For a study, researchers sought to find a link between SARS-CoV-2 infection/coronavirus illness in 2019 and diabetes. They used data from the Veterans Health Administration to conduct a retrospective cohort analysis. Patients without preexisting diabetes who had 1 or more positive nasal swabs for SARS-CoV-2 (March 1, 2020–March 10, 2021; n=126,710) were classified as exposed, while those who had no positive swab but had 1 or more laboratory tests were classified as unexposed (March 1, 2020–March 31, 2021; n=2,651,058). For patients who had been exposed, the index date was the day of the first positive swab, and for those who had not been exposed, a random date was during the month of the qualifying laboratory test. Investigators used sex-stratified logistic regression models to look at the links between SARS-CoV-2 and diabetes within 120 days and the entire follow-up period until June 1, 2021. Only among hospitalized subjects was a subgroup analysis performed to help equalize laboratory surveillance. Men (120 days, odds ratio [OR] 2.56 [95% CI 2.32–2.83]; all time, 1.95 [1.80–2.12]) but not women (120 days, 1.21 [0.88–1.68]; all time, 1.04 [0.82–1.31]) had a higher risk of incident diabetes than women (120 days, 1.21 [0.88–1.68]; all time, 1.04 [0.82–1.31]). SARS-CoV-2 was linked to a greater risk of diabetes in men (OR 1.42 [95% CI 1.22–1.65] and 1.32 [1.16–1.50] after 120 days and at the end of follow-up, respectively), but not in women (0.72 [0.34–1.52] and 0.80 [0.44–1.45]). SARS-CoV-2 interaction P values for both sexes were all less than 0.1. SARS-CoV-2 was linked to a higher incidence of incident diabetes in men but not in women, even when hospitalization-related surveillance was considered.