One instance of the online survey involving choice experiments concerning public health risk reductions. This version is actually one branch within a survey that included either a “prevention” or a “treatment” branch, with the preliminary and following questions being identical across the two versions. This “treatment” variant employed choice tasks concerning public policies that would treat children, adults and seniors in different proportions who have a specific type of illness. Policies varied in terms of how many people would get sick over a specified number of future years, how many would recover fully without and with the policy, how many deaths would occur with and without the policy, and the cost per month to the respondent. There were actually four variations on this survey. The fullest version included information on both illnesses and deaths, and included information about how many people would get recover fully or die, both without and with the policy. The minimal version omitted information about recoveries, mentioning only deaths, and did not provide both the baseline “without policy” counts of illnesses or deaths, just the reduction in counts due to the policy. Both this “treatment” version of the public health risks survey and the associated “prevention” version included a trailing question designed to elicit individual discount rates.