C oronavirus disease (COVID-19) typically causes febrile illness with respiratory symptoms (1,2), and many countries worldwide have been affected. Before characterizing COVID-19 as a pandemic in March 2020 (3), the World Health Organization advised countries to take measures to reduce spread of the virus, including identifying cases and clusters, isolating patients, tracing contacts, and preventing community transmission (4). Several countries have reported on the characteristics of a small number of clusters of COVID-19 cases (5,6). However, few comprehensive reports provide an overview of clusters of COVID-19 cases in communities and the significance of such clusters. We analyzed 61 COVID-19 clusters among various communities in Japan and identified 22 probable primary cases that might have contributed to the disease incidence in clusters. The Study We analyzed COVID-19 cases in Japan reported during January 15-April 4, 2020. All COVID-19 cases confirmed by reverse transcription-PCR in Japan must be reported to the Ministry of Health, Labour and Welfare. Through case interviews, local health authorities collected demographic and epidemiologic information, such as possible source of infection and contact and travel history. During the study period, a total of 3,184 laboratory-confirmed COVID-19 cases, including 309 imported cases, were reported. Among cases of local transmission, 61% (1,760/2,875) had epidemiologic links to known cases (Figure 1, panel A). We excluded 712 cases detected on a cruise that was anchored at Yokohama Port, Japan, from February 3 through March 1 (7). We defined a cluster as >5 cases with primary exposure reported at a common event or venue, excluding within-household transmissions. Our definition also excluded cases with epidemiologic links to secondary transmission. For example, in the following scenario we would exclude cases A and B: boy A is a friend of boy B whose grandmother C contracted nosocomial COVID-19 in a nursing home from which ≥5 cases were reported; although all 3 have symptoms develop and are diagnosed with COVID-19, we would consider only grandmother C part of a cluster from the nursing home. By investigating the epidemiologic links among cases, we identified 61 COVID-19 clusters in various communities. We observed clusters of COVID-19 cases from 18 (30%) healthcare facilities; 10 (16%) care facilities of other types, such as nursing homes and day care centers; 10 (16%) restaurants or bars; 8 (13%) workplaces; 7 (11%) music-related events, such as live music concerts, chorus group rehearsals, and karaoke parties; 5 (8%) gymnasiums; 2 (3%) ceremonial

A c c e p t e d M a n u s c r i p t

The overall coronavirus disease secondary attack rate (SAR) in family members was 19.0% in 10 prefectures of Japan during February 22–May 31, 2020. The SAR was lower for primary cases diagnosed early, within 2 days after symptom onset. The SAR of asymptomatic primary cases was 11.8%.

Detection and isolation of infected people are believed to play an important role in the control of the COVID-19 pandemic. Some countries conduct large-scale screenings for testing, whereas others test mainly people with high prior probability of infection such as showing severe symptoms and/or having an epidemiological link with a known or suspected case or cluster of cases. However, what a good testing strategy is and whether the difference in testing strategy shows a meaningful, measurable impact on the COVID-19 epidemic remain unknown. Here, we showed that patterns of association between effective reproduction number (Rt) and test positivity rate can illuminate differences in testing situation among different areas, using global and local data from Japan. This association can also evaluate the adequacy of current testing systems and what information is captured in COVID-19 surveillance. The differences in testing systems alone cannot predict the results of epidemic containment efforts. Furthermore, monitoring test positivity rates and severe case proportions among the nonelderly can predict imminent case count increases. Monitoring test positivity rates in conjunction with the concurrent Rt could be useful to assess and strengthen public health management and testing systems and deepen understanding of COVID-19 epidemic dynamics.