PSUN229 Transitions of Care and Readmissions in Patients with Diabetes Receiving CGM at Hospital Discharge

Abstract Background Use of inpatient continuous glucose monitoring (CGM) has increased during the COVID-19 pandemic. The objective of this study was to evaluate outpatient follow-up patterns and readmission rates for patients discharged from the hospital with CGM. Research Design and Methods We retrospectively identified patients with diabetes (DM) aged >18 years who were admitted between 3/27/20 and 5/13/21 and received Freestyle Libre or Freestyle Libre 2 14-day sensor at the time of discharge from the hospital. The sensor was placed in the hospital by our diabetic educator. Only the first admission was included. The medical record was reviewed to determine attendance at outpatient follow-up visit, CGM downloading and prescribing at that visit, and readmission within 30 days. Results 87 patient charts were analyzed (31% type 1 DM, 66% type 2 DM, 3% other). The most common reasons for CGM placement were patients new to insulin (40%), DKA (34%), new diabetes diagnosis (30%), or HHS (5%). Follow-up information was not available for 25 patients (29%). The remaining 62 patients had scheduled follow-up within our system. 42 of those patients (68%) attended their follow-up appointment while 14 patients (33%) attended follow-up in an endocrine clinic. Among the 42 patients who attended follow-up, 8 patients (19%) had CGM data downloaded and 18 patients (43%) were prescribed CGM. Of the patients who attended endocrine follow-up, 50% had CGM downloads, compared to only 4% of patients seen in non-endocrine clinics (p = 0.001). Similarly, 79% of patients with endocrine follow-up were prescribed CGM sensors, while 25% of patients seen in non-endocrine clinics received a CGM prescription (p = 0.001). For patients newly diagnosed with DM and available follow-up data, 89% (N=16) attended their appointment. Overall, 16% of patients were readmitted within 30 days. Of the patients readmitted, only 17% attended outpatient follow-up and none had CGM downloaded or prescribed. Backward binary logistic regression was used to assess the relationship between outcomes (follow-up attendance and readmission) and variables (including diagnosis and follow-up location) in the 62 patients with follow-up data. In the final models, a new diagnosis of DM was associated with follow-up attendance, and type 1 DM and follow-up nonattendance were associated with 30-day readmission. Results were verified using exact logistic regression. Conclusions Downloading and prescribing of CGM after hospital discharge occurred in less than half of patients, but occurred more frequently in endocrine clinics. A specialized transition of care clinic may benefit patients discharged with CGM. Follow-up attendance among readmitted patients was poor. Patients with a new diagnosis of DM had good follow-up attendance and may be good candidates for CGM at hospital discharge, which can be examined in future prospective studies. Presentation: Sunday, June 12, 2022 12:30 p.m. - 2:30 p.m.