Publication date: Sep 20, 2023

There is limited evidence that the diabetes in-person consult in hospitalized patients can be replaced by a virtual consult. During COVID-19 pandemic, the diabetes in-person consult service at the University of Miami and Miami Veterans Affairs Healthcare System transitioned to a virtual model. The aim of this study was to assess the impact of telemedicine on glycemic control after this transition. We retrospectively analyzed glucose metrics from in-person consults (In-person) during January 16 to March 14, 2020 and virtual consults during March 15 to May 14, 2020. Data from virtual consults were analyzed by separating patients infected with COVID-19, who were seen only virtually (Virtual-COVID-19-Pos), and patients who were not infected (Virtual-COVID-19-Neg), or by combining the two groups (Virtual-All). Patient-day-weighted blood glucose was not significantly different between In-person, Virtual-All, and Virtual-COVID-19-Neg, but Virtual-COVID-19-Pos had significantly higher mean +/- SD blood glucose (mg/dL) compared with others (206. 7 +/- 49. 6 In-person, 214. 6 +/- 56. 2 Virtual-All, 206. 5 +/- 57. 2 Virtual-COVID-19-Neg, 229. 7 +/- 51. 6 Virtual-COVID-19-Pos; P = .015). A significantly less percentage of patients in this group also achieved a mean +/- SD glucose target of 140 to 180 mg/dL (23. 8 +/- 22. 5 In-person, 21. 5 +/- 20. 5 Virtual-All, 25. 3 +/- 20. 8 Virtual-COVID-19-Neg, and 14. 4+/-18. 1 Virtual-COVID-19-Pos, P = .024), but there was no significant difference between In-person, Virtual-All, and Virtual-COVID-19-Neg. The occurrence of hypoglycemia was not significantly different among groups. In-person and virtual consults delivered by a diabetes team at an academic institution were not associated with significant differences in glycemic control. These real-world data suggest that telemedicine could be used for in-patient diabetes management, although additional studies are needed to better assess clinical outcomes and safety.

Semantics

Original Article