| To compare cost-effectiveness ratios across studies, we must have some assurance that the analyses that produced the ratios used similar methods. Findings of methodological variation have led to calls for greater standardization of cost-effectiveness analyses. However, there is concern that demanding too much methodological rigor could stifle the use of these analyses, by making it difficult to conduct them in a timely and inexpensive manner. This dissertation explores this issue in three related papers. The first uses an audit of 228 cost-utility analyses to create a comprehensive league table that lists costs/quality-adjusted life-year (QALY), and to identify the subset of published studies that come closest to fulfilling the methodological criteria for reference case analyses, as described by the USPHS Panel on Cost Effectiveness in Health and Medicine. The comprehensive league table of 647 interventions provides a useful reference, but ratios may not be comparable because of varying methodologies. The standardized table presents 112 ratios from studies that met basic methodological criteria, allowing comparison with future Reference Case analyses. The second paper compares community preference weights obtained from patients using a generic health-state classification system, those obtained directly from patients, and the same preference weight assessments from a convenience sample of clinical experts. The preference weights estimated by clinicians were significantly lower than those provided by patients with non-small cell lung cancer, unless clinicians were given detailed information on patients' health status. Health state descriptions that included the disease name seemed to bias clinicians' estimated preference weights. The third paper investigates the impact of health-related quality of life adjustment on the results of cost-effectiveness analyses, by comparing ratios from studies that have reported both costs per (unadjusted) life-year and costs per QALY. In a sizable fraction of studies, the use of quality adjustment did not substantially alter the estimated cost-effectiveness of a health-care intervention. The collection of preference weight data should be subjected to the same scrutiny as other data inputs to cost-effectiveness analyses, and should only be undertaken if the value of this information is likely to be greater than the cost of obtaining it. |
