Optimizing Sampling Strategies for Estimating Quality-adjusted Life Years SCOTT D. RAMSEY, MD, PhD, RUTH ETZIONI, ANDREA TROXEL, PhD, NICOLE URBAN, ScD PhD, Accurate estimation of quality of life is critical to cost-effectiveness analysis. Never- theless, development of sampling algorithms to maximize the accuracy and efficiency of estimated quality of life has received little consideration to date. This paper presents a method to optimize sampling strategies for estimating quality-adjusted life years. In particular, the authors address the questions of when to sample and how many ob- servations to sample at each sampling time, assuming realistically that the sample variance of quality of life is not constant over time. The method is particularly useful for the design problems researchers face when time or research budget constraints limit the number of individuals that can be surveyed to estimate quality of life. The article focuses on cross-sectional sampling. The method proposed requires some knowledge of survival in the population of interest, the approximate variances in utilities at various points along the curve, and the general shape of the quality-adjusted survival curve. Such data are frequently available from disease registries, the literature, or previous studies. Key words: health-related quality of life; utility; quality-adjusted life years; variance; survival; cost-effectiveness; sampling; cross-sectional sampling. (Med Decis Making 1997;17:431-438) Cost-effectiveness analysis is a popular approach for consolidating the economic, quality-of-life, and survival outcomes of a particular medical interven- tion into a single summary measure. Several recent articles have highlighted the importance of account- ing for sample size and censoring when calculating costs in clinical and observational trials. l-3 While cost issues are important, incremental effectiveness usually has a greater impact on the outcome of a cost-effectiveness analysis than does the incremen- tal cost. Nevertheless, there is little guidance from the literature for optimizing accuracy and efficiency when designii the outcomes component of a cost- effectiveness analysis. 4 The objective of this study was to develop a strat- egy for maximizing accuracy and precision when es- timating quality-adjusted life years (QALYs). In this article we explore ways of optimizing the QALY study design when restrictions such as a limited research budget or a short timetable force the investigator to “make do” with a smaller-than-ideal sample size. Received February 2, 1996, from the Department of Medicine, University of Washington Medical Center (SDR), and the Fred Hutchinson Cancer Research Center (RE, AT, NU), Seattle, Wash- ington. Revision accepted for publication March 13, 1997. Dr. Urban was supported in part by contract NOl-CN-05230 from the National Cancer Institute and research award DAMD 17-94-J- 4237 from the Department of the Army. Address correspondence and reprint requests to Dr. Ramsey: Department of Health Services, University of Washington, 146 N. Canal Street, Suite 300 (Box 3588521, Seattle, WA 98103. e-mail: (sramsey@u.washington.edu). We address a fairly specific problem, namely esti- mation of QALYs using utility weights from cross- sectional samples of patients at different time points in the history of the disease, along with survival in- formation from an external source such as a disease registry. MEASURING QUALITY-ADJUSTED SURVIVAL An individual’s health-related qualityof life is seen as a continuum, anchored at the top by an optimal level of health and at the bottom by some minimal level, usually death. Utilities are values assigned to these health states, ranging between 1 (optimal Time Death FIGURE 1. Hypothesized quality-of-life curve for an individual li- fespan. The area under the curve equals the number of the qual- ity-adjusted life years for that individual. 431