Application of propensity score methods in observational studies_Chinese Journal of Evidence-Based Medicine

Authors：

 LI Guowei ¹ , ZHANG Ling ² , CHEN Yuming ³ , CHEN Yaolong ⁴ , WANG Xiaojie ³ ,  WU Darong ⁵

1. Center for Clinical Epidemiology and Methodology (CCEM), Guangdong Second Provincial General Hospital, Guangzhou 510317, P.R.China;
2. School of Public Health, Capital Medical University, Beijing 100069, P.R.China;
3. School of Public Health, Sun Yat-sen University, Guangzhou 510080, P.R.China;
4. Evidence-Based Medicine Center, School of Basic Medical Sciences, Lanzhou University, Lanzhou 730000, P.R.China;
5. Guangdong Provincial Hospital of Chinese Medicine, Guangdong Provincial Academy of Chinese Medical Sciences & 2nd Affiliated Hospital of Guangzhou University of Chinese Medicine, Guangzhou 510120, P.R.China;

Corresponding author：

LI Guowei, Email: ligw@gd2h.org.cn; WU Darong, Email: darongwu@gzucm.edu.cn

Keywords：

Propensity score; Observational study; Non-randomization

DOI：

10.7507/1672-2531.202010131

Video：

Export PDF Favorites Scan Get Citation

Abstract Full text Figures/Tables Video References Cited by

Propensity score methods belong to an analytical approach by incorporating the measured covariates and mimicking randomization to enhance the comparability between groups, hence reducing the impact of potential confounding in observational studies. Propensity score methods have been increasingly used in observational studies. This paper illustrates the principle and the methods based on the propensity score, in combination with its application in observational studies. It also compares results from propensity score methods with those from multivariable regression and randomized controlled trials. It was found that currently there has been a lack of recommendations for the selection of propensity score methods. Differences may exist when comparing results from propensity score methods with findings from typical regression analyses and randomized controlled trials.

Citation： LI Guowei, ZHANG Ling, CHEN Yuming, CHEN Yaolong, WANG Xiaojie, WU Darong. Application of propensity score methods in observational studies. Chinese Journal of Evidence-Based Medicine, 2021, 21(4): 469-474. doi: 10.7507/1672-2531.202010131 Copy

1.	Murad MH, Asi N, Alsawas M, et al. New evidence pyramid. Evid Based Med, 2016, 21(4): 125-127.
2.	Yao X, Florez ID, Zhang P, et al. Clinical research methods for treatment, diagnosis, prognosis, etiology, screening, and prevention: a narrative review. J Evid Based Med, 2020, 13(2): 130-136.
3.	Rosenbaum PR, Rubin DB. The central role of the propensity score in observational studies for causal effects. Biometrika, 1983, 70(1): 41-55.
4.	Austin PC. An Introduction to Propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behav Res, 2011, 46(3): 399-424.
5.	Austin PC, Grootendorst P, Anderson GM. A comparison of the ability of different propensity score models to balance measured variables between treated and untreated subjects: a Monte Carlo study. Stat Med, 2007, 26(4): 734-753.
6.	Brookhart MA, Schneeweiss S, Rothman KJ, et al. Variable selection for propensity score models. Am J Epidemiol, 2006, 163(12): 1149-1156.
7.	Wang YH, Wintzell V, Ludvigsson JF, et al. Association between proton pump inhibitor use and risk of fracture in children. JAMA Pediatr, 2020, 174(6): 543-551.
8.	De Ridder KA, Pape K, Cuypers K, et al. High school dropout and long-term sickness and disability in young adulthood: a prospective propensity score stratified cohort study (the Young-HUNT study). BMC Public Health, 2013, 13: 941.
9.	Blok RD, de Jonge J, de Koning MA, et al. Propensity score adjusted comparison of pelviperineal morbidity with and without omentoplasty following abdominoperineal resection for primary rectal cancer. Dis Colon Rectum, 2019, 62(8): 952-959.
10.	Hernandez I, Baik SH, Piñera A, et al. Risk of bleeding with dabigatran in atrial fibrillation. JAMA Intern Med, 2015, 175(1): 18-24.
11.	焦明旭, 张晓, 刘迪, 等. 倾向性评分匹配在非随机对照研究中的应用. 中国卫生统计, 2016, 33(2): 350-352.
12.	Austin PC. Optimal caliper widths for propensity-score matching when estimating differences in means and differences in proportions in observational studies. Pharm Stat, 2011, 10(2): 150-161.
13.	黄丽红, 陈峰. 倾向性评分方法及其应用. 中华预防医学杂志, 2019, (7): 752-756.
14.	Rosenbaum PR. Model-based direct adjustment. J Am Stat Associat, 1987, 82(398): 387-394.
15.	Robins JM, Hernán MA, Brumback B. Marginal structural models and causal inference in epidemiology. Epidemiology, 2000, 11(5): 550-560.
16.	Austin PC. The performance of different propensity score methods for estimating marginal hazard ratios. Stat Med, 2013, 32(16): 2837-2849.
17.	Austin PC, Schuster T. The performance of different propensity score methods for estimating absolute effects of treatments on survival outcomes: a simulation study. Stat Methods Med Res, 2016, 25(5): 2214-2237.
18.	Austin PC. The performance of different propensity score methods for estimating marginal odds ratios. Stat Med, 2007, 26(16): 3078-3094.
19.	Austin PC. The performance of different propensity-score methods for estimating relative risks. J Clin Epidemiol, 2008, 61(6): 537-545.
20.	Stürmer T, Joshi M, Glynn RJ, et al. A review of the application of propensity score methods yielded increasing use, advantages in specific settings, but not substantially different estimates compared with conventional multivariable methods. J Clin Epidemiol, 2006, 59(5): 437-447.
21.	Shah BR, Laupacis A, Hux JE, et al. Propensity score methods gave similar results to traditional regression modeling in observational studies: a systematic review. J Clin Epidemiol, 2005, 58(6): 550-559.
22.	Cepeda MS, Boston R, Farrar JT, et al. Comparison of logistic regression versus propensity score when the number of events is low and there are multiple confounders. Am J Epidemiol, 2003, 158(3): 280-287.
23.	Biondi-Zoccai G, Romagnoli E, Agostoni P, et al. Are propensity scores really superior to standard multivariable analysis? Contemp Clin Trials, 2011, 32(5): 731-740.
24.	Li G, Holbrook A, Jin Y, et al. Comparison of treatment effect estimates of non-vitamin K antagonist oral anticoagulants versus warfarin between observational studies using propensity score methods and randomized controlled trials. Eur J Epidemiol, 2016, 31(6): 541-561.
25.	Dahabreh IJ, Sheldrick RC, Paulus JK, et al. Do observational studies using propensity score methods agree with randomized trials? A systematic comparison of studies on acute coronary syndromes. Eur Heart J, 2012, 33(15): 1893-1901.
26.	Zhang Z, Ni H, Xu X. Observational studies using propensity score analysis underestimated the effect sizes in critical care medicine. J Clin Epidemiol, 2014, 67(8): 932-939.
27.	Lonjon G, Boutron I, Trinquart L, et al. Comparison of treatment effect estimates from prospective nonrandomized studies with propensity score analysis and randomized controlled trials of surgical procedures. Ann Surg, 2014, 259(1): 18-25.

1. Murad MH, Asi N, Alsawas M, et al. New evidence pyramid. Evid Based Med, 2016, 21(4): 125-127.
2. Yao X, Florez ID, Zhang P, et al. Clinical research methods for treatment, diagnosis, prognosis, etiology, screening, and prevention: a narrative review. J Evid Based Med, 2020, 13(2): 130-136.
3. Rosenbaum PR, Rubin DB. The central role of the propensity score in observational studies for causal effects. Biometrika, 1983, 70(1): 41-55.
4. Austin PC. An Introduction to Propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behav Res, 2011, 46(3): 399-424.
5. Austin PC, Grootendorst P, Anderson GM. A comparison of the ability of different propensity score models to balance measured variables between treated and untreated subjects: a Monte Carlo study. Stat Med, 2007, 26(4): 734-753.
6. Brookhart MA, Schneeweiss S, Rothman KJ, et al. Variable selection for propensity score models. Am J Epidemiol, 2006, 163(12): 1149-1156.
7. Wang YH, Wintzell V, Ludvigsson JF, et al. Association between proton pump inhibitor use and risk of fracture in children. JAMA Pediatr, 2020, 174(6): 543-551.
8. De Ridder KA, Pape K, Cuypers K, et al. High school dropout and long-term sickness and disability in young adulthood: a prospective propensity score stratified cohort study (the Young-HUNT study). BMC Public Health, 2013, 13: 941.
9. Blok RD, de Jonge J, de Koning MA, et al. Propensity score adjusted comparison of pelviperineal morbidity with and without omentoplasty following abdominoperineal resection for primary rectal cancer. Dis Colon Rectum, 2019, 62(8): 952-959.
10. Hernandez I, Baik SH, Piñera A, et al. Risk of bleeding with dabigatran in atrial fibrillation. JAMA Intern Med, 2015, 175(1): 18-24.
11. 焦明旭, 张晓, 刘迪, 等. 倾向性评分匹配在非随机对照研究中的应用. 中国卫生统计, 2016, 33(2): 350-352.
12. Austin PC. Optimal caliper widths for propensity-score matching when estimating differences in means and differences in proportions in observational studies. Pharm Stat, 2011, 10(2): 150-161.
13. 黄丽红, 陈峰. 倾向性评分方法及其应用. 中华预防医学杂志, 2019, (7): 752-756.
14. Rosenbaum PR. Model-based direct adjustment. J Am Stat Associat, 1987, 82(398): 387-394.
15. Robins JM, Hernán MA, Brumback B. Marginal structural models and causal inference in epidemiology. Epidemiology, 2000, 11(5): 550-560.
16. Austin PC. The performance of different propensity score methods for estimating marginal hazard ratios. Stat Med, 2013, 32(16): 2837-2849.
17. Austin PC, Schuster T. The performance of different propensity score methods for estimating absolute effects of treatments on survival outcomes: a simulation study. Stat Methods Med Res, 2016, 25(5): 2214-2237.
18. Austin PC. The performance of different propensity score methods for estimating marginal odds ratios. Stat Med, 2007, 26(16): 3078-3094.
19. Austin PC. The performance of different propensity-score methods for estimating relative risks. J Clin Epidemiol, 2008, 61(6): 537-545.
20. Stürmer T, Joshi M, Glynn RJ, et al. A review of the application of propensity score methods yielded increasing use, advantages in specific settings, but not substantially different estimates compared with conventional multivariable methods. J Clin Epidemiol, 2006, 59(5): 437-447.
21. Shah BR, Laupacis A, Hux JE, et al. Propensity score methods gave similar results to traditional regression modeling in observational studies: a systematic review. J Clin Epidemiol, 2005, 58(6): 550-559.
22. Cepeda MS, Boston R, Farrar JT, et al. Comparison of logistic regression versus propensity score when the number of events is low and there are multiple confounders. Am J Epidemiol, 2003, 158(3): 280-287.
23. Biondi-Zoccai G, Romagnoli E, Agostoni P, et al. Are propensity scores really superior to standard multivariable analysis? Contemp Clin Trials, 2011, 32(5): 731-740.
24. Li G, Holbrook A, Jin Y, et al. Comparison of treatment effect estimates of non-vitamin K antagonist oral anticoagulants versus warfarin between observational studies using propensity score methods and randomized controlled trials. Eur J Epidemiol, 2016, 31(6): 541-561.
25. Dahabreh IJ, Sheldrick RC, Paulus JK, et al. Do observational studies using propensity score methods agree with randomized trials? A systematic comparison of studies on acute coronary syndromes. Eur Heart J, 2012, 33(15): 1893-1901.
26. Zhang Z, Ni H, Xu X. Observational studies using propensity score analysis underestimated the effect sizes in critical care medicine. J Clin Epidemiol, 2014, 67(8): 932-939.
27. Lonjon G, Boutron I, Trinquart L, et al. Comparison of treatment effect estimates from prospective nonrandomized studies with propensity score analysis and randomized controlled trials of surgical procedures. Ann Surg, 2014, 259(1): 18-25.

Chinese Journal of Evidence-Based Medicine

Application of propensity score methods in observational studies

Abstract Full text Figures/Tables Video References Cited by

Previous Article

Next Article

Format

Content