Purpose: Describe the relationship among modifiable health behaviors and short-term medical costs.
Design: Prospective study linking cross-sectional survey data that assessed modifiable risk behaviors with insurance claims.
Setting: A large health plan in Minnesota.
Subjects: A stratified, random sample of 10,000 yielded an analytic data set for 7983 members.
Measures: The dependent variable was per-member-per-month insurance payment plus subscriber liability. Eighteen months of medical costs were analyzed. Control variables included subscriber age, sex, type of insurance plan, days of enrollment, chronic disease status, education, and marital status. Independent variables, included self-reported health behaviors of smoking, heavy drinking, nutrition, and physical activity.
Analysis: Linear regression was performed on the natural log of the cost variable, followed by a retransformation to dollars.
Results: Physical inactivity and smoking were significant predictors of higher medical costs. Each day a member did not exercise there was a 2.9% difference in cost. Compared with never smokers, current smokers had 16% higher costs. Former smokers who had quit more than 1 year before taking the survey had 15% higher costs than never smokers. Recent former smokers cost 32% more than never smokers and more frequently experienced smoking-related medical conditions before they quit. Alcohol consumption was nonsignificant. Nutrition also was not significant but was narrowly measured by only fruit and vegetable consumption.
Conclusion: Physical inactivity and smoking were associated with higher short-term medical costs among health plan members.