Comparative study of mortality causes between subjects with bipolar disorder and a control sample between 1995 and 2005

Abstract

Background: Bipolar disorder affects 1% to 2% of the world population. It has been described that it is accompanied by an increase in mortality from both violent and non-violent causes. Few studies have been published in our country about causes of death in this particular population. The objective of our study is to analyze the causes of death of subjects with bipolar disorder and compare them to healthy controls.

Methods: From the pool of subjects with bipolar disorder previously recruited for genetic studies conducted at the Cellular and Molecular Biology Research Centre of the UCR, we recruited the 154 subjects who had died between 1993 and 2005. Controls of the same sex who were born and diedin the same period as the subject, with a difference no greater than three months, were selected. For the statistical analysis we ran a Fisher's exact test with 1000 simulations of the p-value with Monte Carlo.

Results: Of 154 subjects, 50% (n=77) were women. We found no difference in the cause of death between sexes among the cases. A significant difference in the main causes of death was found between cases and controls (p<0.05). A significant difference was observed in the categories of ""self-inflicted lesions"" (χ2 (1)=11.0, p<0.05) and ""cancer"" (χ2 (1)= 4.1, p= 0.04). No difference was documented in cardiovascular disease, (χ2 (1)=0.3, p= 0.61), pneumonia (χ2 (1)=3.1, p= 0.07) nor accidents (χ2 (1)= 2.0, p= 0.16).

Conclusion: Self-inflicted lesions, as has been documented in international literature, are a cause of death of major importance in the population affected with bipolar disorder.