Body-Mass Index (BMI) and Mortality in Diabetic and Non-Diabetic Patients

A new study shows a surprising correlation between the BMI-mortality relationship and diabetes….

Obesity and type 2 diabetes rates are skyrocketing in the US and we have long associated an increased BMI/obesity with increased mortality. Studies have also shown that type 2 diabetics have increased mortality compared to their non-diabetic counterparts. However, most studies showing BMI-mortality relationships have not further stratified and analyzed the data to determine if obesity plus diabetes demonstrates a higher mortality rate as compared to those who are obese and non-diabetic.

In a recently published study, a group of researchers tried to determine those relationships and their findings were fairly surprising. The researchers gathered large amounts of data representative of the adult US population by assessing records from the 1997 to 2002 National Health Interview Survey and then subsequently followed those persons through 2006 for mortality. A total of 74,710 non-Hispanic white and black adults were included in the study as long as they were born in the US; had no history of cancer and/or heart disease; were not pregnant; height, weight, mortality, heart disease, cancer, and smoking status data was all present; they did not fall into extremes of BMI (<15 or >55); and they were not diagnosed with diabetes at an age <25 years old and were not treated with insulin. The primary outcome of the study was all-cause mortality with the groups separated into 5 categories of BMI ([15.02–22.83]; [22.84–25.09]; [25.1–27.46]; [27.47–31.02]; [31.03–54.92]) and then further stratified to indicate those with diabetes vs. those without diabetes. Other secondary measures were also analyzed such as mortality association with smoking vs. non-smoking, etc. The study results showed that the BMI-mortality relationship was significantly related to diabetes and that no matter which quintile of BMI was assessed, the proportion of death rates were higher in the diabetic population as compared to the non-diabetic population. These findings were as expected however, surprisingly, when the researchers ran adjusted Cox models on the data, mortality was negatively associated with increasing BMI in those with diabetes and mortality amongst non-diabetics demonstrated a positive correlation with increasing BMI.

While these findings are a little shocking, the study did rely on patient self-report of diabetes diagnosis and up to 27% of type 2 diabetics are undiagnosed which could introduce some error and skew the data. Also, mortality rates were consistently higher in all BMI quintiles of those with diabetes as compared to those without diabetes, supporting current practice of encouraging weight loss. Future studies assessing obesity/BMI as related to mortality should include information about participant’s diabetes status/diagnosis. Even if analysis is run post hoc, it could show us better trends.

Practice Pearls:

• Persons who are obese have a higher rate of all-cause mortality as compared to normal weight individuals
• Despite the inverse relationship between increasing BMI, mortality and diabetes demonstrated in this study, the study showed mortality rates were higher across all BMI quintiles in those with diabetes as compared to their non-diabetic counterparts
• Both diabetes and increased BMI put a person at increased risk for mortality, development of other co-morbidities, and poorer health-related quality of life

Jackson, C. et al. Body-Mass Index and All-Cause Mortality in US Adults With and Without Diabetes. Journal of General Internal Medicine. Published online August 9, 2013.