To combat COVID-19, focus on social behaviour instead of case counts, says study

Despite lockdowns and restrictions, night-time populations start to swell following reports of declining cases of the novel coronavirus disease (COVID-19), hinting at easing caution on a personal level, according to a Japan study. Such changes in behaviour may in turn lead to an uptick in infections.  

“A tracking system of the night-time population using mobile phone location data would be helpful for policy decision makers to monitor social behaviour dynamics and suppress COVID-19 transmission at an earlier stage,” the researchers said.

Mobile phone location data were used to assess changes in population counts between 10 PM and midnight across seven metropolitan areas in Tokyo. In particular, on-site dining vs stay-at-home and stay-at-work patterns were evaluated, as well as their relation to the number of new cases, symptom onset, and the implementation of social distancing measures by establishments.

After its first peak in late March, Tokyo implemented a 7-week, self-restraint-based approach during the first wave of COVID-19, running from April 7 to May 24. This included the declaration of a state of emergency, as well as restrictions on establishments and gatherings. These public health measures seemed effective, as both counts for confirmed cases and new symptom onsets dropped sharply during this period. [JMIR Mhealth Uhealth 2021;9:e27342]

Mobile phone data showed that even before the implementation of these countermeasures, night-time population started to decrease, reaching its lowest point within a week after the restrictions had taken effect. But despite the self-restraint orders, outdoor activity at night started to grow throughout the rest of April and May, peaking in late June 2020.

According to the researchers, this suggests that “people adjusted their level of mobility according to the number of confirmed case reports and, thus, their perceived risk of their own infection,” adding that while the public could adjust behaviours according to policies, attitudes regarding the restrictions could also change in response to how the outbreak evolves.

Soon after, in late July 2020, the number of confirmed case counts reached its second peak, along with a spike in new onset, which reached its highest count by the start of August 2020.

In terms of viral spread, patterns of change in the effective reproduction number (Rt) mirrored those of night-time population, gradually decreasing before the first-wave countermeasures, then increasing while restrictions were in place.

Granger causality tests pointed to the bidirectional causality between confirmed case reports and the night-time population, symptom onset and night-time population, and Rt and night-time population.

“It is notable that there were considerable time lags between behaviour changes, symptom onsets, and Rt,” suggesting that policy measures need to be anticipative and based on social behaviour dynamics, rather than simply virus case counts, the researchers said.

“A tracking system of the night-time population using mobile phone location data would be helpful for policy decision makers to monitor these dynamics in a real-time manner,” they added. “[A]n automated information system to support strategic policy decision making is of particular importance in metropolitan areas with high population density and mobility, where there is an elevated risk of COVID-19 transmission.”