Responding to COVID-19: Agile Use of Information Technology to Serve Patients With Diabetes

We have learned that penetration of technology such as telehealth does not happen until we encounter a real crisis like the one we are currently facing—the COVID-19 outbreak.

The Japanese government announced a one-month state of emergency from April 7 to May 6 in view of the spread of COVID-19 across the county.^1,2 To avoid cross infection at hospitals and clinics, limitations on telehealth have been temporarily loosened—for the first time, regulations have been significantly relaxed to promote the use of telehealth starting with the initial patient contact^3,4 Unless proper care is provided, a massive wave of patients with more advanced illness may flow into hospitals, causing another medical crisis on top of the COVID-19 crisis. Telehealth offers a good solution for those with the highest need and can help prevent a second wave of medical crisis. However, telehealth usage is limited; health care providers worry that they are unable to obtain reliable and sufficient information from new patients using telehealth, potentially jeopardizing patient safety.⁵

The quality of care in telehealth will be much improved when the information in patients’ personal health records (PHRs) is used in combination with telehealth. Data from PHRs help us to analyze how patients with diabetes are doing under the unique challenges of the COVID-19 situation. Under the state of emergency, patients could suffer from an unhealthy lifestyle, due to decreased levels of exercise and changes in diet.

To examine how the announcement affected patients’ exercise habits, we evaluated the numbers of steps of users who participate in two of our studies. The studies include the use of ICT-based self-management systems where users are asked to measure blood glucose, blood pressure, body weight, and number of steps every day at home; measured data are automatically transferred from each device to the server of the systems via smartphone. The first study uses DialBeticsPlus; it is a prospective, randomized clinical trial.⁶ The second study uses GlucoNote where iPhone users across Japan who have type 2 diabetes or prediabetes are free to download it and to participate in the study after they provided consent electronically on the app.⁷ In both studies, the users were classified into two groups—constant walkers (CWs) and nonconstant walkers (NCWs). Constant walkers are defined as those who measured step counts every day in April, and NCWs are defined as those who missed at least one day of measuring step counts in April. In the first study, only 23 (13 CW and 10 NCW, with a mean observation duration of 304.5 ± 52.2 days) of 131 participants are left for the present analysis because the study is almost at the end of its one-year follow-up period. In the second study, 29 users (6 CW, and 23 NCW, with a mean observation duration of 648.8 ± 559.2 days) who measured the number of steps in April were included in the present analysis. Figure 1 shows the changes in the average number of steps on Sundays in March and April taken by the two groups. The results suggest that patients who have been walking constantly continue to walk during the COVID-19 outbreak, with 6002.5 ± 4228.3 steps on April 19; on the contrary, those who have not walked constantly walk the same as their previous amount, 4870.3 ± 1541.2 steps on April 19. Patients in both groups seemed to hesitate to walk on Sunday, March 29, 2020, the first weekend when new cases of COVID-19 started to increase dramatically in Tokyo.⁸ (It also snowed heavily that day). The overall effect is obvious from comparison of measurement rates between the two groups (Figure 2). The measurement rate of step counts is constant, nearly 100%, in the CW, but it tends to decline from 65.0% in February and 86.0% in March to 50.0% in April in the NCW, suggesting that more than half of NCW quit measuring step counts because they really do not walk. A similar trend was found in the second study (Figure 3); those who have been walking constantly continue to walk during the COVID-19 outbreak, 8077.2 ± 5181.7 steps on April 19. Our data are based on a small number of patients, so it is too early to draw any conclusions, but it seems that patients tend to keep to their daily patterns during the COVID-19 crisis.

Figure 1.

Average number of steps of DialBeticsPlus users on Sundays.

Figure 2.

Comparison of measurement rate.

Figure 3.

Average number of steps of GlucoNote users on Sundays.

In the future, we predict that telehealth will become part of routine practice–patients will be seen in-person or by telehealth. Instead of being lost in the heat of the current crisis, we must draw the right conclusions from our current experiences with telehealth.