Variations in Risk Attitude across Race, Gender, and Education

Allison B. Rosen, Jerry S. Tsai, Stephen M. Downs

Research output: Contribution to journalReview articlepeer-review

86 Scopus citations


Background. Significant disparities in health care utilization exist across gender and race. Little is known about the patient-specific factors that may contribute to this variation. This study examined variations in risk attitude across major sociodemographic groups. Methods. A survey elicited utility measures for health states under risk-insensitive and risk-sensitive conditions (time tradeoff and standard gamble methods, respectively). Risk attitude was modeled assuming constant proportional risk posture, thus the utility function used was a power function. A multivariable linear regression model was used to examine the relationship between risk attitude and sociodemographic factors. Results. Of the 62 study subjects, the mean age was 47.6 years, 47% were female, and 33% were African American. Overall, 37% of respondents were decidedly risk averse, 37% moderately risk averse, 15% moderately risk seeking, and 11% decidedly risk seeking. Significant predictors of increasing risk aversion in multivariate modeling were white race (P < 0.01) and lower education (P < 0.05). Women also tended to be more risk averse (P = 0.07). Conclusions. This study found significant differences in risk attitude across race and educational status, with a smaller difference across gender. Further research is needed to validate these findings and clarify their contribution to racial and gender variations in health care utilization and their future role in decision and cost-effectiveness analyses.

Original languageEnglish (US)
Pages (from-to)511-517
Number of pages7
JournalMedical Decision Making
Issue number6
StatePublished - Nov 1 2003
Externally publishedYes


  • Gender
  • Race
  • Risk attitude
  • Risk aversion
  • Socioeconomic disparities
  • Utility assessment

ASJC Scopus subject areas

  • Health Policy

Fingerprint Dive into the research topics of 'Variations in Risk Attitude across Race, Gender, and Education'. Together they form a unique fingerprint.

Cite this