Body mass index and health status in diabetic and non-diabetic individuals

Nutr Diabetes. 2015 Apr 27;5(4):e152. doi: 10.1038/nutd.2015.2.

Abstract

Background/objectives: There is controversy regarding the existence of a body mass index (BMI) mortality paradox in diabetes, whereby the optimal BMI category is higher than it is in non-diabetic persons. To explore possible pathways to a mortality paradox, we examined the relationship of BMI with physical and mental health status in diabetic and non-diabetic persons.

Subjects/methods: We examined adjusted SF-12 Physical and Mental Component Summary (PCS-12 and MCS-12) scores by BMI (kg m(-2)) category (underweight, <20; normal weight, 20 to <25; overweight, 25 to <30; obese, 30 to <35; severely obese ⩾35) in adult diabetic and non-diabetic respondents to the 2000-2011 United States national Medical Expenditure Panel Surveys (N=119 161). Adjustors were age, sex, race/ethnicity, income, health insurance, education, smoking, comorbidity, urbanicity, geographic region and survey year.

Results: In non-diabetic persons the adjusted mean PCS-12 score was highest (that is, most optimal) in the normal-weight category, whereas for diabetic persons the optimal adjusted mean PCS-12 score was in the overweight category (adjusted difference between non-diabetic and diabetic persons in the difference in PCS-12 means for overweight versus normal-weight category=0.8 points, 95% confidence interval; CI 0.1, 1.6; P=0.03). This paradoxical pattern was not evident for the MCS-12, and the adjusted difference between non-diabetic and diabetic persons in the difference in MCS-12 means for overweight versus obese persons was not significant (-0.3 points, 95% CI -0.9, 0.4; P=0.43). The findings were not significantly moderated by smoking status, cancer diagnosis or time period.

Conclusions: The optimal BMI category for physical health status (but not mental health status) was higher among diabetic than non-diabetic persons. The findings are consistent with a BMI physical health status paradox in diabetes and, in turn, a mortality paradox.