Is Africa really less affected?

Africa represents 17% of the world’s population. Paradoxically, it represents only 5% of COVID-19 cases and 3% of worldwide deaths in July 2020. These figures are well below those in other countries (China, Europe, USA), where seroprevalence (the presence of anti- SARS-CoV-2 antibodies in the blood) was generally between 5 and 11% at the end of the first wave, and where deaths are counted in the ten of thousands. In Kenya for example, since the first declared case on the 12^th of March 2020 through to the 31^st of July 2020, 20 636 cases and 341 deaths were reported. However, there is not much information available about serological data in Africa In order to fill this gap, an Anglo-Kenyan team (Oxford and KEMRI) evaluated the presence of anti-spike antibodies (using ELISA testing) in a cohort of 300 Kenyans. Blood samples came from several regional blood transfusion centers (Kenya National Blood Transfusion Service, KNBTS) and were taken between April and June 2020 in people aged between 15 and 65 years old. Since the cohort was not necessarily representative of the population, the results were adjusted mathematically (age, sex, geographical distribution) to concur with the census of 2019. Amongst the 3 098 participants, an average of 4,3% tested positive, with the majority aged between 35 and 44 years old. Participants from large cities had a higher rate of seroprevalence, with the highest rate of 8% in Mombassa. This study, like others, confirms that sex is not a factor, and that seroprevalence diminishes with age. These results show two things:

• the Kenyan population has had significantly greater exposure to SARS-CoV-2 than was thought.
• prevalence is comparable to that observed in other countries where known cases are much higher.

If applied to the total Kenyan population of 53 million people, this would correspond to 1,3 million people infected amongst the 15-64 age group. But on the 30^th May 2020, only 2 093 cases had been detected, with 71 deaths. How can this divergence between a seroprevalence similar to other countries and a relatively low number of COVID-19  cases be explained? 4 theories have been put forward:

• Seroprevalence may be overestimated because of selection bias in the choice of participants. 43% of the Kenyan population is made up of people younger than 15 and older than 65 who have not been infected by SARS-CoV-2 in large numbers.
• Cases may have been underestimated by health authorities (insufficient PCR testing). However Kenyan hospitals have not been overloaded with severe cases of COVID-19.
• Demographical factors may influence the results: 3,9% of the Kenyan population are aged over 65, against 23,3% in Italy.
• The population is frequently exposed to common coronaviruses that may confer a certain cross-protection, although this has never been shown.

The authors do not think that these hypotheses are sufficient to explain the differences in figures. They suggest that it is more likely, as a previous study suggested, that COVID-19 symptoms may be lessened for reasons that are yet to be clarified. The authors also underline the fact that in the absence of social and economic protection, measures such as lockdown could have potentially catastrophic consequences. It is therefore crucial to increase testing provision and to carry out further studies in order to better understand the course of the epidemic in Africa.