The evaluation of clinical and cost outcomes associated with earlier initiation of insulin in patients with type 2 diabetes mellitus.

BACKGROUND: The treatment for patients with type 2 diabetes mellitus (T2DM) follows a stepwise progression. As a treatment loses its effectiveness, it is typically replaced with a more complex and frequently more costly treatment. Eventually this progression leads to the use of basal insulin typically with concomitant treatments (e.g., metformin, a GLP-1 RA [glucagon-like peptide-1 receptor agonist], a TZD [thiazolidinedione] or a DPP-4i [dipeptidyl peptidase 4 inhibitor]) and, ultimately, to basal-bolus insulin in some forms. As the cost of oral antidiabetics (OADs) and noninsulin injectables have approached, and in some cases exceeded, the cost of insulin, we reexamined the placement of insulin in T2DM treatment progression. Our hypothesis was that earlier use of insulin produces clinical and cost benefits due to its superior efficacy and treatment scalability at an acceptable cost when considered over a 5-year period.

OBJECTIVES: To (a) estimate clinical and payer cost outcomes of initiating insulin treatment for patients with T2DM earlier in their treatment progression and (b) estimate clinical and payer cost outcomes resulting from delays in escalating treatment for T2DM when indicated by patient hemoglobin A1c levels.

METHODS: We developed a Monte Carlo microsimulation model to estimate patients reaching target A1c, diabetes-related complications, mortality, and associated costs under various treatment strategies for newly diagnosed patients with T2DM. Treatment efficacies were modeled from results of randomized clinical trials, including the time and rate of A1c drift. A typical treatment progression was selected based on the American Diabetes Association and the European Association for the Study of Diabetes guidelines as the standard of care (SOC). Two treatment approaches were evaluated: two-stage insulin (basal plus antidiabetics followed by biphasic plus metformin) and single-stage insulin (biphasic plus metformin). For each approach, we analyzed multiple strategies. For each analysis, treatment steps were sequentially and cumulatively removed from the SOC until only the insulin steps remained. Delays in escalating treatment were evaluated by increasing the minimum time on a treatment within each strategy. The analysis time frame was 5 years.

RESULTS: Relative to SOC, the two-stage insulin approach resulted in 0.10% to 1.79% more patients achieving target A1c (<7.0%), at incremental costs of $95 to $3,267. (The ranges are due to the different strategies within the approach.) With the single-stage approach, 0.50% to 2.63% more patients achieved the target A1c compared with SOC at an incremental cost of -$1,642 to $1,177. Major diabetes-related complications were reduced by 0.38% to 17.46% using the two-stage approach and 0.72% to 25.92% using the single-stage approach. Severe hypoglycemia increased by 17.97% to 60.43% using the two-stage approach and 6.44% to 68.87% using the single-stage approach. In the base case scenario, the minimum time on a specific treatment was 3 months. When the minimum time on each treatment was increased to 12 months (i.e., delayed), patients reaching A1c targets were reduced by 57%, complications increased by 13% to 76%, and mortality increased by 8% over 5 years when compared with the base case for the SOC. However, severe hypoglycemic events were reduced by 83%.

CONCLUSIONS: As insulin was advanced earlier in therapy in the two-stage and single-stage approaches, patients reaching their A1c targets increased, severe hypoglycemic events increased, and diabetes-related complications and mortality decreased. Cost savings were estimated for 3 (of 4) strategies in the single-stage approach. Delays in treatment escalation substantially reduced patients reaching target A1c levels and increased the occurrence of major nonhypoglycemic diabetic complications. With the exception of substantial increases in severe hypoglycemic events, earlier use of insulin mitigates the clinical consequences of these delays.