Adjusting Complex Heterogeneity in Treatment Assignment in Observational Studies
DOI:
https://doi.org/10.6000/1929-6029.2014.03.02.13Keywords:
Heterogeneity, partial least squares, propensity scoreAbstract
Treatment assignment in observational studies is complex and can be influenced by many factors that include patient characteristics, physician practices, and health care systems. These influences can present heterogeneity or clustering effects in the treatment assignment. If those heterogeneity or clustering effects are not appropriately adjusted, the estimated treatment effect may be severely biased. Through a series of models that mimic various level of heterogeneity in treatment assignment in observational studies, we evaluate, through simulation study, the performance of several estimators under the impact of different types of heterogeneity. These estimators include propensity score stratification, propensity score inverse probability weighting, propensity score regression and the partial least squares method. Our results suggest that the partial least squares method is most robust while the dummy variable adjustment method in propensity regression also performs fairly consistently. We use the proposed method to analyze a data set from the German Breast Cancer Study Group study.
References
Renehan AG, Tyson M, Egger M, Heller RF, Zwahlen M. Body-mass index and incidence of cancer: a systematic review and meta-analysis of prospective observational studies. Lancet 2008; 371: 569-78. http://dx.doi.org/10.1016/S0140-6736(08)60269-X DOI: https://doi.org/10.1016/S0140-6736(08)60269-X
Silverman SL. From randomized controlled trials to observational studies. AM J Med 2009; 122: 114-20. http://dx.doi.org/10.1016/j.amjmed.2008.09.030 DOI: https://doi.org/10.1016/j.amjmed.2008.09.030
Thompson S, Ekelund U, Jebb S, Lindroos AK, Mander A, Sharp S, Turner R, Wilks D. A proposed method of bias adjustment for meta-analyses of published observational studies. Int J Epidemiol 2011; 40: 765-77. http://dx.doi.org/10.1093/ije/dyq248 DOI: https://doi.org/10.1093/ije/dyq248
Hair JF, Sarstedt MC, Ringle CM, Mean JA. An assessment of the use of partial least squares structural equation modeling in marketing research. J Acad Mark Sci 2012; 40: 414-33. DOI: https://doi.org/10.1007/s11747-011-0261-6
Kucher N, Koo S, Quiroz R, Cooper J, Paterno M, Soukonnikov B, Goldhaber S. Electronic Alerts to Prevent Venous Thromboembolism among Hospitalized Patients. N Engl J Med 2005; 352: 969-77. http://dx.doi.org/10.1056/NEJMoa041533 DOI: https://doi.org/10.1056/NEJMoa041533
Sharpe N. Clinical Trials and the Real World: Selection Bias and Generalizability of Trial Results. Cardiovasc Drugs Therapy 2002; 16: 75-7. DOI: https://doi.org/10.1023/A:1015327801114
Alemayehu D, Ma J, Jones B, Willke R. Statistical issue with the analysis of non- randomized studies in comparative effectiveness research. J Manag Care Pharm 2011; 17: 22-26. DOI: https://doi.org/10.18553/jmcp.2011.17.s9-a.S22
Willke R, Mullins C. “Ten commandments” for conducting comparative effectiveness research using ”Real-World Data”. J Manag Care Pharm 2011; 17: 10-15. DOI: https://doi.org/10.18553/jmcp.2011.17.s9-a.S10
Rosenbaum R, Rubin D. The central role of the propensity score in observational studies for causal effects. Biometrika 1983; 70: 41-55. DOI: https://doi.org/10.1093/biomet/70.1.41
Stukel T, Fisher E, Wennberg D, Alter D, Gottlieb D, Vermeulen M. Analysis of Observational Studies in the Presence of Treatment Selection Bias. JAMA 2007; 297: 278-85. DOI: https://doi.org/10.1001/jama.297.3.278
Chen Y, Lin J, Yu H, Ko W, Jerng J, Chang W, Chen W, Huang S, Chi N, Wang C, Chen L, Tsai P, Wang S, Hwang J, Lin F. Cardiopulmonary resuscitation with assisted extracorporeal life-support versus conventional cardiopulmonary resuscitation in adults with in-hospital cardiac arrest: an observational study and propensity analysis. Lancet 2008; 372: 554-61. http://dx.doi.org/10.1016/S0140-6736(08)60958-7 DOI: https://doi.org/10.1016/S0140-6736(08)60958-7
Hong J, Yu B. Effects of kindergarten retention on children’s social-emotional de- velopment: An application of propensity score method to multivariate, multilevel data. Dev Psychol 2008; 44: 407-21. http://dx.doi.org/10.1037/0012-1649.44.2.407 DOI: https://doi.org/10.1037/0012-1649.44.2.407
Ye Y, Kaskutas L. Using propensity scores to adjust for selection bias when assessing the effectiveness of Alcoholics Anonymous in observational studies. Drug Alcohol Depend 2009; 104: 56-64. http://dx.doi.org/10.1016/j.drugalcdep.2009.03.018 DOI: https://doi.org/10.1016/j.drugalcdep.2009.03.018
Wyse A, Keesler V, Schneider B. Assessing the effects of small school size on mathematics achievement: A propensity score-matching approach. Teachers College Record 2008; 110: 1879-900. DOI: https://doi.org/10.1177/016146810811000901
Staff J, Patrick M, Loken E, Maggs J. Teenage alcohol use and educational attain- ment. J Stud Alcohol Drugs 2008; 69: 848-58. DOI: https://doi.org/10.15288/jsad.2008.69.848
Maciejewski M, Livingston E, Smith V, Kavee A, Kahwati L, Henderson W, Arter- burn D. Survival Among High-Risk Patients After Bariatric Surgery. JAMA 2011; 305: 2419-26. http://dx.doi.org/10.1001/jama.2011.817 DOI: https://doi.org/10.1001/jama.2011.817
Hoskuldsson A. PLS regression methods. J Chemometr 1988; 2: 211-28. http://dx.doi.org/10.1002/cem.1180020306 DOI: https://doi.org/10.1002/cem.1180020306
Frank I, Friedman J. A Statistical View of Some Chemometrics Regression Tools. Technometrics 1993; 35: 109-35. http://dx.doi.org/10.1080/00401706.1993.10485033 DOI: https://doi.org/10.1080/00401706.1993.10485033
Rosipal R, Kramer N. Overview and Recent Advances in Partial Least Squares. Lecture Notes Comp Sci 2006; 3940: 34-51. DOI: https://doi.org/10.1007/11752790_2
Garthwaite P. An interpretation of partial least squares. J Am Statist Assoc 1994; 89: 122-27. http://dx.doi.org/10.1080/01621459.1994.10476452 DOI: https://doi.org/10.1080/01621459.1994.10476452
Rauschecker H, Sauer R, Schauer A, Schumacher M, Olschewski M, Sauerbrei W, Seegenschmiedt M, Schmoor C. Therapy of small breast cancer-four year results of a prospective non-randomized study. Breast Cancer Res Treat 1995; 34: 1-13. DOI: https://doi.org/10.1007/BF00666486
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: 5529-44. DOI: https://doi.org/10.1002/sim.3133
Chen J, Zhang D, Davidian M. A Monte Carlo EM algorithm for generalized linear mixed models with flexible random effects distribution. Biostatistics 2002; 3: 347-60. DOI: https://doi.org/10.1093/biostatistics/3.3.347
Published
How to Cite
Issue
Section
License
Copyright (c) 2014 Jichang Yu, Haibo Zhou, Xianchen Liu, Fei Zou, Richard J. Willke
This work is licensed under a Creative Commons Attribution 4.0 International License.
Policy for Journals/Articles with Open Access
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are permitted and encouraged to post links to their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work
Policy for Journals / Manuscript with Paid Access
Authors who publish with this journal agree to the following terms:
- Publisher retain copyright .
- Authors are permitted and encouraged to post links to their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work .
