67-4

15/90

HbA1c levels in 2019 (%)HbA1c levels in 2020 (%)HbA1c levels in 2019 (%)HbA1c levels in 2020 (%)HbA1c levels in 2019 (%)HbA1c levels in 2020 (%)Median (min, max). Differences in HbA1c levels were calculated by subtracting the values in April, May, or June from the values in March. † Analysis by Wilcoxon rank-sum test.Subjects in 2019 (n)Subjects in 2020 (n)Subjects in 2019 (n)Subjects in 2020 (n)Subjects in 2019 (n)Subjects in 2020 (n)Differences in HbA1c levels higher than 0 were defined as worsened, while differences in HbA1c levels 0 or lower were defined as not worsened. † Analysis by the chi-squared test.Table 1　Changes noted in HbA1c before and after implementation of activity restrictionsDifference in HbA1c levels-0.1 (-4.6, 3.2)0.0 (-3.7, 1.8)0.0 (-4.9, 3.6)0.0 (-5.5, 2.3)0.0 (-7.5, 3.7)-0.1 (-6.3, 3.1)Table 2　Changes noted in diabetes management before and after implementation of activity restrictionsexhibiting worsened HbA1c levels from March to May, [501 (38.1%) in 2019 vs. 396 (35.7%) in 2020 (p=0.23)].4. DiscussionThere was no persistent deterioration of the short-term changes in the HbA1c levels noted during the state of emergency that was declared in Japan, and our results did not appear to show any profound changes. However, the results did show that patients with worsening HbA1c exhibited a slight decrease after a temporary increase. Thus, some diabetic patients may have been affected by the COVID-19 pandemic and the various related restrictions on social activities.Reports of changes in diabetes management before and after the state of emergency lockdowns remain controversial, with studies indicating both worsening and improvement in these patients5-8). A simulation model was reported that predicted worsening diabetes management in conjunction March7.6 (5.3, 15,0)7.5 (4.9, 16.2)7.2 (5.2, 15.0)7.2 (5.1, 14.0)7.2 (5.2, 15.0)7.1 (4.9, 14.0)7.4 (5.2, 13.8)7.5 (5.1, 14.0)March to AprilMarch to MayMarch to JuneAprilMay7.2 (4.5, 14.6)7.2 (5.0, 14.6)Worsened122 (26.0)137 (39.7)501 (38.1)396 (35.7)305 (39.0)240 (31.9)with lockdowns, while another study reported that a lack of access to insulin/glucostrips led to a wors‑ening of the management of type 1 diabetes mellitus5, 6). In contrast, Fernández et al. and Bonora et al. examined lockdowns in Spain and Italy, respectively, and reported finding improved diabetes control in patients with type 1 diabetes mellitus7, 8). They suggested that patients could spend more time on self-management due to a reduction in daily routine activities during the lock‑downs, and thus, this may have been responsible for improving the diabetes control.Although our present results suggest that there was no persistent deterioration in the overall diabetes management, it is possible that some of the diabetic patients may have had either worse or better diabetes control. In Japan, social participation restrictions led to a temporary decrease in daily workloads in addition to a decrease in alcohol intake amounts. Furthermore, as these diabetic patients had more time to spend on diabetes management, June7.1 (5.3, 14.3)7.0 (5.0, 14.0)Not worsened347 (74.0)208 (60.3)814 (61.9)713 (64.3)478 (61.0)512 (68.1)p value†<0.010.930.07p value†<0.010.23<0.01335