(HealthDay News) — In a research article published online Jan. 4 in Clinical Epidemiology, long-term glycemia trajectories, which can be predicted by a combination of factors, are identified for patients with type 2 diabetes.
Piia Lavikainen, PhD, from the University of Eastern Finland in Kuopio, and colleagues extracted data covering primary and specialized health care for 9,631 patients with type 2 diabetes. Six-year hemoglobin A1c (HbA1c) trajectories were examined. To predict trajectory membership, a linear discriminant analysis and neural networks were applied.
Over six years, thee HbA1c trajectories were distinguished: stable, adequate; improving but inadequate; and fluctuating, inadequate glycemic control (86.5, 7.3, and 6.2%, respectively). The researchers found that the most important predictors for the long-term treatment balance were prior glucose levels, type 2 diabetes duration, use of insulin only, use of insulin and oral antidiabetic medications, and use of metformin only. Balanced accuracy was 85% for the prediction model, and the area under the receiver operating characteristic curve was 91%.
“Our findings suggest that heterogeneity in long-term treatment outcomes is predictable with [a] patient’s unique risk factors,” the authors write. “This, in turn, offers a useful tool to support treatment planning in the future. However, future studies are needed to obtain even more accurate and personalized predictions.”
