The purpose of this dissertation was to assess the long-term cost-effectiveness of type 2 diabetes mellitus prevention from a state of impaired glucose tolerance with the use of intensive lifestyle intervention. The analysis was primarily based on findings by the Diabetes Prevention Program (DPP). Additional goals of the analysis were: (1) to examine the effect of obesity on the results; and (2) to examine the effect of the study design of the DPP, in particular the healthcare exclusion criteria, on the analysis. Two basic Markov models were designed for the dissertation. The first was designed to closely approximate the Diabetes Prevention Program study design and results, and the second was designed to attempt to approximate a more generalized U.S. population. These two basic models were then divided into two subcategories: a subcategory that with adjustments intended to reflect the effects of obesity in the analysis; and a subcategory with no adjustments for obesity. This subdivision was then further divided by duration of effect of the three-year intervention: a three-year duration of effect and a lifetime duration. The results showed, in all models examined, the lifestyle intervention arm of the analysis dominated the placebo arm. The incremental cost-effectiveness ratios derived for the models exhibited a wide range (-$551/LY to -$19,496/LY). The variables that appeared to most influence this wide range of ratios included the following: (1) overall direct medical costs; (2) effects of obesity on mortality rates; (3) effect of obesity on control cost of illness; and (4) duration of intervention. In addition, the maximum acceptable cost of intervention for any one of the models also depended on these variables. Overall, diabetes prevention from the state of impaired glucose tolerance appears to dominate non-intervention. However, a large range of values were derived for both incremental cost-effectiveness ratios and maximum acceptable cost of intervention. These values appear to depend on cost inputs as well as obesity-adjustments to the models used for analysis. Researchers and policy makers should consider studying these possible influences in interpreting the study findings, conducting future research, or making decisions regarding implementation of interventions.