Background: Glucocorticoids are prescribed at high rates in the inpatient setting for various autoimmune and inflammatory conditions. A common complication is steroid-induced hyperglycemia (SIHG), diagnosed when blood glucose levels surpass 140 mg/dL (7.8 mmol/l) in response to steroid exposure. Clinical practice guidelines offer strategies for correction of hyperglycemia among hospitalized patients but lack specific best practices for SIHG. The aim of our study was to capture baseline characteristics and trends in insulin prescription for hospitalized patients with SIHG. We sought to identify SIHG treatment patterns and their association with healthcare utilization following discharge.
Methods: We performed a retrospective chart review of adult patients admitted to a tertiary care institution with SIHG listed as an active hospital problem from 01/2018 to 01/2023. We excluded patients who were discharged to post-acute care facilities and only used patients’ index hospitalization during the study period. Data was obtained on demographics, endocrinology consultation and insulin status at discharge (classified as new insulin, continued insulin, or no insulin). Outcomes included unplanned readmission at 30- and 60-days and contact with the healthcare system for glucose-related concerns at 30-days post-discharge. We used multivariable logistic regression adjusting for pre-specified covariates to evaluate the effect of i) discharge insulin status and ii) inpatient endocrinology consults on outcomes.
Results: There were a total of 265 patients with SIHG identified. The median age was 59 years, 58% were male, the majority were white and English-speaking. Almost half of individuals with SIHG were on the Hematology-Oncology service and half had pre-existing diabetes. Unplanned readmission within 60 days occurred in 116 of 265 patients (44%), with 10 (9%) being attributed to hyper- or hypoglycemia. There was no difference in readmission based on discharge insulin status. Of those individuals with SIHG newly prescribed insulin therapy, 59 of 79 patients (75%) had insulin as the only glycemic agent prescribed while 27 (34%) had an anticipated insulin taper mentioned in their discharge summary. Almost half (n=37, 47%) of patients newly prescribed insulin had no up-to-date HbA1c. Among individuals newly prescribed insulin, 37% had glucose-related healthcare contact events within 30 days of discharge. This was significantly higher compared to those discharged without insulin or on continued insulin (7% and 21%, respectively; p < 0.001). Of all individuals with SIHG, 22% received an inpatient Endocrinology consult. Adjusting for discharge insulin status, age, sex, BMI and LACE score, exposure to Endocrinology consult was associated with a 57% reduced rate of 60-day readmissions (OR 0.43, p=0.02), with a similar trend observed in the 30-day readmission group (OR 0.54, p=0.10).
Conclusions: SIHG is commonly encountered in hospital medicine without clear management guidelines. We found an overall high readmission rate in this patient population, in addition to variable treatment patterns and outcomes between patient groups stratified by insulin status. This included significantly higher rates of healthcare contact for patients discharged with new insulin, and lower readmission rates among individuals who received inpatient Endocrinology consultation. Areas for future improvement may target the low rates of HbA1c assessment and limited insulin tapering plans at discharge in order to streamline care.