Prevalence, under-reporting, and epidemiological surveillance of COVID-19 in the Araguaína City of Brazil

Asymptomatic and underreported individuals remain a source of coronafig disease 2019 (COVID-19) transmission to others. Data on the prevalence and epidemiological factors influencing transmission are fundamental for establishing control measures, especially in vulnerable regions such as the Amazon. This study aimed to determine the point prevalence and active infection of COVID-19 among the population in Araguaína, a Brazilian city located in the Amazon region, analyzed the socioeconomic and behavioral variables of a statistically representative sample of this population using an epidemiological survey, and identify the viral genomic diversity in the region. During the sixth epidemiological week of 2021 (February 8 to 12), samples of 497 inhabitants of the municipality asymptomatic for respiratory syndromes underwent reverse transcription-quantitative polymerase chain reaction and serological tests (immunoglobulin M and immunoglobulin G). A questionnaire collated data on socioeconomic factors, prevention measures, and health status history. The active infection rate was 6.2%, and the prevalence was 13.5% of the study population. Active infection cases were under-reported; each reported positive case represented 14–28 under-reported cases. Lineages P.2, P.1, and B.1.1 were detected. Working from home was a protective factor against the infection, and clinical signs of fever, dry cough, and loss of taste or smell were associated with testing positive (p <0.05). A descriptive analysis of the indicators revealed that the entire population was susceptible to the disease. Intensified vaccination strategies are required regardless of socioeconomic factors, health conditions, and preventive measures. Implementation of objective, comprehensive, and efficient management tools to minimize the spread of COVID-19 in this municipality can serve as a model for other regions of Brazil.

Blocks were chosen at random followed by homes.Residents aged ≥18 years were invited to participate.If participation was rejected, the invitation to participate was extended to the next residence.This method was repeated until a predetermined level of acceptance among the residents of these blocks/neighborhoods was achieved.The process of sending out invites to participate in the study was conducted from February 1 to 5, 2021, by endemic agents from the Zoonoses Control Center of the Municipal Secretary of Health from Araguaína (MSHA).
We distributed 608 invitations to residents of selected households, with an estimated nonattendance rate of 18.3%.To ensure the estimated sample size was met, without prejudice to the stratified representativeness, we replaced absent participants with others from the same by region, extending invitations through television media and local websites.Consequently, 497 residents participated in the study, meeting the minimum stipulated sample size for each region, as shown in Table 1. the Federal University of Goiás (UFG), while wearing personal protective equipment.The terms of free and informed consent (TFIC) to participate in this study were explained.Only participants who consented and declared themselves asymptomatic (for respiratory syndromes) were included in the study.Participants who showed symptoms suspicious of COVID-19 were instructed to seek medical care in the city and were excluded from the study.After informed consent was provided, the participants were identified using numeric and non-sequential codes to anonymize the data.The survey questionnaire (S2 File) was subdivided into socioeconomic data, preventive measures adopted, and health history.The estimated time for completing of the questionnaire was 14 min.

Serological tests for COVID-19
After completion of the survey, the participants were referred for blood collection for serological testing to quantify IgM and IgG titers.The nursing technician verified the identification document, completed questionnaire, and collected sample following standard procedures marked with the participants' identifier codes.Blood samples with patient identifier codes were sent to a private clinical laboratory, where the serological examination was performed after serum separation in a biosafety environment.
IgM and IgG titers were quantified by chemiluminescence using commercial kits (SARS-CoV-2 IgM Reagent Kit and SARS-CoV-2 IgG Reagent Kit 6R86; Abbott, Sligo, Ireland) registered with the National Health Surveillance Agency.The sensitivity and specificity of the kits were calculated after symptom onset for IgM (sensitivity 88.14%, specificity 99.56%, from 8 to 14 days) and IgG (sensitivity 100%, specificity 99.63%, from ≥14 days), as reported by the manufacturer.The IgG results were considered reactive if the index was ≥1.00 S/CO, indeterminate between 0.8 and 1.00 S/CO, and non-reactive if the index was ≤0.80 S/CO.The IgM results were considered reactive if the index was >1.00 S/CO, and non-reactive if the index was ≤1.00 S/CO.Serological results were attached individually to the survey questionnaire.

RT-qPCR testing for SARS-CoV-2
After blood collection, the participants were directed to the swab collection area.Two nursing professionals provided by the MSHA verified the participants' details and identified the sample transport tubes using the participant codes.Three swabs (rayon tip with plastic stem) were collected

Genome sequencing and phylogenetic analysis
Of the positive RT-qPCR samples, six were selected for genome sequencing.cDNA was synthesized using the Luna Script RT SuperMix (5X) (New England Biolabs, Ipswich, MA, USA).
Libraries for whole-virus genome sequencing were prepared according to the ARTIC nCoV-2019 sequencing protocol, version 3 (https://artc.network/ncov-2019).The MinION library was prepared using the Ligation Sequencing kit SQK-LSK-109 and Natives Barcoding kits EXP-NBD104 (Oxford Nanopore, Oxford, UK).The resulting library was loaded onto an R9.4 Oxford MinION flow cell (FLO-MIN106) and sequenced using a MinION Mk1B device.High-accuracy base calling was performed after sequencing the FAST5 files using the Oxford Nanopore Guppy tool (version 3.4.5).
Pango lineages were attributed to newly assembled genomes using the Pangolin version 3.1.5software (https://pangolin.cog-uk.io/)[5].All SARS-CoV-2 Brazilian genomes deposited in GISAID until February 31, 2021, were used to construct a phylogenetic tree.To reduce the genome dataset and allow feasible phylogenetic analysis, a genome sampler was used to select the most closely related samples within the same geographic region and period using our dataset of focal sequences.
We then included the reference sequence (NC_045512) to help with the root phylogenies.These steps generated a random sub-sampling of 1,875 SARS-CoV-2 genomes from Brazilian.

Fig 1 .
Fig 1. Maps of the Tocantins state.(A) Brazil, (B) the municipality of Araguaína in Tocantins, and (C) Araguaína divided into regions.
During the sixth epidemiological week of 2021 (February 8 to 12), 497 asymptomatic participants voluntarily visited the study site to complete the epidemiological questionnaire (S2 File Questionnaire).Biological samples were collected during the visit.Participants were verified by their names, and ages, and residence.The survey data were collected by trained interviewers (undergraduate and graduate students in Animal Health and Public Health in the Tropics from the Federal University of North of Tocantins (UFNT) and in Tropical Medicine and Public Health from from the combined bicavitary areas (oropharyngeal and nasopharyngeal), according to Epidemiological Bulletin No. 01/2020 [1].The swabs were immediately placed into a 15 mL Falcon tube containing 2 mL sterile saline solution.The samples were refrigerated and were sent twice a day to the Microbiology Laboratory of the School of Veterinary Medicine and Animal Science at UFNT, Araguaína Campus.In the laboratory, the codes of the participants' samples were crosschecked with those of the signed TFICs.The samples were stored at 7 °C for no more than 24 h before viral ribonucleic acid (RNA) extraction using a commercial kit (QIAamp Viral RNA Mini Kit, Qiagen, Germany) in an NB2 biosafety cabinet.The extracted RNA samples were stored at −20 °C until molecular testing was performed.The extracted RNA products were subjected to RT-qPCR using commercial primers and probes kits from the Centers for Disease Control and Prevention (CDC) (2019NCOV RUO Kit, IDT, USA) targeting two regions of the nucleocapsid phosphoprotein (N) gene and an internal human RNA control in paired uniplex assays on the same plate.A commercial mix (QuantiTect® Probe RT-PCR, Qiagen) with a final volume of 40 µL was used.Amplification reaction was set according to the manufacturer's protocol, and reading in the FAM channel was measured using the QIAquant 96 5plex equipment (Qiagen).Positive controls (2019-nCoV_N and Hs_RPP30 positive controls, IDT) were used for all reactions.The results were interpreted according to the Interim Guidelines for Collecting,

Table 1 .
Sampling and selection of participants according to the residential regions of Araguaína, Tocantins, Brazil, who were asymptomatic for respiratory syndromes.