Health Care Contingent Valuation Studies: A review and classification of the literature

Purpose: The contingent valuation method (CVM) is a survey‐based approach for eliciting consumer's monetary valuations for programme benefits for use in cost‐benefit analysis (CBA). We used the conceptual framework of O'Brien and Gafni (1996) to classify and critically appraise health care CVM studies. Methods: Search of computerized health care and economic citation databases (e.g. MEDLINE, ECONLIT) and manual search for papers published between 1984–1996 reporting primary data valuing health programme benefits in monetary units by CVM using willingness‐to‐pay (WTP) or accept (WTA). We classified studies using both empirical (i.e. who was surveyed and how) and conceptual criteria (i.e. which measure of consumer utility was measured and why). Results: 48 CVM studies were retrieved; the majority (42) undertook money valuation in the context of cost benefit analysis (CBA), with the remainder being pricing/demand studies. Among the 42 CBA studies, the consumer utility being measured (i.e. compensating (CV) vs. equivalent variation (EV) was explicitly stated in only three (7%) studies). WTP was measured in 95% of studies and WTA in 5%. By cross‐tabulation, 42 (91%) studies were designed as WTP/CV, two (4%) were WTP/EV, two (4%) were WTA/CV and no studies used WTA/EV. Most studies were administered by mail (52%) with 38% being in‐person interviews. Value elicitation techniques included open‐ended questions (38%), payment cards (19%) discrete choice questions (26%) or bidding games (29%). Some form of construct validation tests, particularly associations between WTP and income, were done in 21 studies (50%). Conclusions: (i) The number of health care CVM studies is growing rapidly and the majority are done in the context of CBA; (ii) there is wide variation among health care CVM studies in terms of the types of questions being posed and the elicitation formats being used; (iii) classification and appraisal of the literature is difficult because reporting of methods and th(This abstract was borrowed from another version of this item.)