The Toll of Lockdown Against COVID-19 on Diabetes Outpatient Care: Analysis From an Outbreak Area in Northeast Italy

After the spread of severe acute respiratory syndrome coronavirus 2 (SARS-CoV2) from China, Italy became the second most affected country. One of the first outbreaks started in the municipality of Vo’ in the Padua province of the Veneto region. The area was quarantined, and most residents with symptomatic coronavirus disease 2019 (COVID-19) were admitted to the Padua Hospital. Due to escalating numbers of cases, lockdownmeasures were imposed at a national level. Alongwithmass testing, such interventions helped with restraining SARS-CoV-2 diffusion (1). During lockdown, hospitals reorganized to care for COVID-19 patients. From 15 March 2020, outpatient visits were limited to nondeferrable ones, while other appointments were switched to telemedicine, postponed, or cancelled. Diabetes is a key risk factor for severe COVID-19 (2), but the impactof lockdown on diabetes care is less appreciated. We analyzed the outpatient clinic database of the Padua Hospital, containing routine clinical data on demographics, anthropometrics, laboratory results, complications, and therapies. Patients hadprovided written informed consent for the reuse of anonymized data for research purposes. In agreement with national regulation, the local ethics committee (Padua, Italy) was notified of the protocol. Wefirst identified patients forwhoma visit was available during lockdown from 15 March to 14 April 2020 and then identified patients seen in 2018 and 2019 in the same month to match for seasonal variations in access to the clinic. To account for year-to-year variations, we compared patients seen from 15 January–14 February in 2018, 2019, and 2020. We used a generalized estimating equation to compare clinical characteristics of patients seen during lockdown with characteristics of those attending the clinic in the same period of 2018– 2019, adjusting for differences in the prelockdown period. The number of visits (on-site or online) performed in the lockdown period was 47.7% lower than in the same month of the previous 2 years (660 vs. 1,208 and 1,316; P , 0.001), while no substantial reduction was observed in the prelockdown month. The reduction was significantly greater for type 2 diabetes (T2D) (253%) than for type 1 diabetes (T1D) (240%; P , 0.001). During lockdown, on-site visits had high priority due to emerging issues in diabetes management, glucose control, or complications, but most visits (82% for T2D and 95% for T1D) were performed via e-mail, telephone, and other media. Patients received remote consultations on health status, review of laboratory exams and imaging studies, and discussion of issues related to diabetes management including pharmacotherapies. The obliged online approach affected the patients’ ability to contact the clinic andattend the visit, particularly for thosewith T2D, who are older and arguably have less digital skills than patients with T1D. Patients with T2D assisted during lockdown as compared with those seen in previous years were significantly younger, had a shorter diseaseduration, andhada lower prevalence of microangiopathy and heart failurehistory, and theywerenotasoften treated with metformin, sulfonylureas, glucagon-likepeptide1 receptor agonists (GLP-1RA), and antihypertensive, lipidlowering, and antiplatelet medications (Table 1). As a mirror of these characteristics, we infer that aged T2D patients with a heavier complication burden and complex pharmacotherapies could not get in contactwith the clinic for anon-site visit or remote consultation. This means that the toll of lockdown was paid by the most fragile patients, who needed more attention than others during limited functioning of many health care services. Worryingly, the increase in the prescription of GLP-1RA observed prior to lockdown was significantly halted for T2D patients assisted during lockdown. Along