High seroprevalence of anti-SARS-CoV-2 antibodies among Ethiopian healthcare workers

Abstract

Background: COVID-19 pandemic has a devastating impact on the economies and health care system of sub-Saharan Africa. Healthcare workers (HWs), the main actors of the health system, are at higher risk because of their occupation. Serology-based estimates of SARS-CoV-2 infection among HWs represent a measure of HWs' exposure to the virus and could be used as a guide to the prevalence of SARS-CoV-2 in the community and valuable in combating COVID-19. This information is currently lacking in Ethiopia and other African countries. This study aimed to develop an in-house antibody testing assay, assess the prevalence of SARS-CoV-2 antibodies among Ethiopian high-risk frontline HWs.

Methods: We developed and validated an in-house Enzyme-Linked Immunosorbent Assay (ELISA) for specific detection of anti-SARS-CoV-2 receptor binding domain immunoglobin G (IgG) antibodies. We then used this assay to assess the seroprevalence among HWs in five public hospitals located in different geographic regions of Ethiopia. From consenting HWs, blood samples were collected between December 2020 and February 2021, the period between the two peaks of COVID-19 in Ethiopia. Socio-demographic and clinical data were collected using questionnaire-based interviews. Descriptive statistics and bivariate and multivariate logistic regression were used to determine the overall and post-stratified seroprevalence and the association between seropositivity and potential risk factors.

Results: Our successfully developed in-house assay sensitivity was 100% in serum samples collected 2- weeks after the first onset of symptoms whereas its specificity in pre-COVID-19 pandemic sera was 97.7%. Using this assay, we analyzed a total of 1997 sera collected from HWs. Of 1997 HWs who provided a blood sample, and demographic and clinical data, 51.7% were females, 74.0% had no symptoms compatible with COVID-19, and 29.0% had a history of contact with suspected or confirmed patients with SARS-CoV-2 infection. The overall seroprevalence was 39.6%. The lowest (24.5%) and the highest (48.0%) seroprevalence rates were found in Hiwot Fana Specialized Hospital in Harar and ALERT Hospital in Addis Ababa, respectively. Of the 821 seropositive HWs, 224(27.3%) of them had a history of symptoms consistent with COVID-19 while 436 (> 53%) of them had no contact with COVID-19 cases as well as no history of COVID-19 like symptoms. A history of close contact with suspected/confirmed COVID-19 cases is associated with seropositivity (Adjusted Odds Ratio (AOR) = 1.4, 95% CI 1.1-1.8; p = 0.015).

Conclusion: High SARS-CoV-2 seroprevalence levels were observed in the five Ethiopian hospitals. These findings highlight the significant burden of asymptomatic infection in Ethiopia and may reflect the scale of transmission in the general population.

Keywords: Antibodies; COVID-19; ELISA; Ethiopia; RBD; SARS-CoV-2; Seroprevalence.

Fig. 1

Validation of the SARS-CoV-2 RBD specific IgG antibody detection ELISA. The value on the y-axis represents the ratio of OD₄₅₀ nm to the average mean OD₄₅₀ nm of the negative controls. The broken black line represents the cut-off value (2.5). We tested a total of 405 serum/plasms samples collected from cohort of mild and moderate (93.6%) and severe Ethiopian COVID-19 patients confirmed by RT-PCR (represented in red color). Of these 325 samples were collected during 0–7 days post-onset of symptoms (dps); 52 were collected during 8–14 dps, and 17 were collected within 15–28 dps (Additional file 1: Table S2). We also tested serum/plasma samples collected from 365 Ethiopian individuals before the global COVID-19 pandemic, represented in blue color (Additional file 1: Table S1)

Fig. 2

A map of Ethiopia showing the location of the study hospitals with corresponding SARS-CoV-2 seroprevalence. a Shows the location of five hospitals from which a total of 1997 healthcare workers enrolled between December 2020 and February 2021. b Shows the corresponding seroprevalence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). The y-axis of Fig. 2b represents the study hospitals. The x-axis of Fig. 2b shows crude seroprevalence rates (%) with 95% confidence intervals estimated by dividing the number of participants tested seropositive for immunoglobin G (IgG) antibodies elicited against the receptor binding domain (RBD) of the spike protein of SARS-CoV-2 to the total number of participants who provided sera and were tested