key: cord-0750912-il1bk5bs
authors: Kawashima, Takayuki; Nomura, Shuhei; Tanoue, Yuta; Yoneoka, Daisuke; Eguchi, Akifumi; Shi, Shoi; Miyata, Hiroaki
title: The relationship between fever rate and telework implementation as a social distancing measure against the COVID-19 pandemic in Japan
date: 2020-05-22
journal: Public Health
DOI: 10.1016/j.puhe.2020.05.018
sha: 70369f1a70916fa7e6a1790ba59c75189151c178
doc_id: 750912
cord_uid: il1bk5bs

Abstract Objectives On March 28, the Japanese government decided on the "Basic Policies for Novel Coronavirus Disease (COVID-19) Control", and called on the public to thoroughly implement social distance measures (i.e. behavioral restrictions to limit the frequency and intensity of human contact), especially telework. Methods We used population-level questionnaire data from a social networking service (SNS) with 275,560 respondents from March 5 to April 6 to evaluate the relationship between telework implementation and the presence of a fever (body temperature above 37.5°C) within one month as a surrogate indicator of COVID-19 infection, by occupation type and age group. Results Among company employees, a statistical significance was identified in the 15–29 and 30–59 age groups, showing higher fever rates in the non-teleworker group (for 15–29 age group, non-teleworkers: 7.64%; teleworkers: 6.45%; p=0.02; for 30–59 age group, non-teleworkers: 3.46%; teleworkers: 3.14%; p=0.02). Conclusions Telework remains a controversial topic in Japan as the government called for emergency measures. While caution is warranted in interpreting our findings because our data is limited to the voluntary SNS users, they will be essential to push forward with more measures to promote social distancing measures in the midst of Japan's current tense political climate.

We considered respondents aged over 15 years of age during March 5 to April 6. In addition, to 27 correct for confounding effects, the analysis was limited to those who reported implementing basic 28 public health precautions such as hand washing, gargling, mask-wearing, and crowd-avoidance.

The data was separated into five occupations: company employees, part-time workers, civil 30 servants, self-employed, and others. Students and unemployed were excluded from the analysis.

For occupations with a sufficient sample size, the ages were divided into 15-29 years, 30-59 years, 32 and 60+ years old groups. Significant difference tests were performed by the chi-square test or 33 Fisher's exact test (when frequencies were less than five in a two-by-two table).

We used the data of 275,560 respondents during the study period. Figure 1 shows the the 37 percentage of people who reported a fever within one month, among teleworkers and non-38 3 teleworkers, respectively, separated by occupation and age group. Among company employees, a 1 statistical significance was identified in the 15-29 and 30-59 age groups (for 15-29 age group, 2 non-teleworkers: 7.64%; teleworkers: 6.45%; p=0.02; for 30-59 age group, non-teleworkers: 3 3.46%; teleworkers: 3.14%; p=0.02), showing higher fever rates in the non-teleworker group. No 4 statistically significant difference was identified for other occupation types. 

World Health Organization. WHO Director-General's opening remarks at the media 2 briefing on COVID-19 -11

World Health Organization. Coronavirus disease (COVID-19) advice for the public

Getting your workplace ready for COVID-19: How COVID-9 19 spreads

Ministry of Health, Labour and Welfare. Basic Policies for Novel Coronavirus Disease 12 Control by the Government of Japan

Teleworking in the Context of the Covid-19 Crisis

Ministry of Health, Labour and Welfare. LINE Corporation will conduct the second 18 nationwide survey to ask people about their health status

Part-time workers 5. Part-time workers 6. Part-time workers

We would like to thank Kanagawa Prefecture for providing us with data.