Obesity is the strongest modifiable predictor of type 2 diabetes. The public health concern is due primarily to the dramatic increase in the population burden of obesity and diabetes within the past several decades, which is now predicted to further increase morbidity and premature mortality, possibly decrease life expectancy, and lead to unsustainable health costs in developed and developing nations. The Body Mass Index (BMI: weight in kilograms/height in meter squared) is the current metric that characterizes obesity levels and stratifies associated health risk. The BMI, however, is unable to distinguish heterogeneity of risk within obese populations and across ethnic populations. Research has revealed that adipose tissue, which is a function of both adipose cellularity and adipocyte size, is actively involved in energy homeostasis and metabolic regulation, with increasing adipocyte size strongly implicated in obesity-related cellular changes that, if left unaddressed, will eventually lead to the onset of type 2 diabetes. Risk of metabolic disease may depend more on the relative contribution of adipose cellularity and adipocyte size changes than on obesity per se, explaining in part, why heterogeneity of obesity-related diabetes risk occurs within and across populations. The overall objective of this dissertation is to present an alternative obesity-disease paradigm, the Energy Storage Capacity Hypothesis, which may better explain the obesity-disease risk relationship. The specific aims are to highlight the heterogeneity of risk across the BMI continuum and to use The Energy Storage Capacity Hypothesis as the etiologic framework to model diabetes in the US adult population.
In the first study, the 2002 National Hospital Discharge Survey (NHDS) was used to estimate the prevalence of type 2 diabetes among a nationally representative adult morbidly obese (BMI 40 or greater) clinical population. We found that 76% of morbidly obese patient discharge records did not include a diagnosis of type 2 diabetes and that type 2 diabetes diagnosis was reported in only 16% of obesity surgery patient records. The 1999-2002 National Health
and Nutrition Examination Survey(NHANES) was the data source for the second study, for which the Energy Storage Capacity Hypothesis was used to model diabetes in the US adult population. Using Asian race as a surrogate for lowest adipose baseline (genetic) cellularity, while largest weight gain in adulthood was treated as a surrogate for increasing adipocyte size, results of this study revealed that despite lower median obesity levels on all anthropometric measures, phenotypically leaner Asians exhibited the highest burden of age-adjusted diabetes (12.3%) compared to Whites (5.4%), Blacks (11.5%), Mexicans (10.7%), and Hispanics (9.5%). The odds of diabetes burden among Asians was 3.9 times higher than Whites [95% CI: 1.4 - 10.8], after controlling for major risk factors of diabetes. Moreover, among those with diabetes, age-adjusted mean hemoglobin A1c levels were higher in Asians, the non-obese, younger adults, and diabetics who did not report having a family
history of diabetes. The results of this dissertation support the contention that the BMI may have limited utility for obesity classification and risk stratification. In addition, risk stratification based on classic risk factors for diabetes may be contributing to unintended glycemic exposure in apparently low risk populations. New approaches for identifying high risk obesity exposure in individuals and populations is warranted.