Some Methods of Propensity‐Score Matching had Superior Performance to Others: Results of an Empirical Investigation and Monte Carlo simulations
The DOI matches its Crossref record; 42 of the 42 references it cites were found in a registry or an index.
Abstract
Propensity‐score matching is increasingly being used to reduce the impact of treatment‐selection bias when estimating causal treatment effects using observational data. Several propensity‐score matching methods are currently employed in the medical literature: matching on the logit of the propensity score using calipers of width either 0.2 or 0.6 of the standard deviation of the logit of the propensity score; matching on the propensity score using calipers of 0.005, 0.01, 0.02, 0.03, and 0.1; and 5 → 1 digit matching on the propensity score. We conducted empirical investigations and Monte Carlo simulations to investigate the relative performance of these competing methods. Using a large sample of patients hospitalized with a heart attack and with exposure being receipt of a statin prescription at hospital discharge, we found that the 8 different methods produced propensity‐score matched samples in which qualitatively equivalent balance in measured baseline variables was achieved between treated and untreated subjects. Seven of the 8 propensity‐score matched samples resulted in qualitatively similar estimates of the reduction in mortality due to statin exposure. 5 → 1 digit matching resulted in a qualitatively different estimate of relative risk reduction compared to the other 7 methods. Using Monte Carlo simulations, we found that matching using calipers of width of 0.2 of the standard deviation of the logit of the propensity score and the use of calipers of width 0.02 and 0.03 tended to have superior performance for estimating treatment effects (© 2009 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)
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
“Also, calliper matchings with widths of 0.2, 0.02 and 0.03 have been reported to have superior results [49], which prompted us to use this method.”
Citations
Cites
R: A Language and Environment for Statistical Computing
R Core Team (2025) · doi:10.32614/r.manuals · Cited by 352,181
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
Rosenbaum, P. R., & Rubin, D. B. (1985) · The American Statistician · doi:10.2307/2683903 · Cited by 4,774
Reducing Bias in Observational Studies Using Subclassification on the Propensity Score
Rosenbaum, P. R., & Rubin, D. B. (1984) · Journal of the American Statistical Association · doi:10.1080/01621459.1984.10478078 · Cited by 3,082
A critical appraisal of propensity‐score matching in the medical literature between 1996 and 2003
Austin, P. C. (2008) · Statistics in Medicine · doi:10.1002/sim.3150 · Cited by 1,249
Self-citationAustin, P. C., Grootendorst, P., & Anderson, G. M. (2007) · Statistics in Medicine · doi:10.1002/sim.2580 · Cited by 1,137
Self-citationThe performance of different propensity score methods for estimating marginal hazard ratios
Austin, P. C. (2013) · Statistics in Medicine · doi:10.1002/sim.5705 · Cited by 1,006
Self-citationControlling Bias in Observational Studies: A Review
Cochran, W. G. (2006) · Matched Sampling for Causal Effects · doi:10.1017/cbo9780511810725.005 · Cited by 958
Reducing bias in a propensity score matched-pair sample using greedy matching techniques
Parsons, L. (2001) · Medical Entomology and Zoology · Cited by 733
Austin, P. C. (2007) · The Journal of Thoracic and Cardiovascular Surgery · doi:10.1016/j.jtcvs.2007.07.021 · Cited by 589
Self-citation
Cited by
Items 501–600 of 795
Domb, B. G., Kufta, A. Y., Kingham, Y. E., Sabetian, P. W., Harris, W. T., & Perez-Padilla, P. A. (2025) · The American Journal of Sports Medicine · doi:10.1177/03635465241302806 · Cited by 7
Oncological outcome after radical prostatectomy: Marital status does not make a difference
Schiffmann, J., Beyer, B., Tennstedt, P., Böehm, K., Mehring, G., Schlomm, T., Salomon, G., Karakiewicz, P. I., & Graefen, M. (2015) · International Journal of Urology · doi:10.1111/iju.12717 · Cited by 7
Association Between the Part D Coverage Gap and Adverse Health Outcomes
Polinski, J. M., Shrank, W. H., Glynn, R. J., Huskamp, H. A., Roebuck, M. C., & Schneeweiß, S. (2012) · Journal of the American Geriatrics Society · doi:10.1111/j.1532-5415.2012.04073.x · Cited by 7
Vidi, V. D., Matheny, M. E., Agarwal, V., Arora, N., Donnelly, S. M., Bangalore, S., & Resnic, F. S. (2010) · The American Journal of Cardiology · doi:10.1016/j.amjcard.2010.06.047 · Cited by 7
“[17–19] Vidi et al.”
Miner, S. M., Squires, A. P., Ma, C., McDonald, M. V., & Jones, S. A. (2019) · Home Health Care Management & Practice · doi:10.1177/1084822319865546 · Cited by 7
Sadiq, N., Probst, J. C., Martin, A. B., Khan, M. M., & Merchant, A. T. (2020) · Community Dentistry And Oral Epidemiology · doi:10.1111/cdoe.12612 · Cited by 8
Singh, V., Muthusamy, N., Ikwuazom, C. P., Sicat, C. S., Schwarzkopf, R. M., & Rozell, J. C. (2021) · European Journal of Orthopaedic Surgery & Traumatology · doi:10.1007/s00590-021-03071-4 · Cited by 7
“This method of cohort matching has been established by previous literature as an optimal method for estimating differences between treatment groups [28].”
MethodKim, T.-W., Kim, G., Park, H., Kang, E. K., & Baek, S. (2023) · Journal of Clinical Medicine · doi:10.3390/jcm12196156 · Cited by 7
“The nearest neighbor (NN) with a PS caliper, which imposed a tolerance level on the maximum PS distance, was used to match the two groups to reduce the risk of bad matches when the NN was located far away [35].”
MethodAjayi, A. A., Soria, K. M., Dupont, R., & Varma, K. (2023) · Journal of College Student Retention Research Theory & Practice · doi:10.1177/15210251231218268 · Cited by 8
“A caliper of 0.20 is generally recommended because it can eliminate between 98% and 99% of the bias (Austin, 2009, 2011); however, tighter calipers may be useful when there is a stronger association between the matching variable and outcome variables (Raynor, 1983).”
InfluentialThomas, J., Ashkenazi, I., Lawrence, K. W., Davidovitch, R. I., Rozell, J. C., & Schwarzkopf, R. M. (2024) · The Bone & Joint Journal · doi:10.1302/0301-620x.106b3.bjj-2023-0807.r1 · Cited by 7
Desai, R. J., Varma, V. R., Mahesri, M., Lee, S. B., Freedman, A., Gerhard, T., Segal, J. B., Vine, S. M., Ritchey, M. E., Horton, D. B., et al. (2025) · Clinical Pharmacology & Therapeutics · doi:10.1002/cpt.3583 · Cited by 7
“45,46 To ensure comparability of treatment groups, 47 we evaluated PS distributional overlap before and after matching and balance in each individual covariate between two treatment groups using standardized differences.”
Saito, I., Kushiro, T., Matsushita, Y., Sato, Y., Sagawa, K., Tanaka, Y., Tanigawa, M., & Okutani, Y. (2015) · Clinical and Experimental Hypertension · doi:10.3109/10641963.2015.1047949 · Cited by 7
“We then compared the balance in baseline characteristics between the matched cohort based on standardized differences (12,13).”
ResultTiming of enterostomy closure for neonatal isolated intestinal perforation
Bonasso, P. C., Dassinger, M. S., Mehl, S. C., Gokun, Y., Gowen, M. S., Burford, J. M., & Smith, S. D. (2020) · Journal of Pediatric Surgery · doi:10.1016/j.jpedsurg.2019.12.001 · Cited by 7
Kula, A. J., Qu, P., Strub, B., Smith, J. M., & Menon, S. (2022) · Pediatric Nephrology · doi:10.1007/s00467-021-05348-6 · Cited by 7
“These methods for matching and caliper width were selected based on previous work demonstrating optimal matching characteristics with this technique [15, 16].”
MethodJimenez, A. E., Fox, J. D., Monahan, P. F., Lee, M. S., George, T., Maldonado, D. R., Saks, B. R., Lall, A. C., & Domb, B. G. (2022) · The American Journal of Sports Medicine · doi:10.1177/03635465211062910 · Cited by 7
“The match excluded patients who fell outside of this caliper.(7,8) Statistical significance was indicated by P values \.”
BackgroundObesity as a Risk Factor for Postoperative Adverse Events in Skull Base Surgery
Abiri, A., Goshtasbi, K., Birkenbeuel, J. L., Lin, H. W., Djalilian, H. R., Hsu, F. P. K., & Kuan, E. C. (2022) · Annals of Otology Rhinology & Laryngology · doi:10.1177/00034894221074745 · Cited by 8
Jimenez, A. E., Lee, M. S., George, T., Owens, J. S., Maldonado, D. R., Saks, B. R., Lall, A. C., & Domb, B. G. (2022) · Orthopaedic Journal of Sports Medicine · doi:10.1177/23259671221075642 · Cited by 7
“Previous studies have shown that the optimal method for group comparison is greedy matching without replacement.(1,3,4)”
MethodOwens, J. S., Jimenez, A. E., Lee, M. S., Maldonado, D. R., Lall, A. C., & Domb, B. G. (2022) · Orthopaedic Journal of Sports Medicine · doi:10.1177/23259671221099840 · Cited by 7
“Previous studies have shown that the optimal method for group comparison is greedy matching without replacement.(1,5,6) The groups were matched according to age at the time of surgery, sex, BMI, Tönnis grade, LCEA, follow-up time, sport type, and competition level.”
MethodShakeri, A., Chu, C., Stamenova, V., Fang, J., Barker, L. C., Vigod, S. N., Bhatia, R. S., & Tadrous, M. (2022) · Schizophrenia Bulletin Open · doi:10.1093/schizbullopen/sgac046 · Cited by 9
Chen, Z., Tang, W., Feng, N., Lv, M., Meng, F., Wu, H., Zhao, Y., Xu, H., Dai, Y., Xue, J., et al. (2023) · EBioMedicine · doi:10.1016/j.ebiom.2023.104912 · Cited by 7
“32,33 Finally, the covariate balance between the two groups was evaluated using the absolute standardized mean difference (SMD), wherein a covariate with SMD 10% was considered well-balanced (Supplementary Figure S1).”
MethodOrlov, D., McCluskey, S. A., Callum, J., Rao, V. A., Garijo, J. M., & Karkouti, K. (2016) · Journal of Cardiothoracic and Vascular Anesthesia · doi:10.1053/j.jvca.2016.11.022 · Cited by 7
Fröhlich, H., Raman, N., Täger, T., Schellberg, D., Goode, K. M., Kazmi, S., Grundtvig, M., Hole, T., Cleland, J. G. F., Katus, H. A., et al. (2017) · International Journal of Cardiology · doi:10.1016/j.ijcard.2017.03.028 · Cited by 7
“2 of the standard deviation of the total sample (24).”
BackgroundPropensity-Matching Analysis is Not Straightforward
Chevret, S. (2014) · American Journal of Respiratory and Critical Care Medicine · doi:10.1164/rccm.201406-0996ed · Cited by 7
The Efficacy of Propensity Score Matching in Bias Reduction with Limited Sample Sizes
Howarter, S. (2015) · KU ScholarWorks (The University of Kansas) · Cited by 14
Ashkenazi, I., Thomas, J., Katzman, J. L., Meftah, M., Davidovitch, R. I., & Schwarzkopf, R. M. (2024) · The Journal of Arthroplasty · doi:10.1016/j.arth.2024.02.052 · Cited by 6
Rajahraman, V., Ashkenazi, I., Thomas, J., Bosco, J., Davidovitch, R. I., & Schwarzkopf, R. M. (2024) · The Journal of Arthroplasty · doi:10.1016/j.arth.2024.04.065 · Cited by 6
Impact of a Novel Cost-Saving Pharmacy Program on Pregabalin Use and Health Care Costs
Martin, C. K., O'Dell, K. B., Cappelleri, J. C., Bancroft, T., Halpern, R., & Sadosky, A. (2016) · Journal of Managed Care & Specialty Pharmacy · doi:10.18553/jmcp.2016.15180 · Cited by 6
“2 of the standard deviation of the logit of the propensity score.(31,32)”
BackgroundLau, W. C. Y., Douglas, I., Wong, I. C. K., Smeeth, L. L., Lip, G. Y. H., Leung, W. K., Siu, C., Cheung, B. M. Y., Mok, M. T. C., & Chan, E. W. (2019) · Heart Rhythm · doi:10.1016/j.hrthm.2019.07.034 · Cited by 6
“Each DAPT user was intended to be matched with up to two dabigatran users and two warfarin users by PS using the greedy matching algorithm, which has been demonstrated to perform well in both actual and simulation studies.(17) Patients were eligible for inclusion in every cohort where that treatment was being assessed.”
MethodVeys, R., Abdollah, F., Briganti, A., Albersen, M., Poppel, H. V., & Joniau, S. (2018) · Editor-in-Chief s Voice List of Authors is an Important Element in a Scientific Publication · doi:10.5173/ceju.2017.1611 · Cited by 6
Singh, V., Realyvasquez, J., Simcox, T. G., Lajam, C. M., Schwarzkopf, R. M., & Davidovitch, R. I. (2022) · The Journal of Arthroplasty · doi:10.1016/j.arth.2022.02.081 · Cited by 6
Insani, W. N., Whittlesea, C., Ju, C., Man, K. K. C., Alwafi, H. H., Alsharif, A. A., Chapman, S., & Wei, L. (2022) · British Journal of Clinical Pharmacology · doi:10.1111/bcp.15438 · Cited by 7
“In this study, a greedy matching algorithm was used for the matching process, which has been shown to generate adequate performance.(31,32) The covariates at baseline were age, sex, treatment duration between statin therapy initiation date and index date of ADR or assigned index date, comorbidities (recorded at any ti…”
Xu, Z., Chen, J., Chen, S., Lin, H., Zhao, K., Zheng, C.-Y., Liu, H., Chen, Z., Fu, Y., Hong, Q., et al. (2023) · Frontiers in Oncology · doi:10.3389/fonc.2023.1137836 · Cited by 6
Guo, Y., Wu, J., Liang, L., Zhu, K., Zhou, J., Lin, L., Chen, Y., Cao, B., Mingji, H., Lian, H., et al. (2023) · Cancer Imaging · doi:10.1186/s40644-023-00604-4 · Cited by 6
“The propensity score was calculated by a logistic regression model using a caliper of 0.02 with variables of age, ECOG PS (0/1), Child-Pugh class (A/B), α-fetoprotein (AFP), number of tumors, largest tumor size, tumor distribution (bilobar/unilobar) and macrovascular invasion (yea/ no) [32, 33].”
MethodInfluentialZhai, L., Yang, Y., Zhang, J., Hou, W., Yang, Y., Ding, D., Li, C., & Zhu, Y. (2024) · Journal of Rehabilitation Medicine · doi:10.2340/jrm.v56.18372 · Cited by 6
Xiao, H., Lv, X., Zhou, S., Ren, Q., Zhang, Z., & Wang, X. (2024) · Journal of Cardiothoracic Surgery · doi:10.1186/s13019-024-02630-0 · Cited by 6
Katzman, J. L., Thomas, J., Ashkenazi, I., Lajam, C. M., Rozell, J. C., & Schwarzkopf, R. M. (2024) · The Journal of Arthroplasty · doi:10.1016/j.arth.2024.11.059 · Cited by 6
Wei, M., Chen, S., Li, J., Li, B., Shen, J., Peng, Z., Zhou, Q., Zou, Y., He, X., Li, S., et al. (2020) · Annals of Surgical Oncology · doi:10.1245/s10434-020-08499-2 · Cited by 6
Stark, A. P., Blum, M. M., Chiang, Y., Das, P., Minsky, B. D., Estrella, J. S., Ajani, J. A., Badgwell, B. D., Mansfield, P., & Ikoma, N. (2020) · Journal of the Korean Gastric Cancer Association · doi:10.5230/jgc.2020.20.e29 · Cited by 6
“2*standard deviation, without replacement) [22].”
Ökmen, K., & Köprücüoğlu, M. (2020) · Türk anesteziyoloji ve reanimasyon derneği dergisi · doi:10.5152/tjar.2020.471 · Cited by 6
Mensah, N. A., Madden, E. F., & Qeadan, F. (2021) · Advances in Neonatal Care · doi:10.1097/anc.0000000000000892 · Cited by 6
Mironova, M., Ehrlich, A. C., Grinspan, A. M., & Protano, M. (2022) · Journal of Digestive Diseases · doi:10.1111/1751-2980.13134 · Cited by 7
“The caliper of 0.2 standard deviation was chosen as the fraction of a standard deviation of the logit of the propensity score as recommended by Austin.17 The treatment outcomes were not considered during matching.”
MethodInfluentialSugimura, Y., Kalampokas, N., Arikan, M., Rellecke, P., Dalyanoĝlu, H., Tudorache, I., Westenfeld, R., Boeken, U., Lichtenberg, A., Akhyari, P., et al. (2023) · Interdisciplinary CardioVascular and Thoracic Surgery · doi:10.1093/icvts/ivac289 · Cited by 6
Rennert, L., Ma, Z., McMahan, C. S., & Dean, D. (2023) · PLOS Global Public Health · doi:10.1371/journal.pgph.0001111 · Cited by 7
Hagler, M., & Poon, C. Y. S. (2023) · Journal of Community Psychology · doi:10.1002/jcop.23016 · Cited by 6
Vu, M. M., Galiano, R. D., Souza, J. M., Qin, C. D., & Kim, J. Y. S. (2016) · Hernia · doi:10.1007/s10029-015-1455-5 · Cited by 6
Branson, Z. (2018) · Cited by 7
Desai, R. J., Wyss, R., Abdia, Y., Toh, S., Johnson, M., Lee, H., Karami, S., Major, J. M., Nguyen, M., Wang, S. V., et al. (2019) · Pharmacoepidemiology and Drug Safety · doi:10.1002/pds.4784 · Cited by 6
“2 of the standard deviation on the logit scale.(4) A Cox proportional hazards regression model was used to assess the comparative rate of major bleeding between dabigatran and warfarin initiators.”
MethodA Narrative Review of Methods for Causal Inference and Associated Educational Resources
Landsittel, D., Srivastava, A., & Kropf, K. (2020) · Quality Management in Health Care · doi:10.1097/qmh.0000000000000276 · Cited by 6
Li, M., Chen, C., Yan, Y., Shen, J., Yang, L., Bu, X., Liwei, Z., & Dongmei, C. (2020) · Geriatric Orthopaedic Surgery & Rehabilitation · doi:10.1177/2151459320976531 · Cited by 6
“In order to address this issue clearly, a propensity score-matched analysis(PSMA) is used, which is an after randomization method and a tool to balance the known confounders in the observational cohorts.(16,17) We performed a PSMA in our retrospective cohort study for the patients receiving THA in our hospital from 20…”
MethodKuo, G. C., Lee, T. H., Chen, J., Yen, C., Fan, P., Lee, C., & Chang, C. (2021) · Scientific Reports · doi:10.1038/s41598-021-00177-x · Cited by 6
Sabetian, P. W., Monahan, P. F., Fox, J. D., Jimenez, A. E., Maldonado, D. R., Saks, B. R., Ankem, H. K., Lall, A. C., & Domb, B. G. (2022) · The American Journal of Sports Medicine · doi:10.1177/03635465221078620 · Cited by 6
“After patients were given a respective propensity score, patients outside the caliper (‘‘propensity range’’) were excluded from analysis.(3,4) Matching was based on age at time of surgery, sex, body mass index (BMI), labral treatment, capsule treatment, Tönnis grade, and lateral center-edge angle (LCEA).”
MethodZhu, J., Shaoqin, C., Tao, Z., Li, Y., Sheng, Y., Hong, Q., Hexin, L., Yinan, C., Huangdao, Y., Yifu, C., et al. (2025) · World Journal of Surgical Oncology · doi:10.1186/s12957-025-03672-x · Cited by 5
“(3) Quality of Life: Symptoms such as acid reflux and dysphagia were documented, and their severity was assessed using the Visick grading system [13].”
Jennings, W. G., Richards, T. N., Smith, M. D., Bjerregaard, B. E., & Fogel, S. J. (2015) · Social Science Research · doi:10.1016/j.ssresearch.2015.01.007 · Cited by 5
Wang, H., Wei, Y., Hung, C. T., Jiang, X., Li, C., Jia, K. M., Leung, E. Y. M., Yam, C. H. K., Chow, T. Y., Zhao, S., et al. (2023) · The Lancet Regional Health - Western Pacific · doi:10.1016/j.lanwpc.2023.100716 · Cited by 6
“19,20 The propensity score matching was conducted using the R package MatchIt version 4.3.0.”
InfluentialWhat should be done and what should be avoided when comparing two treatments?
Bouvier, F. B., & Porcher, R. (2023) · Best Practice & Research Clinical Haematology · doi:10.1016/j.beha.2023.101473 · Cited by 5
Ha, J., Choi, S., Moon, S., Han, J., Lee, J., Baek, K., Han, S., Park, S., & Min, C. (2024) · Clinical and Experimental Medicine · doi:10.1007/s10238-024-01368-2 · Cited by 5
“The matching was considered balanced and effective when the SMD for each covariate was less than 0.1 [17].”
BackgroundInfluentialMohammad, H. R., Judge, A., & Murray, D. W. (2024) · Journal of Bone and Joint Surgery · doi:10.2106/jbjs.23.01418 · Cited by 5
“Greedy matching without replacement was utilized on the basis of its superior performance in estimating treatment effects 26 .”
Screpis, D., Berardinis, L. D., Qordja, F., Piovan, G., Giannini, E., Gigante, A. P., & Zorzi, C. (2025) · European Journal of Orthopaedic Surgery & Traumatology · doi:10.1007/s00590-025-04244-1 · Cited by 5
“The method, recognized as optimal for assessing differences between treatment groups, was applied to minimize the potential impact of confounding variables [39].”
The Role of Mobile Phones Use on Agricultural Output and Household Income in Rural Rwanda
Musafiri, I. (2016) · International Journal of ICT Research in Africa and the Middle East · doi:10.4018/ijictrame.2016010104 · Cited by 5
Nam, Y. H., Bilker, W. B., Leonard, C. E., Bell, M. L., & Hennessy, S. (2019) · BMJ Open · doi:10.1136/bmjopen-2018-023809 · Cited by 5
Smith, E. G., Austin, K., Kim, H. M., Miller, D. R., Sauer, B. C., & Valenstein, M. M. (2021) · Journal of Psychiatric Research · doi:10.1016/j.jpsychires.2021.12.011 · Cited by 5
Cox, T., O’Connell, M., Leeuwenkamp, O. R., Palimaka, S., & Reed, N. S. (2022) · Current Medical Research and Opinion · doi:10.1080/03007995.2022.2065146 · Cited by 5
Courchesne‐Krak, N. S., Zúñiga, M. L., Chambers, C., Reed, M. B., Smith, L. R., Ballas, J., & Marienfeld, C. B. (2022) · Journal of Addictive Diseases · doi:10.1080/10550887.2022.2082834 · Cited by 6
Minakata, N., Murano, T., Wakabayashi, M., Sasabe, M., Watanabe, T., Mitsui, T., Yamashita, H., Inaba, A., Sunakawa, H., Nakajo, K., et al. (2023) · World Journal of Gastroenterology · doi:10.3748/wjg.v29.i23.3668 · Cited by 5
Allaire, B. T., Zabala, D., Lines, L. M., Williams, C., Halpern, M. T., & Mollica, M. A. (2023) · Journal of Geriatric Oncology · doi:10.1016/j.jgo.2023.101561 · Cited by 5
Kim, M.-K., Lee, K., Ham, S. Y., Choi, Y. Y., Lee, E., Lee, S., Lee, B., Jeon, J., Chin, B. S., Kim, Y., et al. (2023) · Journal of Korean Medical Science · doi:10.3346/jkms.2023.38.e272 · Cited by 5
Zhu, E., Zhang, L., Liu, Y., Ji, T., Dai, J., Tang, R., Wang, J., Hu, C., Chen, K., Yu, Q., et al. (2024) · Clinical & Translational Oncology · doi:10.1007/s12094-024-03459-8 · Cited by 5
“Covariates were balanced after IPTW with a between-group SMDs smaller than 0.1 [29].”
InfluentialSafety of emergent carotid stenting after thrombolysis: a multicenter retrospective matched analysis
Colò, F., Alexandre, A. M., Brunetti, V., Arba, F., Scarcia, L., Pedicelli, A., Piano, M., Ruggiero, M., Gabrieli, J. D., Ros, V. D., et al. (2025) · Neuroradiology · doi:10.1007/s00234-025-03571-8 · Cited by 5
COST-EFFECTIVENESS OF HUMAN LEUKOCYTE ANTIGEN MATCHING IN PENETRATING KERATOPLASTY
Bäumler, M., Sundmacher, L., Reinhard, T., & Böhringer, D. (2014) · International Journal of Technology Assessment in Health Care · doi:10.1017/s0266462313000603 · Cited by 5
“There are several methods for matching (for an overview see Austin (15;16)).”
MethodShin, J. Y., Yoon, J. K., Shin, A. K., & Diaz, A. Z. (2018) · Head & Neck · doi:10.1002/hed.25185 · Cited by 5
“01, which has demonstrated negligible relative bias (relative bias ranging from -2% to 3%) and greater precision for estimating treatment effects.(14) These 2 cohorts (ie, adjuvant RT vs no adjuvant RT) were compared using the log-rank test, and the HR for OS was calculated using Cox regression.”
MethodPhillips, A. Z., Ahern, J., Kerr, W. C., & Rodríguez, H. P. (2020) · Tobacco Control · doi:10.1136/tobaccocontrol-2020-055976 · Cited by 5
Wu, T., Pouwels, K. B., Welbourn, R., Wordsworth, S., Kent, S., & Wong, C. K. H. (2021) · International Journal of Obesity · doi:10.1038/s41366-021-00887-2 · Cited by 5
“2 indicated optimal balance between the two groups [37].”
BackgroundThe role of bank lenders in firm leverage adjustments
Gao, W., Zhu, F., & Chen, K. (2022) · The Journal of Financial Research · doi:10.1111/jfir.12307 · Cited by 5
Zhang, W., Pang, Y., Cheng, W., Wang, T., Li, Y., Li, X., Zhang, J., Xia, X., Zheng, Y., Zhang, R., et al. (2022) · The Science of The Total Environment · doi:10.1016/j.scitotenv.2022.160174 · Cited by 5
An Evaluation of Latent Growth Models for Propensity Score Matched Groups
Leite, W. L., Sandbach, R., Jin, R., MacInnes, J. W., & Jackman, M. G.-A. (2012) · Structural Equation Modeling A Multidisciplinary Journal · doi:10.1080/10705511.2012.687666 · Cited by 5
“Also, calipers can be used to enforce a maximum distance allowed between two matched observations, but mixed evidence has been found with respect to their effectiveness (Austin, 2009).”
BackgroundRay, I. L., Dabakuyo‐Yonli, T. S., Créhange, G., Bardou, M., Arnould, L., Fraisse, J., Fumoleau, P., Coudert, B. P., Causeret, S., Arveux, P., et al. (2012) · European Journal of Cancer · doi:10.1016/j.ejca.2012.03.020 · Cited by 5
Wang, B., Mak, I. L., Liu, K., Choi, E. P. H., Lam, C. L. K., & Wan, E. Y. F. (2025) · Diabetes & Metabolism · doi:10.1016/j.diabet.2025.101607 · Cited by 4
Banerjee, R., Khan, P. K., Goswami, B., Mahapatra, S., Kammilli, N., Saini, R., Sengupta, S., Jena, M., Rathore, S., Manolla, M. L., et al. (2025) · The Indian Journal of Medical Research · doi:10.25259/ijmr_1050_2025 · Cited by 4
“This study applied the nearest neighbour with replacement PSM method based on the common support 16,17”
Wang, B., Chen, T. Y., Lee, M. H., Chan, K. H., Chan, S. S. C., Wong, I. C. K., & Wan, E. Y. F. (2026) · BMC Medicine · doi:10.1186/s12916-025-04600-2 · Cited by 5
“To ensure balance among the three comparison groups, the fine stratification weighting approach was applied [34, 35].”
Covariate Balance for Observational Effectiveness Studies: A Comparison of Matching and Weighting
Kush, J. M., Pas, E. T., Musci, R. J., & Bradshaw, C. P. (2022) · Journal of Research on Educational Effectiveness · doi:10.1080/19345747.2022.2110545 · Cited by 5
“Austin (2009b, 2011) and others have demonstrated that caliper widths of 0.6 and 0.2 standard deviations of the propensity score remove approximately 90% and 99% of bias due to measured confounders, respectively, by enforcing more similar matches.”
BackgroundInfluentialYang, X., Vladmirovich, R. I., Georgievna, P. M., Agakina, Y. S., He, M., Zeng, Z., Qiang, Y., Cao, Y., & Sergeevich, K. T. (2024) · Frontiers in Medicine · doi:10.3389/fmed.2024.1418800 · Cited by 4
Powell, E. M., Gungabissoon, U., Tazare, J., Smeeth, L. L., Baptiste, P. J., Hammad, T. M. B., Wong, A. Y. S., Douglas, I., & Wing, K. (2024) · PLoS Medicine · doi:10.1371/journal.pmed.1004377 · Cited by 4
He, Z., Ho, M. K. J., Lee, W. Y. V., Law, H. Y., Wong, Y. W. V., Leung, T., Mui, W. H., Wong, S.-T., Wong, C. S. F., & Yam, K. Y. (2023) · Clinical and Translational Radiation Oncology · doi:10.1016/j.ctro.2023.100642 · Cited by 4
“The propensity-score analysis with 1 to 1 nearest matching was performed with consideration of sample size, to balance the independent baseline covariates and the prescribed dose [33,34].”
MethodHwang, J., Moon, J., Kim, K.-Y., Park, S. H., Cho, G. S., Kim, Y. M., Hyung, W. J., & Kim, H. (2025) · European Journal of Surgical Oncology · doi:10.1016/j.ejso.2025.110137 · Cited by 4
Is There a Drawback of Converting a Laparoscopic Colectomy in Colon Cancer?
Leijssen, L. G. J., Dinaux, A. M., Kunitake, H., Bordeianou, L., & Berger, D. L. (2018) · Journal of Surgical Research · doi:10.1016/j.jss.2018.07.052 · Cited by 4
Hollis, M. E., Jennings, W. G., & Hankhouse, S. (2019) · American Journal of Criminal Justice · doi:10.1007/s12103-019-9472-z · Cited by 4
“Consistent with statistical convention (Austin, 2009; Rosenbaum, 2002) and criminological evaluation research that has applied PSM previously (for example, see Jennings et al., Jennings, Richards, Smith, Bjerregaard, & Fogel, 2014, Jennings, Richards, Tomsich, & Gover, 2015, Jennings et al., 2017;…”
BackgroundInfluentialTrinh, T. D., Jorgensen, S. C. J., Zasowski, E. J., Claeys, K. C., Lagnf, A. M., Estrada, S. J., Delaportes, D. J., Huang, V., Klinker, K. P., Kaye, K. S., et al. (2019) · Antimicrobial Agents and Chemotherapy · doi:10.1128/aac.01007-19 · Cited by 4
“1 to indicate meaningful imbalance (60, 61).”
Li, J., Du, Y., Qu, F., Jing, H., Yan, H., & Dang, S. (2022) · Journal of Developmental Origins of Health and Disease · doi:10.1017/s2040174421000714 · Cited by 4
Kurapatti, M., Patel, V., Arraut, J., Oakley, C. T., Rozell, J. C., & Schwarzkopf, R. M. (2022) · Hip International · doi:10.1177/11207000221082251 · Cited by 4
Nagpal, S. V., Scierka, L. E., Castro‐Dominguez, Y., Kansal, D., Kunnirickal, S. J., Hussain, Y., Love, K., Aboian, E., Smolderen, K. G. E., & Mena‐Hurtado, C. (2022) · Catheterization and Cardiovascular Interventions · doi:10.1002/ccd.30405 · Cited by 4
Arraut, J., Thomas, J., Oakley, C. T., Umeh, U. O., Furgiuele, D. L., & Schwarzkopf, R. M. (2023) · Arthroplasty Today · doi:10.1016/j.artd.2023.101155 · Cited by 4
“This method of cohort matching has been established by previous literature as an optimal method for estimating differences between treatment groups [29].”
MethodLuño, J., Varas, J., Ramos, R., Merello, I., Aljama, P., MartinMalo, A., Pascual, J., Praga, M., Aljama, P., Canaud, B., et al. (2017) · Kidney International Reports · doi:10.1016/j.ekir.2017.03.001 · Cited by 4
“2 of the pooled SD of the logit of the propensity scores.(25) To evaluate the quality of the different PSM models, we assessed the balance in covariates based on the absolute difference before and after matching between the groups, and after making the appropriate bivariate comparisons.”
MethodPerformance of propensity score methods when comparison groups originate from different data sources
Hammill, B. G., Curtis, L. H., & Setoguchi, S. (2012) · Pharmacoepidemiology and Drug Safety · doi:10.1002/pds.3251 · Cited by 4
Gayat, É., & Porcher, R. (2011) · Réanimation · doi:10.1007/s13546-011-0422-4 · Cited by 5
“Les performances des différentes méthodes d’appariement sur le score de propension ont fait récemment l’objet d’une étude [17].”
BackgroundLong-Term Determinants of Agricultural Output in Smallholder Farmers in RwandaPreprint
Musafiri, I., & Mirzabaev, A. (2014) · RePEc: Research Papers in Economics · doi:10.22004/ag.econ.169867 · Cited by 4
A finest balancing score algorithm to avoid common pitfalls of propensity score matchingPreprint
Bestehorn, F., Bestehorn, M., Bestehorn, M., & Kirches, C. (2018) · arXiv (Cornell University) · Cited by 4
Chen, M., Zhang, X., Liu, C., Chen, H., Wang, D., & Liu, C. (2023) · Frontiers in Pharmacology · doi:10.3389/fphar.2023.1110653 · Cited by 3
“The caliper of matching was set as 0.03 (Austin, 2009).”
InfluentialWong, C. K. H., Low, M. C. H., Kwok, A. C. Y., Lui, A. Y. C., Lau, K. T. K., Au, I. C. H., Xiong, X., Chung, M. S. H., Kwan, M. Y. W., Lau, E. H. Y., et al. (2022) · Pediatric Drugs · doi:10.1007/s40272-022-00500-7 · Cited by 4
“To minimize outcome bias due to discrepancy in baseline characteristics between the exposure and control, Inverse Probability of Treatment Weights (IPTW) using the propensity score was applied to balance covariates across treatment groups [31].”
MethodHuang, L. Y., Yu, Q., Peng, H., & Zhen, Z. (2022) · Medical Science Monitor · doi:10.12659/msm.937763 · Cited by 3
“Propensity scores were estimated by logistic regression model using the following data as covariates: age, sex, etiology of cirrhosis, Child-Pugh classification, spleen size, diameter of the portal vein system, variceal hemorrhage history, operation, the splenic pedicle division method, operation period, and preoperat…”
MethodYang, G., Chen, H., & Yue, J. (2024) · American Journal of Cancer Research · doi:10.62347/trno3190 · Cited by 4
Metadata
- Authors
- Austin, P. C.
- Affiliations
- University of Toronto (CA)
- Corresponding author
- Austin, Peter C.
- ORCID
- Austin, Peter C.
- Venue
- Biometrical Journal
- Publisher
- Wiley
- Published
- 2009-02
- Volume
- 51
- Issue
- 1
- Pages
- 171-184
- ISSN
- 0323-3847 (ISSN-L)
- 1521-4036
- OpenAlex
- W2016403661
- Cited by
- 769
- References
- 42
- Citations by year
- 2026: 29 · 2025: 50 · 2024: 54 · 2023: 61 · 2022: 78 · 2021: 72 · 2020: 62 · 2019: 48 · 2018: 53 · 2017: 42 · 2016: 48 · 2015: 34 · 2014: 40 · 2013: 33 · 2012: 35
- FWCI
- 18.03
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Biometrical Journal (publishedVersion)
- PubMed (publishedVersion)
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Advanced Causal Inference Techniques
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods in Clinical Trials
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods in Epidemiology
- Keywords
- propensity score matching, treatment selection bias, observational data, statin exposure, myocardial infarction, mortality reduction, method comparison
- MeSH terms
- Algorithms, Biometry, Canada, Data Interpretation, Statistical, Humans, Monte Carlo Method, Myocardial Infarction, Bias, Confounding Factors, Epidemiologic, Incidence, Survival Rate, Proportional Hazards Models, Survival Analysis, Hydroxymethylglutaryl-CoA Reductase Inhibitors, Prescriptions
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.