Severe Acute Respiratory Syndrome Coronavirus-2 Seropositivity in South-Central Uganda, During 2019 – 2021

Background: Globally, key subpopulations such as healthcare workers (HCWs) have a higher risk of contracting SARS-CoV-2. In Uganda, limited access to personal protective equipment amidst lack of clarity on the extent and pattern of the community disease burden may exacerbate this situation. We assessed SARS-CoV-2 antibody seroprevalence among high-risk sub-populations in South-central Uganda, including HCWs, persons within the general population previously reporting experiencing key COVID-19 like symptoms (fever, cough, loss of taste and smell) and archived plasma specimens collected between October 2019 – 18th March 2020, prior to confirmation of COVID-19 in Uganda. Methods: From November 2020 - January 2021, we collected venous blood from HCWs at selected health facilities in South-Central Uganda and from population-cohort participants who reported specific COVID-19 like symptoms in a prior phone-based survey conducted (between May to August 2020) during the first national lockdown. Pre-lockdown plasma collected (between October 2019 and March 18th, 2020) from individuals considered high risk for SARS-CoV-2 infection was retrieved. Specimens were tested for antibodies to SARS-CoV-2 using the CoronaChek™ rapid COVID-19 IgM/IgG lateral flow test assay. IgM only positive samples were confirmed using a chemiluminescent microparticle immunoassay (CMIA) (Architect AdviseDx SARS-CoV-2 IgM) which targets the spike protein. SARS-CoV-2 exposure was defined as either confirmed IgM, both IgM and IgG or sole IgG positivity. Results: The seroprevalence of antibodies to SARS-CoV-2 in HCWs was 21.1% [95%CI: 18.2–24.2]. Of the phone-based survey participants, 11.9% [95%CI: 8.0–16.8] had antibodies to SARS-CoV-2. Among 636 pre-lockdown plasma specimens, 1.7% [95%CI: 0.9–3.1] were reactive. Conclusions: Findings suggest a high seroprevalence of antibodies to SARS-CoV-2 among HCWs and substantial exposure in persons presenting with specific COVID-19 like symptoms in the general population of South-central Uganda. Based on current limitations in serological test confirmation, it remains unclear whether pre-lockdown seropositivity implies prior SARS-CoV-2 exposure in Uganda.

Conclusions: Findings suggest a high seroprevalence of antibodies to SARS-CoV-2 among HCWs and substantial exposure in persons presenting with speci c COVID-19 like symptoms in the general population of South-central Uganda. Based on current limitations in serological test con rmation, it remains unclear whether pre-lockdown seropositivity implies prior SARS-CoV-2 exposure in Uganda.

Background
It is over a year since the Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) emerged(1) as a global pandemic and as of the 2nd of August 2021, nearly two hundred million cases were reported globally with >4,000,000 fatalities (2). Transmission occurs by respiratory droplets, aerosols, and via fomites and is higher in con ned or congested spaces(3). SARS-CoV-2 infection can be asymptomatic(4) with estimates ranging from 5% -80% while symptoms are largely nonspeci c and include features of u-like illness (5). Diagnosis of asymptomatic and mild cases may be missed due to prioritization of screening/con rmatory tests for individuals with moderate to severe symptoms. However, asymptomatic and pre-symptomatic persons can be highly contagious and contribute greatly to epidemic spread(6, 7).
As of the 3rd of August 2021, more than 94,000 cases with 2,710 deaths were documented in Uganda(2). Community transmission is on the rise(8) despite earlier control measures that included a phased nationwide lockdown between March and August 2020 (9). The SARS-CoV-2 diagnostic testing landscape in Uganda prioritizes testing for symptomatic persons. It is unknown how many infected asymptomatic persons are missed due to this symptom-based testing approach and what impact this has on community transmission. HCWs in particular are at a higher risk of contracting SARS-CoV-2(10, 11) and inadvertently transmitting it to their patients, some of whom may be immunocompromised. According to the World Health Organization (WHO), they account for 10% of the global SARS-CoV-2 burden (12). This risk may be higher in countries like Uganda, due to shortage of Personal Protective Equipment (PPE) amidst unquanti ed community disease burden. Notably, several HCWs in Uganda have been infected and a number have died (13).
Due to the limited testing capacity, there are likely to be many undetected community infections fueling the epidemic. It is also unknown if SARS-CoV-2 importation or exposure in Uganda might have occurred earlier than the rst (o cial) case reported on the 21st of March 2020. We aimed at determining the prevalence of antibodies to SARS-CoV-2 among selected high-risk sub-populations in South-central Uganda, including HCWs, persons who previously reported speci c Coronavirus disease 2019 (COVID-19) like symptoms (fever, cough, loss of taste and smell) in the preceding 30 days, between May and August 2020. Additionally, we aimed at exploring the possibility of prior SARS-CoV-2 importation/exposure in South-Central Uganda before con rmation of the rst (o cial) case on the 21st of March 2020.

Study design and setting
This study was cross-sectional and was conducted at the Rakai Health Sciences Program (RHSP) with participants recruited from within and outside the Rakai Community Cohort Study (RCCS) in four districts of South-central Uganda (Masaka, Kyotera, Rakai and Lyantonde). The RCCS is an open, populationbased cohort in 40 communities in these districts with surveys conducted ~ every 18 months among 2 3,000 adults, resident in shing, agrarian, or peri-urban/trading community settings (14).

Study population and sample size
A total of 980 participants including 753 HCWs and 227 individuals from the RCCS phone-based survey were recruited into the study. Participants from the cohort had previously reported experiencing COVID-19 like symptoms (fever, cough, loss of taste and/or loss of smell) in the preceding 30 days during an earlier phone-based survey conducted between May and August 2020. HCWs were identi ed from health facilities in the region, prioritizing high volume facilities located near the Uganda-Tanzania border or along the Kampala-Mutukula highway serving mobile persons who may be at higher risk of SARS-CoV-2 acquisition. At the selected health facilities, all available, willing HCWs were recruited into the study.
Additionally, we retrieved 636 archived plasma specimens collected between October 2019 and March 18th, 2020, before the rst national lockdown took effect. were positive on both IgM and IgG. Following retesting of the initially IgM only reactive samples using the ARCHITECT assay, 7/37 were con rmed positive. The overall seroprevalence of antibodies to SARS-CoV-2 in this population was 11.9% [95%CI: 8.0-16.8] (27/227). There was nearly no difference in seropositivity among HIV positive and negative participants ( Table 2).

Discussion
These ndings suggest a relatively high SARS-CoV-2 seroprevalence among HCWs at almost all the selected health facilities (24/26) in South-central Uganda and substantial seroprevalence in persons previously reporting speci c COVID-19 like symptoms within the general population. There was also potentially a spike in transmission a few weeks prior to this evaluation with predominance of IgM only antibodies in most of the participants.
There are challenges interpreting SARS-CoV-2 rapid serology in regions with high malaria endemicity as infection with Plasmodium species was shown to induce cross-reactive antibodies to carbohydrate epitopes on the SARS-CoV-2 spike protein (17,18). It is thus unclear whether seropositivity in prelockdown plasma specimens implies prior SARS-CoV-2 or other related coronavirus exposure or malaria in Uganda.
HCWs are minimally protected by face masks and only a few had accesses to other PPE (face shields, gowns, aprons etc.) and this, coupled with likelihood of improper face mask use or lack of N95-level protection, could explain the positive COVD-19 antibody results observed even among participants reporting face mask use. Several undetected cases among HCWs in this region is a potential driver of nosocomial spread. A moderate concordance between reported RT-PCR COVID-19 positives and antibody test outcome may re ect waning antibody levels as reported in several publications (19,20).

Conclusions
Findings suggest a high seroprevalence of antibodies to SARS-CoV-2 among HCWs and substantial exposure in persons presenting with speci c COVID-19 like symptoms in the general population of Southcentral Uganda. Based on current limitations in serological test con rmation, it remains unclear whether pre-lockdown seropositivity implies prior SARS-CoV-2 exposure in Uganda.