New Algorithm That Recommends Insulin Dose Adjustments Shows Promise

Participants who used the algorithm reported being in a safe blood sugar range much more often than those who self-monitored their blood sugar.

Matt MacKenzie

March 26, 2026

Key Highlights

The algorithm increased patients' time in a safe blood sugar range from 54.1% to 75.3%, compared to 50.2% to 55.3% with self-monitoring.
Participants using the algorithm received weekly insulin dose recommendations, leading to better blood sugar control.
Insulin titration remains challenging, but this new approach shows promise for more effective diabetes management.
Further research is needed to confirm long-term benefits, but initial findings are encouraging for AI-assisted diabetes care.

A new algorithm developed by researchers at the University of Virginia Center for Diabetes Technology was able to help users “better manage their type 2 diabetes by recommending insulin-dose adjustments.”

Participants in a clinical trial were randomly assigned to “make insulin adjustments for 16 weeks based either on weekly recommendations from the algorithm and glucose monitor or by self-monitoring their blood sugar levels. Participants who used the algorithm saw their average time spent in a safe blood-sugar range increase from 54.1% to 75.3%. Participants who self-monitored their blood sugar saw their average time spent in a safe blood-sugar range increase only from 50.2% to 55.3%.”

The effectiveness of medications used to lower blood sugar tends to wane over time, leading to the need for insulin in patients. Insulin titration, or the process of adjusting insulin doses by self-monitoring blood sugar levels, can be time-consuming and challenging, and there is no standard titration process. Longer clinical trials will be needed to confirm the effectiveness of this algorithm, but the researchers are “encouraged” by these initial findings.