Health risk aversion, health risk affinity, and socio-economic position in the USA: The demographics of multiple risk

Abstract

Understanding the distribution of behavioural risk factors in the population can improve health promotion. This article reports on a research project which analysed the distribution of numbers of behavioural risk factors among US adults, by race/Hispanic origin, sex, and age. Income, education, and region were examined as potential confounders in observed patterns. The B ehavioral Risk Factor Surveillance System (BRFSS) data for 1993 were used to assess the distribution of numbers of risk factors, i.e. smoking, heavy drinking, overweight, inadequate seatbelt use, Papanicolaou (pap) smear screening, mammography, colorectal screening, and influenza and pneumonia vaccination. Two hypotheses were examined: (1) given the distribution of each risk factor in the US population by age and sex, prevalences both of low and of high numbers of risk behaviours (but not of moderate numbers of risk behaviours) are greater than expected within each race/Hispanic origin-sex-age group; and (2) differences in socio-economic position among these groups account for the differences between observed and expected prevalences of numbers of risk factors. The second hypothesis was assessed both graphically by adjustment for income and education, and by multiple linear regression. The research found higher than expected prevalences of both low and high numbers of risk factors among whites and, possibly, among Hispanics. Among Asians prevalences were less than expected with greater numbers of risk factors. Among blacks and American Indians prevalences were lower than expected for low numbers of risk factors and were greater than expected with greater numbers of risk factors. Adjustment for income or education reduced differences between observed and expected prevalences. It appears that risk aversion and risk affinity vary substantially by race/Hispanic origin and are only partially explained by socio-economic position. Exploration of the causes of high and low risk behaviour may improve risk behaviour interventions.

Keywords:

• Health
• Ethnicity
• Gender
• Class
• Behaviour
• Risk