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 101–200 of 795
Abdollah, F., Sun, M., Schmitges, J., Tian, Z., Jeldres, C., Briganti, A., Shariat, S. F., Perrotte, P., Montorsi, F., & Karakiewicz, P. I. (2011) · European Urology · doi:10.1016/j.eururo.2011.06.039 · Cited by 77
Temporary intravascular shunt use improves early limb salvage after extremity vascular injury
Polcz, J. E., White, J. M., Ronaldi, A. E., DuBose, J. J., Grey, S. F., Bell, D., White, P. W., & Rasmussen, T. E. (2020) · Journal of Vascular Surgery · doi:10.1016/j.jvs.2020.08.137 · Cited by 72
The quality‐of‐life benefits of breast reconstruction do not diminish with age
Sisco, M., Johnson, D., Wang, C., Rasinski, K. A., Rundell, V. L. M., & Yao, K. A. (2015) · Journal of Surgical Oncology · doi:10.1002/jso.23864 · Cited by 72
“Women who underwent PMBR were then propensity score matched to women who did not undergo PMBR using the caliper matching method [28].”
MethodPasternak, B., Wintzell, V., Eliasson, B., Svensson, A.-M., Franzén, S., Gudbjörnsdottir, S., Hveem, K., Jonasson, C., Melbye, M., Svanström, H., et al. (2020) · Diabetes Care · doi:10.2337/dc19-2088 · Cited by 69
Propensity Score Matching: should we use it in designing observational studies?
Wan, F. (2025) · BMC Medical Research Methodology · doi:10.1186/s12874-025-02481-w · Cited by 71
“While numerous studies have meticulously examined and validated the properties of PSM through simulations and theoretical investigations [3–9], a more recent study by King and Nielson presents a contrasting view on PSM [10].”
Treatment of Right Heart Thrombi Associated with Acute Pulmonary Embolism
Barrios, D., Chavant, J., Jiménez, D., Bertoletti, L., Rosa‐Salazar, V., Muriel, A., Viallon, A., Fernández‐Capitán, C., Yusen, R. D., Monréal, M., et al. (2016) · The American Journal of Medicine · doi:10.1016/j.amjmed.2016.11.027 · Cited by 72
Zakrison, T. L., Austin, P. C., & McCredie, V. A. (2018) · European Journal of Trauma and Emergency Surgery · doi:10.1007/s00068-017-0786-6 · Cited by 69
“Matching has been the recommended method for pairing control and treated participants in order to attempt to provide unbiased estimation of the treatment effect [16].”
MethodSelf-citationSun, Y., Jin, L., Dian, Y., Shen, M., Zeng, F., Chen, X., & Deng, G. (2023) · EClinicalMedicine · doi:10.1016/j.eclinm.2023.101981 · Cited by 72
Hasegawa, H., Tsukada, Y., Wakabayashi, M., Nomura, S., Sasaki, T., Nishizawa, Y., Ikeda, K., Akimoto, T., & Ito, M. (2020) · International Journal of Colorectal Disease · doi:10.1007/s00384-019-03490-0 · Cited by 68
Abdallah, F. W., MacLean, D. B., Madjdpour, C., Cil, T. D., Bhatia, A., & Brull, R. (2017) · Anesthesia & Analgesia · doi:10.1213/ane.0000000000001975 · Cited by 68
An overview of propensity score matching methods for clustered data
Langworthy, B., Wu, Y., & Wang, M. (2022) · Statistical Methods in Medical Research · doi:10.1177/09622802221133556 · Cited by 67
Automated Surveillance to Detect Postprocedure Safety Signals of Approved Cardiovascular Devices
Resnic, F. S., Gross, T. P., Marinac‐Dabic, D., Loyo‐Berríos, N., Donnelly, S. M., Normand, S. T., & Matheny, M. E. (2010) · JAMA · doi:10.1001/jama.2010.1633 · Cited by 63
“05 using a greedy matching algorithm.(21) The relative imbalance between the exposed and unexposed groups was assessed using the absolute standardized difference (percentile) incovariatemeans and proportions, with values greater than 10% considered severely imbalanced.”
Brachytherapy improves survival in primary vaginal cancer
Orton, A., Boothe, D. L., Williams, N. L., Buchmiller, T., Huang, Y.-H. J., Suneja, G., Poppe, M. M., & Gaffney, D. K. (2016) · Gynecologic Oncology · doi:10.1016/j.ygyno.2016.03.011 · Cited by 64
Prieto‐Alhambra, D., Lalmohamed, A., Abrahamsen, B., Arden, N., de Boer, A., Vestergaard, P., & de Vries, F. (2014) · Arthritis & Rheumatology · doi:10.1002/art.38789 · Cited by 62
Yan, J., Liu, Y., Cao, L., Zheng, Y., Li, W., & Huang, G. (2017) · Nutrients · doi:10.3390/nu9111206 · Cited by 71
Shi, R., Yin, Q., Yuan, Y., Lai, F., & Luo, X. R. (2024) · International Journal of Operations & Production Management · doi:10.1108/ijopm-08-2023-0674 · Cited by 62
Effect of Hydroxyethyl Starch on Postoperative Kidney Function in Patients Having Noncardiac Surgery
Kashy, B. K., Podolyak, A., Makarova, N., Dalton, J. E., Sessler, D. I., & Kurz, A. M. (2014) · Anesthesiology · doi:10.1097/aln.0000000000000375 · Cited by 59
Reitz, K. M., Althouse, A. D., Meyer, J., Arya, S., Goodney, P. P., Shireman, P. K., Hall, D. E., & Tzeng, E. (2021) · JAMA Cardiology · doi:10.1001/jamacardio.2021.3979 · Cited by 59
Kudenchuk, P. J., Newell, C., White, L. L. Y., Fahrenbruch, C. E., Rea, T. D., & Eisenberg, M. S. (2013) · Resuscitation · doi:10.1016/j.resuscitation.2013.05.022 · Cited by 58
Acute Kidney Injury in the Critically Ill: Is Iodinated Contrast Medium Really Harmful?*
Ehrmann, S., Badin, J., Savath, L., Pajot, O., Garot, D., Pham, T. O., Capdevila, X., Perrotin, D., & Lakhal, K. (2013) · Critical Care Medicine · doi:10.1097/ccm.0b013e318275871a · Cited by 58
“2 × sd of the logit of the propensity score) (21).”
BackgroundLee, L., Erkan, A., Alhassan, N. S., Kelly, J. J., Nassif, G. J., Albert, M. R., & Monson, J. R. T. (2018) · Surgical Oncology · doi:10.1016/j.suronc.2018.05.031 · Cited by 61
Machine Learning in Policy Evaluation: New Tools for Causal Inference
Kreif, N., & DiazOrdaz, K. (2019) · Oxford Research Encyclopedia of Economics and Finance · doi:10.1093/acrefore/9780190625979.013.256 · Cited by 57
Khavanin, N., Mlodinow, A. S., Kim, J. Y. S., Halen, J. P. V., Antony, A. K., & Samant, S. (2014) · Annals of Surgical Oncology · doi:10.1245/s10434-014-3785-4 · Cited by 59
Ayad, S. S., Babazade, R., Elsharkawy, H. A., Nadar, V., Lokhande, C., Makarova, N., Khanna, R., Sessler, D. I., & Turan, A. (2016) · PLoS ONE · doi:10.1371/journal.pone.0153675 · Cited by 57
Propensity score methodology for confounding control in health care utilization databases
Patorno, E., Grotta, A., Bellocco, R., & Schneeweiß, S. (2022) · Epidemiology Biostatistics and Public Health · doi:10.2427/8940 · Cited by 60
“Commonly used matching algorithms for 1:1 matching are greedy matching and nearest neighbor matching [47].”
MethodBorah, B. J., Moriarty, J. P., Crown, W. H., & Doshi, J. A. (2013) · Journal of Comparative Effectiveness Research · doi:10.2217/cer.13.89 · Cited by 61
Evaluation of subset matching methods and forms of covariate balance
Resa, M. D. L. A., & Zubizarreta, J. R. (2016) · Statistics in Medicine · doi:10.1002/sim.7036 · Cited by 66
“**We decided to fix the number of treated and control subjects, as in [14], instead of randomly assigning to treatment each unit by means of another Bernoulli random variable, as in [16], to have a fixed number of maximum possible pairs and be able to better compare the number of units matched across simulations.”
MethodValidating the Use of AP Exam Scores for College Course Placement
Patterson, B. F., & Ewing, M. (2013) · Cited by 55
Chen, Y. C., Peng, C., Jeng, W., Chien, R., & Liaw, Y. (2015) · Alimentary Pharmacology & Therapeutics · doi:10.1111/apt.13409 · Cited by 57
“) on the estimated PS of continuing and discontinuing ETV treatment patients.(24)”
BackgroundDomb, B. G., Owens, J. S., Glein, R. M., Jimenez, A. E., & Maldonado, D. R. (2023) · Journal of Bone and Joint Surgery · doi:10.2106/jbjs.22.00340 · Cited by 54
D’Ascenzo, F., Cavallero, E., Biondi‐Zoccai, G., Moretti, C., Omedè, P., Bollati, M., Castagno, D., Modena, M. G., Gaïta, F., & Sheiban, I. (2012) · Journal of Interventional Cardiology · doi:10.1111/j.1540-8183.2012.00753.x · Cited by 54
Prieto‐Alhambra, D., Javaid, M. K., Judge, A., Maskell, J., Cooper, C. C., Arden, N., Mullee, M., Rafferty, J., Carr, A., Price, A., et al. (2014) · Annals of the Rheumatic Diseases · doi:10.1136/annrheumdis-2013-204043 · Cited by 54
Feldman, C. H., Liu, J., Feldman, S., Solomon, D. H., & Kim, S. C. (2016) · Lupus · doi:10.1177/0961203316672928 · Cited by 54
Søgaard, M., Skjøth, F., Jensen, M., Kjældgaard, J. N., Lip, G. Y. H., Larsen, T. B., & Nielsen, P. B. (2019) · Journal of the American Heart Association · doi:10.1161/jaha.118.011358 · Cited by 53
Varma, V. R., Desai, R. J., Navakkode, S., Wong, L., Anerillas, C., Loeffler, T., Schilcher, I., Mahesri, M., Chin, K., Horton, D. B., et al. (2022) · Molecular Psychiatry · doi:10.1038/s41380-022-01912-0 · Cited by 53
“025 on the natural scale of the PS was used to ensure similarity between the matched patients [24].”
MethodAltieri, M. S., Yang, J., Telem, D. A., Zhu, J., Halbert, C. A., Talamini, M. A., & Pryor, A. D. (2015) · Surgical Endoscopy · doi:10.1007/s00464-015-4327-2 · Cited by 52
“20 standard deviations of the logit of the estimated propensity score [14].”
BackgroundMichl, U. H. G., Molfenter, F., Graefen, M., Tennstedt, P., Ahyai, S., Beyer, B., Budäus, L., Haese, A., Hans Heinzer, A., Oh, S. J., et al. (2014) · The Journal of Urology · doi:10.1016/j.juro.2014.08.111 · Cited by 52
Khor, S., Beca, J. M., Krahn, M. D., Hodgson, D. C., Lee, L., Crump, M., Bremner, K. E., Luo, J., Mamdani, M., Bell, C. M., et al. (2014) · BMC Cancer · doi:10.1186/1471-2407-14-586 · Cited by 52
“2 standard deviations of the logit of the propensity score was used [21].”
MethodLemiale, V., Resche‐Rigon, M., Mokart, D., Pène, F., Rabbat, A., Kouatchet, A., Vincent, F., Bruneel, F., Nyunga, M., Lebert, C., et al. (2015) · Annals of Intensive Care · doi:10.1186/s13613-015-0070-z · Cited by 52
“Then, two patients with identical propensity score value but in the two different treatment groups (NIV versus oxygen only) can be considered as comparable, and matching on the propensity score has been shown as one of the most efficient methods for treatment effect assessment [26, 27].”
BackgroundMethodAttributable Risk and Time Course of Colistin-Associated Acute Kidney Injury
Miano, T. A., Lautenbach, E., Wilson, F. P., Guo, W., Borovskiy, Y., & Hennessy, S. (2018) · Clinical Journal of the American Society of Nephrology · doi:10.2215/cjn.06980717 · Cited by 52
Abdallah, F. W., Cil, T. D., MacLean, D. B., Madjdpour, C., Escallón, J., Semple, J. L., & Brull, R. (2018) · Regional Anesthesia & Pain Medicine · doi:10.1097/aap.0000000000000768 · Cited by 52
“2 SDs of the logit of the propensity score, as suggested by Austin.(26) We excluded patients in whom primary outcomes data were missing, or if the prespecified predictors of acute pain data were lacking.”
ResultInfluentialChen, Q., Zhong, Q., Liu, Z.-Y., Li, P., Wang, J., Lin, J., Lu, J., Cao, L., Lin, M., Tu, R., et al. (2021) · Annals of Surgery · doi:10.1097/sla.0000000000004764 · Cited by 52
Lip, G. Y. H., Skjøth, F., Nielsen, P. B., Kjældgaard, J. N., & Larsen, T. B. (2017) · JAMA Cardiology · doi:10.1001/jamacardio.2017.1883 · Cited by 55
Chan, E. W., Lau, W. C. Y., Siu, C. W., Lip, G. Y. H., Leung, W. K., Anand, S., Man, K. K. C., & Wong, I. C. K. (2016) · Heart Rhythm · doi:10.1016/j.hrthm.2016.03.049 · Cited by 51
Kim, K., Lee, Y., Kim, T.-H., Uhm, J., Pak, H., Lee, M., & Joung, B. (2018) · Korean Circulation Journal · doi:10.4070/kcj.2017.0328 · Cited by 54
Benier, K. H. (2017) · International Review of Victimology · doi:10.1177/0269758017693087 · Cited by 60
“This score is then applied using nearest neighbour matching to find the most similar score in the other group (Austin, 2009; King et al., 2011; Stuart, 2010).”
MethodGauss, T., Gayat, É., Harrois, A., Raux, M., Follin, A., Daban, J.-L., Cook, F., Hamada, S., Attias, A., Ausset, S., et al. (2018) · British Journal of Anaesthesia · doi:10.1016/j.bja.2018.02.032 · Cited by 49
“considered to support the assumption of balance between the groups.(27) A Cox proportional hazards regression model was then used to estimate the average treatment effect among the treated [i.”
MethodBauer, A. Z., Gore, R. J., Sama, S. R., Rosiello, R., Garber, L., Sundaresan, D., McDonald, A., Arruda, P., & Kriebel, D. (2020) · Journal of Clinical Hypertension · doi:10.1111/jch.14101 · Cited by 63
Bengesai, A. V., Amusa, L. B., & Makonye, F. (2021) · PLoS ONE · doi:10.1371/journal.pone.0252413 · Cited by 57
The Effects of Coaching on Situational Judgment Tests in High‐stakes Selection
Lievens, F., Buyse, T., Sackett, P. R., & Connelly, B. S. (2012) · International Journal of Selection and Assessment · doi:10.1111/j.1468-2389.2012.00599.x · Cited by 50
“We used the nearest-neighbor matching approach as stratification approaches result in somewhat more distant matches between treatment and control subjects (Austin, 2009).”
BackgroundMethodInfluentialMitzman, B. A., Lutfi, W., Wang, C., Krantz, S. B., Howington, J. A., & Kim, K.-W. (2017) · Seminars in Thoracic and Cardiovascular Surgery · doi:10.1053/j.semtcvs.2017.03.007 · Cited by 48
Maduell, F., Varas, J., Ramos, R., Martín‐Malo, A., Pérez‐García, R., Berdud, I., Moreso, F., Canaud, B., Stuard, S., Gauly, A., et al. (2017) · American Journal of Nephrology · doi:10.1159/000481669 · Cited by 48
Schreyögg, J., Stargardt, T., & Tiemann, O. (2010) · Health Economics · doi:10.1002/hec.1568 · Cited by 49
“A comparison of different calliper widths found that a width of 0.2 of the standard deviation had superior performance for estimating treatment effects in order to reduce conditional bias (Austin, 2009).”
InfluentialXiong, X., Wong, C. K. H., Au, I. C. H., Lai, F. T. T., Li, X., Wan, E. Y. F., Chui, C. S. L., Chan, E. W., Cheng, F. W. T., Lau, K. T. K., et al. (2022) · Thyroid · doi:10.1089/thy.2021.0684 · Cited by 49
“The maximum absolute standardized mean difference was used to assess the difference in each covariate among the three groups, of which a value of less than 0.2 was considered acceptable (23).”
InfluentialFractures in users of antidepressants and anxiolytics and sedatives: effects of age and dose
Vestergaard, P., Prieto‐Alhambra, D., Javaid, M. K., & Cooper, C. C. (2012) · Osteoporosis International · doi:10.1007/s00198-012-2043-5 · Cited by 48
“2 standard deviation calliper, as recommended by some authors [19].”
MethodFröhlich, H., Torres, L., Täger, T., Schellberg, D., Corletto, A., Kazmi, S., Goode, K. M., Grundtvig, M., Hole, T., Katus, H. A., et al. (2017) · Clinical Research in Cardiology · doi:10.1007/s00392-017-1115-0 · Cited by 46
“The beta-blocker agent was the dependent variable [25].”
BackgroundOn the use of propensity scores in case of rare exposure
Hajage, D., Tubach, F., Steg, P. G., Bhatt, D. L., & Rycke, Y. D. (2016) · BMC Medical Research Methodology · doi:10.1186/s12874-016-0135-1 · Cited by 48
“Several methods have previously been described and extensively compared [11–16]: adjustment on PS [8, 12], stratification on PS [11, 17], matching on PS [8, 14, 18], and PS-weighting estimation [15, 19].”
MethodDesai, R. J., Varma, V. R., Gerhard, T., Segal, J., Mahesri, M., Chin, K., Horton, D. B., Kim, S. C., Schneeweiss, S., & Thambisetty, M. (2022) · JAMA Network Open · doi:10.1001/jamanetworkopen.2022.6567 · Cited by 45
Charlton, B. M., Mølgaard‐Nielsen, D., Svanström, H., Wohlfahrt, J., Pasternak, B., & Melbye, M. (2016) · BMJ · doi:10.1136/bmj.h6712 · Cited by 45
What type of clinical evidence is needed to assess medical devices?
Tarricone, R., Boscolo, P. R., & Armeni, P. (2016) · European Respiratory Review · doi:10.1183/16000617.0016-2016 · Cited by 45
Sequential intravenous-to-oral outpatient antibiotic therapy for MRSA bacteraemia: one step closer
Jorgensen, S. C. J., Lagnf, A. M., Bhatia, S., Shamim, M.-D., & Rybak, M. J. (2018) · Journal of Antimicrobial Chemotherapy · doi:10.1093/jac/dky452 · Cited by 45
Cheng, G., Bondeelle, L., Gooley, T. A., He, Q., Jamani, K., Krakow, E. F., Flowers, M. E. D., de Latour, R. P., Michonneau, D., Socié, G., et al. (2019) · Transplantation and Cellular Therapy · doi:10.1016/j.bbmt.2019.10.025 · Cited by 45
“20 standard deviation of the logit of the propensity score, with and without replacement [21].”
BackgroundMaldonado, D. R., Chen, S. L., Walker-Santiago, R., Shapira, J., Rosinsky, P. J., Lall, A. C., & Domb, B. G. (2020) · The American Journal of Sports Medicine · doi:10.1177/0363546519898716 · Cited by 45
“Propensity score matching was performed using RStudio.4,5,17,38 Continuous variables were assessed for normality using the Shapiro-Wilk test and assessed for equal variance using the F test.”
MethodJimenez, A. E., Monahan, P. F., Miecznikowski, K. B., Saks, B. R., Ankem, H. K., Sabetian, P. W., Lall, A. C., & Domb, B. G. (2021) · The American Journal of Sports Medicine · doi:10.1177/03635465211021001 · Cited by 45
A Multivariate Regression Analysis of Panniculectomy Outcomes: Does Plastic Surgery Training Matter?
Mioton, L. M., Buck, D. W., Gart, M. S., Hanwright, P. J., Wang, E., & Kim, J. Y. S. (2013) · Plastic & Reconstructive Surgery · doi:10.1097/prs.0b013e3182818f1f · Cited by 44
Wong, C. K. H., Lau, K. T. K., Au, I. C. H., Xiong, X., Lau, E. H. Y., & Cowling, B. J. (2021) · Clinical Infectious Diseases · doi:10.1093/cid/ciab631 · Cited by 51
Littnerová, S., Jarkovský, J., Pařenica, J., Pavlík, T., Špinar, J., & Dušek, L. (2013) · Cor et Vasa · doi:10.1016/j.crvasa.2013.04.001 · Cited by 44
Jennings, W. G., Park, M., Richards, T. N., Tomsich, E. A., Gover, A. R., & Powers, R. A. (2014) · Child Abuse & Neglect · doi:10.1016/j.chiabu.2014.08.014 · Cited by 46
Maduell, F., Ramos, R., Varas, J., Martín‐Malo, A., Molina, M., Pérez‐García, R., Marcelli, D., Moreso, F., Aljama, P., & Merello, J. I. (2016) · Kidney International · doi:10.1016/j.kint.2016.08.022 · Cited by 44
Matharu, G. S., Judge, A., Murray, D. W., & Pandit, H. G. (2018) · Clinical Orthopaedics and Related Research · doi:10.1007/s11999.0000000000000029 · Cited by 43
“We used greedy matching (that is, each ARMD revision patient was matched to the nearest non-ARMD revision patient) without replacement (once a match is made, that specific patient was no longer available as a potential match for subsequent patients), which has been shown to have superior performance for estimating tre…”
MethodMellado, M., Pijoán, J. I., Jiménez, D., Muriel, A., Aujesky, D., Bertoletti, L., Decousus, H., Barrios, D., Clarà, A. B. I., Yusen, R. D., et al. (2016) · JACC: Cardiovascular Interventions · doi:10.1016/j.jcin.2016.08.039 · Cited by 45
“2 times the SD of the logit of the propensity score (17).”
BackgroundChaudhary, N. S., Donnelly, J. P., Moore, J. X., Baddley, J. W., Safford, M. M., & Wang, H. E. (2017) · Critical Care · doi:10.1186/s13054-017-1767-1 · Cited by 43
“In addition, we repeated the analysis using 1:1 nearest neighbor propensity score matching within fixed caliper width [30].”
MethodNavigating non-positivity in neighbourhood studies: an analysis of collective efficacy and violence
Ahern, J., Cerdá, M., Lippman, S. A., Tardiff, K. J., Vlahov, D., & Galea, S. (2012) · Journal of Epidemiology & Community Health · doi:10.1136/jech-2012-201317 · Cited by 45
“2 SD of the propensity score logit.(46) Callipers 0.”
BackgroundAbdollah, F., Sun, M., Schmitges, J., Thuret, R., Bianchi, M., Shariat, S. F., Briganti, A., Claudio Jeldres, A., Perrotte, P., Montorsi, F., et al. (2012) · The Journal of Urology · doi:10.1016/j.juro.2012.03.005 · Cited by 46
“All available covariates were used to match patients treated with RP to those treated with observation, as previously described.(6,10) The standardized difference measure was used to test how well controls (patients on observation) matched cases (patients with RP).”
Performance of propensity score matching to estimate causal effects in small samples
Andrillon, A., Pirracchio, R., & Chevret, S. (2020) · Statistical Methods in Medical Research · doi:10.1177/0962280219887196 · Cited by 49
“2 of the standard deviation of the PS logit was recommended.(9,10) There is no further consensus regarding the matching procedure especially concerning the use of matching with or without replacement due to the trade-off between bias and variance.”
BackgroundRovers, K. P. B., Bakkers, C., van Erning, F. N., Burger, J. W. A., Nienhuijs, S. W., Simkens, G. A. A. M., Creemers, G., Hemmer, P. H. J., Punt, C. J. A., Lemmens, V. E. P. P., et al. (2020) · JAMA Oncology · doi:10.1001/jamaoncol.2020.2701 · Cited by 43
Aung, Y., Khor, L., Tran, N. V., Akester, M. J., & Zeller, M. (2021) · Aquaculture Economics & Management · doi:10.1080/13657305.2021.2011988 · Cited by 41
“In more recent studies, Austin (2009) suggested to use one-fifth of the standard deviation (caliper ¼ 0.2 sd).”
MethodInfluentialProton Pump Inhibitor Usage and the Risk of Mortality in Hemodialysis Patients
de Francisco, A. L. M., Varas, J., Ramos, R., Merello, J. I., Canaud, B., Stuard, S., Pascual, J., Aljama, P., Aljama, P., Canaud, B., et al. (2017) · Kidney International Reports · doi:10.1016/j.ekir.2017.11.001 · Cited by 41
“2 of the pooled SD of the logit of the propensity scores.(25) After conducting the PSM, there were 1422 unmatched patients, leaving 820 matched patients for statistical analysis (410 in each group).”
MethodMetcalfe, D., Judge, A., Perry, D. C., Gabbe, B. J., Zogg, C. K., & Costa, M. L. (2019) · BMC Musculoskeletal Disorders · doi:10.1186/s12891-019-2590-4 · Cited by 46
Maldonado, D. R., Kyin, C., Walker-Santiago, R., Rosinsky, P. J., Shapira, J., Lall, A. C., & Domb, B. G. (2019) · Hip International · doi:10.1177/1120700019881937 · Cited by 41
Matharu, G. S., Judge, A., Murray, D. W., & Pandit, H. G. (2017) · The Journal of Arthroplasty · doi:10.1016/j.arth.2017.08.036 · Cited by 41
“Primary THAs with TM and non-TM implants were matched for multiple potential patient 161 and surgical confounding factors using propensity score techniques [17, 18].”
MethodPropensity Score Matching for Education Data: Worked Examples
Powell, M. G., Hull, D. M., & Beaujean, A. A. (2020) · The Journal of Experimental Education · doi:10.1080/00220973.2018.1541850 · Cited by 40
“When using a PS matching method, matching is typically done using a one-to-one method without replacement (Austin, 2009).”
MethodLui, D. T. W., Chung, M. S. H., Lau, E. H. Y., Lau, K. T. K., Au, I. C. H., Lee, C., Woo, Y. C., Wong, C. K. H., & Cowling, B. J. (2023) · JAMA Network Open · doi:10.1001/jamanetworkopen.2023.14393 · Cited by 41
Kim, S. C., Glynn, R. J., Liu, J., Everett, B. M., & Goldfine, A. B. (2014) · Acta Diabetologica · doi:10.1007/s00592-014-0663-2 · Cited by 41
“025 on the PS at a fixed ratio of 1:1 [22, 23].”
BackgroundJennings, W. G., Richards, T. N., Smith, M. D., Bjerregaard, B. E., & Fogel, S. J. (2014) · Journal of Criminal Justice · doi:10.1016/j.jcrimjus.2014.05.004 · Cited by 44
Trazodone use and risk of dementia: A population-based cohort study
Brauer, R., Lau, W. C. Y., Hayes, J. F., Man, K. K. C., Osborn, D. P. J., Howard, R., Kim, J., & Wong, I. C. K. (2019) · PLOS Medicine · doi:10.1371/journal.pmed.1002728 · Cited by 35
Alexandre, A. M., Colò, F., Brunetti, V., Valente, I., Frisullo, G., Pedicelli, A., Scarcia, L., Rollo, C., Falcou, A., Milonia, L., et al. (2022) · Journal of NeuroInterventional Surgery · doi:10.1136/jnis-2022-019557 · Cited by 39
Yang, J. Y., Wang, T., Pate, V., Gower, E. W., Crowley, M. J., Buse, J. B., & Stürmer, T. (2019) · Diabetes Obesity and Metabolism · doi:10.1111/dom.13647 · Cited by 39
“Covariate balance was assessed using the standardized mean difference.(34) For all outcomes, we estimated crude incidence rates (IRs) using Poisson regression, and crude and adjusted HR (aHR) and 95% CIs using weighted Cox proportional hazards models.”
MethodKanti, F. S., Alari, A., Chaix, B., & Benmarhnia, T. (2022) · SSRN Electronic Journal · doi:10.2139/ssrn.4173697 · Cited by 38
Bugiardini, R., Dorobanţu, M., Vasiljević, Z., Kedev, S. A., Knežević, B., Miličić, D., Câlmâc, L., Trninić, D., Daullxhiu, I., Cenko, E., et al. (2015) · Atherosclerosis · doi:10.1016/j.atherosclerosis.2015.04.794 · Cited by 36
Santos, R. G., & Santos, R. B. (2015) · Evaluation Review · doi:10.1177/0193841x15602818 · Cited by 34
“…that nonvictimized places near places that have been victimized are much more likely to be victimized, exists for residential burglary (Bowers & Johnson, 2005; Johnson & Bowers, 2004; Townsley, Homel, & Chasel-ing, 2003) and has been shown to occur rapidly (Johnson, Summers, & Pease, 2007, 2009).”
BackgroundWater-pocket endoscopic submucosal dissection for superficial gastric neoplasms (with video)
Harada, H., Murakami, D., Suehiro, S., Nakahara, R., Ujihara, T., Sagami, R., Katsuyama, Y., Hayasaka, K., & Amano, Y. (2018) · Gastrointestinal Endoscopy · doi:10.1016/j.gie.2018.04.2331 · Cited by 36
Maldonado, D. R., Chen, J. W., Yelton, M. J., Rosinsky, P. J., Shapira, J., Brayboy, C., Lall, A. C., & Domb, B. G. (2020) · The American Journal of Sports Medicine · doi:10.1177/0363546520914604 · Cited by 36
Desai, R. J., Mahesri, M., Lee, S. B., Varma, V. R., Loeffler, T., Schilcher, I., Gerhard, T., Segal, J. B., Ritchey, M. E., Horton, D. B., et al. (2022) · Brain Communications · doi:10.1093/braincomms/fcac247 · Cited by 38
Roiss, M., Schiffmann, J., Tennstedt, P., Kessler, T. M., Blanc, I. F., Goetz, A. E., Schlomm, T., Graefen, M., & Reuter, D. A. (2014) · European Journal of Surgical Oncology · doi:10.1016/j.ejso.2014.02.223 · Cited by 36
Tóth, R., Szanto, P., Prodàn, Z., Lex, D. J., Sápi, E., Szatmári, A., Gál, J., Szanto, T., & Székely, A. D. (2013) · Interactive Cardiovascular and Thoracic Surgery · doi:10.1093/icvts/ivt267 · Cited by 38
“the propensity score [11].”
Labour Migration, Food Expenditure, and Household Food Security in Eastern Indonesia
Hasanah, A., Mendolia, S., & Yerokhin, O. (2017) · Economic Record · doi:10.1111/1475-4932.12344 · Cited by 40
Matharu, G. S., Garriga, C., Whitehouse, M. R., Rangan, A., & Judge, A. (2020) · The Journal of Arthroplasty · doi:10.1016/j.arth.2020.04.088 · Cited by 36
Cai, X., Zhang, M., Liang, C., Xu, Y., & Yu, W. (2020) · BMC Surgery · doi:10.1186/s12893-020-00809-5 · Cited by 35
“02 to prevent bias from distant matches [21].”
BackgroundBlood Transfusion and the Risk of Atrial Fibrillation after Cardiac Surgery
Alameddine, A. K., Visintainer, P., Alimov, V. K., & Rousou, J. A. (2014) · Journal of Cardiac Surgery · doi:10.1111/jocs.12383 · Cited by 35
“03.(23) The Wilcoxon rank-sum test was used to identify variables thatwere used to generate propensity scores were balanced.”
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.