Survival Analysis Methods for Recurrent Medical Cost Data
Yee, Laura M.
MetadataShow full item record
Cost data are being collected more frequently in randomized clinical trials in order to assess the cost-effectiveness of experimental treatments. As such, the goal of this dissertation is to study three separate topics which advance the analysis of medical cost data. First, time between recurrent medical events may be correlated with the cost incurred at each event. As a result, it may be of interest to describe the relationship between recurrent events and recurrent medical costs by estimating a joint distribution. In this paper, we therefore formulate a nonparametric estimator for the joint distribution of recurrent events and recurrent medical costs in right-censored data. We also derive the asymptotic variance of our estimator, and present simulation studies to demonstrate the performance of our point and variance estimators. Our estimator is shown to perform well for a range of levels of correlation, demonstrating that our estimators can be employed in a variety of situations when the correlation structure may be unknown in advance. We apply our methods to hospitalization events and their corresponding costs in the second Multicenter Automatic Defibrillator Implantation Trial (MADIT-II), which was a randomized clinical trial studying the effect of implantable cardioverter-defibrillators in preventing ventricular arrhythmia. Next, as the costs of medical care increase, more studies are evaluating cost in addition to effectiveness of treatments. Cost-effectiveness in randomized clinical trials has typically been evaluated only at the end of follow-up. However, cost-effectiveness may change over time. We therefore propose a nonparametric estimator to assess the incremental cost-effectiveness ratio over time. We also derive the asymptotic variance of our estimator and present implementation of simultaneous confidence bands. Simulation studies demonstrate the performance of our proposed methods. We also illustrate our methods using data from a randomized clinical trial, the second Multicenter Automatic Defibrillator Implantation Trial (MADIT-II). This trial studied the effects of implantable cardioverter-defibrillators on patients at high risk for cardiac arrhythmia. Results show that our estimator performs well in large samples, indicating promising future directions in the field of cost-effectiveness. Finally, in randomized clinical trials that study cost as well as effectiveness, a common complication is often noncompliance to assigned treatment. In situations where compliance in the trial may differ from compliance rates in the population, it may be of interest to study complier average cost-effectiveness. In this paper, we relate the standard intention-to-treat parameters to the complier average causal effects of two well-known measures of cost-effectiveness, the incremental net benefit (INB) and the incremental cost-effectiveness ratio (ICER). In particular, we show that the intention-to-treat effects are proportional to the complier average effect in the case of the INB, but that the intention-to-treat effect can be interpreted as the complier average effect for the ICER. We outline the assumptions required for these relationships to hold and we also present simulation studies confirming these properties. This work provides some incentive for employing the ICER over the INB when researchers are interested in complier average cost-effectiveness.
- Biostatistics