key: cord-0817474-ze7ihp3v
authors: Mizumoto, K.; Shimakawa, Y.; Aizawa, Y.; Butcher, C.; Chibana, N.; Collins, M.; Kameya, K.; Kim, T. G.; Koyama, S.; Matsuyama, R.; Matthews, M. M.; Mori, T.; Nagamoto, T.; Narita, M.; Omori, R.; Shibata, N.; Shibata, S.; Shiiki, S.; Takakura, S.; Toyozato, N.; Tsuchiya, H.; Wolf, M.; Yokoyama, S.; Yonaha, S.; Takayama, Y.
title: SARS-CoV-2 IgG seroprevalence in the Okinawa Main Island and remote islands in Okinawa, Japan, 2020-2021.
date: 2022-03-03
journal: nan
DOI: 10.1101/2022.03.02.22271759
sha: 215bb063cab397e454f7a25a983de5eb1c814033
doc_id: 817474
cord_uid: ze7ihp3v

We estimated the seroprevalence of anti-SARS-COV-2 IgG in different island groups in Okinawa and described its changes over time. A cross-sectional sero-survey was repeated in three distinct periods between July 2020 and February 2021. A total of 2683 serum samples were collected from six referral medical centers, each covering a separate region in Okinawa. Patients who visited the emergency department for any reason and underwent blood collection were eligible for the study. Samples were analyzed using an FDA-authorized two-step enzyme-linked immunosorbent assay (ELISA) protocol. The case detection ratio was computed by dividing the seroprevalence by the attack rate obtained from publicly available surveillance data. In the main island, the seroprevalence was 0.0% (0/392, 95% CI: 0.0-0.9), 0.6% (8/1448, 0.2-1.1), and 1.4% (8/582, 0.6-2.7) at the 1 st , 2 nd , and 3 rd sero-survey, respectively. In the remote islands, the seroprevalence was 0.0% (0/144, 95% CI: 0.0-2.5) and 1.6% (2/123, 0.2-5.8) at the 2 nd and 3 rd survey, respectively. The overall case detection ratios at the 3 rd survey were 2.7 (95% CI: 1.3-5.3) in the main island and 2.8 (0.7-11.1) in the remote islands. The highest age-specific case detection ratio was observed in people aged 20-29 years (8.3, 95% CI: 3.3-21.4) in the main island and in those aged 50-59 years (14.1, 2.1-92.7) in the remote islands. The low seroprevalence at the latest survey suggested that a large-scale epidemic had not yet occurred in Okinawa by February 2021. The case detection ratios imply that the cumulative number of incident cases in Okinawa should be 2-3 times higher than that reported by routine surveillance. The ratio was particularly high in young people probably due to a frequent asymptomatic/mild COVID-19 disease in this age group. To accurately measure the scale of the COVID-19 epidemic, it is crucially important to conduct a sero-survey targeting the young.

Since the first case of COVID-19 in Japan was reported on February 14, 2020, 82 Okinawa Prefecture has faced a total of six waves of COVID-19 as of January 10, 2022, 83 with a total of 58209 cases and 398 deaths, compared to 395 patients and seven deaths as CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted March 3, 2022. ; https://doi.org/10.1101/2022.03.02.22271759 doi: medRxiv preprint obtain a complete picture of the disease dynamics and burden caused by COVID-19. The 99 proportion of previously infected people is a useful indicator to assess the unreported and 100 ascertainment bias regarding the number of cases notified through a surveillance system, 101 the true incidence rate of severe disease and mortality, and the degree of herd immunity. 102 Large-scale sero-epidemiological surveys have been conducted worldwide, including in 104 Spain and Switzerland, since the first wave of the COVID-19 pandemic [7] [8] . In Japan, 105 when this sero-epidemiological survey was originally planned for 2020, there were 106 already a few serosurveys that have been conducted. But these were limited to urban 107 settings (Tokyo, Miyagi, and Osaka) [9] or to a single hospital or private company [10- 108 11] . 109 To the best of our knowledge, this is the first sero-epidemiological survey 110 targeting a wide rural population in Japan. The objective was to determine the 111 seroprevalence of SARS-COV-2 infection in different island groups in Okinawa and to 112 capture the increase in the number of infected residents over time using statistical methods.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Sample size calculation 156 We defined the sample size for each study period to ensure the precision of an is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted March 3, 2022. ; https://doi.org/10.1101/2022.03.02.22271759 doi: medRxiv preprint the Southern Public Health Center. The population size by age group and municipality 186 as of January 1, 2020, was obtained from the public source and categorized into 187 Okinawa Main Island and remote islands and by age group [14] .

Statistical analysis 189 The attack rate was calculated by the number of cumulative confirmed cases 190 divided by the total number of the population. Case detection ratio was defined as a 191 ratio of seroprevalence to attack rate [9] . In order to take into account a latency period 

The number of serum samples collected in the Main Island was 392, 1442, and 201 582, at the 1 st , 2 nd, and 3 rd survey period, respectively. In remote islands, a total of 144 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted March 3, 2022. ; https://doi.org/10.1101/2022.03.02.22271759 doi: medRxiv preprint and 123 serum samples were collected at the 2 nd and 3 rd survey period, respectively.

There were no duplicate samples collected from the same individual across the study 204 periods. Table 1 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. In Main Island, seroprevalence was taking the average value of zero with 228 different upper 95%CI for all subgroups in age group, age category, and gender for the . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. 

In remote islands, seroprevalence was taking the average value of zero with . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted March 3, 2022. ; https://doi.org/10.1101/2022.03.02.22271759 doi: medRxiv preprint for selection bias due to a difference in the risk of infection between the general 320 population and those visiting in the emergency departments.

The prefectural-level case detection ratios that ranged from 2.2 to 2.8 in our 322 study were similar to those obtained in other parts of Japan during the same period (2.6-323 8.7) and considerably lower than those reported in Switzerland (≈20-50) [9, 17] . These 324 results suggest that the surveillance system in Japan may capture the actual number of 325 incident cases better than other countries. The expansion of the PCR testing capacity 326 and monitoring system during a pandemic represents a key intervention, and we believe 327 that the Okinawa Prefecture was moderately well prepared in the PCR screening and 328 monitoring system to grasp the actual status of the epidemic. In the 2 nd survey period, 329 when the spread of the epidemic was still limited, the case detection ratios of middle-330 aged and older people were higher than those in younger age groups. However, in the 331 3 rd survey period, when the epidemic spread more widely, the detection ratios of people 332 in the age group of 20-29 years were much higher than those in other age-groups. Since 333 the vast majority of young people have asymptomatic or mild disease, it is likely that 334 many of them do not come for testing. To get a complete picture of the epidemic, PCR 335 testing and monitoring systems need to be expanded, especially to improve access to 336 testing and care for young people.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. shown that as the epidemic progresses, the case detection ratio would be higher among 370 younger people, who have a higher proportion of asymptomatic and mild cases and do 371 not actively seek testing. The Omicron variant, which is currently spreading globally, 372 apparently has an even higher rate of mild illness than the previous strains, particularly . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) 379 The data that support the findings of this study are available from the corresponding CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted March 3, 2022. ; https://doi.org/10.1101/2022.03.02.22271759 doi: medRxiv preprint in blood by rapid and simple detection method (immunochromatography). National 439 Institute of Infectious Disease, Japan; 2020 2020/4/1.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted March 3, 2022. ; https://doi.org/10.1101/2022.03.02.22271759 doi: medRxiv preprint

Seroprevalence of antibodies to SARS-CoV-424 2 in healthcare workers in a nonepidemic region

Preliminary results of antibody tests

Antigen Production, and Test Setup

Robust 433 neutralizing antibodies to SARS-CoV-2 infection persist for months

Basic Resident Ledger Population by Age

Authors appreciate the following institutions for their contribution to support this study:

is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted March 3, 2022.