[Perceived stress and cardiovascular disease mortality. The Ohsaki Cohort Study]

Previous studies have indicated that stress can affect the circulatory system. Although prospective studies have examined the association between perceived stress and cardiovascular disease (CVD) mortality, the results are still controversial. The purpose of the present study was to elucidate the relationship with stratified analyses by alcohol intake category and smoking status.

The prospective Ohsaki Cohort Study covered all National Health Insurance beneficiaries aged 40 to 79 years living in the precinct of Ohsaki Public Health Center, Miyagi, Japan. A total of 45,293 Japanese (21,552 men and 23,741 women), without a history of cancer, ischemic heart disease or stroke, and who answered all items related to stress level at the baseline in 1994, were followed prospectively. Over 12 years of follow-up, 1,751 deaths from CVD occurred (994 men and 757 women). We used Cox proportional hazards models to calculate the hazard ratios (HR) and 95% confidence intervals (CI) for CVD mortality according to the perceived stress categories. The low stress category was used as the reference in all analyses.

Perceived stress demonstrated a significant positive association with CVD mortality for men; the multivariate adjusted HR for high versus low stress was 1.43 (95% CI: 1.19, 1.87, P = 0.006). No significant relationship was noted for women. With current smokers, perceived high stress versus low had a pronounced association for both men (HR = 1.76, 95% CI: 1.28, 2.41, P = 0.001) and women (HR = 1.61, 95% CI: 1.20, 2.16, P = 0.004), and a similar tendency was noted for current drinking (HR = 1.56, 95% CI: 1.16, 2.09, P = 0.006, HR = 1.42, 95% CI: 1.08, 1.87, P = 0.001). Additionally, for both smoking and drinking men, those reporting high stress had 2 times the risk of CVD mortality of their low stress counterparts (P for trend < 0.001). The interaction of perceived stress with smoking for CVD mortality was of borderline statistical significance only for men (P for interaction = 0.04).

The results suggest that the percentage of current smoking and drinking are factors that distinguish between sexes with regard to the effects of perceived stress on the incidence of CVD mortality. Furthermore our present findings indicate that smoking and drinking habit are not the way to relieve one's stress. A review of these should be conducted and we need to enhance support for stress management as well as control over smoking and drinking habits.