PURPOSE: Health policies often depend on difficult value judgments. Members of the general public ought to have input into these decisions. However, many experts question whether large-scale public surveys can usefully capture people's values, given how little time they afford people to think about the policies in question. Studies which engage people in deep deliberation about these issues may yield better responses, but are often plagued by small sample sizes. We tested an intermediate approach: a web-based survey with a built-in deliberation exercise, to elicit people's views about whether patient age should be considered when prioritizing healthcare services.

METHODS: We presented five allocation scenarios to a representative (age and gender) sample of US and Canadian citizens in an Internet panel. The scenarios elicited preferences for funding a program targeted to younger patients (average age 35) versus older patients (average age 65). We randomized half of the 2007 respondents to a survey with an embedded deliberation exercise, and half to a survey without deliberation. In the deliberation exercise, respondents selected which 3 out of 10 possible moral considerations they deemed most important before making the allocation decision for each scenario. The scenarios included 3 life-saving programs (liver and lung transplants and coronary bypass surgery), depression treatment, and palliative care. People then expressed their allocation preferences on a sliding scale from -5 (target younger patients) to +5 (target older patients), with 0 indicating no preference between the two age groups.

RESULTS: Canadian and US respondents did not differ in their preferences (all p's>0.12). Overall, people gave slightly higher priority to programs targeting younger patients for all of the health conditions (p's <0.001) except for palliative care (p=0.12). However, people who completed the deliberation exercise placed significantly less importance on age in allocating scarce resources than did non-deliberators (p's<0.001). Responses from people with the deliberation exercise exhibited less variance than those without (p's<0.001). The two most commonly selected considerations for all scenarios were to “treat everyone equally” and “relieve pain and suffering.”

CONCLUSIONS: People's allocation preferences are influenced by whether they undergo a prior deliberation exercise. In this study, a simple and replicable web-based deliberation exercise caused people to place less importance on patient age in treatment allocation decisions, and led to greater consistency in their allocation preferences.