Imputation approaches for potential outcomes in causal inference
Open accessThe DOI matches its Crossref record; 47 of the 47 references it cites were found in a registry or an index.
Abstract
BACKGROUND: The fundamental problem of causal inference is one of missing data, and specifically of missing potential outcomes: if potential outcomes were fully observed, then causal inference could be made trivially. Though often not discussed explicitly in the epidemiological literature, the connections between causal inference and missing data can provide additional intuition. METHODS: We demonstrate how we can approach causal inference in ways similar to how we address all problems of missing data, using multiple imputation and the parametric g-formula. RESULTS: We explain and demonstrate the use of these methods in example data, and discuss implications for more traditional approaches to causal inference. CONCLUSIONS: Though there are advantages and disadvantages to both multiple imputation and g-formula approaches, epidemiologists can benefit from thinking about their causal inference problems as problems of missing data, as such perspectives may lend new and clarifying insights to their analyses.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
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?
In the citing paper
“While multiple imputation has mainly been applied to impute missing covariates, it can also be used to infer counterfactual responses (Westreich et al., 2015).”
From Customer price sensitivities in competitive insurance markets (2022) · cites this paper as Westreich et al., 2015
Citations
Cites
Multiple Imputation for Nonresponse in Surveys
Rubin, D. B. (1987) · Wiley Series in Probability and Statistics · doi:10.1002/9780470316696 · Cited by 20,701
Statistical Analysis With Missing Data
Lahiff, M., Little, R. J. A., & Rubin, D. B. (1989) · Journal of the American Statistical Association · doi:10.2307/2289883 · Cited by 17,265
RUBIN, D. B. (1976) · Biometrika · doi:10.1093/biomet/63.3.581 · Cited by 9,925
Multiple imputation using chained equations: Issues and guidance for practice
White, I. R., Royston, P., & Wood, A. M. (2011) · Statistics in Medicine · doi:10.1002/sim.4067 · Cited by 9,792
Estimating causal effects of treatments in randomized and nonrandomized studies
Rubin, D. B. (1974) · Journal of Educational Psychology · doi:10.1037/h0037350 · Cited by 9,436 · cited 1×
“The fundamental problem of causal inference, and thus of analytical epidemiology, is therefore one of missing data, or the inability to observe all but one potential outcome.(4,5) Yet, when epidemiologists analyse data for the purposes of causal inference, in the absence of explicitly missing data they typically do no…”
BackgroundStatistical Analysis with Missing Data
Little, R. J. A., & Rubin, D. B. (2002) · Wiley Series in Probability and Statistics · doi:10.1002/9781119013563 · Cited by 8,473
Marginal Structural Models and Causal Inference in Epidemiology
Robins, J. M., Hernán, M. Á., & Brumback, B. (2000) · Epidemiology · doi:10.1097/00001648-200009000-00011 · Cited by 5,581
A generalization of sampling without replacement from a finite universe
Horvitz, D. G., & Thompson, D. J. (1952) · Journal of the American Statistical Association · doi:10.1080/01621459.1952.10483446 · Cited by 4,293
Schafer, J. L. (1999) · Statistical Methods in Medical Research · doi:10.1177/096228029900800102 · Cited by 3,949
A Generalization of Sampling Without Replacement From a Finite Universe
Horvitz, D. G., & Thompson, D. J. (1952) · Journal of the American Statistical Association · doi:10.2307/2280784 · Cited by 3,011
Constructing Inverse Probability Weights for Marginal Structural Models
Cole, S. R., & Hernan, M. A. (2008) · American Journal of Epidemiology · doi:10.1093/aje/kwn164 · Cited by 2,796
Self-citationBayesian inference for causal effects: The role of randomization
Rubin, D. B. (1978) · The Annals of Statistics · doi:10.1214/aos/1176344064 · Cited by 2,503
Robins, J. (1986) · Mathematical Modelling · doi:10.1016/0270-0255(86)90088-6 · Cited by 2,451
On the Application of Probability Theory to Agricultural Experiments. Essay on Principles. Section 9
Splawa-Neyman, J., Dabrowska, D. M., & Speed, T. P. (1990) · Statistical Science · doi:10.1214/ss/1177012031 · Cited by 1,627
The Table 2 Fallacy: Presenting and Interpreting Confounder and Modifier Coefficients
Westreich, D., & Greenland, S. (2013) · American Journal of Epidemiology · doi:10.1093/aje/kws412 · Cited by 1,276 · cited 1×
“The use of such techniques—though more technically complex than regression approaches even in pointexposure settings—often results in the concentration on a single causal model for a single causal effect of interest, and thus may help avoid problems such as the misinterpretation of the regression coefficients for cova…”
BackgroundSelf-citationEstimating causal effects from epidemiological data
Hernán, M. A., & Robins, J. M. (2006) · Journal of Epidemiology and Community Health · doi:10.1136/jech.2004.029496 · Cited by 1,028
A critical look at methods for handling missing covariates in epidemiologic regression analyses.
Greenland, S., & Finkle, W. D. (1995) · American Journal of Epidemiology · doi:10.1093/oxfordjournals.aje.a117592 · Cited by 891
Kang, J. D. Y., & Schafer, J. L. (2007) · Statistical Science · doi:10.1214/07-sts227 · Cited by 797
Toward Causal Inference With Interference
Hudgens, M. G., & Halloran, M. E. (2008) · Journal of the American Statistical Association · doi:10.1198/016214508000000292 · Cited by 762
Westreich, D., Lessler, J., & Funk, M. J. (2010) · Journal of Clinical Epidemiology · doi:10.1016/j.jclinepi.2009.11.020 · Cited by 502
Self-citationMarginal Structural Models to Estimate the Joint Causal Effect of Nonrandomized Treatments
Hernán, M. A., Brumback, B., & Robins, J. M. (2001) · Journal of the American Statistical Association · doi:10.1198/016214501753168154 · Cited by 493
A definition of causal effect for epidemiological research
Hernán, M. A. (2004) · Journal of Epidemiology and Community Health · doi:10.1136/jech.2002.006361 · Cited by 462
The Consistency Statement in Causal Inference: A Definition or an Assumption?
Cole, S. R., & Frangakis, C. E. (2009) · Epidemiology · doi:10.1097/ede.0b013e31818ef366 · Cited by 413
Self-citationInvited Commentary: Positivity in Practice
Westreich, D., & Cole, S. R. (2010) · American Journal of Epidemiology · doi:10.1093/aje/kwp436 · Cited by 412
Self-citationConcerning the Consistency Assumption in Causal Inference
VanderWeele, T. J. (2009) · Epidemiology · doi:10.1097/ede.0b013e3181bd5638 · Cited by 366
Data, Design, and Background Knowledge in Etiologic Inference
Robins, J. M. (2001) · Epidemiology · doi:10.1097/00001648-200105000-00011 · Cited by 358 · cited 1×
“In contrast, in an observational setting, we can relax the unconditional exchangeability assumption by assuming exchangeability conditional on a set of observed confounders, chosen based on background knowledge of the investigative team.(27) The conditional exchangeability condition can be stated explicitly in terms o…”
BackgroundIntervening on risk factors for coronary heart disease: an application of the parametric g-formula
Taubman, S. L., Robins, J. M., Mittleman, M. A., & Hernán, M. A. (2009) · International Journal of Epidemiology · doi:10.1093/ije/dyp192 · Cited by 341
Direct and Indirect Causal Effects via Potential Outcomes*
Rubin, D. B. (2004) · Scandinavian Journal of Statistics · doi:10.1111/j.1467-9469.2004.02-123.x · Cited by 324
Berkson's Bias, Selection Bias, and Missing Data
Westreich, D. (2012) · Epidemiology · doi:10.1097/ede.0b013e31823b6296 · Cited by 322
Self-citationRobins, J. (1987) · Journal of Chronic Diseases · doi:10.1016/s0021-9681(87)80018-8 · Cited by 305
Compound Treatments and Transportability of Causal Inference
Hernán, M. A., & VanderWeele, T. J. (2011) · Epidemiology · doi:10.1097/ede.0b013e3182109296 · Cited by 297
Multiple Imputation With Large Data Sets: A Case Study of the Children's Mental Health Initiative
Stuart, E. A., Azur, M., Frangakis, C., & Leaf, P. (2009) · American Journal of Epidemiology · doi:10.1093/aje/kwp026 · Cited by 268
Distinguishing “Missing at Random” and “Missing Completely at Random”
Heitjan, D. F., & Basu, S. (1996) · The American Statistician · doi:10.1080/00031305.1996.10474381 · Cited by 259
Using causal diagrams to guide analysis in missing data problems
Daniel, R. M., Kenward, M. G., Cousens, S. N., & De Stavola, B. L. (2012) · Statistical Methods in Medical Research · doi:10.1177/0962280210394469 · Cited by 155 · cited 1×
“In passing, we note that a complete case analysis is unbiased under MCAR, but may be biased under either MAR or MNAR, depending on which variables are incomplete and how these variables are related to the outcome.(31,32) There is a one-to-one relationship between the missingness of the potential outcomes and the excha…”
On the Consistency Rule in Causal Inference: Axiom, Definition, Assumption, or Theorem?
Pearl, J. (2010) · Epidemiology · doi:10.1097/ede.0b013e3181f5d3fd · Cited by 151
Westreich, D. J., Cole, S. R., Young, J. G., Palella, F. J., Tien, P. C., Kingsley, L., Gange, S. J., & Hernán, M. Á. (2012) · Statistics in Medicine · doi:10.1002/sim.5316 · Cited by 132
Self-citationLyles, R. H., Tang, L., Superak, H. M., King, C. C., Celentano, D. D., Lo, Y., & Sobel, J. D. (2011) · Epidemiology · doi:10.1097/ede.0b013e3182117c85 · Cited by 105
Distinguishing "Missing at Random" and "Missing Completely at Random"
Heitjan, D. F., & Basu, S. (1996) · The American Statistician · doi:10.2307/2684656 · Cited by 102
Gruber, S., & van der Laan, M. J. (2010) · The International Journal of Biostatistics · doi:10.2202/1557-4679.1182 · Cited by 90
Edwards, J. K., Cole, S. R., & Westreich, D. (2015) · International Journal of Epidemiology · doi:10.1093/ije/dyu272 · Cited by 71
Self-citationEstimation of causal effects of binary treatments in unconfounded studies
Gutman, R., & Rubin, D. B. (2015) · Statistics in Medicine · doi:10.1002/sim.6532 · Cited by 61
Gutman, R., & Rubin, D. B. (2012) · Statistics in Medicine · doi:10.1002/sim.5627 · Cited by 49
Petersen, M. L. (2011) · Epidemiology · doi:10.1097/ede.0b013e3182126127 · Cited by 45
The Case-Control Study as Data Missing by Design: Estimating Risk Differences
Wacholder, S. (1996) · Epidemiology · doi:10.1097/00001648-199603000-00007 · Cited by 41
Gutman, R., & Rubin, D. B. (2015) · Statistical Methods in Medical Research · doi:10.1177/0962280215570722 · Cited by 23
Perrott, D. R. (1970) · The Journal of the Acoustical Society of America · doi:10.1121/1.1911946 · Cited by 18
Patel, M. R., Westreich, D. J., Yotebieng, M., Nana, M., Eron, J. J., Behets, F. M. T. F., & Rie, A. V. (2015) · American Journal of Epidemiology · doi:10.1093/aje/kwu338 · Cited by 8
Self-citation
Cited by
Stacked generalization: an introduction to super learning
Naimi, A. I., & Balzer, L. B. (2018) · European Journal of Epidemiology · doi:10.1007/s10654-018-0390-z · Cited by 430
“Indeed, some researchers have advocated for the use of data-adaptive methods, including Super Learner, for effect estimation via singly-robust methods, depending on estimation of either the conditional mean outcome or the propensity score [26–32].”
Challenges in Obtaining Valid Causal Effect Estimates With Machine Learning Algorithms
Naimi, A. I., Mishler, A., & Kennedy, E. H. (2021) · American Journal of Epidemiology · doi:10.1093/aje/kwab201 · Cited by 144
“3,4,42 But, as we have shown, for singly robust estimators this protection may not always be worth the price.”
Background“Truly free consent”? Clarifying the nature of police legitimacy using causal mediation analysis
Pósch, K., Jackson, J., Bradford, B., & MacQueen, S. (2020) · Journal of Experimental Criminology · doi:10.1007/s11292-020-09426-x · Cited by 124
Reflection on modern methods: when worlds collide—prediction, machine learning and causal inference
Blakely, T., Lynch, J., Simons, K., Bentley, R., & Rose, S. (2021) · International Journal of Epidemiology · doi:10.1093/ije/dyz132 · Cited by 98
“3,25,29 Given that, for this simple example at least, we could use a standard regression model to estimate an effect size for the exposure (cid:2) outcome association to undertake prediction of potential outcomes, why bother?”
BackgroundMethodLearning from data with structured missingness
Mitra, R., McGough, S. F., Chakraborti, T., Holmes, C. C., Copping, R., Hagenbuch, N., Biedermann, S., Noonan, J., Lehmann, B., Shenvi, A., et al. (2023) · Nature Machine Intelligence · doi:10.1038/s42256-022-00596-z · Cited by 139
“Extended to SM, there may be much to learn about how to estimate structures of missingness via causal inference methods such as inverse probability weighting and imputation (67; 68; 69).”
Metadata
- Authors
- Westreich, D., Edwards, J. K., Cole, S. R., Platt, R. W., Mumford, S. L., & Schisterman, E. F.
- Affiliations
- University of North Carolina at Chapel Hill (US)
- McGill University (CA)
- National Institutes of Health (US)
- Corresponding author
- Westreich, Daniel
- ORCID
- Venue
- International Journal of Epidemiology
- Publisher
- Oxford University Press (OUP)
- Published
- 2015-07-25
- Volume
- 44
- Issue
- 5
- Pages
- 1731-1737
- ISSN
- 0300-5771 (ISSN-L)
- 1464-3685
- PMC
- PMC4707196
- OpenAlex
- W2278756228
- Cited by
- 64
- References
- 45
- Citations by year
- 2026: 2 · 2025: 6 · 2024: 9 · 2023: 7 · 2022: 9 · 2021: 9 · 2020: 8 · 2019: 7 · 2018: 5 · 2017: 2
- FWCI
- 2.08
- Type
- journal article
- Language
- en
- Open access
- green
- www.ncbi.nlm.nih.gov
- Locations
- International Journal of Epidemiology (publishedVersion)
- PubMed Central (publishedVersion)
- PubMed (publishedVersion)
- Funders
- National Institutes of Health
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Advanced Causal Inference Techniques
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Bayesian Inference
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods in Epidemiology
- Keywords
- potential outcomes, causal inference, missing data, multiple imputation, g-formula
- MeSH terms
- Epidemiology, Humans, Research Design, Models, Statistical, Causality
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
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.