Methods that have been proposed for the analysis of the (various types of) uncertainty surrounding the estimates derived from a health economics model are based on particular assumptions.

Models that calculate cost-effectiveness of health interventions are subjected to uncertainty analysis. It has been suggested that the input of such analyses should be based strictly on data. Although this is tempting, the paper, published in Value in Health, shows that methods that have been proposed for this purpose, still are based on particular assumptions. The paper proposes to use slightly different assumptions and advocates discussing assumptions rather than suggesting that methods are free of assumptions.

Hendriek Boshuizen, first author of the paper and head of the department of Statistics and Mathematical modeling of the Dutch National Institute of Public Health and the Environment says: "all statistical analyses implicitly assume some prior information. When data are ample, the data outvote this prior information, and these assumptions are not very relevant. The current trend towards modeling many subgroups, however, means that modeling situations based on small amounts of data will become more common. In that case some commonly made assumptions are not as 'uninformative' or 'vague' as often believed. We propose some alternative assumptions we think will work better".

Value in Health (ISSN 1098-3015) publishes papers, concepts, and ideas that advance the field of pharmacoeconomics and outcomes research and help health care leaders to make decisions that are solidly evidence-based. The journal is published bi-monthly and has a regular readership of over 4,000 clinicians, decision-makers, and researchers worldwide.

Source
ISPOR