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 201–300 of 795
Noji, T., Tsuchikawa, T., Okamura, K., Tanaka, K., Nakanishi, Y., Asano, T., Nakamura, T., Shichinohe, T., & Hirano, S. (2016) · Journal of Hepato-Biliary-Pancreatic Sciences · doi:10.1002/jhbp.363 · Cited by 34
“2 standard deviations of the logit of the estimated propensity score [19].”
BackgroundElectroconvulsive therapy is associated with lower readmission rates in patients with schizophrenia
Ying, Y.-B., Jia, L.-N., Wang, Z.-Y., Jiang, W., Zhang, J., Wang, H., Yang, N.-Q., Wang, R.-W., Ren, Y., Gao, F., et al. (2021) · Brain stimulation · doi:10.1016/j.brs.2021.05.010 · Cited by 34
“14 matching has been used to achieve balance between treatment groups in pretreatment 344 covariates, thereby controlling for confounding bias in estimating treatment effects ([32-36]).”
Balancing Treatment and Control Groups in Quasi‐Experiments: An Introduction to Propensity Scoring
Connelly, B. S., Sackett, P. R., & Waters, S. D. (2013) · Personnel Psychology · doi:10.1111/peps.12020 · Cited by 38
Solomon, D. H., Ruppert, K. M., Zhao, Z., Lian, Y. J., Kuo, I.-H., Greendale, G. A., & Finkelstein, J. S. (2015) · Osteoporosis International · doi:10.1007/s00198-015-3332-6 · Cited by 38
“The propensity score was then used to match a woman in the user group with a nonuser [19].”
MethodWertz, D. A., Chang, C., Sarawate, C. A., Willey, V. J., Cziraky, M. J., & Bohn, R. L. (2010) · Circulation Cardiovascular Quality and Outcomes · doi:10.1161/circoutcomes.109.911461 · Cited by 35
Real-world evidence on sodium-glucose cotransporter-2 inhibitor use and risk of Fournier’s gangrene
Yang, J. Y., Wang, T., Pate, V., Buse, J. B., & Stürmer, T. (2020) · BMJ Open Diabetes Research & Care · doi:10.1136/bmjdrc-2019-000985 · Cited by 35
Pérez‐García, R., Varas, J., Cives, A., Martín‐Malo, A., Aljama, P., Ramos, R., Pascual, J., Stuard, S., Canaud, B., Merello, J. I., et al. (2017) · Nephrology Dialysis Transplantation · doi:10.1093/ndt/gfx269 · Cited by 36
“2 of the pooled SD of the logit of the propensity scores [31].”
BackgroundBrachytherapy Improves Survival in Stage III Endometrial Cancer With Cervical Involvement
Bingham, B. S., Orton, A., Boothe, D. L., Stoddard, G., Huang, Y.-H. J., Gaffney, D. K., & Poppe, M. M. (2017) · International Journal of Radiation Oncology*Biology*Physics · doi:10.1016/j.ijrobp.2016.12.035 · Cited by 32
The effect of anticoagulants on delayed bleeding after colorectal endoscopic submucosal dissection
Harada, H., Nakahara, R., Murakami, D., Suehiro, S., Nagasaka, T., Ujihara, T., Sagami, R., Katsuyama, Y., Hayasaka, K., Tounou, S., et al. (2019) · Surgical Endoscopy · doi:10.1007/s00464-019-07101-5 · Cited by 32
Dragomir, A., Nazha, S., Wood, L. A., Rendon, R., Finelli, A., Hansen, A. R., So, A., Kollmannsberger, C., Basappa, N. S., Pouliot, F., et al. (2020) · Urologic Oncology Seminars and Original Investigations · doi:10.1016/j.urolonc.2020.07.021 · Cited by 32
Kim, S. C., Schneeweiß, S., Liu, J., Karlson, E. W., Katz, J. N., Feldman, S., & Solomon, D. H. (2016) · Arthritis & Rheumatology · doi:10.1002/art.39689 · Cited by 33
“05 on the PS with a fixed ratio of 1:1 in the Medicaid and commercial insurance separately (26,27).”
BackgroundCai, M., Liang, L., Zhang, J., Chen, N., Huang, W., Guo, Y., Hong, X., Lin, L., Liu, Y., Dan, C., et al. (2024) · International Journal of Surgery · doi:10.1097/js9.0000000000001819 · Cited by 31
“To reduce potential selection bias and confounding between the two groups, a 1:1 propensity score matching (PSM) analysis was performed using nearest-neighbor method and a caliper width of 0.2 standard deviation without replacement [31] .”
InfluentialNeoadjuvant Chemoradiation Shows No Survival Advantage to Chemotherapy Alone in Stage IIIA Patients
Krantz, S. B., Mitzman, B. A., Lutfi, W., Kuchta, K. M., Wang, C., Howington, J. A., & Kim, K. W. (2018) · The Annals of Thoracic Surgery · doi:10.1016/j.athoracsur.2017.10.056 · Cited by 31
“A 1-to-1 propensity score matching for nine preoperative patient, hospital, and clinical variables between NCRT and NCT was conducted using the caliper-matching method (Table 1) [4].”
MethodAndersson, N. W., Olsen, R. H., & Andersen, J. T. (2021) · BMJ · doi:10.1136/bmj.n107 · Cited by 31
Qin, C. D., Antony, A. K., Aggarwal, A., Jordan, S. W., GUTOWSKI, K. A., & Kim, J. Y. S. (2015) · Annals of Surgical Oncology · doi:10.1245/s10434-015-4407-5 · Cited by 31
Zhou, W., Huang, C., & Yuan, N. (2018) · International Journal of Surgery · doi:10.1016/j.ijsu.2018.10.038 · Cited by 31
Zheng, C., Fillmore, N. R., Ramos‐Cejudo, J., Brophy, M. T., Osorio, R. S., Gurney, M. E., Qiu, W. Q., Au, R., Perry, G. L., Dubreuil, M., et al. (2021) · Alzheimer s & Dementia · doi:10.1002/alz.12465 · Cited by 30
Mirtazapine exposure in pregnancy and fetal safety: A nationwide cohort study
Ostenfeld, A., Petersen, T. S., Pedersen, L. H., Westergaard, H. B., Løkkegaard, E., & Andersen, J. T. (2022) · Acta Psychiatrica Scandinavica · doi:10.1111/acps.13431 · Cited by 35
Employment effects of hospital privatization in Germany
Heimeshoff, M., Schreyögg, J., & Tiemann, O. (2013) · The European Journal of Health Economics · doi:10.1007/s10198-013-0520-1 · Cited by 30
“This caliper width was identified as the most appropriate in previous studies [32].”
Abdollah, F., Sun, M., Schmitges, J., Thuret, R., Tian, Z., Shariat, S. F., Briganti, A., Claudio Jeldres, A., Perrotte, P., Montorsi, F., et al. (2012) · International Journal of Radiation Oncology*Biology*Physics · doi:10.1016/j.ijrobp.2011.11.034 · Cited by 33
A Closer Examination of the Academic Benefits of AP
McKillip, M. E. M., & Rawls, A. (2013) · The Journal of Educational Research · doi:10.1080/00220671.2012.692732 · Cited by 37
“These differences were defined by Austin (2009) as d = (xtr eatment − xcontr ol )/√((s 2tr ea tment + s 2contr ol )/2), where x is the mean, and s is the standard deviation of continuous variables, or d = (ptr eatment − pcontr ol )/√((pt (1 − pt ) + pc(1 − pc))/2), where p is the percent of nominal…”
BackgroundFrancis, S. R., Orton, A., Thorpe, C. S., Stoddard, G., Lloyd, S. A. J., & Anker, C. J. (2017) · International Journal of Radiation Oncology*Biology*Physics · doi:10.1016/j.ijrobp.2017.06.2457 · Cited by 30
Leite, W. L., Stapleton, L. M., & Bettini, E. A. (2018) · Structural Equation Modeling A Multidisciplinary Journal · doi:10.1080/10705511.2018.1522591 · Cited by 33
“Studies comparing matching strategies (e.g. Austin, 2009, 2014; Cepeda, Boston, Farrar, & Strom, 2003; Diamond & Sekhon, 2013; Gu & Rosenbaum, 1993) have shown that optimal, genetic algorithms or the use of a caliper produce the best covariate balance and reduction in bias in most conditions.”
Overview of Propensity Score Methods
He, H., Hu, J., & He, J. (2016) · ICSA book series in statistics · doi:10.1007/978-3-319-41259-7_2 · Cited by 31
“We may check that covariates are balanced across treatment and control groups, and then analysis can be performed based on new sample [2].”
MethodLau, W. C. Y., Li, X., Wong, I. C. K., Man, K. K. C., Lip, G. Y. H., Leung, W. K., Siu, C., & Chan, E. W. (2017) · Journal of Thrombosis and Haemostasis · doi:10.1111/jth.13780 · Cited by 29
“Dabigatran and warfarin cohorts were 1:1 matched by the greedy matching algorithm, which has been demonstrated to perform well in both actual and simulation studies.(35) Standardized differences were calculated to assess the balance on”
MethodGao, G., Liao, H., Jiang, Q., Liu, D., & Li, T. (2022) · World Journal of Surgical Oncology · doi:10.1186/s12957-022-02778-w · Cited by 29
Smolderen, K. G. E., Romain, G., Provance, J., Scierka, L. E., Mao, J., Goodney, P. P., Henke, P. K., Sedrakyan, A., & Mena‐Hurtado, C. (2023) · JACC: Cardiovascular Interventions · doi:10.1016/j.jcin.2022.09.022 · Cited by 29
Czobor, N. R., Róth, G., Prodàn, Z., Lex, D. J., Sápi, E., Ablonczy, L., Gergely, M., Székely, E., Gál, J., & Székely, A. D. (2017) · Journal of Thoracic Disease · doi:10.21037/jtd.2017.07.88 · Cited by 29
“2 of the SD of the logit of the propensity score (19).”
BackgroundLee, L., Althoff, A. L., Edwards, K., Albert, M. R., Atallah, S. B., Hunter, I. A., Hill, J., & Monson, J. R. T. (2018) · Diseases of the Colon & Rectum · doi:10.1097/dcr.0000000000000962 · Cited by 29
Matharu, G. S., Judge, A., Murray, D. W., & Pandit, H. G. (2018) · Journal of Bone and Joint Surgery · doi:10.2106/jbjs.17.00718 · Cited by 29
Lee, J., Lee, Y., Park, K., Hwang, D.-W., Lee, J. H., & Song, K. B. (2016) · World Journal of Surgery · doi:10.1007/s00268-016-3421-5 · Cited by 28
Gao, H., & Zhang, Z. (2015) · Medicine · doi:10.1097/md.0000000000001245 · Cited by 31
“Then a 1:1 match between the laparoscopy and laparotomy groups was performed by using the nearest available neighbor matching.(23) After matching, data of patient baseline characteristics, operation outcome and incidence of intraoperative and postoperative complication were compared using t test, Chi-square test, or F…”
MethodOutcomes of outborn very-low-birth-weight infants in Japan
Hirata, K., Kimura, T., Hirano, S., Wada, K., Kusuda, S., & Fujimura, M. (2020) · Archives of Disease in Childhood Fetal & Neonatal · doi:10.1136/archdischild-2019-318594 · Cited by 28
Zasowski, E. J., Trinh, T. D., Claeys, K. C., Lagnf, A. M., Bhatia, S., Klinker, K. P., Veve, M. P., Estrada, S., Johns, S. T., Sawyer, A., et al. (2021) · Open Forum Infectious Diseases · doi:10.1093/ofid/ofab606 · Cited by 31
Owens, J. S., Jimenez, A. E., Lee, M. S., Monahan, P. F., Maldonado, D. R., & Domb, B. G. (2022) · Arthroscopy The Journal of Arthroscopic and Related Surgery · doi:10.1016/j.arthro.2022.08.023 · Cited by 28
Mussa, E. C., Palermo, T., Ángeles, G., Kibur, M., Otchere, F., Amhara ISNP Evaluation Team, Gavrilovic, M., Valli, E., Waidler, J., Quiñones, S., et al. (2023) · BMC Health Services Research · doi:10.1186/s12913-023-09024-3 · Cited by 31
Maiolino, G., Pedon, L., Cesari, M., Frigo, A. C., Wolfert, R. L., Barisa, M., Pagliani, L., Rossitto, G., Seccia, T. M., Zanchetta, M., et al. (2012) · PLoS ONE · doi:10.1371/journal.pone.0048171 · Cited by 28
“2 standard deviations of the logit of the propensity score [39].”
BackgroundHauser, R. G., Mugglin, A. S., Friedman, P. A., Kramer, D. B., Kallinen, L. M., McGriff, D. M., & Hayes, D. L. (2012) · Circulation Cardiovascular Quality and Outcomes · doi:10.1161/circoutcomes.111.962621 · Cited by 28
Maldonado, D. R., Laseter, J. R., Kyin, C., Lall, A. C., & Domb, B. G. (2019) · JAAOS Global Research and Reviews · doi:10.5435/jaaosglobal-d-19-00118 · Cited by 28
Täger, T., Fröhlich, H., Grundtvig, M., Seiz, M., Schellberg, D., Goode, K. M., Kazmi, S., Hole, T., Katus, H. A., Atar, D., et al. (2019) · International Journal of Cardiology · doi:10.1016/j.ijcard.2019.01.109 · Cited by 27
“The loop diuretic was the dependent variable [25].”
BackgroundDesai, R. J., Varma, V. R., Gerhard, T., Segal, J., Mahesri, M., Chin, K., Nonnenmacher, E., Gabbeta, A., Mammen, A. M., Varma, S., et al. (2020) · Alzheimer's & Dementia: Translational Research & Clinical Interventions · doi:10.1002/trc2.12095 · Cited by 27
Ekeke, C. N., Kuiper, G. M., Luketich, J. D., Ruppert, K. M., Copelli, S. J., Baker, N., Levy, R. M., Awais, O., Christie, N. A., Dhupar, R., et al. (2022) · Journal of Thoracic and Cardiovascular Surgery · doi:10.1016/j.jtcvs.2022.11.027 · Cited by 27
Li, L., Huang, D., Xu, J., Li, M., Zhao, J., Shi, Q., & Guo, Q. (2023) · BMC Pregnancy and Childbirth · doi:10.1186/s12884-023-05503-x · Cited by 27
“The major strength of our study was the robustness of the study result, which was supported by univariate analysis, multivariate regression analysis adjusted for propensity score, and 1:1 propensity score-matched cohort study [39, 40], the Table 3 Comparison of PE treatment associated with outcomes in AFLP patients”
MethodAlter, D. A., Wijeysundera, H. C., Franklin, B. A., Austin, P. C., Chong, A. S., OH, P. I., Tu, J. V., & Stukel, T. A. (2012) · BMC Health Services Research · doi:10.1186/1472-6963-12-238 · Cited by 34
“2 standard deviation of the logit of the propensity score [32,33].”
BackgroundSelf-citationPrieto‐Alhambra, D., Javaid, M. K., Judge, A., Maskell, J., Kiran, A., Cooper, C. C., & Arden, N. (2010) · Osteoporosis International · doi:10.1007/s00198-010-1368-1 · Cited by 29
“[28]; secondly, using the Matching package in R (available at http://cran.”
MethodSmith, E. G., Austin, K., Kim, H. M., Eisen, S. V., Kilbourne, A. M., Miller, D. R., Zivin, K., Hannemann, C. M., Sauer, B. C., & Valenstein, M. M. (2015) · The British Journal of Psychiatry · doi:10.1192/bjp.bp.113.138685 · Cited by 28
Fröhlich, H., Zhao, J., Täger, T., Cebola, R., Schellberg, D., Katus, H. A., Grundtvig, M., Hole, T., Atar, D., Agewall, S., et al. (2015) · Circulation Heart Failure · doi:10.1161/circheartfailure.114.001701 · Cited by 30
“2 of the SD of the total sample.(23) In a 1-pass procedure starting with a given patient receiving carvedilol, the closest match of a patient receiving metoprolol succinate was identified.”
BackgroundRosemann, T. J. (2016) · International Journal of Integrated Care · doi:10.5334/ijic.2455 · Cited by 36
“Propensity score matching is increasingly being used to reduce the impact of selection bias when estimating causal treatment effects using observational data [19].”
BackgroundLee, L., Wong-Chong, N., Kelly, J. J., Nassif, G. J., Albert, M. R., & Monson, J. R. T. (2018) · Surgical Endoscopy · doi:10.1007/s00464-018-6319-5 · Cited by 27
Rassen, J. A., Solomon, D. H., Glynn, R. J., & Schneeweiß, S. (2011) · Pharmacoepidemiology and Drug Safety · doi:10.1002/pds.2121 · Cited by 32
“We employ a 1:1 greedy matching algorithm, although nearest neighbor matching can be substituted.(11,12) 1:1 greedy matching on propensity score pairs patients, one from each of two exposure groups, at decreasing levels of precision; in a typical configuration, those patients who have propensity scores that are equiva…”
MethodOmichi, K., Shindoh, J., Cloyd, J. M., Mizuno, T., Chun, Y. S., Conrad, C. H., Aloia, T. A., Tzeng, C.-W. D., & Vauthey, J. (2017) · European Journal of Surgical Oncology · doi:10.1016/j.ejso.2017.11.006 · Cited by 26
“unilobar CLM was the dependent variable.[30, 31] The matching algorithm was based on logistic regression and included the following clinically relevant covariates: sex, age, body mass index, primary tumor site (colon or rectum), primary tumor lymph node status, time from primary diagnosis to CLM diagnosis, RAS status,…”
MethodZayed, S., Chen, H., Ali, E., Rodrigues, G., Warner, A., Palma, D. A., & Louie, A. V. (2020) · International Journal of Radiation Oncology*Biology*Physics · doi:10.1016/j.ijrobp.2020.06.008 · Cited by 26
“2 (standard deviation) logit(propensity score), respectively.(22-25) The caliper distance of 0.”
Wong, C. K. H., Wan, E. Y. F., Luo, S., Ding, Y., Lau, E. H. Y., Ling, P., Hu, X., Lau, E. C. H., Wong, J., Zheng, X., et al. (2021) · EClinicalMedicine · doi:10.1016/j.eclinm.2021.100743 · Cited by 27
“2 implied sufficient balance between the groups [12].”
BackgroundAndersson, N. W., Skov, L., & Andersen, J. T. (2021) · JAMA Dermatology · doi:10.1001/jamadermatol.2021.1090 · Cited by 26
Glein, R. M., Jimenez, A. E., Miecznikowski, K. B., Saks, B. R., Ankem, H. K., Sabetian, P. W., Maldonado, D. R., Lall, A. C., & Domb, B. G. (2021) · The American Journal of Sports Medicine · doi:10.1177/03635465211039834 · Cited by 28
“Previous studies have shown this to be the optimal method for propensity score matching.(6) Via the propensity score matching process, 73 elite female athletic hips (68 patients) were matched to 73 elite male athletic hips (66 patients).”
MethodMohammad, H. R., Liddle, A. D., Judge, A., & Murray, D. W. (2021) · The Journal of Arthroplasty · doi:10.1016/j.arth.2021.09.018 · Cited by 26
Quigley, M. R., Bello, N. R., Jho, D. H., Fuhrer, R., Karlovits, S., & Buchinsky, F. J. (2015) · Neurosurgery · doi:10.1227/neu.0000000000000707 · Cited by 26
Young, G. J., Flaherty, S. F., Zepeda, E. D., Mortelé, K. J., & Griffith, J. L. (2020) · Journal of General Internal Medicine · doi:10.1007/s11606-019-05621-3 · Cited by 28
Andersson, N. W., Thomsen, S. F., & Andersen, J. T. (2020) · JAMA Dermatology · doi:10.1001/jamadermatol.2020.0142 · Cited by 26
High-Value Work and the Rise of Women: The Cotton Revolution and Gender Equality in ChinaPreprint
Xue, M. M. (2014) · SSRN Electronic Journal · doi:10.2139/ssrn.2389218 · Cited by 39
Austin, P. C., & Fine, J. P. (2025) · Statistics in Medicine · doi:10.1002/sim.70009 · Cited by 29
Self-citationA cohort study of digoxin exposure and mortality in men with prostate cancer
Flahavan, E. M., Sharp, L., Bennett, K., & Barron, T. I. (2013) · British Journal of Urology · doi:10.1111/bju.12287 · Cited by 26
Smith, E. G., Austin, K., Kim, H. M., Miller, D. R., Eisen, S. V., Christiansen, C. L., Kilbourne, A. M., Sauer, B. C., McCarthy, J. F., & Valenstein, M. M. (2014) · BMC Psychiatry · doi:10.1186/s12888-014-0357-x · Cited by 25
“2 standard deviations of the propensity score logit) [27,28], achieving 98.”
BackgroundChiang, A., Thibault, I., Warner, A., Rodrigues, G., Palma, D. A., Soliman, H., Jain, S., Poon, I., & Cheung, P. (2016) · Radiotherapy and Oncology · doi:10.1016/j.radonc.2015.12.026 · Cited by 26
“10) settings [13].”
Søgaard, M., Nielsen, P. B., Skjøth, F., Kjældgaard, J. N., Coleman, C. I., & Larsen, T. B. (2018) · The American Journal of Medicine · doi:10.1016/j.amjmed.2018.01.041 · Cited by 27
Closed‐form variance estimator for weighted propensity score estimators with survival outcome
Hajage, D., Chauvet, G., Belin, L., Lafourcade, A., Tubach, F., & Rycke, Y. D. (2018) · Biometrical Journal · doi:10.1002/bimj.201700330 · Cited by 31
“…methods have been reported in the literature: adjustment (Austin, Grootendorst, Normand, & Anderson, 2007), stratification (Lunceford & Davidian, 2004; Rosenbaum & Rubin, 1984), matching (Abadie & Imbens, 2016; Austin, 2009; Rubin & Thomas, 1996), and weighting (Austin, 2013; Rosenbaum, 1987).”
Lv, Y., Chen, L., Wu Yu, J., Xiang, Q., Tang, Q. S., Wang, F. D., Cai, H., Zou, K., Wei, Q. D., Zhou, H. Z., et al. (2019) · Journal of Infection and Public Health · doi:10.1016/j.jiph.2019.01.069 · Cited by 28
Carbon Trading Mechanism, Low-Carbon E-Commerce Supply Chain and Sustainable Development
Shen, L. Z., Wang, X., Liu, Q., Wang, Y., Lv, L., & Tang, R. (2021) · Mathematics · doi:10.3390/math9151717 · Cited by 25
“Among such observational studies, scholars tend to choose the Propensity Score Matching (PSM) to overcome the selective bias in causality assessment [60].”
BackgroundAn evaluation of a distributed medical device safety surveillance system: The DELTA network study
Vidi, V. D., Matheny, M. E., Donnelly, S. M., & Resnic, F. S. (2011) · Contemporary Clinical Trials · doi:10.1016/j.cct.2011.02.001 · Cited by 30
“05 of the propensity score was chosen to reduce bias and optimize matching proportion [19].”
MethodYu, E. Y. T., Wan, E. Y. F., Mak, I. L., Chao, D. V. K., Ko, W. W., Leung, M. K. W., Li, Y. C., Liang, J., Luk, W., Wong, M. M., et al. (2023) · JAMA Network Open · doi:10.1001/jamanetworkopen.2023.15064 · Cited by 29
Predialysis Nephrology Care and Costs in Elderly Patients Initiating Dialysis
Stroupe, K. T., Fischer, M. J., Kaufman, J. S., O’Hare, A. M., Sohn, M., Browning, M. M., Huo, Z., & Hynes, D. M. (2011) · Medical Care · doi:10.1097/mlr.0b013e31820192ba · Cited by 24
“Sensitivity Analyses To examine potential biases from nonrandom patient assignment into intensity of predialysis nephrology care groups, we used propensity scores to match patients with no predialysis nephrology visits to those in each of the 3 groups with predialysis nephrology care, using nearest-neighbor matching w…”
MethodMaiolino, G., Pedon, L., Cesari, M., Frigo, A. C., Barisa, M., Rossitto, G., Seccia, T. M., Zanchetta, M., & Rossi, G. P. (2012) · International Journal of Cardiology · doi:10.1016/j.ijcard.2012.09.165 · Cited by 24
Freischlag, K., Ezekian, B., Schroder, P. M., Mulvihill, M. S., Cox, M. L., Hartwig, M. G., & Knechtle, S. J. (2018) · Transplantation · doi:10.1097/tp.0000000000002529 · Cited by 26
“The propensity match used a 1:2 nearest-neighbor matching algorithm to form matched pairs of LLT and single-organ recipients.(20-23) Calipers of width 0.”
MethodDe-escalating chemotherapy for stage II colon cancer?
Fu, J., Wu, L., Ge, C., Xu, T., Li, D., Fu, W., Wang, L., & Du, J. (2019) · Therapeutic Advances in Gastroenterology · doi:10.1177/1756284819867553 · Cited by 25
Yoo, D., Song, K. B., Lee, J. W., Hwang, K., Hong, S., Shin, D., Hwang, D. W., Lee, J. H., Lee, W., Kwon, J., et al. (2020) · Journal of Clinical Medicine · doi:10.3390/jcm9072214 · Cited by 24
Randomization Tests to Assess Covariate Balance When Designing and Analyzing Matched Datasets
Branson, Z. (2021) · Observational Studies · doi:10.1353/obs.2021.0031 · Cited by 30
“For example, a ruleof-thumb is that standardized covariate mean differences of a matched dataset should be below 0.1 (Normand et al., 2001; Austin, 2009b; Zubizarreta, 2012; Resa & Zubizarreta, 2016).”
BackgroundMethodInfluentialJorda, A., Ensle, D., Eser, H., Glötzl, F., Riedl, B., Széll, M., Valipour, A., Zoufaly, A., Wenisch, C., Haider, D., et al. (2024) · Clinical Microbiology and Infection · doi:10.1016/j.cmi.2024.10.026 · Cited by 23
“Standardized mean differences 0.1 were interpreted as appropriate group balance [11].”
InfluentialArraut, J., Lygrisse, K. A., Singh, V., Fiedler, B., Schwarzkopf, R. M., & Rozell, J. C. (2022) · Archives of Orthopaedic and Trauma Surgery · doi:10.1007/s00402-022-04696-8 · Cited by 24
“This method of cohort matching has been established by previous literature as an optimal method for estimating differences between treatment groups [18].”
MethodComparison of Calipers for Matching on the Disease Risk Score
Connolly, J. G., & Gagne, J. J. (2016) · American Journal of Epidemiology · doi:10.1093/aje/kwv302 · Cited by 23
“In a simulation study, Austin (5) showed that PS calipers between 0.005 and 0.03 on the natural scale reduced bias more than a caliper of 0.2 times the standard deviation of the logit of the PS but that the latter yielded lower mean squared error by including more matched pairs in the analysis.”
MethodInfluentialLi, Z., Wang, Y., Ying, X., Shan, F., Wu, Z., Zhang, L., Li, S., Jia, Y., Ren, H., & Ji, J. (2019) · BMC Cancer · doi:10.1186/s12885-019-5283-3 · Cited by 24
Velkers, C., Berger, M., Gill, S. S., Eckenhoff, R. G., Stuart, H. L., Whitehead, M., Austin, P. C., Rochon, P. A., & Seitz, D. P. (2020) · Journal of the American Geriatrics Society · doi:10.1111/jgs.16834 · Cited by 23
Self-citationMaldonado, D. R., Diulus, S. C., Shapira, J., Rosinsky, P. J., Kyin, C., Ankem, H. K., Lall, A. C., & Domb, B. G. (2020) · The American Journal of Sports Medicine · doi:10.1177/0363546520969985 · Cited by 23
“0; R Foundation for Statistical Computing) was used to perform propensity score matching, decreasing the influence of potentially confounding variables and enhancing the study’s power.(2,29) Aside from propensity score matching, all statistical analyses were conducted using Microsoft Excel (Microsoft Corp) and the Rea…”
MethodMatharu, G. S., Shah, A., Hawley, S. J., Johansen, A., Inman, D., Moppett, I. K., Whitehouse, M. R., & Judge, A. (2022) · BMC Medicine · doi:10.1186/s12916-022-02517-8 · Cited by 23
“10 or more for any covariate were suggestive of imbalance [32, 33], with only one covariate (pre-operative cognitive state) having evidence of imbalance (Table 1).”
BackgroundManzi, J. E., Dennis, E. R., Black, G. G., Trauger, N., Quan, T., Mcelheny, K., & Dines, J. S. (2022) · Journal of Orthopaedics · doi:10.1016/j.jor.2022.02.025 · Cited by 22
Maldonado, D. R., Kyin, C., Rosinsky, P. J., Shapira, J., Diulus, S. C., Lall, A. C., & Domb, B. G. (2021) · The American Journal of Sports Medicine · doi:10.1177/03635465211013006 · Cited by 22
“determining differences between groups is greedy matching without replacement.(2)”
BackgroundMethodBariatric Surgery and Breast Cancer Incidence: a Population-Based, Matched Cohort Study
Doumouras, A. G., Lovrics, O., Paterson, J. M., Sutradhar, R., Paszat, L. F., Sivapathasundaram, B., Tarride, J., & Anvari, M. (2022) · Obesity Surgery · doi:10.1007/s11695-022-05946-9 · Cited by 23
“Greedy nearest neighbor matching was done in a one-to-one fashion [28].”
MethodMaldonado, D. R., Kyin, C., Simpson, J. R., Annin, S., Jimenez, A. E., Saks, B. R., Lall, A. C., & Domb, B. G. (2022) · The American Journal of Sports Medicine · doi:10.1177/03635465221091192 · Cited by 22
“Evidence has shown that greedy matching without replacement is the best available method for assessing and determining differences between groups.(2)”
MethodHombrados, J. G., Devisscher, M., & Martínez, M. H. (2015) · Journal of Development Effectiveness · doi:10.1080/19439342.2015.1105850 · Cited by 26
“The selected caliper width of 0.01 is commonly used in the epidemiological and economics literature (see Oakes and Kaufman 2006) and has been assessed as adequate in simulation analysis in Austin (2009).”
MethodInfluentialWan, B., Fang, N., Guan, W., Ding, H., Wang, Y., Ge, X., Liang, H., Li, X., & Zhan, Y. (2019) · Journal of Diabetes Research · doi:10.1155/2019/1341963 · Cited by 23
“03 [30] and 1 : 1 matched pair was utilized in this study.”
MethodZhong, Q., Chen, Q., Xu, Y., Zhao, G., Cai, L., Li, G., Xu, Z., Yan, S., Wu, Z., Xue, F., et al. (2020) · Gastric Cancer · doi:10.1007/s10120-020-01110-3 · Cited by 22
Maldonado, D. R., Kyin, C., Shapira, J., Rosinsky, P. J., Meghpara, M. B., Yelton, M. J., Lall, A. C., & Domb, B. G. (2020) · The American Journal of Sports Medicine · doi:10.1177/0363546520969878 · Cited by 22
“To minimize the potential effect of confounding variables between the groups, propensity scoring techniques were used.(2,18) Patients in the borderline dysplasia group were matched in a 1:3 ratio using greedy matching without replacement to patients in the control group with similar sex, age, body mass index, and foll…”
MethodWu, C., Lei, W., Wang, J., Lin, C., Chen, C., & Wen, S. (2020) · PLoS ONE · doi:10.1371/journal.pone.0242975 · Cited by 22
Maldonado, D. R., Ouyang, V. W., Owens, J. S., Jimenez, A. E., Saks, B. R., Sabetian, P. W., Lall, A. C., & Domb, B. G. (2021) · The American Journal of Sports Medicine · doi:10.1177/03635465211046915 · Cited by 23
“This method has been established in the literature as the optimal method for estimating differences between 2 groups.(3) All propensity score matching was conducted using R (Version 3.”
MethodSun, H., Samra, N. S., Kalakoti, P., Sharma, K., Patra, D. P., Dossani, R. H., Thakur, J. D., Disbrow, E. A., Phan, K. H., Veeranki, S. P., et al. (2017) · World Neurosurgery · doi:10.1016/j.wneu.2017.05.108 · Cited by 21
“2 of the standard deviation of the logit of the propensity score.(38,39,46,47) Using the MatchIt package, we created matched patient cohorts following balancing covariates to minimize the risk of confounding by indication.”
MethodQuezada, A., Muriel, A., Jiménez, D., Bikdeli, B., Aramberri, M., López‐Jiménez, L., Sahuquillo, J. C., Otero, R., Porfidia, A., Monréal, M., et al. (2020) · Thrombosis and Haemostasis · doi:10.1055/s-0040-1710589 · Cited by 21
Saks, B. R., Fox, J. D., Owens, J. S., Maldonado, D. R., Jimenez, A. E., Ankem, H. K., Lall, A. C., & Domb, B. G. (2021) · The American Journal of Sports Medicine · doi:10.1177/03635465211043510 · Cited by 21
Al-Hindi, B., Mohammed, M. A., Mangantig, E., & Martini, N. D. (2023) · Journal of the American Pharmacists Association · doi:10.1016/j.japh.2023.10.010 · Cited by 22
Wray, L. O., Wade, M. J., Beehler, G. P., Hershey, L. A., & Vair, C. L. (2013) · American Journal of Geriatric Psychiatry · doi:10.1016/j.jagp.2013.04.018 · Cited by 22
Fujii, T., Yamada, S., Murotani, K., Sugimoto, H., Hattori, M., Kanda, M., Takami, H., Nakayama, G., Nomoto, S., Fujiwara, M., et al. (2015) · Journal of Gastrointestinal Surgery · doi:10.1007/s11605-015-2980-3 · Cited by 21
“2 standard deviations of the logit of the estimated propensity score.(38) After propensity score matching, multivariable linear regressions for the length of the postoperative hospital stay, length of drain placement, and drainage fluid amylase level on POD 1, 3, and 5 were performed with adjustment for propensity sco…”
MethodYeh, R. W., Chandra, M., McCulloch, C. E., & Go, A. S. (2011) · BMC Medicine · doi:10.1186/1741-7015-9-78 · Cited by 21
“0001 using the STATA program psmatch2 (Leuven and Sianesi, 2003) [43].”
Method
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.