859-P: Impact of a Prediabetes Clinical Decision Support System on CV Risk Factor Control in Primary Care: Randomized Trial

Purpose: We designed, implemented, and evaluated impact of a prediabetes clinical decision support system on identification of prediabetes and control of major cardiovascular (CV) risk factors in adults with prediabetes receiving care in primary care clinics.

Methods: We randomized 34 primary care clinics with 18,229 study-eligible patients age 40-75 years at index encounter with laboratory evidence of prediabetes to usual care (UC) or a prediabetes clinical decision support (PreD-CDS) intervention condition. Study-eligible patients and their primary care clinician (PCC) were given patient-specific treatment suggestions to support diagnosis of PreD and control of systolic BP (SBP), low-density lipoprotein cholesterol (LDL), smoking, and Body Mass Index (BMI, Kg/m2). Electronic medical record (EHR) and other data were analyzed using linear and logistic mixed models to account for nested data structures.

Results: At index visit study subjects were 50.7% female, 97.1% white, mean age 60.9 years, 56.7% not diagnosed with PreD, 17.2% current tobacco smokers, and mean age 60.9 years, BMI 33.6 Kg/m2, SBP of 151 mmHg, and LDL 118 mg/dl. Compared to UC clinics, those receiving care at PreD-CDS clinics had no significant change in likelihood of PreD diagnosis, mean SBP, mean LDL, mean BMI, or smoking (all p >.05).

Conclusions: This point-of-care EHR-linked web-based clinical decision support system had no statistically significant impact on diagnosis of prediabetes or CV risk factor control. Alternative methods of improving care for these vulnerable patients are needed.