Temporal trends in glycaemia and hospital outcomes in three metropolitan Australian hospitals, 2017–2024

BACKGROUND: Longitudinal data on in-hospital glycaemia from Australian hospitals are limited. AIMS: The aim of this study was to evaluate the efficacy and safety of current inpatient management strategies in New South Wales hospitals and to compare hospital-acquired complication (HAC) rates in patients with and without diabetes. We also evaluated trends in diabetes admissions, glycaemia and HAC between 2017 and 2024. METHODS: This was a retrospective, observational study of adult medical and surgical admissions to three hospitals over an 8-year period. RESULTS: Diabetes was observed in 21% of admissions (n = 84 864/398 803), with the proportion increasing over time (OR 1.008 per year (95% CI 1.005, 1.011), P < 0.001). Among diabetes admissions, the mean blood glucose (BG) of the hospital admission was 9.17 ± 2.82 mmol/L. The odds of BG 15 mmol/L (OR 0.999 per quarter (CI 0.998, 0.999), P = 0.012) during the admission decreased over time. The odds of hospital-acquired infection (HAI) (OR 0.977 per year (95% CI 0.962, 0.992), P < 0.001) and hospital-acquired cardiac complication (OR 0.973 per year (95% CI 0.947, 0.999), P < 0.039) decreased over time. However, the age-standardised HAI rate ratio increased in diabetes compared with non-diabetes admissions (Exp(B) 1.028 per year (95% CI 1.004, 1.032), P = 0.020). The composite HAC rate decreased in non-diabetes admissions (OR 0.982 per year (95% CI 0.975, 0.989), P < 0.001) but increased in diabetes admissions (OR 1.020 (95% CI 1.009, 1.032), P < 0.001). CONCLUSION: We demonstrate marginal improvements in glycaemia. While some HACs improved, the gains were of greater magnitude in those without diabetes. These findings highlight both progress and persisting inequities in hospital outcomes for people with diabetes.