Diabetes and cause-specific mortality in a prospective cohort of one million U.S. adults

Abstract

Objective: Diabetes is a major predictor of death from heart disease and stroke; its impact on nonvascular mortality, including specific cancers, is less understood. We examined the association of diabetes with cause-specific mortality, including deaths from specific cancers.

Research design and methods: A prospective cohort of 1,053,831 U.S. adults, without cancer at baseline, enrolled in the Cancer Prevention Study-II in 1982 and was followed for mortality until December 2008. At baseline, participants completed a self-administered questionnaire that included information on diabetes, smoking, physical activity, height, and weight. Multivariable-adjusted relative risks (RRs) (95% CI) were estimated using Cox proportional hazards regression.

Results: During 26 years of follow-up, 243,051 men and 222,109 women died. In multivariable models that controlled for age, BMI, and other variables, diabetes was associated with higher risk of all-cause mortality (women RR 1.90 [95% CI 1.87-1.93]; men 1.73 [1.70-1.75]). Among women, diabetes was associated with higher risk of death from cancers of the liver (1.40 [1.05-1.86]), pancreas (1.31 [1.14-1.51]), endometrium (1.33 [1.08-1.65]), colon (1.18 [1.04-1.33]), and breast (1.16 [1.03-1.29]). Among men, diabetes was associated with risk of death from cancers of the breast (4.20 [2.20-8.04]), liver (2.26 [1.89-2.70]), oral cavity and pharynx (1.44 [1.07-1.94]), pancreas (1.40 [1.23-1.59]), bladder (1.22 [1.01-1.47]), colon (1.15 [1.03-1.29]), and (inversely) prostate (0.88 [0.79-0.97]). Diabetes was also associated with higher risks of death involving the circulatory system, respiratory system, digestive system, genitourinary system, and external causes/accidental deaths.

Conclusions: Diabetes is associated with higher risk of death for many diseases, including several specific forms of cancer.

Figure 1

RRs (95% CI) for deaths from noncancer outcomes comparing female (A) and male (B) participants with diabetes (DM) with female and male participants without diabetes (No DM) at baseline, adjusting for age, education, BMI, smoking, alcohol intake, vegetable intake, red meat intake, physical activity, and aspirin use, in the CPS-II, 1982–2008. (Continued on p. 1841.)

Figure 1

RRs (95% CI) for deaths from noncancer outcomes comparing female (A) and male (B) participants with diabetes (DM) with female and male participants without diabetes (No DM) at baseline, adjusting for age, education, BMI, smoking, alcohol intake, vegetable intake, red meat intake, physical activity, and aspirin use, in the CPS-II, 1982–2008. (Continued on p. 1841.)