Addressing Extreme Propensity Scores via the Overlap Weights
The DOI matches its Crossref record; 19 of the 19 references it cites were found in a registry or an index.
Abstract
The popular inverse probability weighting method in causal inference is often hampered by extreme propensity scores, resulting in biased estimates and excessive variance. A common remedy is to trim patients with extreme scores (i.e., remove them from the weighted analysis). However, such methods are often sensitive to the choice of cutoff points and discard a large proportion of the sample. The implications for bias and the precision of the treatment effect estimate are unclear. These problems are mitigated by a newly developed method, the overlap weighting method. Overlap weights emphasize the target population with the most overlap in observed characteristics between treatments, by continuously down-weighting the units in the tails of the propensity score distribution. Here we use simulations to compare overlap weights to standard inverse probability weighting with trimming, in terms of bias, variance, and 95% confidence interval coverage. A range of propensity score distributions are considered, including settings with substantial nonoverlap and extreme values. To facilitate practical implementation, we further provide a consistent estimator for the standard error of the treatment effect estimated using overlap weighting.
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
“Overlap weighting emphasizes patients for whom either strategy was clinically plausible and limits the influence of extreme weights (17, 18).”
From Day-4 antipseudomonal and anti-MRSA de-escalation was not associated with higher mortality in critically ill patients with sepsis (2026) · cites this paper as [17]
Citations
Cites
The central role of the propensity score in observational studies for causal effects
ROSENBAUM, P. R., & RUBIN, D. B. (1983) · Biometrika · doi:10.1093/biomet/70.1.41 · Cited by 31,241
Matching Methods for Causal Inference: A Review and a Look Forward
Stuart, E. A. (2010) · Statistical Science · doi:10.1214/09-sts313 · Cited by 5,478
Austin, P. C., & Stuart, E. A. (2015) · Statistics in Medicine · doi:10.1002/sim.6607 · Cited by 4,495
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
Lunceford, J. K., & Davidian, M. (2004) · Statistics in Medicine · doi:10.1002/sim.1903 · Cited by 1,674
Hirano, K., & Imbens, G. W. (2001) · Health Services and Outcomes Research Methodology · doi:10.1023/a:1020371312283 · Cited by 1,039
Dealing with limited overlap in estimation of average treatment effects
Crump, R. K., Hotz, V. J., Imbens, G. W., & Mitnik, O. A. (2009) · Biometrika · doi:10.1093/biomet/asn055 · Cited by 1,007
Balancing Covariates via Propensity Score Weighting
Li, F., Morgan, K. L., & Zaslavsky, A. M. (2018) · Journal of the American Statistical Association · doi:10.1080/01621459.2016.1260466 · Cited by 962
Self-citationRothman, K. J. (1988) · Cited by 690
Weight Trimming and Propensity Score Weighting
Lee, B. K., Lessler, J., & Stuart, E. A. (2011) · PLoS ONE · doi:10.1371/journal.pone.0018174 · Cited by 500
Cited by
Items 301–400 of 1,000
Deep causal learning for robotic intelligence
Li, Y. (2023) · Frontiers in Neurorobotics · doi:10.3389/fnbot.2023.1128591 · Cited by 11
“1)[41], the overlap weights algorithm down-weighting the units in the tails of the propensity score distribution to emphasizes the target population with the most overlap in observed characteristics to overcome the extreme propensity score problem [42].”
BackgroundMethodFriedman, D. J., Holmes, D. N., Curtis, A. B., Ellenbogen, K. A., Frankel, D. S., Knight, B. P., Russo, A. M., Matsouaka, R., Turakhia, M. P., Lewis, W. R., et al. (2020) · Journal of Cardiovascular Electrophysiology · doi:10.1111/jce.14858 · Cited by 11
Boissiere-O’Neill, T., van Gelder, M. M. H. J., Engjom, H. M., & Nordeng, H. (2025) · BJOG An International Journal of Obstetrics & Gynaecology · doi:10.1111/1471-0528.18155 · Cited by 10
“Overlap weights were then computed, weighting each pregnancy by the probability of being assigned to the opposite treat-ment group [43].”
Rozenblat, D., Serret-Larmande, A., Maillard, A., Arrestier, R., Benghanem, S., Charpentier, J., Darmon, M., Das, V., Dépret, F., Donay, J.-L., et al. (2025) · Annals of Intensive Care · doi:10.1186/s13613-025-01469-5 · Cited by 10
“We employed propensity score weighting with overlap weights to account for potential confounding by indication [33].”
Celestin, F., Maltais‐Landry, G., Dubeux, J. C. B., Mylavarapu, R. S., & Yang, L. (2025) · Geoderma · doi:10.1016/j.geoderma.2025.117330 · Cited by 11
Wee, L. E., Ho, R. W. L., Lim, J. T., Chiew, C. J., Lye, D. C., & Tan, K. B. (2025) · International Journal of Infectious Diseases · doi:10.1016/j.ijid.2025.107946 · Cited by 10
Zhou, Y., Ren, X., Yabe, T., & Guan, C. (2026) · Landscape and Urban Planning · doi:10.1016/j.landurbplan.2026.105579 · Cited by 10
Sample size and power calculations for causal inference in observational studies
Liu, B., Yang, C., & Li, F. (2026) · The Annals of Statistics · doi:10.1214/26-aos2649 · Cited by 10
“The Hájek estimator for the WATE is: We can adapt the procedures in Section 3 to derive sample size calculation formulas for the ATT and ATO estimands.”
InfluentialSelf-citationShin, J., & Grams, M. E. (2023) · American Journal of Kidney Diseases · doi:10.1053/j.ajkd.2023.07.024 · Cited by 10
TNF-α inhibitor use during pregnancy and the risk of preeclampsia: population-based cohort study
Adomi, M., McElrath, T. F., Hernández–Díaz, S., Vine, S. M., & Huybrechts, K. F. (2024) · Journal of Hypertension · doi:10.1097/hjh.0000000000003747 · Cited by 10
Takano, H., Kanda, N., Wakimoto, Y., Ohbe, H., & Nakamura, K. (2025) · Frontiers in Immunology · doi:10.3389/fimmu.2024.1511481 · Cited by 10
Wang, P., Hong, H., Jeon, K., & Thomas, L. E. (2026) · Contemporary Clinical Trials · doi:10.1016/j.cct.2026.108220 · Cited by 10
Self-citationZhang, F., Ma, Y., Yu, Y., Sun, M., Li, H., Lou, J., Cao, J., Liu, Y., Niu, M., Wang, L., et al. (2022) · Frontiers in Aging Neuroscience · doi:10.3389/fnagi.2021.810050 · Cited by 10
“We excluded the patients whose estimated PS was beyond the range (10–90%) (Li et al., 2019).”
BackgroundDesign and analysis of partially randomized preference trials with propensity score stratification
Wang, Y., Li, F., Bláha, O., Meng, C., & Esserman, D. A. (2022) · Statistical Methods in Medical Research · doi:10.1177/09622802221095673 · Cited by 10
Self-citationLong-term survival of female versus male patients after coronary artery bypass grafting
Abreu, A., Máximo, J., & Leite-Moreira, A. (2022) · PLoS ONE · doi:10.1371/journal.pone.0275035 · Cited by 7
“The properties of overlap weights relative to inverse probability weighting include improved covariate balance and increased precision of effect measures estimates [12].”
BackgroundMethodInfluentialArrestier, R., Chaba, A., Mabrouki, A., Saccheri, C., Canet, E., de Chambrun, M. P., Stoclin, A., Picard, M., Wallet, F., Perier, F., et al. (2023) · Annals of Intensive Care · doi:10.1186/s13613-023-01125-w · Cited by 10
“This model has demonstrated high stability and the ability to obtain precise adjustment in various situations, emphasizing the proportion of the population where the most treatment equipoise exists [11, 12].”
BackgroundKim, D. H., Lee, S. B., Park, C. M., Levin, R., Metzger, E. D., Bateman, B. T., Ely, E. W., Pandharipande, P. P., Pisani, M. A., Jones, R. N., et al. (2023) · Annals of Internal Medicine · doi:10.7326/m22-3021 · Cited by 11
“This procedure estimates the average treatment effect in a population with the most overlap in clinical characteristics across the treatment groups by down-weighting patients with extreme propensity scores (35, 36).”
Using public clinical trial reports to probe non-experimental causal inference methods
Steinberg, E., Ignatiadis, N., Yadlowsky, S., Xu, Y., & Shah, N. H. (2023) · BMC Medical Research Methodology · doi:10.1186/s12874-023-02025-0 · Cited by 10
Instrumental variable estimation of truncated local average treatment effects
Choi, B. Y. (2021) · PLoS ONE · doi:10.1371/journal.pone.0249642 · Cited by 10
“Various simulation studies have showed that these methods perform much better than the IPW estimator [16, 17].”
MethodImproving the Generation and Evaluation of Synthetic Data for Downstream Medical Causal Inference
Amad, H., Qian, Z., Frauen, D., Piskorz, J., Feuerriegel, S., & van der Schaar, M. (2025) · neural information processing systems · doi:10.52202/085713-1021 · Cited by 9
Cho, Y., Kang, D., Lim, H., Lee, H., Choi, E.-Y., Shin, J., & Choi, K. H. (2026) · Circulation · doi:10.1161/circulationaha.125.078919 · Cited by 9
Schubert, C. C., Perkins, A. J., Myers, L. J., Damush, T. M., Penney, L., Zhang, Y., Schwartzkopf, A. L., Preddie, A. K., Riley, S., Menen, T., et al. (2022) · Journal of the American Geriatrics Society · doi:10.1111/jgs.18013 · Cited by 9
“Including only the very few VA-GRACE patients for whom there was a matched control in the analysis would have markedly altered our target population and would have limited our assessment of the effectiveness of VA-GRACE in a routine clinical setting, which was our primary objective.(16,17) Therefore, instead of using…”
BackgroundMethodInfluentialLiver transplantation for elderly patients with early-stage hepatocellular carcinoma
Endo, Y., Sasaki, K., Moazzam, Z., Lima, H. A., Alaimo, L., Munir, M. M., Shaikh, C. F., Schenk, A. D., Kitago, M., & Pawlik, T. M. (2023) · British journal of surgery · doi:10.1093/bjs/znad243 · Cited by 9
Chatelet, F., Vérillaud, B., & Chevret, S. (2023) · BMC Medical Research Methodology · doi:10.1186/s12874-023-02071-8 · Cited by 9
“Otherwise, overlap weighting has been shown to preserve a higher proportion of the sample with a reduction in bias [24] and to provide close performances to that of g-computation [8].”
BackgroundMethodPropensity score weighting with survey weighted data when outcomes are binary: a simulation study
Chen, Y., Cuerden, M. S., Zhang, W., Aldridge, M. D., & Li, L. (2023) · Health Services and Outcomes Research Methodology · doi:10.1007/s10742-023-00317-y · Cited by 9
“Depending upon the results of prelimi-nary analysis, researchers may decide whether additional procedures such as trimming extreme scores or overlap weighting are needed (Li et al. 2019).”
Giebel, M., & Kraft, K. (2023) · Industrial and Corporate Change · doi:10.1093/icc/dtad071 · Cited by 9
Linder, M., Ekbom, A., Brobert, G., Vogtländer, K., Balabanova, Y., Becattini, C., Carrier, M., Cohen, A. T., Coleman, C. I., Khorana, A. A., et al. (2024) · Journal of Thrombosis and Thrombolysis · doi:10.1007/s11239-024-02992-1 · Cited by 9
“The comparisons of the independent outcomes rVTE, major bleeding, and all-cause mortality used propensity score (PS) overlap weights [24].”
MethodGraham, A. J., Platt, A. C., Knutsen, K., Fletcher, E., & Gallagher, D. M. (2024) · BMJ Open Quality · doi:10.1136/bmjoq-2023-002737 · Cited by 9
“Across all arms, discharged patients were a median of 61 years old (IQR=45.0–74.0), relatively balanced by gender (50.8% female vs 49.2% male), predominantly white (49.2”
BackgroundInfluentialHypertension-Associated Expenditures Among Privately Insured US Adults in 2021
Kumar, A., He, S., Pollack, L. M., Lee, J. S., Imoisili, O. E., Wang, Y., Kompaniyets, L., Luo, F., & Jackson, S. L. (2024) · Hypertension · doi:10.1161/hypertensionaha.124.23401 · Cited by 9
“38,44 We presented results for 3 different specifications using the overlap propensity score-weighted sample: model 1: unadjusted; model 2: adjusted for demographic characteristics and BMI category (sex, age group, race, BMI category, interaction of hypertension with sex, age group, race, and BMI…”
Atrial Fibrillation and Retinal Stroke
Lusk, J. B., Nalawade, V. V., Wilson, L. E., Song, A., Schrag, M., Biousse, V., Dumitrașcu, O., Poli, S., Piccini, J. P., Hammill, B. G., et al. (2025) · JAMA Network Open · doi:10.1001/jamanetworkopen.2024.53819 · Cited by 9
Comparative Efficacy of Nonsteroid Immunosuppressive Medications in Childhood Nephrotic Syndrome
Robinson, C., Aman, N. F., Banh, T. H. M., Brooke, J. A., Chanchlani, R., Cuthbertson, B. H., Dhillon, V., Fan, E., Langlois, V., Levin, L., et al. (2025) · Archives of Pediatrics and Adolescent Medicine · doi:10.1001/jamapediatrics.2024.5286 · Cited by 9
Discrimination and Oral Health Impact: Moderating Role of Sex and Sexuality
Soares, G. H., Sethi, S., Jessani, A., & Jamieson, L. (2025) · Journal of Dental Research · doi:10.1177/00220345241310223 · Cited by 9
“We used propensity score (PS) overlap weighting to adjust for confounding between experience of discrimination and oral health impact (Li et al. 2019).”
Yuan, L., Peng, X., Wu, D., Chen, S., Tang, W., Wang, H., & Zhou, J. (2025) · Scientific Reports · doi:10.1038/s41598-025-91263-x · Cited by 9
“(cid:31) For participants interviewed in 2020 and 2023, the IPW was defined as w = 1/(1-(cid:31) e i ) and w = 1/ e i , respectively 41–43 .”
A generalizability score for aggregate causal effect
Chen, R., Chen, G., & Yu, M. (2021) · Biostatistics · doi:10.1093/biostatistics/kxab029 · Cited by 9
Endo, Y., Sasaki, K., Munir, M. M., Woldesenbet, S., Resende, V., Rawicz‐Pruszyński, K., Waqar, U., Mavani, P. T., Katayama, E., Yang, J., et al. (2023) · Journal of Gastrointestinal Surgery · doi:10.1007/s11605-023-05821-7 · Cited by 9
“22 An overlap weight propensity score for treatment allocation was estimated from a multi-variate model containing the patient’s age, gender, year of diagnosis, race, median income, insurance status, education levels, facility location and types, Charlson-Deyo comorbidity score, and fibrosis.”
MethodUnrecognized opportunities: The landscape of pediatric kidney‐paired donation in the United States
Verbesey, J. E., Thomas, A. G., Waterman, A. D., Karhadkar, S. S., Cassell, V., Segev, D. L., Hogan, J., & COOPER, M. M. (2024) · Pediatric Transplantation · doi:10.1111/petr.14657 · Cited by 9
“32,33 To account for potential center-level differences in the baseline hazard, we performed Cox regression with strata for each center.”
MethodEstimating the effect of cesarean delivery on long-term childhood health across two countries
Keshet, A., Rossman, H., Shilo, S., Barbash‐Hazan, S., Amit, G., Bivas‐Benita, M., Yanover, C., Girshovitz, I., Akiva, P., Ben‐Haroush, A., et al. (2022) · PLoS ONE · doi:10.1371/journal.pone.0268103 · Cited by 9
“To adjust for baseline confounders we either added baseline selected variables to the logistic model (standardization) [17], or performed weighting using either IPW [18] or OW [19] (see Methods).”
MethodMoving Toward Best Practice When Using Propensity Score Weighting in Survey Observational Studies
Zeng, Y., Li, F., & Tong, G. (2026) · Statistics in Medicine · doi:10.1002/sim.70555 · Cited by 8
“Importantly, these different tilting functions target subpopulations within the original external population from which the survey sample is drawn, and hence, the actual target population can differ from that original survey population.”
Self-citationIshida, K., Nishimoto, Y., Ohbe, H., Ikeda, N., Sugiura, T., Suda, R., Tanabe, N., Mo, M., Kimura, Y., Matsui, H., et al. (2025) · Journal of Thoracic and Cardiovascular Surgery · doi:10.1016/j.jtcvs.2025.07.039 · Cited by 8
Zhang, D. “., Lee, J. S., Therrien, N. L., Pollack, L. M., Jackson, S. L., Dong, X., Rajan, A., Hong, K., Moran, A. E., & Luo, F. (2025) · JACC Advances · doi:10.1016/j.jacadv.2025.102091 · Cited by 8
Lai, S., Liu, J., Lai, C., Seery, S., Hu, H., Wang, M., Hu, H., & Xu, T. (2023) · International Journal of Surgery · doi:10.1097/js9.0000000000000790 · Cited by 8
“Propensity score weighting using overlapping weights (PSOW) is a novel statistical method which estimates the treatment effect in an “ arti fi cial ” population consisting of patients with the highest mutual overlap of propensity score between two cohorts [24,25] .”
MethodChan, Y., Chao, T., Chen, S., Lee, H., Li, P., Yeh, Y., Kuo, C.-T., See, L., & Lip, G. Y. H. (2024) · The Journal of Clinical Endocrinology & Metabolism · doi:10.1210/clinem/dgae157 · Cited by 8
Sumida, K., Shrestha, P., Mallisetty, Y., Thomas, F., Gyamlani, G. G., Streja, E., Kalantar‐Zadeh, K., & Kovesdy, C. P. (2024) · JAMA Network Open · doi:10.1001/jamanetworkopen.2024.6822 · Cited by 8
Takahashi, T., Morita, K., Uda, K., Matsui, H., Yasunaga, H., & Nakagami, G. (2024) · Expert Review of Medical Devices · doi:10.1080/17434440.2024.2346191 · Cited by 9
Hu, W., Chen, S., Cai, J., Yang, Y., Yan, H., & Chen, F. (2024) · BMC Medical Research Methodology · doi:10.1186/s12874-024-02254-x · Cited by 8
“OW is likely to be beneficial in the presence of extreme tail weights [23, 39].”
BackgroundMethodInfluentialTreatment extent of femoropopliteal disease and clinical outcomes following endovascular therapy
Yoon, Y., Lee, J., Hwang, W., Park, H. W., Roh, J., Lee, S., Ko, Y., Ahn, C., Yu, C. W., Lee, S., et al. (2024) · EuroIntervention · doi:10.4244/eij-d-24-00037 · Cited by 8
Hasan, M., Meador, K. J., Brothers, T. N., Wang, S., Lewkowitz, A. K., Ward, K. E., Slaughter, J. L., Zhang, Y., & Wen, X. (2024) · Pharmacotherapy The Journal of Human Pharmacology and Drug Therapy · doi:10.1002/phar.4616 · Cited by 8
“21 Standardized differences of characteristics between treatment groups are presented prior to PS-OW because this approach establishes an exact balance (a standardized difference of 0) for all covariates included in the PS model.”
Sasaki, K., Tagami, T., Obinata, H., Tanaka, C., Otake, K., Yoshino, Y., Watanabe, A., Shibata, A., Kuwamoto, K., Inoue, J., et al. (2025) · Critical Care · doi:10.1186/s13054-025-05364-0 · Cited by 8
Kitano, N., Jindo, T., Yoshiba, K., Yamaguchi, D., Fujii, Y., Wakaba, K., Maruo, K., Kai, Y., & Arao, T. (2025) · Scandinavian Journal of Work Environment & Health · doi:10.5271/sjweh.4224 · Cited by 9
“Hence, to examine the effect of our multicomponent program on the outcomes, PS weighting using overlap weights (control group: PS; intervention group: 1-PS) was applied to examine the average treatment effect for the overlapping population (ATO) (34).”
Gaudino, M. F. L., Alexander, J. H., Egorova, N., Kurlansky, P. A., Lamy, A. L., Bakaeen, F. G., Hameed, I., Franco, A. D., Demetres, M. R., Robinson, N. B., et al. (2020) · Journal of Cardiac Surgery · doi:10.1111/jocs.14903 · Cited by 8
“The analysis will be restricted only to those subjects with PSs that overlap with those of the other group (common support).(32) The distributions of the PSs in the original and matched groups and PS overlap will be displayed using mirrored histograms, while a Love plot of the standardized differences of means will be…”
MethodIshikawa, Y., Nishikawa, K. K. N., Takahashi, K., Kurogochi, T., Yuda, M., Tanaka, Y., Matsumoto, A., Tanishima, Y., Yano, F., Eto, K., et al. (2022) · Annals of Surgical Oncology · doi:10.1245/s10434-022-12737-0 · Cited by 8
Ida, T., Ikeda, K., Ohbe, H., Nakamura, K., Furuya, H., Iwamoto, T., Furuta, S., Miyamoto, Y., Nakajima, M., Sasabuchi, Y., et al. (2023) · Rheumatology · doi:10.1093/rheumatology/kead343 · Cited by 8
Association of oral iron replacement therapy with kidney failure and mortality in CKD patients
Paul, S., Shrestha, P., Sumida, K., Thomas, F., Surbhi, S., Naser, A. M., Streja, E., Rhee, C. M., Kalantar‐Zadeh, K., & Kovesdy, C. P. (2023) · Clinical Kidney Journal · doi:10.1093/ckj/sfad190 · Cited by 8
Zhu, X., Guo, Z., Tian, L., Zeng, G., Zhou, P., Yuan, Z., Sang, H., Qiu, Z., Yang, Q., Wu, Y., et al. (2023) · Journal of Stroke and Cerebrovascular Diseases · doi:10.1016/j.jstrokecerebrovasdis.2023.107526 · Cited by 8
Prioritizing Variables for Observational Study Design using the Joint Variable Importance Plot
Liao, L. D., Zhu, Y., Ngo, A. L., Chehab, R. F., & Pimentel, S. D. (2024) · The American Statistician · doi:10.1080/00031305.2024.2303419 · Cited by 8
“…is best understood not as an estimate of an average treatment effect on the treated (Stuart 2010), but as an average effect on a “marginal” population consisting of individuals for whom treatment by either arm is reasonably likely (Rosenbaum 2012; Li, Thomas, and Li 2019; Greifer and Stuart 2021).”
Hamada, S., Nakajima, M., Kaszynski, R. H., Otaka, S., Goto, H., Matsui, H., Fushimi, K., Yamaguchi, Y., & Yasunaga, H. (2021) · European Journal of Clinical Microbiology & Infectious Diseases · doi:10.1007/s10096-021-04376-2 · Cited by 8
Kang, H., Kim, S. Y., Kim, J. H., Kim, E. R., Choi, H. G., Lim, H. S., Kim, J., Park, H. Y., Kim, N. Y., Hong, S., et al. (2022) · Pharmaceuticals · doi:10.3390/ph15111385 · Cited by 8
“Propensity scores, where migraine patients and comparison members were weighted by the probability of a 1-propensity score and the probability of a propensity score, respectively, were subject to overlap weighting calculated between 0 and 1 and reflected the fulfillment of accurate balance and optimized precision [46,…”
BackgroundA tool for empirical equipoise assessment in multigroup comparative effectiveness research
Yoshida, K., Solomon, D. H., Haneuse, S., Kim, S. C., Patorno, E., Tedeschi, S. K., Lyu, H., Hernández–Díaz, S., & Glynn, R. J. (2019) · Pharmacoepidemiology and Drug Safety · doi:10.1002/pds.4767 · Cited by 8
Li, F. (2020) · Statistical Science · doi:10.1214/20-sts774 · Cited by 8
Self-citationLiao, Y.-N., Chen, H.-Y., Yang, C.-W., Lee, P.-W., Hsu, C.-Y., Huang, Y.-T., & Yang, T.-H. (2022) · Frontiers in Pharmacology · doi:10.3389/fphar.2022.978814 · Cited by 7
“…is a propensity score (PS)-based statistical method widely used in observational studies to mimic randomized clinical trials, especially when considering case imbalance or potential biases caused by prominently different initial status (Li, Thomas, and Li 2019; Thomas, Li, and Pencina 2020).”
MethodStraub, L., Hernández–Díaz, S., Bateman, B. T., Zhu, Y., Mogun, H., Wisner, K. L., Gray, K. J., Lester, B. M., McDougle, C. J., Pennell, P. B., et al. (2026) · BMJ · doi:10.1136/bmj-2025-085725 · Cited by 7
Combination of Radiotherapy and Immunotherapy in Advanced Non–Small Cell Lung Cancer
Zhou, H., Wang, S.-C., Lee, T. T. L., Ye, Z., Fang, C.-B., Wai, A. K. C., Hu, C., & Kong, F. (2026) · JAMA Oncology · doi:10.1001/jamaoncol.2026.0392 · Cited by 8
Identification of in-sample positivity violations using regression trees: The PoRT algorithm
Danelian, G., Foucher, Y., Léger, M., Leborgne, F., & Chatton, A. (2023) · Journal of Causal Inference · doi:10.1515/jci-2022-0032 · Cited by 7
Cao, Z., Cho, Y., & Li, F. (2024) · Pharmaceutical Statistics · doi:10.1002/pst.2354 · Cited by 8
“2,9,36,37 In this article, we focus on evaluating a simpler strategy that ignores the uncertainty in estimating the parametric sampling score model, censoring model as well as the outcome model, and instead focus on addressing the key component of the estimating function that drives the uncertainty…”
Self-citationMokart, D., Boutaba, M., Servan, L., Bertrand, B., Baldési, O., Lefebvre, L., Gonzalez, F., Bisbal, M., Pastène, B., Duclos, G., et al. (2024) · Annals of Intensive Care · doi:10.1186/s13613-024-01333-y · Cited by 8
“…the influence of EAF therapy on the death rate at 90 days and the occurrence of the composite criteria at 30 days, we performed an overlap propensity score weighting analysis, in which each patient’s weight determined one’s probability of being assigned to the opposite treatment group [23, 24].”
MethodKutsuna, S., Ohbe, H., Matsui, H., & Yasunaga, H. (2023) · Journal of Infection and Chemotherapy · doi:10.1016/j.jiac.2023.01.022 · Cited by 7
Nguyen, J. K., & Sanghavi, P. (2023) · Health Services Research · doi:10.1111/1475-6773.14148 · Cited by 7
“To adjust for potential confounding in these comparisons, we next constructed propensity-score-based overlap weights, which assign heavier weights to observations within each group that are more similar in characteristics (as summarized by the propensity score) to the comparison group.(29,30) In addition, this approac…”
MethodLane, N. E., Bai, L., Seitz, D. P., Juurlink, D. N., Paterson, J. M., Guan, J., & Stukel, T. A. (2024) · Journal of the American Geriatrics Society · doi:10.1111/jgs.18930 · Cited by 7
“26 We first estimated a propensity score for receipt of serum electrolyte testing using a multivariable logistic regression model including age group-sex interactions, demographic characteristics listed above, CIHI Population Grouper comorbidity index, 27 medical comorbidities, concurrent…”
MethodRobust evaluation of longitudinal surrogate markers with censored data
Agniel, D. M., & Parast, L. M. (2024) · Journal of the Royal Statistical Society Series B (Statistical Methodology) · doi:10.1093/jrsssb/qkae119 · Cited by 7
Correlates of refusal of radical cystectomy in patients with muscle-invasive bladder cancer
Elshabrawy, A., Wang, H., Satsangi, A., Wheeler, K. M., Ramamurthy, C., Pruthi, D. K., Kaushik, D., Liss, M. A., Gelfond, J., Fernandez, R. A., et al. (2021) · Urologic Oncology Seminars and Original Investigations · doi:10.1016/j.urolonc.2020.11.023 · Cited by 7
The Risk of Postkidney Transplant Outcomes by Induction Choice Differs by Recipient Age
Ahn, J. B., Bae, S., Chu, N. M., Wang, L., Kim, J. Y., Schnitzler, M. A., Hess, G. P., Lentine, K. L., Segev, D. L., & McAdams‐DeMarco, M. (2021) · Transplantation Direct · doi:10.1097/txd.0000000000001105 · Cited by 7
“Age (y) 37 (24, 49) 38 (24, 50) 48 (33, 57) 48 (33, 57)”
Subclassification estimation of the weighted average treatment effect
Choi, B. Y. (2021) · Biometrical Journal · doi:10.1002/bimj.202000310 · Cited by 7
“This can be achieved by using the balancing weights approach (Li, Thomas & Li, 2018) that estimates the weighted average treatment effect (WATE) (Hirano, Imbens & Ridder, 2003) in a unified way.”
MethodPossible Incidental Parkinson’s Disease following Asthma: A Nested Case–Control Study in Korea
Kwon, M. J., Kim, J., Kang, H. S., Lim, H. S., Kim, M.-J., Kim, N. Y., Kim, S. H., Choi, H. G., & Kim, E. S. (2023) · Journal of Personalized Medicine · doi:10.3390/jpm13050718 · Cited by 7
Information content of stepped wedge designs under the working independence assumption
Tian, Z., & Li, F. (2023) · Journal of Statistical Planning and Inference · doi:10.1016/j.jspi.2023.106097 · Cited by 7
Self-citationSchein, J. R., Cloutier, M., Gauthier‐Loiselle, M., Catillon, M., Xu, C., Qu, A., & Childress, A. C. (2024) · Current Medical Research and Opinion · doi:10.1080/03007995.2024.2373883 · Cited by 7
“Weight capping defines a maximum weight based on the observed distribution of weights, and replaces the extreme weights by the weight cap 37,39–44 .”
Miao, W., Liu, J., Liu, J., Hao, Y., Yang, N., Liu, T., Smith, S. C., Huo, Y., Fonarow, G. C., Ge, J., et al. (2022) · Frontiers in Cardiovascular Medicine · doi:10.3389/fcvm.2022.828614 · Cited by 7
“patients had substantial effect on the relationship between the early treatment and in-hospital outcomes (14).”
BackgroundAndrew, B. Y., Habib, A. S., & Taicher, B. M. (2023) · Anesthesia & Analgesia · doi:10.1213/ane.0000000000006543 · Cited by 7
Rebordosa, C., Farkas, D. K., Montonen, J., Laugesen, K., Voß, F., Aguado, J., Bothner, U., Rothman, K. J., Zint, K., Mines, D. I., et al. (2022) · Pharmacoepidemiology and Drug Safety · doi:10.1002/pds.5432 · Cited by 7
“, the population with greatest treatment equipoise.(22) In addition, we performed a series of analyses to assess the robustness of results with respect to the analysis techniques used to evaluate allcause mortality (Tables S7 and S8).”
MethodTong, G., Li, F., & Allen, A. S. (2019) · Springer eBooks · doi:10.1007/978-3-319-52677-5_117-1 · Cited by 7
Self-citationCovariate-adjusted win statistics in randomized clinical trials with ordinal outcomesPreprint
Cao, Z., Zuo, S., Baumann, M. R., Davis‐Plourde, K. L., Heagerty, P. J., Tong, G., & Li, F. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2508.20349 · Cited by 7
Self-citationWeight Clipping for Robust Conformal Inference under Unbounded Covariate ShiftsPreprint
Wang, J., & Goel, S. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2605.02072 · Cited by 7
“In medical studies, extreme density ratios are also common ((Li et al., 2019), (Gao et al., 2021)).”
Preconception SGLT2 or DPP4 inhibitor use and adverse pregnancy outcomes
Ray, J. G., Harel, Z., Gilbert, R. E., Wald, R., Berger, H. M., & Park, A. L. (2023) · Diabetes Research and Clinical Practice · doi:10.1016/j.diabres.2023.110946 · Cited by 6
Inverse probability weighting for causal inference in hierarchical data
Hu, L., Yu, J., Yang, C., Chen, M., Tang, Z., Liao, R., Jike, C., Wang, J., Wang, R., Liao, Q., et al. (2025) · BMC Medical Research Methodology · doi:10.1186/s12874-025-02627-w · Cited by 6
“Model mis-specification can lead to extreme weights, affecting effect estimation [19].”
Wee, L. E., Loy, E. X., Ngiam, J. N., Lim, J. T., Chiew, C. J., Li, R. J., Chia, Y. W., Lim, S. L., Yap, J., Wong, R., et al. (2025) · European Journal of Preventive Cardiology · doi:10.1093/eurjpc/zwaf714 · Cited by 6
Neurological Events Associated With Acute Dengue Infection
Wee, L. E., Tan, W. Z., Chow, J. Y., Lim, J. T., Chiew, C. J., Chia, P. Y., Dickens, B. S. L., Ng, L. C., Ong, B. K. C., Leo, Y., et al. (2025) · JAMA Neurology · doi:10.1001/jamaneurol.2025.4608 · Cited by 6
Budhiraja, P., Lopez, R., Arrigain, S., & Schold, J. D. (2025) · Transplant International · doi:10.3389/ti.2025.14934 · Cited by 6
“The overlap propensity score weighting method assigns each patient’s weight based on the probability of that patient receiving the alternative transplant type [16] and has been shown to outperform inverse probability of treatment weighting (IPTW) in cases of minimal overlap [17].”
Nguyen, X.-M. T., Li, Y., Czernichow, S., Rassy, N., Houghton, S. C., Lu, B., Ho, Y.-L., CHAREST, B. R., Wang, D. D., GAGNON, D. R., et al. (2026) · The Lancet Diabetes & Endocrinology · doi:10.1016/s2213-8587(25)00395-x · Cited by 7
Deshpande, A., Schulte, R. J., Walker, R. E., Pallotta, A. M., Wang, M., Kadri, S. S., Haessler, S., Klompas, M., Lindenauer, P. K., & Rothberg, M. B. (2026) · Clinical Infectious Diseases · doi:10.1093/cid/ciag165 · Cited by 6
Ishikawa, Y., Zhao, L., Carrott, P. W., Chang, A. C., Lin, J., Orringer, M. B., Lynch, W. R., Lagisetty, K. H., Wakeam, E., & Reddy, R. M. (2023) · Surgical Endoscopy · doi:10.1007/s00464-023-10191-x · Cited by 6
“The main analysis was to investigate whether GIP significantly reduced post-operative anastomotic leaks after esophagectomy using the R package “PSweight” [15].”
Kamgang, K. S., Rhedin, S., Almqvist, C., & Wintzell, V. (2024) · Thorax · doi:10.1136/thorax-2023-220742 · Cited by 6
Lazarus, B., Kotwal, S. S., Gallagher, M., Gray, N. A., Coggan, S. E., Talaulikar, G. S., Polkinghorne, K. R., & on behalf of the REDUCCTION investigators (2024) · Clinical Journal of the American Society of Nephrology · doi:10.2215/cjn.0000000000000495 · Cited by 6
“35 Overlap and inverse probability of censoring weights were used to compare complication-free catheter survival in two balanced pseudopopulations, 36 to emulate the hypothetical randomized trial in which treatment assignment would be independent of baseline covariates.”
Hong, H., Liu, L., & Stuart, E. A. (2025) · Statistical Methods in Medical Research · doi:10.1177/09622802241307642 · Cited by 6
“Alternatively, recently proposed weighting approaches pursuing exact balance such as the overlap weighting 45,46 and entropy weighting 47 approaches can be considered, and this remains as future work.”
Xu, J., Shuxian, W., linlin, Y., Ling, L., & Chen, W. L. (2021) · Contraception · doi:10.1016/j.contraception.2021.09.011 · Cited by 6
Chinen, T., Sasabuchi, Y., Matsui, H., Yamaguchi, H., & Yasunaga, H. (2022) · BMC Cancer · doi:10.1186/s12885-022-09581-6 · Cited by 6
“Overlap weighting created a pseudo-population that included all the analysed participants by down-weighting at both ends of the propensity score distribution as many overlaps in covariates between the two groups [33].”
MethodClinical outcomes and cumulative healthcare costs of TAVR vs. SAVR in Asia
Tan, E. C.-H., Lee, Y., Kuo, Y. C., Tsao, T., Lee, K.-C., Hsiung, M., Wei, J., Lin, K., & Yin, W. (2022) · Frontiers in Cardiovascular Medicine · doi:10.3389/fcvm.2022.973889 · Cited by 6
Hirano, T., Nakajima, M., Ohbe, H., Kaszynski, R. H., Iwasaki, Y., Arakawa, Y., Sasabuchi, Y., Fushimi, K., Matsui, H., & Yasunaga, H. (2022) · Resuscitation Plus · doi:10.1016/j.resplu.2022.100308 · Cited by 6
“The target of overlap weighting focuses on patients with the most pronounced overlap in observed characteristics and its corresponding estimand is the average effect of treatment in the overlap population.(29) Second, considering the possibility of the minimal involvement of immortal time bias and the effect of the re…”
MethodZerbit, J., Detroit, M., Chevret, S., Pène, F., Luyt, C. E., Ghosn, J., Eyvrard, F., Martin‐Blondel, G., Sarton, B., Clère-Jehl, R., et al. (2022) · Journal of Clinical Medicine · doi:10.3390/jcm11216545 · Cited by 6
“These weightings estimate the average effects of the treatment on different populations: ATE (average treatment effect) on the population corresponding to the sample [26], ATT (average effect in treated), or ATO (on the population balanced in prognostic terms) [27,28].”
BackgroundApenteng, P. N., Virdone, S., Camm, A. J., Fox, K. A. A., Bassand, J., Turpie, A. G. G., Oh, S., Brodmann, M., Cools, F. J. S., Barretto, A. C. P., et al. (2023) · Open Heart · doi:10.1136/openhrt-2023-002275 · Cited by 6
Guha, S., & Reiter, J. P. (2023) · Journal of Statistical Planning and Inference · doi:10.1016/j.jspi.2023.07.004 · Cited by 6
Marozoff, S., Tan, J., Lu, N., Kirmani, A., Loree, J. M., Xie, H. W., Lacaille, D. V., Kopec, J. A., Esdaile, J. M., Corradetti, B., et al. (2023) · ACR Open Rheumatology · doi:10.1002/acr2.11620 · Cited by 6
“Details of the overlap weighting method are provided elsewhere (29,30), but brie fl y, weights are assigned to each participant proportional to their probability of being in the opposite exposure group.”
Method
Metadata
- Authors
- Li, F., & Thomas, L. E.
- Affiliations
- Duke University (US)
- Corresponding author
- Li, Fan
- Venue
- American Journal of Epidemiology
- Publisher
- Oxford University Press (OUP)
- Published
- 2018-09-05
- Volume
- 188
- Issue
- 1
- Pages
- 189-190
- ISSN
- 0002-9262 (ISSN-L)
- 1476-6256
- OpenAlex
- W2907932858
- Cited by
- 694
- References
- 19
- Citations by year
- 2026: 97 · 2025: 124 · 2024: 145 · 2023: 119 · 2022: 83 · 2021: 73 · 2020: 35 · 2019: 15 · 2018: 1
- FWCI
- 25.73
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model (am)
- Locations
- American Journal of Epidemiology (publishedVersion)
- PubMed (publishedVersion)
- Funders
- Agency for Healthcare Research and Quality (SES-1424688)
- National Science Foundation
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Advanced Causal Inference Techniques
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods in Clinical Trials
- Social Sciences › Decision Sciences › Statistics, Probability and Uncertainty › Meta-analysis and systematic reviews
- Keywords
- inverse probability weighting, treatment effect estimation, bias and variance, confidence interval coverage, standard error estimation, causal inference
- MeSH terms
- Epidemiologic Methods, Humans, Models, Statistical, Bias, Causality, Propensity Score
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.