Objective: To evaluate the association between hyperglycemia treatment and mortality in patients with diabetes and COVID-19 in a Peruvian hospital. Methods: A retrospective cohort study was conducted between March and July 2020. Individual-level data were extracted from an implemented virtual platform. We included patients with type 2 diabetes hospitalized with COVID-19. The assessed outcome was in-hospital mortality. The Independent variable of interest was hyperglycemic treatment. We used Poisson regressions with robust variance to obtain crude and adjusted relative risks (RR) and their 95% confidence intervals (95% CI). Results: Out of 1635 patients hospitalized for COVID-19 during the study period, 248 patients with diabetes mellitus were included. The majority were men (66.9%), the median age was 62 years. Ninety-seven patients died in the hospital (39.1%). The median glycemia on admission was 222.5 mg/dL. At 48 h after hospital admission, 125 patients (50.4%) received sliding scale insulin alone (SSI), 46 (18.5%) received a fixed-dose insulin regimen. In the adjusted analysis, the group with SSI at 48 h of hospitalization had higher mortality than those with fixed-dose insulin (adjusted RR: 1.69; 95% CI: 1.01 – 2.83), and those and who continued with SSI in the following days had higher mortality compared to the group that switched to fixed-dose insulin (adjusted RR: 1.64; 95% CI: 1.17 – 2.32). Conclusion: Among assessed patients with diabetes and COVID-19, more than a third died during hospitalization. Early and continuous use of the sliding scale was associated with higher mortality compared to fixed-dose insulin regimens.