88317.cb · Cited by 464 · see citing works
Effect of Flexible Sigmoidoscopy Screening on Colorectal Cancer Incidence and Mortality
Holme, Ø., Løberg, M., Kalager, M., Bretthauer, M., Hernán, M. Á., Aas, E., Eide, T. J., Skovlund, E., Schneede, J., Tveit, K. M., et al. (2014) · JAMA · doi:10.1001/jama.2014.8266 · Cited by 424 · see citing works
Bosco, J. L. F., Silliman, R. A., Thwin, S. S., Geiger, A. M., Buist, D. S. M., Prout, M. N., Yood, M. U., Haque, R., Wei, F., & Lash, T. L. (2009) · Journal of Clinical Epidemiology · doi:10.1016/j.jclinepi.2009.03.001 · Cited by 393 · see citing works
Brookhart, M. A., Patrick, A. R., Dormuth, C. R., Avorn, J., Shrank, W. H., Cadarette, S. M., & Solomon, D. H. (2007) · American Journal of Epidemiology · doi:10.1093/aje/kwm070 · Cited by 386 · see citing works
Randomization, Statistics, and Causal Inference
Greenland, S. (1990) · Epidemiology · doi:10.1097/00001648-199011000-00003 · Cited by 385 · see citing works
Postmenopausal Estrogen and Prevention Bias
Barrett-Connor, E. L. (1991) · Annals of Internal Medicine · doi:10.7326/0003-4819-115-6-455 · Cited by 378 · see citing works
Glymour, M. M., Tchetgen Tchetgen, E. J., & Robins, J. M. (2012) · American Journal of Epidemiology · doi:10.1093/aje/kwr323 · Cited by 292 · see citing works
The need for randomization in the study of intended effects
Miettinen, O. S. (1983) · Statistics in Medicine · doi:10.1002/sim.4780020222 · Cited by 277 · see citing works
Control of Confounding in the Assessment of Medical Technology
Greenland, S., & Neutra, R. (1980) · International Journal of Epidemiology · doi:10.1093/ije/9.4.361 · Cited by 201 · see citing works
Commentary: How to Report Instrumental Variable Analyses (Suggestions Welcome)
Swanson, S. A., & Hernán, M. A. (2013) · Epidemiology · doi:10.1097/ede.0b013e31828d0590 · Cited by 197 · cited 3× · see citing works
“Moreover, they show how much “information” that needs to be provided by a fourth assumption in order to obtain a point estimate [24].”
BackgroundRassen, J. A., Brookhart, M. A., Glynn, R. J., Mittleman, M. A., & Schneeweiss, S. (2009) · Journal of Clinical Epidemiology · doi:10.1016/j.jclinepi.2008.12.005 · Cited by 164 · see citing works
Who is the marginal patient? Understanding instrumental variables estimates of treatment effects.
Harris, K. M., & Remler, D. K. (1998) · PubMed · Cited by 159 · see citing works
Identifiability, exchangeability and confounding revisited
Greenland, S., & Robins, J. M. (2009) · Epidemiologic Perspectives & Innovations · doi:10.1186/1742-5573-6-4 · Cited by 155 · see citing works
Issues in the Reporting and Conduct of Instrumental Variable Studies
Davies, N. M., Smith, G. D., Windmeijer, F. A. G., & Martin, R. M. (2013) · Epidemiology · doi:10.1097/ede.0b013e31828abafb · Cited by 146 · see citing works
Petitti, D. B. (1994) · Annals of Epidemiology · doi:10.1016/1047-2797(94)90056-6 · Cited by 109 · see citing works
Think Globally, Act Globally: An Epidemiologist’s Perspective on Instrumental Variable EstimationPreprint
Swanson, S. A., Hernán, M. Á., & Hernán, M. Á. (2016) · Cited by 51 · see citing works
Definition and Evaluation of the Monotonicity Condition for Preference-based Instruments
Swanson, S. A., Miller, M., Robins, J. M., & Hernán, M. Á. (2015) · Epidemiology · doi:10.1097/ede.0000000000000279 · Cited by 50 · see citing works
Davies, N. M., Smith, G. D., Windmeijer, F. A. G., & Martin, R. M. (2013) · Epidemiology · doi:10.1097/ede.0b013e318289e024 · Cited by 40 · see citing works
Think Globally, Act Globally: An Epidemiologist’s Perspective on Instrumental Variable Estimation
Swanson, S. A., & Hernán, M. Á. (2014) · Statistical Science · doi:10.1214/14-sts491 · Cited by 37 · see citing works
Bounding the per-protocol effect in randomized trials: an application to colorectal cancer screening
Swanson, S. A., Holme, Ø., Løberg, M., Kalager, M., Bretthauer, M., Hoff, G. S., Aas, E., & Hernán, M. Á. (2015) · Trials · doi:10.1186/s13063-015-1056-8 · Cited by 27 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Sanderson, E., Glymour, M. M., Holmes, M. V., Kang, H., Morrison, J., Munafò, M. R., Palmer, T., Schooling, C. M., Wallace, C., Zhao, Q., et al. (2022) · Nature Reviews Methods Primers · doi:10.1038/s43586-021-00092-5 · Cited by 2,229 · see citing works
Causal inference with observational data: the need for triangulation of evidence
Hammerton, G., & Munafò, M. R. (2021) · Psychological Medicine · doi:10.1017/s0033291720005127 · Cited by 262 · see citing works
“More detailed descriptions of the instrumental variable approach, including the underlying assumptions and potential pitfalls, are available else-where (Labrecque & Swanson, 2018; Lousdal, 2018).”
BackgroundKiratipaisarl, W., Surawattanasakul, V., & Sirikul, W. (2024) · BMC Medical Education · doi:10.1186/s12909-024-06195-3 · Cited by 37 · see citing works
“Randomized studies, such as preference-based trials [99], and non-randomized studies with targeted trial frameworks, incorporating adequate baseline and time-varying confounder control methods like regression and inverse probability weighting can enhance the effectiveness of outcomes [34, 100, 101].”
BackgroundThe challenge of dementia prevention trials and the role of quasi-experimental studies
Lindhout, J. E., van Dalen, J. W., van Gool, W. A., & Richard, E. (2023) · Alzheimer's & Dementia · doi:10.1002/alz.13029 · Cited by 6 · see citing works
Sakellariou, C. (2023) · Child Indicators Research · doi:10.1007/s12187-023-10029-x · Cited by 4 · see citing works
- Authors
- Lousdal, M. L.
- Affiliations
- Aarhus University (DK)
- Corresponding author
- Lousdal, Mette Lise
- ORCID
- Lousdal, Mette Lise
- Venue
- Emerging Themes in Epidemiology
- Published
- 2018-01-22
- Volume
- 15
- Issue
- 1
- Pages
- 1-1
- ISSN
- 1742-7622 (ISSN-L)
- PMC
- PMC5776781
- OpenAlex
- W2788469717
- Cited by
- 173
- References
- 26
- Citations by year
- 2026: 15 · 2025: 20 · 2024: 36 · 2023: 25 · 2022: 28 · 2021: 18 · 2020: 20 · 2019: 9 · 2018: 2
- FWCI
- 10.17
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Locations
- Emerging Themes in Epidemiology (publishedVersion, cc-by)
- Europe PMC (PubMed Central) (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- Aarhus Universitet
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Advanced Causal Inference Techniques
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods in Epidemiology
- Social Sciences › Decision Sciences › Statistics, Probability and Uncertainty › Meta-analysis and systematic reviews
- Keywords
- instrumental variables, causal inference, unmeasured confounding, randomization, complier average causal effect, exchangeability, instrument validation, generalizability, observational studies, causal assumptions
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
