Final Report: Valuation of Risks to Human Health Insensitivity to Magnitude?

EPA Grant Number: R825312
Title: Valuation of Risks to Human Health Insensitivity to Magnitude?
Investigators: Hammitt, James K. , Graham, John , Corso, Phaedra
Institution: Harvard University
EPA Project Officer: Chung, Serena
Project Period: October 15, 1996 through October 14, 1999
Project Amount: $377,584
RFA: Decision-Making and Valuation for Environmental Policy (1996) RFA Text |  Recipients Lists
Research Category: Environmental Justice


The objectives of this project were to: (1) determine whether (and to what extent) insensitivity to magnitude variation is a barrier to eliciting valid estimates of willingness to pay (WTP) for reduction of risks to human health; (2) develop and test tools for the contingent valuation (CV) practitioner that enhance respondent understanding of the nature and magnitude of the health risk reduction to be offered; (3) perform rigorous, external (split sample) tests of tools designed to address the problem of insensitivity to magnitude variation; and (4) offer constructive guidance to CV practitioners, based on results from the proposed research and the existing literature, on what steps can be taken in the design of CV-health studies to reduce the problem of insensitivity to magnitude variation.

Summary/Accomplishments (Outputs/Outcomes):

We have conducted two nationwide random-digit-dial telephone surveys with approximately 1,000 respondents each. A third survey used a telephone-mail-telephone format to test the effect of different visual aids.

In the first survey, WTP for an automobile safety device that would reduce mortality risk from vehicle accidents was elicited using standard double-bounded dichotomous-choice questions. In the second survey, WTP was elicited for reductions in risks associated with automobile accidents, bacterial food contamination, and blood transfusion. In the third survey, WTP was elicited for reductions in risks associated with automobile accidents, blood transfusion, and pneumonia at advanced ages.

In each case, subsets of respondents were presented with different numerical magnitudes of risk reduction, allowing split-sample estimation of sensitivity to scope. In the second survey, half the respondents were asked to rate the helpfulness of alternative analogies in understanding probabilities (e.g., minutes in a year, inches in a mile, empty seats in a football stadium) and were presented with a minutes per year analogy for each risk (e.g., a 20/100,000 risk is like 105 minutes in a year). In addition, we tested an "indifference-risk" elicitation method in which respondents are asked how large a risk reduction would be required for them to purchase a safety device at a stated price.

In the third survey, respondents were randomly assigned to one of three treatment groups or a control group. Respondents in each treatment group were presented with one of three visual aids used to communicate the risk reductions: a logarithmic risk ladder, a hierarchical linear risk ladder, and a page with 25,000 dots. These visual aids were based on risk communication aids previously used in contingent valuation.

Our results reflect substantial variation between topics and details of the elicitation process, with some variations yielding results that are consistent with the theoretically prescribed proportionality between WTP and probability change. With respect to the pure telephone surveys (no visual aids), in the automobile context, we consistently find statistically significant differences in WTP using the indifference-risk approach (eliciting indifference risk reductions for specified prices in the second survey) but not using the conventional format (eliciting WTP for specified risk changes in the first survey). Similar conclusions were obtained using both external (between subsample) and internal (within subsample) comparisons. In the indifference-risk elicitation, estimates based on the subsample of respondents indicating high confidence in their answers were roughly proportional to the price difference; estimates based on less-confident respondents were alternatively inadequately and excessively sensitive to the price difference. We found no difference in WTP to reduce the risk of food contamination, despite an order-of-magnitude variation in the probability increment.

The telephone-mail-telephone survey revealed substantial differences in sensitivity of WTP to magnitude of risk reduction depending on the visual aid used. We found that WTP depends on the magnitude of risk reduction for three groups of respondents presented with alternative visual aids, but not for the subsample that did not receive any visual aid. Estimated WTP is consistent with the theoretically predicted proportionality to risk change for the subgroups presented with a logarithmic scale or an array of 25,000 dots, but not for the subgroups receiving a linear scale or no visual aid. These results suggest that CV can provide valid estimates of WTP for mortality risk reduction, if appropriate methods are used to communicate the risk change to respondents.

In addition to the primary goals of this research, we included questions on the final survey to elicit respondents' preferences between lotteries on lifespan. By asking respondents to choose between lotteries having a common life expectancy but differences in risk (i.e., one lottery second-order stochastically dominates the other), we can determine whether respondents act in a risk averse, risk seeking, or risk neutral manner with respect to each choice. Preliminary analysis of these results suggests people do not hold a constant risk posture across pairs of lotteries, but rather their preferences depend upon life expectancy and other factors. This work, which was presented by Phaedra Corso at the 1999 Society for Medical Decision Making Annual Conference, was awarded the Lee B. Lusted Student Prize for best research presentation by a student. A manuscript reporting this work is in preparation.

Finally, with funding from the Chiang Chin-Kuo Foundation for International Scholarly Exchange (USA), we conducted similar CV studies in Taiwan to compare the sensitivity of estimated WTP to magnitude of risk reduction in the two societies. Without visual aids, we found a similar lack of sensitivity in the two populations.

In addition to publications, this work has been reported at the NCER investigator conferences as well as at numerous professional conferences and university seminars, including presentations at the American Economics Association/Association of Environmental and Resource Economists conferences, the Society for Risk Analysis conferences, the National Bureau of Economic Research Summer Institute, and the Taipei International Conference on Health Economics.

Journal Articles on this Report : 3 Displayed | Download in RIS Format

Other project views: All 21 publications 3 publications in selected types All 3 journal articles
Type Citation Project Document Sources
Journal Article Corso PS, Hammitt JK, Graham JD. Valuing mortality-risk reduction: using visual aids to improve the validity of contingent valuation. Journal of Risk and Uncertainty 2001;23(2):165-184. R825312 (1999)
R825312 (Final)
  • Abstract: Springer - Abstract HTML
  • Journal Article Hammitt JK, Graham JD. Willingness to pay for health protection: inadequate sensitivity to probability? Journal Risk and Uncertainty 1999;18(1):33-62. R825312 (1999)
    R825312 (Final)
  • Abstract: SpringerLink
  • Journal Article Hammitt JK, Liu J-T, Lin W-C. Sensitivity of willingness to pay to the magnitude of risk reduction:a Taiwan-United States comparison. Journal of Risk Research 2000;3(4):305-320. R825312 (1999)
    R825312 (Final)
  • Full-text: Informa World PDF
  • Abstract: Informa World
  • Other:
  • Supplemental Keywords:

    willingness to pay, contingent valuation, sensitivity to scope, risk communication, health risk., RFA, Economic, Social, & Behavioral Science Research Program, Scientific Discipline, Health Risk Assessment, Economics, decision-making, Ecology and Ecosystems, Economics & Decision Making, contingent valuation, ecosystem valuation, policy analysis, social psychology, surveys, risk preferences, valuation, decision analysis, economic benefits, valuing environmental quality, environmental values, preference formation, standards of value, cost benefit, environmental policy, psychological attitudes, public values, magnitude variation, public policy, stated preference, willingness to pay, interviews

    Progress and Final Reports:

    Original Abstract
  • 1997 Progress Report
  • 1998 Progress Report