Abstract
OBJECTIVE: To establish whether health utility (time trade-off, TTO) and willingness to pay (WTP) values reflect clinical health outcome as evaluated by the Western Ontario McMaster Universities Osteoarthritis Index (WOMAC) in hip and knee osteoarthritis (OA). METHODS: One hundred twenty-eight patients with OA attending a specialized arthritis clinic were interviewed about their socioeconomic characteristics and administered the TTO technique and the WOMAC. Their WTP for 2 hypothetical anti-osteoarthritic drugs was also investigated: the first drug was said to provide a significant improvement in WOMAC dimensions and the second a complete cure of the disease. WTP was elicited by both discrete-choice and bidding game methods. Results. Answer rates were 89.1% for TTO, 98.4% for discrete-choice WTP for both scenarios, and 89.8% and 85.2% for bidding game WTP in the relief and the cure scenario, respectively. The mean TTO utility value was 0.84 (standard deviation 0.20). In discrete-choice, those accepting the bid had higher monthly income (EURO 1536.5 vs EURO 1060.1, p < 0.001, for the relief scenario and EURO 1449.3 vs EURO 1071.6, p < 0.001, for the cure scenario). With the bidding game format, WTP was positively correlated with income in both scenarios (r = 0.56, r = 0.55, p < 0.001). WTP measures differed equally between education and socioeconomic groups with those in favored groups consistently reporting higher WTP (Kruskal-Wallis tests statistics ranging from p < 0.01 to p < 0.001). Except for stiffness, WOMAC dimensions were correlated in the expected direction with TTO values (r = -0.27, p < 0.01 for pain and r = -0.36, r = -0.34, p < 0.001 for physical function and total score, respectively). CONCLUSION: Whereas they showed good feasibility, WTP measures poorly reflected clinical condition and were mainly related to economic status and ability to pay. TTO was correlated with the WOMAC dimensions and may be considered closer to clinical situations than WTP. However, concern arises regarding the homogeneity of the study sample in terms of clinical severity, which may have precluded the identification of a relationship between WTP and clinical status.
