Propensity Score Analysis in Health Science Research
Article Sidebar
Published:
Feb 1, 2018
Keywords:
propensity score confounding observational study bias
Main Article Content
Abstract
Research in health sciences mainly takes the form of observational studies. The major problem encountered is the management of confounding variables and many covariate variables. This affects the reliability of the data analysis on a number of variables in the model at the same time. Currently, propensity score is a statistical method that is well-known and used increasingly in the design of observational studies. Propensity score has been used to summarize data of several variables as one covariate; the propensity score estimates the probability of covariates based on each individual of received intervention. Furthermore, the propensity score has been used with other analysis techniques such as matching, stratification and regression analysis. The propensity score is not appropriate for poor research or potentially confounding variables. On the other hand, the propensity score is a good tool for good research management.
Article Details
How to Cite
Mongkolsomlit, S., & Homkham, N. (2018). Propensity Score Analysis in Health Science Research. Vajira Medical Journal : Journal of Urban Medicine, 62(1), 33–42. Retrieved from https://he02.tci-thaijo.org/index.php/VMED/article/view/195791
Section
Original Articles
References
1. Austin PC. An Introduction to Propensity Score Methods for Reducing the Effects of Confounding in Observational Studies. Multivariate behavioral research. 2011;46(3):399-424.
2. Olthof DC, Joosse P, Bossuyt PM, de Rooij PP, Leenen LP, Wendt KW, et al. Observation Versus Embolization in Patients with Blunt Splenic Injury After Trauma: A Propensity Score Analysis. World J Surg. 2016;40(5):1264-71.
3. Hur M, Koo CH, Lee HC, Park SK, Kim M, Kim WH, et al. Preoperative aspirin use and acute kidney injury after cardiac surgery: A propensity-score matched observational study. PLoS One. 2017;12(5):e0177201.
4. Trojano M, Pellegrini F, Paolicelli D, Fuiani A, Di Renzo V. observational studies: propensity score analysis of non-randomized data. Int MS J / MS Forum. 2009;16(3):90-7.
5. Wang Z. Propensity score methods to adjust for confounding in assessing treatment effects: bias and precision. Internet J Epidemiol.2008;7(2):1-7.
6. Mongkolsomlit S, Patumanond J, Rawdaree P. How to detect and handle confounding factors. Vajira Med J. 2010;54:223-35.
7. Rossi RJ. Applied Biostatistics for the health sciences: John Wiley & Sons, Inc; 1996.
8. Fitzmaurice G. Confounding: propensity score adjustment. Nutrition. 2006;22 (11-12):1214-6.
9. Sturmer T, Joshi M, Glynn RJ, Avorn J, Rothman KJ, Schneeweiss S. 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):437-47.
10. Rosenbaum PR, Rubin DB. The Central Role of the Propensity Score in Observational Studies for Causal Effects. Biometrika. 1983;70(1):41-55.
11. Marshall MJ, Paul RR. Invited commentary:Propensity score. Am J Epidemiol. 1999;150: 327-33.
12. Mongkolsomlit S, Rawdaree P, Komoltri C, Tawichasri C, Patumanond J. Effect of Angiotensin-Converting Enzyme Inhibitors and/ or Angiotensin Receptor Blockers on the Prevention of Death in Patients with Type 2 Diabetes and Undetermined Nephropathy : Five-Year Survival Data. J Diabetes Metab.2012;3:188.
13. Justus J. Randolph, Falbe K, Manuel AK, Joseph L. Balloun. A Step-by- Step Guide to Propensity Score Matching in R. Practical Assessment, Research & Evaluation. 2014;19(18).
14. Duhamel A, Labreuche J, Gronnier C, Mariette C. Statistical Tools for Propensity Score Matching. Ann Surg. 2017;265(6):E79-E80.
15. Gray E, Pasta DJ, Norris S, O'Leary A, Irish Hepatitis CO, Research N. Effectiveness of triple therapy with direct-acting antivirals for hepatitis C genotype 1 infection: application of propensity score matching in a national HCV treatment registry. BMC Health Serv Res. 2017;17(1):288.
16. Yu HS, Hwang JE, Chung HS, Cho YH, Kim MS, Hwang EC, et al. Is preoperative chronic kidney disease status associated with oncologic outcomes in upper urinary tract urothelial carcinoma? a multicenter propensity score-matched analysis. Oncotarget. 2017;8(39):66540-9.
17. Desai RJ, Rothman KJ, Bateman BT, Hernandez-Diaz S, Huybrechts KF. A Propensity-score-based Fine Stratification Approach for Confounding Adjustment When Exposure Is Infrequent. Epidemiology. 2017;28(2):249-57.
18. Linden A, Adams J, Roberts N. Using propensity scores to construct comparable control groups for disease management program evaluation. Disease Management and Health Outcomes. 2005;13(2):107-15.
19. Linden A. Improving causal inference with a doubly robust estimator that combines propensity score stratification and weighting. J Eval Clin Pract. 2017;23(4):697-702.
20. Senn S, Graf E, Caputo A. Stratification for the propensity score compared with linear regression techniques to assess the effect of treatment or exposure. Stat Med.2007;26(30):5529-44.
21. Bohn J, Eddings W, Schneeweiss S. Conducting Privacy-Preserving Multivariable Propensity Score Analysis When Patient Covariate Information Is Stored in Separate Locations. Am J Epidemiol. 2017;185(6):501-10.
22. Anderson EJ, Carosone-Link P, Yogev R, Yi J, Simoes EAF. Effectiveness of Palivizumab in High-risk Infants and Children: A Propensity Score Weighted Regression Analysis. Pediatr Infect Dis J. 2017;36(8):699-704.
23. Caron C, Wasser T, Eisenberg D. The Use of Propensity Score Matching Does Not Protect Against Regression Artifacts (Regression Towards The Mean). Value Health.2015;18(7):A721.
24. Westreich D, Lessler J, Funk MJ. Propensity score estimation: neural networks, support vector machines, decision trees (CART), and meta-classifiers as alternatives to logistic regression. J Clin Epidemiol. 2010;63(8):826-33.
2. Olthof DC, Joosse P, Bossuyt PM, de Rooij PP, Leenen LP, Wendt KW, et al. Observation Versus Embolization in Patients with Blunt Splenic Injury After Trauma: A Propensity Score Analysis. World J Surg. 2016;40(5):1264-71.
3. Hur M, Koo CH, Lee HC, Park SK, Kim M, Kim WH, et al. Preoperative aspirin use and acute kidney injury after cardiac surgery: A propensity-score matched observational study. PLoS One. 2017;12(5):e0177201.
4. Trojano M, Pellegrini F, Paolicelli D, Fuiani A, Di Renzo V. observational studies: propensity score analysis of non-randomized data. Int MS J / MS Forum. 2009;16(3):90-7.
5. Wang Z. Propensity score methods to adjust for confounding in assessing treatment effects: bias and precision. Internet J Epidemiol.2008;7(2):1-7.
6. Mongkolsomlit S, Patumanond J, Rawdaree P. How to detect and handle confounding factors. Vajira Med J. 2010;54:223-35.
7. Rossi RJ. Applied Biostatistics for the health sciences: John Wiley & Sons, Inc; 1996.
8. Fitzmaurice G. Confounding: propensity score adjustment. Nutrition. 2006;22 (11-12):1214-6.
9. Sturmer T, Joshi M, Glynn RJ, Avorn J, Rothman KJ, Schneeweiss S. 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):437-47.
10. Rosenbaum PR, Rubin DB. The Central Role of the Propensity Score in Observational Studies for Causal Effects. Biometrika. 1983;70(1):41-55.
11. Marshall MJ, Paul RR. Invited commentary:Propensity score. Am J Epidemiol. 1999;150: 327-33.
12. Mongkolsomlit S, Rawdaree P, Komoltri C, Tawichasri C, Patumanond J. Effect of Angiotensin-Converting Enzyme Inhibitors and/ or Angiotensin Receptor Blockers on the Prevention of Death in Patients with Type 2 Diabetes and Undetermined Nephropathy : Five-Year Survival Data. J Diabetes Metab.2012;3:188.
13. Justus J. Randolph, Falbe K, Manuel AK, Joseph L. Balloun. A Step-by- Step Guide to Propensity Score Matching in R. Practical Assessment, Research & Evaluation. 2014;19(18).
14. Duhamel A, Labreuche J, Gronnier C, Mariette C. Statistical Tools for Propensity Score Matching. Ann Surg. 2017;265(6):E79-E80.
15. Gray E, Pasta DJ, Norris S, O'Leary A, Irish Hepatitis CO, Research N. Effectiveness of triple therapy with direct-acting antivirals for hepatitis C genotype 1 infection: application of propensity score matching in a national HCV treatment registry. BMC Health Serv Res. 2017;17(1):288.
16. Yu HS, Hwang JE, Chung HS, Cho YH, Kim MS, Hwang EC, et al. Is preoperative chronic kidney disease status associated with oncologic outcomes in upper urinary tract urothelial carcinoma? a multicenter propensity score-matched analysis. Oncotarget. 2017;8(39):66540-9.
17. Desai RJ, Rothman KJ, Bateman BT, Hernandez-Diaz S, Huybrechts KF. A Propensity-score-based Fine Stratification Approach for Confounding Adjustment When Exposure Is Infrequent. Epidemiology. 2017;28(2):249-57.
18. Linden A, Adams J, Roberts N. Using propensity scores to construct comparable control groups for disease management program evaluation. Disease Management and Health Outcomes. 2005;13(2):107-15.
19. Linden A. Improving causal inference with a doubly robust estimator that combines propensity score stratification and weighting. J Eval Clin Pract. 2017;23(4):697-702.
20. Senn S, Graf E, Caputo A. Stratification for the propensity score compared with linear regression techniques to assess the effect of treatment or exposure. Stat Med.2007;26(30):5529-44.
21. Bohn J, Eddings W, Schneeweiss S. Conducting Privacy-Preserving Multivariable Propensity Score Analysis When Patient Covariate Information Is Stored in Separate Locations. Am J Epidemiol. 2017;185(6):501-10.
22. Anderson EJ, Carosone-Link P, Yogev R, Yi J, Simoes EAF. Effectiveness of Palivizumab in High-risk Infants and Children: A Propensity Score Weighted Regression Analysis. Pediatr Infect Dis J. 2017;36(8):699-704.
23. Caron C, Wasser T, Eisenberg D. The Use of Propensity Score Matching Does Not Protect Against Regression Artifacts (Regression Towards The Mean). Value Health.2015;18(7):A721.
24. Westreich D, Lessler J, Funk MJ. Propensity score estimation: neural networks, support vector machines, decision trees (CART), and meta-classifiers as alternatives to logistic regression. J Clin Epidemiol. 2010;63(8):826-33.