AN EXTENSION OF ESTIMATING EQUATIONS TO MODEL LONGITUDINAL MEDICAL COST TRAJECTORY WITH MEDICARE CLAIMS DATA LINKED TO SEER CANCER REGISTRY

• 成果类型: Article
• 署名作者: Wang, Shikun; Ning, Jing; Xu, Ying; Shih, Ya-Chen Tina; Shen, Yu; Li, Liang
• 署名单位: Columbia University; University of Texas System; UTMD Anderson Cancer Center; University of Texas System; UTMD Anderson Cancer Center
• 刊物名称: ANNALS OF APPLIED STATISTICS
• ISSN/ISSBN: 1932-6157
• DOI: 10.1214/22-AOAS1659
• 发表日期: 2023
• 页码: 881-899
• 关键词: regression; spline; care; asymptotics; inference; survival

摘要

Insurance claims' data is an increasingly important health policy research resource, given its longitudinal assessment of cancer care clinical outcomes. Population-level information on medical cost trajectory from disease diagno-sis to terminal events, such as death, specifically interests policy makers. Es-timating the mean cost trajectory has statistical challenges. The shape of the trajectory is usually highly nonlinear with varying durations, depending on the diagnosis-to-death population time distribution. The terminal event may be right censored, resulting in missing subsequent costs. Medical costs often have skewed distributions with zero inflation and heteroscedasticity which may not fit well with the commonly used parametric family of distributions. In this paper we propose a flexible semiparametric model to address chal-lenges without imposing a cost data distributional assumption. The estimation procedure is based on generalized estimating equations with censored covari-ates. The proposed model adopts a bivariate surface that quantifies the interre-lationship between longitudinal medical costs and survival, and results in the nonlinear population mean cost trajectories given survival time. We develop a novel generalized estimating equations algorithm to accommodate covariates subject to right censoring without fully specifying the joint distribution of the cost and survival data. We provide theoretical and simulation-based justifi-cation for the proposed approach and apply the methods to estimate prostate cancer patient cost trajectories from the Surveillance, Epidemiology, and End Results (SEER)-Medicare linked database.