Evaluating large-scale propensity score performance through real-world and synthetic data experiments
Open accessThe DOI matches its Crossref record; 13 of the 13 references it cites were found in a registry or an index.
Abstract
Background: Propensity score adjustment is a popular approach for confounding control in observational studies. Reliable frameworks are needed to determine relative propensity score performance in large-scale studies, and to establish optimal propensity score model selection methods. Methods: We detail a propensity score evaluation framework that includes synthetic and real-world data experiments. Our synthetic experimental design extends the 'plasmode' framework and simulates survival data under known effect sizes, and our real-world experiments use a set of negative control outcomes with presumed null effect sizes. In reproductions of two published cohort studies, we compare two propensity score estimation methods that contrast in their model selection approach: L1-regularized regression that conducts a penalized likelihood regression, and the 'high-dimensional propensity score' (hdPS) that employs a univariate covariate screen. We evaluate methods on a range of outcome-dependent and outcome-independent metrics. Results: L1-regularization propensity score methods achieve superior model fit, covariate balance and negative control bias reduction compared with the hdPS. Simulation results are mixed and fluctuate with simulation parameters, revealing a limitation of simulation under the proportional hazards framework. Including regularization with the hdPS reduces commonly reported non-convergence issues but has little effect on propensity score performance. Conclusions: L1-regularization incorporates all covariates simultaneously into the propensity score model and offers propensity score performance superior to the hdPS marginal screen.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Austin, P. C. (2011) · Multivariate Behavioral Research · doi:10.1080/00273171.2011.568786 · Cited by 12,242 · cited 1×
“Propensity scores are estimated with a classification model, often logistic regression, using pretreatment baseline patient covariates such as demographics and indicators for medical conditions, procedures and drug exposures.(6) Recent research has questioned the reliability of expert opinion in the traditional approa…”
MethodAustin, P. C. (2009) · Statistics in Medicine · doi:10.1002/sim.3697 · Cited by 6,758
Causal Inference for Statistics, Social, and Biomedical Sciences: An Introduction
Imbens, G. W., & Rubin, D. B. (2015) · doi:10.1017/cbo9781139025751 · Cited by 4,210
Variable Selection for Propensity Score Models
Brookhart, M. A., Schneeweiss, S., Rothman, K. J., Glynn, R. J., Avorn, J., & Stürmer, T. (2006) · American Journal of Epidemiology · doi:10.1093/aje/kwj149 · Cited by 2,336
Negative Controls: A Tool for Detecting Confounding and Bias in Observational Studies
Lipsitch, M., Tchetgen Tchetgen, E., & Cohen, T. (2010) · Epidemiology · doi:10.1097/ede.0b013e3181d61eeb · Cited by 1,583 · cited 1×
“In addition to simulations under known hazard ratios, we perform negative outcome control experiments using sets of outcomes believed to be unrelated to the compared treatments, thus having a presumed true hazard ratio of 1.(28,29) Negative control outcomes entirely use real-world data, and when properly specified the…”
MethodGeorge, H., Duke Jon, D., Shah Nigam, H., Reich Christian, G., Vojtech, H., Schuemie Martijn, J., Suchard Marc, A., Woong, P. R., Kei, W. I. C., Rijnbeek Peter, R., et al. (2015) · Studies in Health Technology and Informatics · doi:10.3233/978-1-61499-564-7-574 · Cited by 1,542
Schneeweiss, S., Rassen, J. A., Glynn, R. J., Avorn, J., Mogun, H., & Brookhart, M. A. (2009) · Epidemiology · doi:10.1097/ede.0b013e3181a663cc · Cited by 1,128 · cited 2×
“The ‘high-dimensional propensity score’ (hdPS) is a PS model selection approach that selects a fixed number of covariates with highest marginal association with the study outcome.(15) Despite relying on univariate associations in high-dimensional data with likely nonindependent covariates, the hdPS has gained widespre…”
BackgroundMethodSparse data bias: a problem hiding in plain sight
Greenland, S., Mansournia, M. A., & Altman, D. G. (2016) · BMJ · doi:10.1136/bmj.i1981 · Cited by 898
For objective causal inference, design trumps analysis
Rubin, D. B. (2008) · The Annals of Applied Statistics · doi:10.1214/08-aoas187 · Cited by 831
Graham, D. J., Reichman, M. E., Wernecke, M., Zhang, R., Southworth, M. R., Levenson, M., Sheu, T.-C., Mott, K., Goulding, M. R., Houstoun, M., et al. (2015) · Circulation · doi:10.1161/circulationaha.114.012061 · Cited by 644
Validation of a common data model for active safety surveillance research
Overhage, J. M., Ryan, P. B., Reich, C. G., Hartzema, A. G., & Stang, P. E. (2012) · Journal of the American Medical Informatics Association · doi:10.1136/amiajnl-2011-000376 · Cited by 548
Characterizing treatment pathways at scale using the OHDSI network
Hripcsak, G., Ryan, P. B., Duke, J. D., Shah, N. H., Park, R. W., Huser, V., Suchard, M. A., Schuemie, M. J., DeFalco, F. J., Perotte, A., et al. (2016) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1510502113 · Cited by 326
Self-citationInterpreting observational studies: why empirical calibration is needed to correct p‐values
Schuemie, M. J., Ryan, P. B., DuMouchel, W., Suchard, M. A., & Madigan, D. (2014) · Statistics in Medicine · doi:10.1002/sim.5925 · Cited by 234
Self-citation
Research works printed but not linked: 35
- Brookhart, 2010, Confounding control in healthcare database research: challenges and potential approaches, Med Care, 48, S114, 10.1097/MLR.0b013e3181dbebe3 · DOI · Paper page
- Ryan, 2012, Empirical assessment of methods for risk identification in healthcare data: results from the experiments of the Observational Medical Outcomes Partnership, Stat Med, 31, 4401, 10.1002/sim.5620 · DOI · Paper page
- Rubin, 1997, Estimating causal effects from large data sets using propensity scores, Ann Intern Med, 127, 757, 10.7326/0003-4819-127-8_Part_2-199710151-00064 · DOI · Paper page
- Rosenbaum, 1983, The central role of the propensity score in observational studies for causal effects, Biometrika, 70, 41, 10.1093/biomet/70.1.41 · DOI · Paper page
- Rubin, 1996, Matching using estimated propensity scores: relating theory to practice, Biometrics, 52, 249, 10.2307/2533160 · DOI · Paper page
- Franklin, 2017, Comparing the performance of propensity score methods in healthcare database studies with rare outcomes, Stat Med, 36, 1946
- Austin, 2013, The performance of different propensity score methods for estimating marginal hazard ratios, Stat Med, 32, 2837, 10.1002/sim.5705 · DOI · Paper page
- Schneeweiss, 2017, Variable selection for confounding adjustment in high-dimensional covariate spaces when analyzing healthcare databases, Epidemiology, 28, 237, 10.1097/EDE.0000000000000581 · DOI · Paper page
- Franklin, 2015, Regularized regression versus the high-dimensional propensity score for confounding adjustment in secondary database analyses, Am J Epidemiol, 182, 651, 10.1093/aje/kwv108 · DOI
- Rassen, 2011, Covariate selection in high-dimensional propensity score analyses of treatment effects in small samples, Am J Epidemiol, 173, 1404, 10.1093/aje/kwr001 · DOI · Paper page
- Imai, 2014, Covariate balancing propensity score, J R Stat Soc B, 76, 243, 10.1111/rssb.12027 · DOI · Paper page
- Shortreed, 2017, Outcome-adaptive lasso: variable selection for causal inference, Biometrics, 73, 1111, 10.1111/biom.12679 · DOI · Paper page
- Mansournia, 2018, Separation in logistic regression–causes, consequences, and control, Am J Epidemiol, 187, 864, 10.1093/aje/kwx299 · DOI · Paper page
- Tibshirani, 1996, Regression shrinkage and selection via the lasso, J R Stat Soc B, 57, 267, 10.1111/j.2517-6161.1996.tb02080.x · DOI · Paper page
- Ryan, 2013, Empirical performance of a new user cohort method: Lessons for developing a risk identification and analysis system, Drug Saf, 36, 59, 10.1007/s40264-013-0099-6 · DOI
- Greenland, 2007, Invited commentary: variable selection versus shrinkage in the control of multiple confounders, Am J Epidemiol, 167, 523, 10.1093/aje/kwm355 · DOI · Paper page
- Garbe, 2013, High-dimensional versus conventional propensity scores in a comparative effectiveness study of coxibs and reduced upper gastrointestinal complications, Eur J Clin Pharmacol, 69, 549, 10.1007/s00228-012-1334-2 · DOI
- Bender, 2005, Generating survival times to simulate Cox proportional hazards models, Stat Med, 24, 1713, 10.1002/sim.2059 · DOI · Paper page
- Franklin, 2014, Plasmode simulation for the evaluation of pharmacoepidemiologic methods in complex healthcare databases, Comput Stat Data Anal, 72, 219, 10.1016/j.csda.2013.10.018 · DOI
- Vaughan, 2009, The use of plasmodes as a supplement to simulations: A simple example evaluating individual admixture estimation methodologies, Comput Stat Data Anal, 53, 1755, 10.1016/j.csda.2008.02.032 · DOI
- Voss, 2017, Accuracy of an automated knowledge base for identifying drug adverse reactions, J Biomed Inform, 66, 72, 10.1016/j.jbi.2016.12.005 · DOI · Paper page
- Walker, 1991, Observation and Inference: An introduction to the Methods of Epidemiology
- Zhou, 2017, Sentinel modular program for propensity score–matched cohort analyses: application to glyburide, glipizide, and serious hypoglycemia, Epidemiology, 28, 838, 10.1097/EDE.0000000000000709 · DOI
- Suchard, 2013, Massive parallelization of serial inference algorithms for a complex generalized linear model, ACM Trans Model Comput Simul, 23, 1., 10.1145/2414416.2414791 · DOI · Paper page
- Schuemie, CohortMethod: New-User Cohort Method with Large Scale Propensity and Outcome Models
- Austin, 2008, Assessing balance in measured baseline covariates when using many-to-one matching on the propensity-score, Pharmacoepidemiol Drug Saf, 17, 1218, 10.1002/pds.1674 · DOI · Paper page
- Franklin, 2014, Metrics for covariate balance in cohort studies of causal effects, Stat Med, 33, 1685, 10.1002/sim.6058 · DOI · Paper page
- Gail, 1984, Biased estimates of treatment effect in randomized experiments with nonlinear regressions and omitted covariates, Biometrika, 71, 431, 10.1093/biomet/71.3.431 · DOI · Paper page
- Weitzen, 2005, Weaknesses of goodness-of-fit tests for evaluating propensity score models: The case of the omitted confounder, Pharmacoepidemiol Drug Saf, 14, 227, 10.1002/pds.986 · DOI · Paper page
- Austin, 2007, A comparison of the ability of different propensity score models to balance measured variables between treated and untreated subjects: a Monte Carlo study, Stat Med, 26, 734, 10.1002/sim.2580 · DOI · Paper page
- Westreich, 2011, The role of the c-statistic in variable selection for propensity score models, Pharmacoepidemiol Drug Saf, 20, 317, 10.1002/pds.2074 · DOI · Paper page
- Kumamaru, 2016, Comparison of high-dimensional confounder summary scores in comparative studies of newly marketed medications, J Clin Epidemiol, 76, 200, 10.1016/j.jclinepi.2016.02.011 · DOI
- Ding, 2017, Instrumental variables as bias amplifiers with general outcome and confounding, Biometrika, 104, 291, 10.1093/biomet/asx009 · DOI
- Westreich, 2010, Propensity score estimation: machine learning and classification methods as alternatives to logistic regression, J Clin Epidemiol, 63, 826., 10.1016/j.jclinepi.2009.11.020 · DOI · Paper page
- Greenland, 2015, Penalization, bias reduction, and default priors in logistic and related categorical and survival regressions, Stat Med, 34, 3133, 10.1002/sim.6537 · DOI · Paper page
Cited by
Suchard, M. A., Schuemie, M. J., Krumholz, H. M., You, S. C., Chen, R., Pratt, N., Reich, C. G., Duke, J., Madigan, D., Hripcsak, G., et al. (2019) · The Lancet · doi:10.1016/s0140-6736(19)32317-7 · Cited by 359
Self-citationLane, J. C. E., Weaver, J., Kostka, K., Duarte-Salles, T., Abrahao, M. T. F., Alghoul, H., Alser, O., Alshammari, T. M., Biedermann, P., Banda, J. M., et al. (2020) · The Lancet Rheumatology · doi:10.1016/s2665-9913(20)30276-9 · Cited by 258
“29 Predictor variables included were based on all observed patient characteristics as available in each data source, including conditions, procedures, visits, observations, and measurements.”
Self-citationChen, R., Suchard, M. A., Krumholz, H. M., Schuemie, M. J., Shea, S., Duke, J., Pratt, N., Reich, C. G., Madigan, D., You, S. C., et al. (2021) · Hypertension · doi:10.1161/hypertensionaha.120.16667 · Cited by 190
Self-citationSemaglutide and Nonarteritic Anterior Ischemic Optic Neuropathy
Cai, C. X., Hribar, M., Baxter, S., Goetz, K., Swaminathan, S. S., Flowers, A., Brown, E. N., Toy, B., Xu, B., Chen, J., et al. (2025) · JAMA Ophthalmology · doi:10.1001/jamaophthalmol.2024.6555 · Cited by 129
Self-citationComparative Effectiveness of Second-Line Antihyperglycemic Agents for Cardiovascular Outcomes
Khera, R., Aminorroaya, A., Dhingra, L. S., Thangaraj, P. M., Pedroso Camargos, A., Bu, F., Ding, X., Nishimura, A., Anand, T. V., Arshad, F., et al. (2024) · Journal of the American College of Cardiology · doi:10.1016/j.jacc.2024.05.069 · Cited by 47
Self-citation
Metadata
- Authors
- Tian, Y., Schuemie, M. J., & Suchard, M. A.
- Venue
- International Journal of Epidemiology
- Publisher
- Oxford University Press (OUP)
- Published
- 2018-06-22
- Volume
- 47
- Issue
- 6
- Pages
- 2005-2014
- ISSN
- 0300-5771 (ISSN-L)
- 1464-3685
- OpenAlex
- W2808880546
- Cited by
- 169
- References
- 13
- FWCI
- 10.88
- Type
- journal article
- Language
- en
- Open access
- bronze
- Licence
- academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model (vor)
- Funders
- National Science Foundation
- Division of Information and Intelligent Systems (IIS 1251151)
- National Institutes of Health
- National Library of Medicine (1F31LM012636-01)
- Paul and Daisy Soros Fellowships for New Americans
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.