Background: Although growing concern has been expressed about premature medical mortality in persons with mental illness, limited data are available quantifying the extent and correlates of this problem using population-based, nationally representative samples.
Methods: The study used data from the 1989 National Health Interview Survey mental health supplement, with mortality data through 2006 linked through the National Death Index (80,850 participants, 16,435 deaths). Multivariable models adjusting for demographic factors assessed the increased hazard of mortality adding socioeconomic status, healthcare variables, clinical factors first separately, and then together.
Results: Persons with mental disorders died an average of 8.2 years younger than the rest of the population (P < 0.001). Adjusting for demographic factors, presence of a mental illness was associated with a significant risk of excess mortality, (hazard ratio=2.06, 95% confidence interval=1.71-2.40), with 95.4% of deaths owing to medical rather than unnatural causes. Adding socioeconomic variables to the model, the hazard ratio was 1.77 (P < 0.001); adding health system factors, it was 1.80 (P < 0.001)); adding baseline clinical characteristics, the hazard ratio was 1.32 (P < 0.001). After adding all the 3 groups of variables simultaneously, the association was reduced by 82% from baseline and became statistically nonsignificant (hazard ratio=1.19, P=0.053).
Conclusions: The results of the study underscore the complex causes and high burden of medical mortality among persons with mental disorders in the United States. Efforts to address this public health problem will need to address the socioeconomic, healthcare, and clinical risk factors that underlie it.
*Rollins School of Public Health, Department of Health Policy and Management, Emory University, Atlanta, GA
†Department of Health Systems, Management and Policy University of Colorado Denver School of Public Health, Aurora, CO
‡School of Social Work, University of Pennsylvania School of Social Policy and Practice, Philadelphia, PA
Supported by K24MH075867 from the National Institute of Mental Health.
None of the authors report any competing interests.
Reprints: Benjamin G. Druss, MD, MPH, Rollins School of Public Health, Department of Health Policy and Management, Emory University, 1518 Clifton Rd, NE Room 606, Atlanta, GA 30322. E-mail: email@example.com.