Contrast-specific propensity scores for causal inference with multiple interventions.

Fanwen Meng, Melvin Khee-Shing Leow, Shasha Han, Joel Goh, Donald B Rubin

Journal: Statistical methods in medical research 2024;33(5):825-837

PMID: 38499338

Abstract

Existing methods that use propensity scores for heterogeneous treatment effect estimation on non-experimental data do not readily extend to the case of more than two treatment options. In this work, we develop a new propensity score-based method for heterogeneous treatment effect estimation when there are three or more treatment options, and prove that it generates unbiased estimates. We demonstrate our method on a real patient registry of patients in Singapore with diabetic dyslipidemia. On this dataset, our method generates heterogeneous treatment recommendations for patients among three options: Statins, fibrates, and non-pharmacological treatment to control patients' lipid ratios (total cholesterol divided by high-density lipoprotein level). In our numerical study, our proposed method generated more stable estimates compared to a benchmark method based on a multi-dimensional propensity score.

Address: School of Population Medicine and Public Health, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, China.; NUS Business School, National University of Singapore, Singapore.; Global Asia Institute, National University of Singapore, Singapore.; Institute of Operations Research and Analytics, National University of Singapore, Singapore.; Department of Health Services & Outcomes Research, National Healthcare Group, Singapore.; Cardiovascular & Metabolic Disorders Programme, Duke-NUS Medical School, Singapore.; Department of Endocrinology, Tan Tock Seng Hospital, Singapore.; Lee Kong Chian School of Medicine, Nanyang Technological University, Singapore.; Department of Statistics, Harvard University, Cambridge, MA, USA.; Department of Statistical Science, Fox Business School, Temple University, Philadelphia, PA, USA.; Yau Mathematical Center, Tsinghua University, Beijing, China.

DOI: 10.1177/09622802241236952

Link outs

Free resources

Medical:

https://medlineplus.gov/cholesterolmedicines.html

Subscription / membership required

Full Text Sources:

MeSH Terms: Propensity Score, Humans, Dyslipidemias, Hydroxymethylglutaryl-CoA Reductase Inhibitors, Singapore, Causality, Models, Statistical, Fibric Acids, Hypolipidemic Agents