67-4

14/90

334restrictions on diabetes management remains unclear. This preliminary investigation was designed to evaluate the impact of activity restric‑tions associated with COVID-19 on diabetes management. Diabetic patients being treated at Juntendo University Hospital were examined by using data collected from electronic medical records, along with information on the changes in the hemoglobin A1c (HbA1c) levels.2. Materials and MethodsThis retrospective, observational study evalu‑ated data collected from all patients who visited the Diabetes and Endocrine Clinics of Juntendo Univer‑sity Hospital and who were prescribed antidiabetic drugs and/or were diagnosed with diabetes mellitus from March to June 2019 and/or from March to June 2020. Juntendo University Hospital is located in Tokyo, the capital of Japan, with most of the patients living in metropolitan areas. HbA1c levels measured during above observation period in the eligible patients were anonymized and extracted through the Data WareHouse attached to the electronic medical record system in collabo‑ration with the Center for Promotion of Data Science. The facility’s website published the research plan. Prior to enrollment, all subjects were informed that any clinical data obtained during their medical treatment would be retrospectively analyzed and published. The study protocol was approved by the Ethics Review Board of Juntendo University (No. 20-152).In order to determine the impact of the state of emergency that was declared on April 7, 2020, this study enrolled 345 patients whose HbA1c levels were measured in both March and April, 1,109 patients whose HbA1c levels were measured in both March and May, and 752 patients whose HbA1c levels were measured in both March and June. Additional subjects enrolled for the purpose of comparison were selected from the 2019 data utilizing the same conditions. The subjects selected for the 2019 comparison group included 469 patients whose HbA1c levels were measured in both March and April, 1,315 patients whose HbA1c levels were measured in both March and May, and 783 patients whose HbA1c levels were measured in both March and June. If the same subject was seen more than once within one month and HbA1c levels were measured multiple times, the mean values of the measurements were calculated and then used. To assess the impact of the restricted activity during the state of emergency on the diabetes manage‑ment, the HbA1c levels in April, May, and June minus HbA1c levels in March were compared between 2019 and 2020. A Wilcoxon rank-sum test was used to analyze the differences in the HbA1c levels. It has been previously reported that the risk of “adverse outcome” caused by diabetes increases linearly with increasing HbA1c3). In addition, a 1 kg weight loss has been estimated to reduce HbA1c by 0.1%4). Therefore, during this comparison, subjects exhibiting a difference in the HbA1c levels greater than 0 were defined as “worsened”, while subjects exhibiting a difference in the HbA1c levels ≤0 were defined as “not worsened”. A chi-squared test was used to analyze the deterioration rate of HbA1c levels in 2019 and 2020. All statistical anal‑yses were performed using SAS 9.4 for Windows (SAS Institute Inc., Cary, NC, USA).Table 1 presents the comparison of the changes in the HbA1c levels, while Table 2 presents the changes in the diabetes management between 2019 and 2020. Continuous variables are described as the median, (minimum, maximum). Changes in HbA1c levels from March to April were worse in 2020 as compared to 2019, with an HbA1c difference of -0.1 (-4.6, 3.2) in 2019 vs. a difference of 0.0 (-3.7, 1.8) in 2020 (p<0.01). In addition, worsening HbA1c levels were observed in a greater number of patients from March to April in 2020 as compared to that seen in 2019, [(26.0%) in 2019 vs. 137 (39.7%) in 2020 (p<0.01)]. In 2020, there was an improvement trend in the changes in HbA1c levels from March to June as compared to 2019, with an HbA1c differ‑ence of 0.0 (-7.5, 3.7) in 2019 vs. a difference of -0.1 (-6.3, 3.1) in 2020 (p=0.07). Furthermore, there fewer patients exhibiting worsened HbA1c levels from March to June in 2020 as compared to 2019, [305 (39.0%) in 2019 vs. 240 (31.9%) in 2020 (p<0.01)]. There was no significant difference between 2020 and 2019 for the HbA1c levels from March to May, with an HbA1c difference of 0.0 (-4.9, 3.6) in 2019 vs. a difference of 0.0 (-5.5, 2.3) in 2020 (p=0.93). Moreover, there was also no signifi‑cant difference between 2020 and 2019 for patients 3. Results