Mikkel B. Christensen

Aim To explore the impact of type 2 diabetes (T2D), glycaemic control and use of glucose‐lowering medication on clinical outcomes in hospitalized patients with COVID‐19. Materials and Methods For all patients admitted to a hospital in the Capital Region of Denmark (1 March 2020 to 1 December 2021) with confirmed COVID‐19, we extracted data on mortality, admission to intensive care unit (ICU), demographics, comorbidities, medication use and laboratory tests from the electronic health record system. We compared patients with T2D to patients without diabetes using Cox proportional hazards models adjusted for available confounding variables. Outcomes were 30‐day mortality and admission to an ICU. For patients with T2D, we also analysed the association of baseline haemoglobin A1c (HbA1c) levels and use of specific glucose‐lowering medications with the outcomes. Results In total, 4430 patients were analysed, 1236 with T2D and 2194 without diabetes. The overall 30‐day mortality was 19% (n = 850) and 10% (n = 421) were admitted to an ICU. Crude analyses showed that patients with T2D both had increased mortality [hazard ratio (HR) 1.37; 95% CI 1.19‐1.58] and increased risk of ICU admission (HR 1.28; 95% CI 1.04‐1.57). When adjusted for available confounders, this discrepancy was attenuated for both mortality (adjusted HR 1.13; 95% CI 0.95‐1.33) and risk of ICU admission (adjusted HR 1.01; 95% CI 0.79‐1.29). Neither baseline haemoglobin A1c nor specific glucose‐lowering medication use were significantly associated with the outcomes. Conclusion Among those hospitalized for COVID‐19, patients with T2D did not have a higher risk of death and ICU admission, when adjusting for confounders.