Free access

Research Article

21 June 2018

Application and comparison of generalized propensity score matching versus pairwise propensity score matching

Authors: Zhanglin L Cui [email protected], Lisa M Hess, Robert Goodloe, and Doug FariesAuthor Info & Affiliations

Publication: J. Comp. Eff. Res.

Volume 7, Number 9

https://doi.org/10.2217/cer-2018-0030

PDF OtherFormats

Abstract

Aim: A comparison of conventional pairwise propensity score matching (PSM) and generalized PSM method was applied to the comparative effectiveness of multiple treatment options for lung cancer. Materials & methods: Deidentified data were analyzed. Covariate balances between compared treatments were assessed before and after PSM. Cox proportional hazards regression compared overall survival after PSM. Results & conclusion: The generalized PSM analyses were able to retain 61.2% of patients, while the conventional PSM analyses were able to match from 24.1 to 77.1% of patients from each treatment comparison. The generalized PSM achieved statistical significance (p < 0.05) in 8/10 comparisons, whereas conventional pairwise PSM achieved 1/10. The noted differences arose from different matched patient samples and the size of the samples.

Supplementary Material

File (suppl_materials.docx)

Download
766.04 KB

References

1.

Rosenbaum PR, Rubin DB. The central role of the propensity score in observational studies for causal effects. Biometrika 70(1), 41–55 (1983).

2.

Stürmer T, Wyss R, Glynn RJ, Brookhart MA. Propensity scores for confounder adjustment when assessing the effects of medical interventions using nonexperimental study designs. J. Intern. Med. 275(6), 570–580 (2014).

3.

Rosenbaum PR. Observational Studies (2nd Edition). Springer-Verlag, NY, USA (2002).

4.

Austin PC. An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behav. Res. 46(3), 399–424 (2011).

5.

Rassen JA, Shelat AA, Franklin JM, Glynn RJ, Solomon DH, Schneeweiss S. Matching by propensity score in cohort studies with three treatment groups. Epidemiology 24(3), 401–409 (2013).

6.

Yang S, Imbens GW, Cui Z, Faries DE, Kadziola Z. Propensity score matching and subclassification in observational studies with multi-level treatments. Biometrics 72(4), 1055–1065 (2016).

7.

Siegel RL, Miller KD, Jemal A. Cancer statistics, 2015. CA Cancer J. Clin. 65(1), 5–29 (2015).

8.

Karve SJ, Price GL, Davis KL, Pohl GM, Smyth EN, Bowman L. Comparison of demographics, treatment patterns, health care utilization, and costs among elderly patients with extensive-stage small cell and metastatic non-small-cell lung cancers. BMC Health Serv. Res. 14(1), 555 (2014).

9.

Davis KL, Goyal RK, Able SL, Brown J, Li L, Kaye JA. Real-world treatment patterns and costs in a US Medicare population with metastatic squamous non-small-cell lung cancer. Lung Cancer 87(2), 176–185 (2015).

10.

NCCN. NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines) Version 2.2017. www.nccn.org/professionals/physician_gls/pdf/nscl.pdf.

11.

Hess LM, Michael D, Mytelka DS, Beyrer J, Liepa AM, Nicol S. Chemotherapy treatment patterns, costs, and outcomes of patients with gastric cancer in the United States: a retrospective analysis of electronic medical record (EMR) and administrative claims data. Gastric Cancer 19(2), 607–615 (2016).

12.

Mealing NM, Dobbins TA, Pearson SA. Validation and application of a death proxy in adult cancer patients. Pharmacoepidemiol. Drug Saf. 21(7), 742–748 (2012).

13.

Wong SL, Ricketts K, Royle G, Williams M, Mendes R. A methodology to extract outcomes from routine healthcare data for patients with locally advanced non-small-cell lung cancer. BMC Health Serv. Res. 18(1), 278 (2018).

14.

Wade RL, Korytowsky B, Singh P, Dev P, Bobiak S, Cariola P. The novel generation and validation of survival curves in oncology utilizing electronic medical records linked to point of service claims data. Value Health 20(9), A421 (2017).

15.

de Rooij M. Transitional modeling of experimental longitudinal data with missing values. Adv. Data Anal. Classif. 12(1), 107–130 (2018).

16.

Oken MM, Creech RH, Tormey DC et al. Toxicity and response criteria of the Eastern Cooperative Oncology Group. Am. J. Clin. Oncol. 5(6), 649–655 (1982).

17.

Rosenbaum PR. Observational Studies. Springer, NY, USA (1995).

18.

Austin PC. Optimal caliper widths for propensity-score matching when estimating differences in means and differences in proportions in observational studies. Pharm. Stat. 10(2), 150–161 (2011).

19.

Normand ST, Landrum MB, Guadagnoli E et al. Validating recommendations for coronary angiography following acute myocardial infarction in the elderly: a matched analysis using propensity scores. J. Clin. Epidemiol. 54(4), 387–398 (2001).

20.

Crump R, Hotz VJ, Imbens G, Mitnik O. Dealing with limited overlap in estimation of average treatment effects. Biometrika 96, 187–199 (2009).

21.

Austin PC, Stuart EA. Moving towards best practice when using inverse probability of treatment weighting (IPTW) using the propensity score to estimate causal treatment effects in observational studies. Stat. Med. 34(28), 3661–3679 (2015).

22.

Bang H, Robins JM. Doubly robust estimation in missing data and causal inference models. Biometrics 61(4), 962–973 (2005).

|