Secondary Logo

Journal Logo

Mental Illness: Brief Report

Body Mass Index and Risk of Suicide Among One Million US Adults

Mukamal, Kenneth J.a; Rimm, Eric B.b,c,d; Kawachi, Ichirod,e; O'Reilly, Eilis J.b; Calle, Eugenia E.f†; Miller, Matthewg

Author Information
doi: 10.1097/EDE.0b013e3181c1fa2d
  • Free


Obesity and depression have a complex relationship. Given the social stigma associated with obesity, it is not surprising that obesity has been associated with depressive symptoms in several studies.1–3 Other studies have the found the opposite,4–6 supporting the “jolly fat” hypothesis.

If obesity is positively associated with depression and anxiety symptoms, it might be hypothesized that overweight and obese individuals would have a higher risk of suicide. A series of studies (with generally few cases of suicide) have suggested exactly the opposite,7–10 particularly among men,11–13 although an inverse association has not been detected in all populations.14,15 The lack of large, detailed studies of middle-aged and older adults (in whom suicide rates are highest) has precluded definitive assessment of this relationship.

To determine the association of body mass index (BMI) with risk of suicide, we studied over 1 million participants of the American Cancer Society Cancer Prevention Study II, who have been followed for mortality for over 20 years.


Cancer Prevention Study II is a prospective study of mortality among 1,184,657 men and women that was begun by the American Cancer Society in 1982.16 Participants were identified and enrolled by over 77,000 volunteers in all 50 states, the District of Columbia, and Puerto Rico. Those enrolled were usually friends, neighbors, or relatives of the volunteers. Families were enrolled if at least one household member was 45 years of age or older, and all enrolled members were at least 30 years of age. In 1982, 1,184,507 participants completed a confidential questionnaire that was mailed to them; the questionnaire is available online.17 Participants provided information on demographic characteristics, personal and family history of diseases, various aspects of behavior, environmental and occupational exposures, and diet. We excluded 26,818 participants missing information on height or weight, 2200 with BMI values beyond the extreme 0.1 or 99.9 percentiles, and 526 missing information on medical history, leaving 1,154,963 participants for analysis. Analyses of deidentified data were approved for exemption from continuing review by the Beth Israel Deaconess Medical Center Committee on Clinical Investigation.

Anthropomorphic Measures

On the questionnaire, participants reported their current weight, their weight 1 year previously, and their height (without shoes). We categorized BMI (the weight in kilograms divided by the square of the height in meters) as underweight (<18.5), normal weight (18.5–22.9 and 23.0–24.9), grade 1 overweight (25.0–29.9), grade 2 overweight (30.0–39.9), and grade 3 overweight (40.0 or higher); the category of 18.5–22.9 was set as the referent. We defined stable weight as individuals with a weight change (gain or loss) of 5 pounds or less in the preceding year.

Assessment of Suicide Death

Deaths that occurred between enrollment and December 2004 were ascertained through detailed individual inquiries by American Cancer Study volunteers in September 1984, September 1986, and September 1988 and then through linkage with the National Death Index beginning in 1989. Death certificates or codes for cause of death were requested for all identified deceased participants and were obtained for 99% of all deaths.18 For deaths occurring from 1989 through 2004, we defined suicide as International Classification of Diseases, 9th Revision (ICD-9) codes E950–E959 and ICD-10 codes X60–X84 and Y87.0. Prior to 1989, a single code was used for all suicides.

Statistical Analysis

We calculated person-years for each participant from the date of return of the 1982 questionnaire to the date of death, or 31 December 2004. In Cox proportional hazards analyses, we controlled for age (linear and centered quadratic terms), race, smoking, comorbidity (each of 6 prevalent conditions), geographical region, physical activity, marital status, religion, educational attainment, alcohol consumption, psychoactive medication use (each of 3 classes), and presence of a recent stressful life event; men and women had independent baseline hazard functions. We evaluated the proportional hazards assumption with visual inspection of plots of scaled Schoenfeld residuals and –log(–log(survival probability)) with time. We created indicator variables for individuals with missing information for covariates, but exclusion of such individuals had no qualitative effect on our findings. For tests of trend, we included BMI as a continuous variable.


Selected characteristics of the cohort according to BMI are shown in Table 1. BMI tended to be positively associated with diabetes and inversely associated with current smoking, white race, and previous cancer.

Selected Characteristics of Cancer Prevention Study II Participants, According to Body Mass Index in 1982

A total of 2231 participants died of suicide during 21,632,509 years of follow-up, a rate of 10.3/100,000 person-years. The median age at death from suicide was 69 years. Table 2 shows incidence rates and hazard ratios for suicide according to BMI. In age- and sex-adjusted analyses, there was generally lower suicide mortality with increasing BMI. Additional adjustment for a number of sociodemographic and clinical factors modestly attenuated the results but a nearly 2-fold gradient in risk remained across the spectrum of BMI.

Association of BMI in 1982 With Subsequent Suicide

Stratified analyses are shown in Table 3. Exclusion of those with extreme BMI did not influence the linear association of BMI with risk. BMI appeared to be inversely associated with risk only among married adults, although the unmarried subgroup represented less than 20% of the cohort. Among unmarried participants, the hazard ratios for suicide mortality per 5 kg/m2 increment in BMI were all close to the null for single, divorced, and widowed participants. The heterogeneity of ratio scale effects for BMI across strata of marital status was similar among men (hazard ratio per 5 kg/m2 increment in BMI = 0.87 among married and 1.03 among unmarried persons) and women (0.73 among married and 0.99 among unmarried persons).

Adjusted Hazard Ratios for Suicide Mortality Among Cancer Prevention Study II Participants per 5 kg/m2 Increase in Body Mass Index


An emerging body of literature supports an inverse relationship of BMI with risk of suicide death,7–13,19 although suicide attempts (rather than completed suicides) have been positively associated with BMI in several samples worldwide.20–22 In Norway, BMI was found to be inversely associated with risk of suicide despite a generally positive association with depression.9 The seemingly contradictory inverse association of BMI with risk of completed suicide but positive association with depression or suicide attempts suggest that depression and suicide attempts have distinct risk-factor precursors from completed suicide.

Our findings differed little between men and women. This contrasts with the particularly adverse relationship of heavier BMI with mental health in women. For example, in the 1992 National Longitudinal Alcohol Epidemiologic Survey, heavier BMI was associated with increased odds of suicide attempts among women, but decreased odds among men.23 However, in the HUNT Study,9 heavier measured BMI was associated with lower risk of completed suicide in both sexes, as we observed.

The apparent multiplicative interaction of BMI with marital status has not been observed previously.13 In unmarried persons, the postulated beneficial effects of higher BMI may be counterbalanced by the well-known stigma associated with overweight and obesity.24 This stigma may be attenuated in the setting of marriage, where spouses are commonly BMI-concordant.25

Studies of BMI and health outcomes are prone to several problems.26 These include confounding by disorders and behaviors (such as smoking27) that themselves lead to weight loss but also increase risk of suicide. To evaluate this possibility, we excluded participants who died in the first 2 years of follow-up (when increased risk due to pre-existing conditions tends to be strongest26), and those who reported recent weight change, major chronic diseases, current or former smoking, or use of sedatives. We also used a referent category of normal-weight individuals (although underweight was not associated with higher risk). The consistency of our results in these analyses tends to support our findings, as do studies from cohorts with repeated measures of weight13 or teenage populations (where the likelihood of chronic disease is low).7,12

Neither hypnotics/benzodiazepine use nor social support appeared to confound the association of BMI with risk of suicide death to any substantial degree, although both were themselves associated with risk. However, we did not have systematic information on depression or antidepressants. We also did not have information on use of substances other than alcohol and tobacco, although this has been associated both positively and inversely with obesity in population surveys.3,28 Future studies ideally need to incorporate repeated measures of depression and measured BMI, in samples with large numbers of outcomes, to evaluate whether depression acts as a mediator or a confounder of the relationship between BMI and completed suicide.29

The Cancer Prevention Study II is subject to other limitations. Participants were a convenience sample of acquaintances of American Cancer Society volunteers and were predominately white, although the sample is geographically diverse and more representative than populations defined by region or occupation. Nonetheless, studies of BMI and suicide among minority populations, in which the epidemic of obesity is particularly profound, are still needed.

We relied on self-reported measures of height and weight, although all assessments were made prospectively. Self-reports of height and weight tend to provide valid rank ordering of individuals,30 albeit with underreporting of weight that increases at greater levels of BMI.31 Furthermore, we observed the expected associations of BMI with other outcomes.32

In conclusion, higher BMI was strongly associated with a lower risk of suicide mortality among US adults. Given this relationship, research to identify its mechanisms could lead to important insights into prevention of this important cause of death.


1. Scott KM, Bruffaerts R, Simon GE, et al. Obesity and mental disorders in the general population: results from the world mental health surveys. Int J Obes (Lond). 2008;32:192–200.
2. Roberts RE, Deleger S, Strawbridge WJ, Kaplan GA. Prospective association between obesity and depression: evidence from the Alameda County Study. Int J Obes Relat Metab Disord. 2003;27:514–521.
3. Simon GE, Von Korff M, Saunders K, et al. Association between obesity and psychiatric disorders in the US adult population. Arch Gen Psychiatry. 2006;63:824–830.
4. Crisp AH, McGuiness B. Jolly fat: relation between obesity and psychoneurosis in general population. Br Med J. 1976;1:7–9.
5. Jasienska G, Ziomkiewicz A, Gorkiewicz M, Pajak A. Body mass, depressive symptoms and menopausal status: an examination of the “Jolly Fat” hypothesis. Womens Health Issues. 2005;15:145–151.
6. Palinkas LA, Wingard DL, Barrett-Connor E. Depressive symptoms in overweight and obese older adults: a test of the “jolly fat” hypothesis. J Psychosom Res. 1996;40:59–66.
7. Paffenbarger RS Jr, Asnes DP. Chronic disease in former college students. 3. Precursors of suicide in early and middle life. Am J Public Health Nations Health. 1966;56:1026–1036.
8. Golomb BA, Tenkanen L, Alikoski T, et al. Insulin sensitivity markers: predictors of accidents and suicides in Helsinki Heart Study screenees. J Clin Epidemiol. 2002;55:767–773.
9. Bjerkeset O, Romundstad P, Evans J, Gunnell D. Association of adult body mass index and height with anxiety, depression, and suicide in the general population: the HUNT study. Am J Epidemiol. 2008;167:193–202.
10. Osler M, Nybo Andersen AM, Nordentoft M. Impaired childhood development and suicidal behaviour in a cohort of Danish men born in 1953. J Epidemiol Community Health. 2008;62:23–28.
11. Gasse C, Derby LE, Vasilakis C, Jick H. Risk of suicide among users of calcium channel blockers: population based, nested case-control study. BMJ. 2000;320:1251.
12. Magnusson PK, Rasmussen F, Lawlor DA, Tynelius P, Gunnell D. Association of body mass index with suicide mortality: a prospective cohort study of more than one million men. Am J Epidemiol. 2006;163:1–8.
13. Mukamal KJ, Kawachi I, Miller M, Rimm EB. Body mass index and risk of suicide among men. Arch Intern Med. 2007;167:468–475.
14. Jiang GX, Rasmussen F, Wasserman D. Short stature and poor psychological performance: risk factors for attempted suicide among Swedish male conscripts. Acta Psychiatr Scand. 1999;100:433–440.
15. Allebeck P, Bergh C. Height, body mass index and mortality: do social factors explain the association? Public Health. 1992;106:375–382.
16. Stellman SD, Garfinkel L. Smoking habits and tar levels in a new American Cancer Society prospective study of 1.2 million men and women. J Natl Cancer Inst. 1986;76:1057–1063.
17. American Cancer Society. Study Questionnaires. Available at:
18. Calle EE, Terrell DD. Utility of the National Death Index for ascertainment of mortality among cancer prevention study II participants. Am J Epidemiol. 1993;137:235–241.
19. Kaplan MS, McFarland BH, Huguet N. The relationship of body weight to suicide risk among men and women: results from the US National Health Interview Survey Linked Mortality File. J Nerv Ment Dis. 2007;195:948–951.
20. Brunner J, Bronisch T, Pfister H, Jacobi F, Hofler M, Wittchen HU. High cholesterol, triglycerides, and body-mass index in suicide attempters. Arch Suicide Res. 2006;10:1–9.
21. Eaton DK, Lowry R, Brener ND, Galuska DA, Crosby AE. Associations of body mass index and perceived weight with suicide ideation and suicide attempts among US high school students. Arch Pediatr Adolesc Med. 2005;159:513–519.
22. Dong C, Li WD, Li D, Price RA. Extreme obesity is associated with attempted suicides: results from a family study. Int J Obes (Lond). 2006;30:388–390.
23. Carpenter KM, Hasin DS, Allison DB, Faith MS. Relationships between obesity and DSM-IV major depressive disorder, suicide ideation, and suicide attempts: results from a general population study. Am J Public Health. 2000;90:251–257.
24. Puhl R, Brownell KD. Bias, discrimination, and obesity. Obes Res. 2001;9:788–805.
25. Jacobson P, Torgerson JS, Sjostrom L, Bouchard C. Spouse resemblance in body mass index: effects on adult obesity prevalence in the offspring generation. Am J Epidemiol. 2007;165:101–108.
26. Manson JE, Bassuk SS, Hu FB, Stampfer MJ, Colditz GA, Willett WC. Estimating the number of deaths due to obesity: can the divergent findings be reconciled? J Womens Health (Larchmt). 2007;16:168–176.
27. Miller M, Hemenway D, Bell NS, Yore MM, Amoroso PJ. Cigarette smoking and suicide: a prospective study of 300,000 male active-duty Army soldiers. Am J Epidemiol. 2000;151:1060–1063.
28. Petry NM, Barry D, Pietrzak RH, Wagner JA. Overweight and obesity are associated with psychiatric disorders: results from the National Epidemiologic Survey on Alcohol and Related Conditions. Psychosom Med. 2008;70:288–297.
29. Hernan MA, Cole SR, Margolick J, Cohen M, Robins JM. Structural accelerated failure time models for survival analysis in studies with time-varying treatments. Pharmacoepidemiol Drug Saf. 2005;14:477–491.
30. Rimm EB, Stampfer MJ, Colditz GA, Chute CG, Litin LB, Willett WC. Validity of self-reported waist and hip circumferences in men and women. Epidemiology. 1990;1:466–473.
31. Stevens J, Keil JE, Waid LR, Gazes PC. Accuracy of current, 4-year, and 28-year self-reported body weight in an elderly population. Am J Epidemiol. 1990;132:1156–1163.
32. Calle EE, Thun MJ, Petrelli JM, Rodriguez C, Heath CW Jr. Body-mass index and mortality in a prospective cohort of US adults. N Engl J Med. 1999;341:1097–1105.
© 2010 Lippincott Williams & Wilkins, Inc.