Weight management in basal-bolus insulin therapy for type 2 diabetes mellitus: The deep south diabetes program

Abstract

Background: Weight gain is considered to be a standard clinical complication of insulin therapy. This complication bears heavily on patient populations at risk for obesity, such as those in the southeastern United States. Objective: This study was designed to evaluate the weight changes associated with intensive basal-bolus insulin therapy in the Deep South Diabetes Program. Effectiveness was assessed by evaluating the relationship between changes in glycosylated hemoglobin (A1C) and changes in body weight that occurred during therapy. Methods: The clinical setting was a safety-net, rural community health center for the uninsured and underinsured population in Hardin County, Tennessee. The patients were sick adults with significant, often disabling disease, typically on treatments that were ineffective or produced significant clinical toxicities. This study occurred during a period of retrenchment in the state health insurance program. In this retrospective observational study, information on body weight and A1C measurements was collected over a period of 4 years and analyzed using proprietary and customized database and analysis tools. All patients in the Deep South Diabetes Program who elected intensive basal-bolus insulin therapy and who sustained the treatment for up to 4 years were included in the study. Insulin glargine was used as the primary basal insulin, and insulin aspart was used as the primary bolus insulin. The correlations between net weight change and net A1C change, the duration of treatment, and the degree of reduction in A1C required to achieve normoglycemia and near-normoglycemia were analyzed. Glycemic variability and psychosocial variables were outside the scope of this study. Results: The mean weight for the study population did not change during intensive basal-bolus insulin therapy. When we examined the relationship between weight gain and decreases in A1C, no net weight gain for the patient population as a whole could be associated with lowering of A1C; 20% of the patients experienced no net weight change and 47% experienced a net weight change of <7.5 lb during the 4-year study. An investigation of weight loss with treatment duration showed that patients in treatment <1 year lost a mean of ~4 lb, whereas patients in treatment >2 years gained a mean of ~2 lb. Significant variations in weight among the patients were attributed to the all-inclusive nature of the study; patients with weight-sensitive conditions, including cancer, depression, and congestive heart failure, were not excluded from the study. Conclusions: Providers and patients worked together, using a treatment algorithm and shared medical visits, to create a culture in which normoglycemia was an expected outcome. Results of this study showed that weight gain was not an inescapable clinical complication of basal-bolus insulin therapy.