What is perfect health to an 85-year-old?: evidence for scale recalibration in subjective health ratings.

If an 85-year-old man rates his health as 90 on a scale in which 100 represents "perfect health," would his rating mean the same thing as a 90 rating from a 25-year-old? We conducted a randomized trial of 3 different ways of eliciting subjective health ratings from participants in the Health and Retirement Study to test whether the meaning of perfect health changes as people age, causing people to recalibrate their self-reported health ratings to account for their age.The Health and Retirement Study (HRS) is a nationally representative, prospective study of 22,000 persons born in 1947 or earlier. The data analyzed in this study come from the self-assessed health utilities module administered in 2002 to 1031 randomly selected HRS respondents. Respondents were randomized to receive one of 3 versions of a subjective health rating task. In the perfect health version, they were asked how they would rate their "current health on a scale from 0 to 100, in which 0 represents death and 100 represents perfect health." In the your-age version, the phrase "for someone your age" was added to the end of the question to encourage people to recalibrate their responses based on age, and in the 20-year-old version, the phrase "for a 20-year-old" was added to minimize recalibration.A total of 1015 subjects responded to the rating task (98% response rate). Health ratings varied significantly across versions, with subjects responding to the 20-year-old version reporting lower health (mean rating 66 of 100) than those responding to the your-age version (mean rating of 73, P<0.001) or the perfect health version (mean rating of 73, P<0.001). This result suggests that subjects interpret perfect health to mean "perfect health for someone your age." However, additional analysis showed that the interpretation of the phrase perfect health lies somewhere between the other 2 versions. For example, responses to the perfect health and 20-year-old versions varied significantly by respondent age (both P's<0.075), whereas responses to the your-age scale did not (P=0.8).The phrase "perfect health" is ambiguous, causing some people to recalibrate their responses based on their age. Such ambiguity threatens the validity of common subjective health ratings, thereby reducing the comparability of responses across people of different ages or different circumstances.

Duke Scholars

Author Peter A Ubel Fuqua School of Business