Prevent Diabetes Mellitus (PreDM) Clinical Decision Support Intervention in Community Health Centers

Prediabetes affects 88 million adults in the United States, up to 50 percent of whom will develop type 2 diabetes (T2D). Intensive lifestyle interventions (ILI) and metformin are safe and effective treatments to prevent or delay diabetes in adults with prediabetes and obesity, and yet these treatments are used in less than 1 percent of eligible US adults. Clinical decision support (CDS) is a possible approach for increasing adoption of ILI and metformin in primary care. While CDS is known to improve the delivery of other preventive health services, its impact for ILI and metformin has been poorly studied. To address this problem, researchers developed and evaluated the Prevent Diabetes Mellitus Clinical Decision Support (PreDM CDS) intervention.

The specific aims of the research were as follows:

• Design the PreDM CDS.
• Test the effectiveness of the PreDM CDS at inducing weight loss and improving other cardiometabolic markers.
• Examine the reach, adoption, implementation, maintenance, and costs of the PreDM CDS.

Researchers conducted a retrospective observational study to examine the feasibility of implementing the tool in a large community health center. The tool appears as a passive button in the electronic health record appearing under the assessment and plan for patients with prediabetes. If clinicians choose to click on the button, they see the patient’s last three measurements for weight, body mass index, hemoglobin A1c (HbA1c), fasting glucose, random glucose, and creatinine. The tool provides functionality that allows providers to: add a prediabetes diagnosis code to the problem list; order metformin, order HbA1c, or refer to a health educator for counseling about lifestyle changes and ILI. 

The tool launched approximately 1 week before the COVID-19 pandemic led to statewide stay-at-home orders, interrupting usual workflows and clinical priorities at the health center. Therefore, the originally planned provider trainings, technical assistance, and reminders could not occur, and instead providers received an email that described the tool with a video displaying the tools functions. 

Of 7,424 eligible patients seen during the study period, providers only used the tool for 1.5 percent of patients. Researchers speculate that low use was due to the tool being a passive and not an interruptive alert, as well as the significant impact to workflows and clinical priorities during the pandemic. However, use of the tool was associated with higher rates of HbA1c orders, as well as prescriptions for metformin. There was also a small, nonsignificant weight loss among those patients for whom the tool was used. 

This research showed that it was feasible to develop and implement the PreDM CDS. The tool led to improvements in the processes of diabetes care, but with no significant differences in ILI participation or weight change. The researchers concluded that future research is needed that prioritizes strengthened linkages to persistent engagement in effective ILI programs.