Background: Decision analytic models, as used in economic evaluations, require data on several clinical parameters. The gold standard approach is to conduct a systematic review of the relevant clinical literature, although reviews of economic evaluations indicate that this is rarely done. Technology appraisals for the National Institute for Health and Clinical Excellence (NICE), which are fully funded, represent the best case scenario for the close integration of economic evaluations and systematic reviews. The objective of this study was to assess the extent to which the systematic review of the clinical literature informs the economic evaluation in NICE technology appraisals.

Methods: All NICE technology assessment reports (TARs) published between January 2003 and July 2006 were considered. Data were abstracted on the TAR topics, the primary measure of clinical effectiveness, the approach to pooling in the clinical review, the measure of economic benefit and the use, or non-use, of the systematic review in the economic evaluation.

Results: Forty-one TARs were published in the period studied, all of which contained a systematic review. Most of the economic evaluations (85 percent) were cost-utility analyses, reflecting NICE's guidelines for economic evaluation. In seventeen cases, the clinical data were not pooled in the review, owing to heterogeneity in the clinical data or the limited number of studies. In these cases, the economists used alternative approaches for estimating the key effectiveness parameter in the model. The results of the review (when pooled) were always used when the primary clinical effectiveness measure corresponded with the measure of economic benefit (e.g., survival). However, because preference-based quality of life measures are rarely included in clinical trials, the results of the systematic review were never directly used in the cost-utility analyses. Nevertheless, the outputs of the systematic review were used when the data were useful in estimating components of the quality-adjusted life-year (QALY) (e.g., the life-years gained, or the frequencies of health states to which QALYs could be assigned). Problems occurred mainly when the clinical data were not pooled, or when the measure of clinical benefit could not be converted into health states to which QALYs could be assigned.

Conclusions: Economic evaluations can benefit from systematic reviews of the clinical literature. However, such reviews are not a panacea for conducting a good economic evaluation. Much of the relevant data for estimating QALYs are not contained in such reviews and the chosen method for summarizing the clinical data may inhibit the assessment of economic benefit. Problems would be reduced if those undertaking the technology assessments discussed the data requirements for the economic model at an early stage.