Changing drinking pattern does not influence health perception: a longitudinal study of the atherosclerosis risk in communities study.

OBJECTIVE: To investigate if dynamic changes in the pattern of alcoholic beverages consumption are associated with modifications in health perception. DESIGN, SETTING, AND PARTICIPANTS: This study investigated 12 332 middle aged men and women from the atherosclerosis risk in communities study who reported drinking status and perceived health triennially from 1987 to 1995. Crude and adjusted risks for change in health perception between visits two and three by change in drinking status between visits one and two were computed. In the multivariate analysis the sample was restricted to participants with stable drinking status between visit two and three and stable health perception between visits one and two, to assure that exposure and outcome were not temporary. Covariates included age, sex, race, income, smoking status, educational level, and obesity. RESULTS: Health for persons who stopped or started drinking, or continued to abstain was more likely to decline than was health for persons who continued to drink even after adjustment and restrictions (drinking cessation: OR = 1.6, 95% CI = 1.1, 2.3; started drinking; OR = 1.4, 95% CI = 0.9, 2.2; continued abstaining from alcohol: OR = 1.5, 95% CI = 1.3, 1.9). Among participants with poor perceived health, starting, stopping, or continuing to abstain from alcohol did not improve health in relation to participants that continued to drink. CONCLUSION: Increasing and decreasing drinking patterns and continuous abstinence were associated with declining health perception in comparison with continuous drinking, while starting or stopping drinking did not improve health perception of persons with poor perceived health. These findings suggest that change in health perception was not biologically related to alcohol consumption.

Anger temperament is modestly associated with the risk of type 2 diabetes mellitus: the Atheroslcerosis Risk in Communities Study.

OBJECTIVE: Trait anger has been shown to predict coronary heart disease; however, there are no prior studies evaluating anger as a risk factor for type 2 diabetes. The objective of this study was to determine if anger proneness predicted type 2 diabetes using prospective analyses. METHODS: We analyzed data on trait anger, assessed by questionnaire, in a longitudinal, bi-racial cohort study of 11,615 initially non-diabetic adults aged 48-67, who were subsequently followed for 6 years for the development of type 2 diabetes. RESULTS: There was no relation between overall trait anger and the subsequent risk of diabetes; however, individuals in the highest tertile of trait anger temperament scores had a 34% increased risk of developing diabetes compared to those in the lowest tertile (Relative hazard [RH] = 1.34; 95% confidence interval: 1.10, 1.62), following adjustment for age, race, gender, and education. The relation between anger temperament and diabetes remained significant following adjustment for behavioral factors (smoking, physical activity, and caloric intake) (RH = 1.31; 95% CI: 1.08, 1.60) but was non-significant after additional adjustment for waist-to-hip ratio and body-mass index (RH = 1.18; 95% CI: 0.97, 1.44). CONCLUSIONS: While we found no relation between overall trait anger and incident diabetes, persons who had higher anger temperament subscale scores had a slightly increased risk of type 2 diabetes, which was largely explained by adiposity. Anger temperament may deserve further attention as a potential risk factor for type 2 diabetes.