Date of Award


Document Type


Degree Name

Master of Medicine (MMed)

First Supervisor/Advisor

Daniel Maina

Second Supervisor/Advisor

Geoffrey Omuse


Pathology (East Africa)


Background: Tests for antibodies against severe acute respiratory syndrome coronavirus 2, the aetiologic agent of coronavirus disease 2019, provide evidence of past infection or immunization. At the individual level, they are useful as markers of protection against severe disease and in cases where the diagnosis by nucleic acid amplification is inconclusive. At the population level, they are an indirect measure of exposure and may be useful in monitoring changes in viral infection or uptake of immunization over time. The interpretation and optimal use of antibody assays on populations requires an appreciation of the wide variability in individual antibody responses, the timing of testing, and individual assay performance. Direct comparison of assays in representative populations is one way of providing information that will aid in optimization of their use.

Objectives: The primary objective in this pre-vaccine study was to determine the difference in estimates of prevalence of anti-SARS-CoV-2 antibodies in a population of health care workers tested using two qualitative immunoassay platforms targeting two different SARS-CoV-2 proteins. Secondary objectives were: (i) to determine the proportion of discordant results, (ii) to investigate sex, age, and time since exposure as predictors of discordance, and (iii) describe the changes in antibody level with time for each assay.

Methods: This was a secondary analysis of data collected from an earlier cross-sectional study that determined seroprevalence of antibodies against nucleocapsid and spike proteins of SARSCoV-2 in sera collected from the study population. Antibodies against SARS-CoV-2 were detected using two different assays detecting anti-nucleocapsid and anti-spike antibodies in samples collected at a single time point for each participant. The proportion with a positive result by each assay was determined, as was the proportion of discordant results. Age, sex, and time as predictors of discordance were explored using a binomial logistic regression model. Plots of antibody index against time were prepared for each assay.

Results:A total of 1477 serum samples from 643 (43.5%) males and 834 females were analysed. This study found a prevalence of 18.3% (95% CI, 16.4% – 20.3%) and 26.5% (95% CI, 24.2% – 28.7%), respectively, of antibodies against nucleocapsid and spike. Overall, 162 (10.9%) participants had discordant antibody results. Of the 107 individuals with a prior PCR positive result, discordance was found in 27 (25.2%). Discordance was found to increase with v time since positive PCR; 100% of results of samples collected at 28 weeks post PCR were discordant. One hundred forty-one (87.1%) of participants with discordant results had a positive spike result and negative nucleocapsid result; this figure was 74.1% in those with a prior positive PCR result. Age (OR = 0.94, 95% CI 0.89 – 1.00, P

Conclusion: Following natural infection by SARS-CoV-2, the seroprevalence of anti-spike antibodies is higher than that of anti-nucleocapsid antibodies in a population naïve to the SARS CoV-2 vaccine. Discordance in antibody result is more likely the longer the interval between exposure and testing. Antibodies against spike remain detectable for a longer duration than those against nucleocapsid. Assays detecting anti-spike antibodies would result in higher estimates of seroprevalence in surveillance studies.

Included in

Pathology Commons