April 10, 1995

Serum Cholesterol Level and Mortality Due to Suicide and Trauma in the Honolulu Heart Program

Author Affiliations

From the Institute for Prevention Research, Department of Preventive Medicine (Drs Iribarren and Dwyer) and the Atherosclerosis Research Institute, Department of Medicine (Dr Dwyer), University of Southern California School of Medicine, Alhambra; the Buck Center for Research in Aging, Novato, Calif (Dr Reed); the Honolulu Heart Program, Kuakini Medical Center, Honolulu, Hawaii (Dr Wergowske); and the National Heart, Lung, and Blood Institute, Bethesda, Md (Dr Burchfiel).

Arch Intern Med. 1995;155(7):695-700. doi:10.1001/archinte.1995.00430070045005

Background:  Recent results from cholesterol level— lowering trials and some, but not all, observational studies support an intriguing link between low or lowered serum cholesterol levels and violent death. The reasons behind this relationship are far from clear.

Methods:  In this report, we further investigate this issue by assessing the relationship of baseline serum cholesterol levels with long-term risk of mortality due to trauma and suicide in a cohort of 7309 middle-aged Japanese-American men.

Results:  After 23 years of follow-up, a total of 75 traumatic fatalities and 24 deaths by suicide were documented. Rather than an inverse relation, a positive association between serum cholesterol level and risk of suicide death was observed. After controlling for potential confounders, the relative risk of suicide associated with an increment of 0.98 mmol/L (38 mg/dL) in serum cholesterol level (1 SD) was 1.46 (95% confidence interval, 1.04 to 2.05; P=.02). Multivariate analysis of traumatic mortality failed to detect a relation with serum cholesterol level (relative risk=0.89; 95% confidence interval, 0.70 to 1.13; P=.44). Heavy alcohol consumption (>1200mL of alcohol per month, top quintile) was an independent risk factor for trauma death relative to abstinence (relative risk=1.86; 95% confidence interval, 1.07 to 3.22; P=.02).

Conclusions:  These findings contradict the hypothesis of an inverse relation between serum cholesterol level and suicide, but they support the hypothesis that heavy alcohol consumption is a risk factor for traumatic fatal events.(Arch Intern Med. 1995;155:695-700)