A large-scale study conducted at Oxford University Hospitals NHS Foundation Trust reveals hospitalized patients treated with systemic glucocorticoids have a significantly increased risk of developing new-onset diabetes compared to those not receiving such treatment.

The research, presented at the European Association for the Study of Diabetes (EASD) Annual Meeting in Madrid, Spain, found that glucocorticoid-exposed patients were 2.6 times more likely to develop diabetes during their hospital stay.

The study, led by Rajna Golubic, MD, PhD, DSc, MPhil, MRCP, from the Diabetes Trials Unit at the University of Oxford, analyzed data from 451,606 adult inpatients admitted between January 1, 2013, and October 1, 2023.

Key findings included:

1. Study population characteristics:
   • Median age: 52 years
   • Sex distribution: 55% female
   • Ethnicity: 69% white
   • Body composition: 57% overweight or obese
2. Exposure and outcomes:
   • 17,258 patients (3.8%) received systemic glucocorticoids during hospitalization
   • Primary outcome: New-onset diabetes, defined as new glucose-lowering therapy, coded diagnosis of new diabetes, or random venous blood glucose ≥11.1 mmol/L during hospitalization
   • Exposed group: 316/17,258 (1.8%) developed new-onset diabetes
   • Unexposed group: 3,430/434,348 (0.8%) developed new-onset diabetes
3. Incidence rates:
   • Exposed group: 51.9 per 100 person-years
   • Unexposed group: 18.5 per 100 person-years
4. Risk ratios:
   • Unadjusted incidence rate ratio (IRR): 2.8 (95% CI, 2.4-3.1)
   • Age- and sex-adjusted IRR: 2.6 (95% CI, 2.4-2.9)
5. Length of hospital stay:
   • Exposed group: Median 3 days (IQR, 2-8)
   • Unexposed group: Median 1 day (IQR, 1-3)
6. Indications for glucocorticoid use:
   • Most common: Autoimmune/inflammatory diseases/infections (65.3%)

The researchers employed a matched cohort design to address potential biases:

1. Exclusion criteria: Known diabetes or systemic glucocorticoid use on admission
2. Exposure definition: Prescription of prednisolone >5mg daily or equivalent during admission
3. Follow-up: Patients were considered unexposed until first glucocorticoid prescription
4. Exposure period: Date of last prescription plus 15 days or discharge date, whichever came first
5. Matching: Follow-up for unexposed periods started on the prescription date of the matched exposed individual
6. Censoring: Occurrence of new diabetes, discharge, transfer, or death

Statistical analysis involved Poisson regression models to estimate IRR. The study utilized data from electronic healthcare records of the Oxford University Hospitals NHS Foundation Trust, encompassing 528,787 adult inpatients initially. After exclusions, the final analysis dataset consisted of 451,606 adults free from diabetes at baseline.

The study defined exposure as having a prescription of prednisolone >5 mg daily or equivalent during admission. The outcome of new-onset diabetes was determined by documentation of new glucose-lowering therapy, a coded diagnosis of new diabetes, or a random venous blood glucose ≥11.1 mmol/L at any time during hospitalization.

"These latest results give clinical staff a better estimate of how likely new diabetes is to occur and could prompt doctors to plan clinical care more effectively to detect and manage new diabetes," Dr. Golubic states in a press release announcing the study.

The study's strengths included its large sample size and the use of real-world data from electronic health records. However, limitations included potential confounding factors and the inability to assess long-term outcomes beyond hospital discharge.

The authors declared no conflict of interest.