Quality Adjusted Survival Estimation With Periodic Observations

Abstract

Quality adjusted survival is a measure that integrates both longevity and quality of life information. The analysis of quality adjusted survival in a clinical study with data collected at periodic intervals encounters difficulties due to incomplete information. We propose an estimator of the expected quality adjusted survival time under a Markovian assumption on patient's health status. Based on observed time points, we partition the time axis into a set of disjoint time intervals. The expected quality adjusted survival can be estimated as the summed product of the quality of life and its mean sojourn time of each health state within partitioned intervals. It is shown that the estimator is asymptotically normal with simple variance calculation. A simulation study is conducted to investigate the behavior of the estimator and a stroke study illustrates the use of the estimator.