Diabetes and hyperglycaemia among hospitalised patients with COVID-19 in Western Sydney: a retrospective cohort study

Abstract

Background: Diabetes has been recognised as a major risk factor for COVID-19 mortality and hospital complications in earlier studies.

Aims: To examine the characteristics of hospitalised COVID-19 patients with diabetes and the impact of diabetes and hyperglycaemia on hospital outcomes.

Methods: This was a retrospective cohort study. Admission glucose levels, HbA1c, diabetes status and hospital outcomes were determined for subjects admitted from June to November 2021 by matching a pathology data set, a clinical data set and the hospital administrative database. The outcomes of interest were death, intensive care unit (ICU) admission and length of stay (LOS).

Results: There were 1515 individuals admitted with COVID-19 with 49 deaths (3.2%) and 205 (13.5%) ICU admissions. The median length of hospital stay was 3.7 days. Three hundred and ten patients (20%) had diabetes, with 46 (15%) newly diagnosed. Patients with diabetes had a higher mortality than patients who did not have diabetes (8% vs 2%, P < 0.001), were more likely to be admitted to ICU (20% vs 12%, P = 0.001) and have longer median LOS stay (6.6 (interquartile range (IQR) 2.9-12.5) vs 2.9 (IQR 0.5-7.1) days, P < 0.001). In multivariate models, neither diabetes nor admission glucose predicted death. Admission glucose level but not diabetes was an independent predictor of ICU admission and LOS.

Conclusions: There is a high prevalence of diabetes among patients hospitalised with COVID-19, with worse outcomes. In contrast to previous studies, the association of diabetes with mortality was not significant when adjusted for other variables. This is possibly related to the benefits of vaccination and current medical and ICU interventions.

Keywords: COVID-19; HbA1c; diabetes mellitus; hyperglycaemia; inpatients.