This article is more than 5 years old. Information may no longer be current.
Personalized care recommendations improve HbA1c in type 2 diabetes
Click Here to Manage Email Alerts
An algorithm that generated personalized approaches for managing type 2 diabetes significantly improved HbA1c outcomes compared with standard care, study data show.
“Evidence suggests that the response to blood glucose regulation agents can differ among population subgroups. … Tailoring glycemic management for specific subpopulations can be critical,” Dimitris Bertsimas, PhD, of the operations research center of Massachusetts Institute of Technology in Cambridge, and colleagues wrote in the study background. “A personalized treatment recommendation using a quantitative approach could readily incorporate different glycemic targets and contraindications and thus allow for more systematic management of subgroups.”
Bertsimas and colleagues used the medical records of 10,806 patients with type 2 diabetes who received care at Boston Medical Center from 1999 to 2014 to model outcomes of 13 different pharmacologic therapies. Researchers then analyzed outcomes under alternative care using a k-nearest neighbor approach for each patient visit. The algorithm recommended switching regimens only if alternative treatments surpassed a specific threshold, Bertsimas and colleagues wrote.
The study included 48,140 patient visits. In most visits (68.2%), the algorithm’s recommendation was the same as the observed standard of care, the researchers reported. In visits where the algorithmic recommendation deviated from the standard of care, the model predicted that the algorithm’s recommendation produced a decrease in posttreatment HbA1c to a mean 7.93%, compared with a predicted mean HbA1c of 8.37% decrease under the standard of care (63.2 mmol/mol vs. 68 mmol/mol; P < .001).
The researchers acknowledged several limitations in the study, including that patients were not randomly assigned into different treatment groups. The study also did not take into account socioeconomic factors or patient preferences, and Bertsimas and colleagues noted that the population in the study may not be representative of the U.S. population at large.
“Despite these limitations, the study establishes strong evidence of the benefit of individualizing diabetes care,” the researchers wrote. “The success of this data-driven approach invites further testing using datasets from other hospital and care settings. … Our work is a key step toward a fully patient-centered approach to diabetes management.” – by Andy Polhamus
Disclosure: The researchers report no relevant financial disclosures.
Click Here to Manage Email Alerts