MetSense Pilot and Feasibility

Overview

People with serious mental illness have high risk for type 2 diabetes due to multiple risk factors, including the metabolic side effects of psychotropic medications that are used to treat these conditions. Type 2 diabetes is preventable through lifestyle and pharmacological interventions, but many people with serious mental illness do not receive regular screening for type 2 diabetes risk. In many health care settings, clinical pharmacists are increasingly managing patients with serious mental illness and have expertise in monitoring the metabolic side effects of psychotropic medications. This study evaluates the feasibility and acceptability of using a diabetes prediction model that is based on electronic health record data (the MetSense risk flag) to alert clinical pharmacists about patients who are at high diabetes risk, prompting these clinicians to prioritize diabetes risk management services.

MetSense is a clinical decision support tool that is being developed to help clinicians prioritize diabetes risk management services for patients with serious mental illness who have high diabetes risk. This study seeks to evaluate the feasibility, acceptability, and preliminary effectiveness of using the MetSense risk flag to prompt further evaluation of metabolic risk factors and to prioritize other diabetes risk management services, including diabetes risk screening, care coordination with patients' primary care physicians, and pharmacological management of diabetes risk. In this pragmatic, cluster-randomized trial, patients with serious mental illness will be allocated to 1 of 2 trial arms based on which clinical pharmacist manages their care. The clinical pharmacists will be randomized to view the MetSense risk flag (intervention arm) or to not view the MetSense risk flag (control arm, i.e., usual care). The primary outcome is completion of hemoglobin A1c laboratory testing. The study will also examine differences in clinician evaluation of metabolic risk factors, body mass index measurement, prediabetes diagnosis, diabetes diagnosis, primary care visits, metformin medication dispensing, hemoglobin A1c result, body mass index result, and clinician-reported acceptability and feasibility.