151-LB: A Pilot Randomized Controlled Trial to Reduce Hospital Readmission Risk of Patients with Diabetes: 90-Day Outcomes

Unplanned hospital readmission is a high-priority quality measure and target for cost reduction. Patients with diabetes are at higher risk of readmission than patients without diabetes. We previously presented results of a pilot randomized controlled trial (RCT) of an intervention designed to reduce readmission risk (the Diabetes Transition of Hospital Care [DiaTOHC] program) with outcomes assessed 30 days after hospital discharge. Here we present secondary outcomes assessed 90 days after discharge.

Patients predicted to be high risk (>=27%) for readmission based on a validated readmission risk tool (DERRI^TM) were randomized 1:1 to the intervention (INT) or usual care (UC). The intervention consisted of inpatient diabetes education, coordination of care, post-discharge support by a nurse practitioner, adjustment of diabetes therapy, and weekly phone calls for 30 days after discharge.

There were 45 INT and 46 UC patients randomized and analyzed by intention-to-treat. Twenty-one INT and 23 UC patients had a readmission (46.7% vs. 50%) while 25 INT and 27 UC patients had a readmission or Emergency Department (ED) visit (55.6% vs. 58.7%). The ratio of the mean estimated cost of readmissions, ED visits, and the intervention in the INT group was 0.51 (0.25-1.02)95%CL the cost of readmissions and ED visits in the UC group.

Among the 69 patients with an admission A1C >7%, 14 INT and 17 UC patients had a readmission (41.2% vs. 48.6%), and 18 INT and 21 UC patients had a readmission or ED visit (52.9% vs. 60.0%), yielding relative risk reductions of 15.2% and 11.8%. The INT:UC group ratio of the mean estimated cost was 0.50 (0.22-1.12)95%CL. No differences were statistically significant in this pilot study.

The DiaTOHC intervention may modestly reduce readmission risk and cut costs by half within 90 days after discharge among patients with an admission A1C >7%. This merits further investigation in a larger RCT.