Assessing preferences for prevention versus treatment using willingness to pay

Background, Rising health care costs and limited resources necessitate trade-offs between resources allocated toward prevention and those toward treatment. Information from opinion polls suggests citizens favor spending a higher proportion of all health care dollars on prevention rather than treatment. Objectives. To assess the policy implications of willingness to pay (WTP) for use in cost-benefit analysis (CBA) as a method for capturing individual preferences for prevention and treatment in the context Of resource allocation decisions. Methods. The authors recruited a random sample of 1456 US residents age 18 years and greater by telephone using random-digit dialing. The survey was designed as a 3-stage (phone-mail-phone) process and was conducted between December 1998 and March 1999. For all persons completing the survey(N = 1104), the authors 1st collected respondents' opinions about the costs and effectiveness of prevention versus treatment programs in general. Half of respondents were then asked to state their WTP for a hypothetical prevention scenario and half were asked to state their WTP for a hypothetical treatment scenario. Both scenarios were specific to the same health context and included an identical reduction in mortality risk. Results. WTP for treatment was significantly greater than WTP for prevention, $665 and $223, respectively. Prior opinions on the relative effectiveness afforded by preventive and treatment interventions moderately influenced the WTP estimates for persons randomized to either scenario. Prior opinions on costs had no significant effect on WTP estimates for either scenario. WTP significantly increased with age and household income in the full sample but was not significantly affected by gender or educational attainment. Conclusions. The aggregated WTP responses from the prevention and treatment scenarios presented in our study would imply that treatment is more strongly preferred by society than prevention when the health context is the same and benefits of each are held constant. A better understanding is needed of the discrepancy between citizens' stated preferences for prevention (e.g., through polling) and our findings that they were willing to pay substantially in ore for treatment than for prevention.