BMI and Mortality: The Diabetes-Obesity Paradox Examined in a Large US Cohort.

Abstract

Background/objectives: BMI is a major risk factor for diabetes incidence, but a controversial predictor of mortality among those with diabetes.

Subjects/methods: We conducted a mortality follow-up (2002-2019) of participants aged 40-79 with young-onset (diagnosed < age 30, n = 1335), older-onset (diagnosed ≥ 30, n = 15,194), and without (n = 62,295) diabetes at cohort entry. Cox analysis with age as the time scale assessing mortality according to BMI after adjusting for multiple potential confounding factors was used.

Results: Mean baseline age and diabetes duration at cohort entry were 50.1 and 29.4 years and 55.3 and 7.7 years among those with young- and older-onset diabetes, respectively. During an average of 12.3 years of follow-up, 47% of the young-onset, 40% of the older-onset diabetes, and 22.6% of those without diabetes at cohort entry died. In multivariable adjusted analyses, compared to a BMI of 18.5-<25 kg/m², HRs (95% CIs) were 4.10 (1.65-10.18), 0.69 (0.54-0.88), 0.81 (0.63-1.05), 0.64 (0.48-0.86) and 0.64 (0.54-0.77) for BMI categories <18.5, 25-<30 30-<35, 35-<40, 40+ kg/m² in those with young-onset diabetes. Corresponding HRs (95% CIs) were 2.02 (1.54-2.67), 0.74 (0.68-0.80), 0.74 (0.68-0.80), 0.83 (0.75-0.91) and 1.09 (0.99-1.19) in those with older-onset diabetes, and 1.50 (1.36-1.67), 0.76 (0.73-0.79), 0.73 (0.70-0.77), 0.83 (0.78-0.89) and 1.03 (0.95-1.10) in those without diabetes. Results were generally similar in analyses stratified by smoking status, gender, race and among those on insulin therapy.

Conclusion: Among this low socioeconomic status population with diabetes, overweight and obesity tend to be inversely associated with mortality. Risk factors for complications of diabetes other than BMI may be more clinically relevant when treating patients with diabetes.