A commonly accepted hypothesis relates to many cases and deaths being undetected and unreported in this population due to a lack of efficient healthcare systems and limited access to resources in healthcare facilities (18). However, many African countries, such as Uganda, Kenya, Rwanda, Cameroon, Zambia, South Africa, and Botswana, made early efforts supported by the Africa Centre for Disease Control (Africa CDC) to strengthen their capacities in the face of the pandemic. These efforts included training healthcare personnel, procuring testing kits to enhance screening and isolation programs, and investing in basic preventative solutions, such as hand sanitizers, which were made readily available to the public. Emphasis was also put on effectively communicating good hygiene and safety guidelines through various social media channels and displaying of posters in public spaces (19).

The hypothesis of case underreporting was initially tested via in silico models. An epidemiological model revealed notable differences in levels of case underreporting across various countries in Africa. Whilst the average rate of SARS-CoV-2 case reporting in Africa was estimated to be 5.37%, Libya had the highest rate of 30.41%, and São Tomé and Príncipe had the lowest rate of 0.02% (20). Population-based models estimated that actual infection rates in Africa were no less than the rest of the world, and could potentially be higher (21). These results were supported by a meta-analysis which pooled 153 SARS-CoV-2 seroprevalence studies done in African populations. It reported that despite reports of substantially low case numbers, population levels of SARS-CoV-2 specific antibodies had significantly increased during the first year of the pandemic (22). This implied that a higher proportion of the population had been infected than reported in Africa, and challenged the notion of the African population being less susceptible to SARS-CoV-2 infection. Additionally, another study estimating excess mortality due to COVID-19 claimed that death underreporting was especially greater in sub-Saharan Africa than in other regions (23). However, the WHO Africa Regional Office and their collaborators disagreed with these findings (24). Their observations and analyses indicated that the under-reporting rate in Africa was comparable to the rest of the world.

However, it is also important to consider that results from serological tests can vary by commercial assay kits used and may not be directly comparable (25). Additionally, in silico modelling studies have a major limitation in that their predictions can be significantly skewed by a low accuracy of training parameters and baseline figures used, and the relative contributions of confounding variables accounted for (26).

Nonetheless, a statistical analysis that controlled for underreporting showed that Africa still saw fewer COVID-19 cases and deaths by a factor of 2.7 and 1.4 respectively, compared to all other continents (27). Due to a lack of data on excess mortality in Africa during the pandemic resulting from weak death registration systems, it is difficult to conclusively interpret SARS-CoV-2 mortality in the continent to ascertain whether the disease burden was genuinely lower, particularly in the older population.

Africa was the last continent to see its first case of SARS-CoV-2 infection, which was reported from Egypt on the 14th of February 2020 (28). It hence had the advantage of learning from the experiences of regions affected earlier, and imposing mitigation strategies in advance, unlike many other parts of the world which were caught unprepared. Moreover, controlling infectious disease epidemics was not a novelty for the African continent, as it had in recent years experienced managing multiple outbreaks, including those caused by Ebola (29) and chikungunya viruses (30).

Many African countries seized the early opportunity and adapted previous outbreak management plans to contain SARS-CoV-2. They began early surveillance at borders, restricted travel, and upon detection of the first few cases, imposed mandatory quarantines, national lockdowns, and curfews to limit social gatherings (28). Africa CDC played a key role in monitoring and coordinating responses across the continent. It also supported capacity building and response planning to enhance preparedness at the face of the pandemic (19). There were however a few countries, such as Tanzania, which did not implement recommended protective measures to protect its community throughout the pandemic, as political leaders questioned scientific guidance and were hesitant to believe in the existence of SARS-CoV-2 (31).

Regardless, as the pandemic progressed, many African countries found it difficult to sustain stringent restrictions for prolonged periods of time due to various political, economic, and social reasons. By the second wave in late 2020, restrictions had been relaxed in most regions (32). Thus, mitigation strategies were likely effective in the early phase of the pandemic, but had limited contributions in maintaining an overall reduced burden in the long run, particularly in the face of fragmented health systems (27).

Furthermore, a modelling study has reported that the stringency or timing of government-imposed restrictions was not predictive of COVID-19 mortality during the first wave of the pandemic across different countries in Africa (33).

Africa may have benefited from its warmer climate which could have limited transmission of SARS-CoV-2. Higher temperatures have been noted to restrict transmission of other respiratory viruses, like influenza, whose cases have been observed for years to peak during winters (34).

Initial studies carried out towards the beginning of the pandemic found inverse correlations between temperature and number of COVID-19 cases. This suggested that colder climates favoured SARS-CoV-2 transmission (35, 36). These findings were supported by lab studies showing greater persistence and increased stability of human coronaviruses in lower temperatures and more humid conditions (37).

Contrastingly, more recent studies have found that temperature and climate in isolation are not sufficient to significantly reduce SARS-CoV-2 transmission. It is a more complex interplay of weather, human behaviour and mobility that influences transmission rates. Individuals tend to stay indoors to keep warm in winter and minimize natural ventilation, this significantly enhances and favours virus transmission cycles (38, 39). Nonetheless widespread transmission was observed even in tropical regions with warmer climate.

Despite widespread transmission of SARS-CoV-2 across all regions, epidemiological studies have reported more cases being observed from urban areas compared to rural regions (33, 40). A study conducted in Kenya has also shown that living in urban areas compared to rural regions was associated with higher levels of pro-inflammatory cytokines and chemokines (41). Although access to healthcare resources and facilities may be limited in rural areas, it may be easier to disrupt transmission cycles in these regions which tend to have a lower population density (40). Moreover, a key aspect of the lifestyle of rural populations is that individuals spend the majority of their time outdoors (42), which further reduces chances of viral spreading (43). As Africa is the least urbanized continent with more than half of its population living in rural counties (44), it may have benefited from this geo-epidemiological profile.

Additionally, due to the more physically demanding nature of jobs in rural areas, and the consumption of traditional African heritage diets comprising of locally produced fresh foods, rural African populations may be at a lower risk of developing non-communicable diseases such as hypertension, diabetes, obesity, respiratory diseases, and cardiovascular diseases compared to Westernised populations, which have a more sedentary lifestyle and consume mostly ultra-processed foods (45, 46). However, food insecurity is still a significant issue in many African regions (47) with a high prevalence of malnutrition that increases vulnerability to infections (48). Nonetheless, these are important considerations to examine when assessing susceptibility of the African population to severe COVID-19, since non-communicable diseases have been described to be strong predictors of severe SARS-CoV-2 infection outcomes (49, 50).

The variation in immune-related host genetics within and between African populations may have also contributed to the heterogeneity in susceptibility to COVID-19, as has previously been noted with multiple infectious diseases including malaria, HIV, and schistosomiasis (51, 52). Severe COVID-19 has been associated with multiple genetic variants through genome-wide association studies (GWAS) (53). One such variant relates to the ABO blood group O having the lowest risk (54). In a study conducted in Nairobi on blood donors, this blood group was found to be more prevalent amongst indigenous African donors (49%) compared to the general donor population (45%) and Asian donors (34%) (55).

Importantly, SARS-CoV-2 internalises into host cells by binding to angiotensin-converting enzyme 2 (ACE2) receptors, but ACE inhibitors have been shown to be less effective in treating high blood pressure in patients of African descent (56) due to reduced expression of the ACE2 protein (57). This may subsequently have naturally reduced the likelihood of these individuals becoming infected, as well as limited progression of infection. Variations in human leukocyte antigens (HLA) have also been shown to influence the nature of immune responses, with certain types of HLA inducing activation of inflammatory immune pathways, while others reduce inflammatory responses (58). To the best of our knowledge, HLA types have not been mapped at the population level in Africa and so the impacts of prevalent HLA types in the context of SARS-CoV-2 infection in the African population are unclear.

Contrastingly, some studies have found that individuals of African descent residing in Europe or the United States were at an increased risk of progressing to severe stages of SARS-CoV-2 infection (59, 60). These findings imply that genetic factors are unlikely to have contributed significantly to reducing susceptibility of the African population to severe COVID-19, as if this were the case, then African descendants residing outside the continent would similarly have benefited from a reduced susceptibility. However, in England and Wales, this higher risk for African descendants has been correlated with increased exposure due to other factors that can enhance transmission of SARS-CoV-2, such as occupation and overcrowding in larger households (61). Therefore, whether immune-related host genetics played a key role in influencing outcomes of SARS-CoV-2 infection is still ambiguous.

In the early days of the pandemic, advanced age was quickly identified as a significant risk factor for severe COVID-19 (62). In contrast to the ageing population in many westernised countries, the majority of Africa’s population is youth. Hence, the population overall may have been less susceptible to severe disease, with a high proportion of infections being asymptomatic or only causing mild disease (40, 63).

Modelling studies, which aimed to understand the role of Africa’s demographics in shaping outcomes of SARS-CoV-2 infection, have strongly supported the hypothesis that low reports of severe disease and deaths from Africa are a result of its younger population (27). That is despite high infection rates, implying that the primary advantage was that of attenuated disease severity not reduced susceptibility (40). The observed outcomes of SARS-CoV-2 infection were even lower than that predicted by age-based models (6). This is because the calculations were derived from infection fatality rates determined from higher income countries. Since the health systems in these countries were overwhelmed, the calculations would have predicted a higher number of severe cases than was observed. This subsequently implies that age alone cannot explain the reduced burden of disease in Africa.

Data curated by the WHO in 2021 revealed that infectious diseases, particularly lower respiratory infectious diseases, malaria, and tuberculosis, were amongst the top five leading causes of death in Africa (64). This highlights the constant exposure to endemic pathogens in the African continent. Hence, through regularly battling with endemic infections, which can result in the acquisition of cross-reactive protection and broadly neutralising antibodies, the population’s immunity may have been better trained to recognise and rapidly clear new infections such as those caused by SARS-CoV-2.

Seroprevalence studies found cross-reactive antibodies to SARS-CoV-2 from previous exposures to human coronaviruses in both African and European cohorts (65, 66). However, the diversity of heterologous immunity may be wider in the African population due to their greater exposure to endemic pathogens (67). For example, patients infected with SARS-CoV-2 had reactivation of Plasmodium falciparum specific B- and T- cells depending on the severity of disease (68). This may be attributed to pre-existing cross-reactive protection (67), but may also be a result of pre-trained regulation of regulatory, pro-, and anti-inflammatory immune responses from previous malaria infections (69). This is because malaria also overstimulates the immune system (70). Thus, individuals from endemic regions may have a higher tolerance to pro-inflammatory immune responses, and may be less susceptible to succumb to severe immunopathology induced by progressive SARS-CoV-2 infection. Further research is needed to elucidate this theory.

It is also essential to take into account the fact that serology studies primarily tested for antibody titers to SARS-CoV-2 spike proteins. Although the highest homology of the SARS-CoV-2 spike protein is to the spike proteins from SARS-CoV-1 (76%) and MERS-CoV (35%) (71), these viruses have not been found to circulate in Africa (72). Hence any pre-existing cross-reactive antibodies in this population are likely from their constant exposure to other known and unknown endemic pathogens (67).

Furthermore, live vaccines such as the Bacillus Calmette-Guerin (BCG) vaccine are mandated in many African countries to protect against tuberculosis. Such routine vaccinations could also lead to the development of heterologous immunity by inducing long-lived non-specific responses that can offer broad protection against more infectious agents including SARS-CoV-2 (73). Hygiene and sanitation practices in low- and middle-income countries also tend to be less rigorous than in westernised regions, which further exposes individuals to more microbes and expands commensal microbial communities throughout their lifetime. As implied by the “hygiene hypothesis,” this contributes to enhancing the diversity of immune repertoires and priming immune responses to detect new antigens (74).

As mentioned earlier, the African continent carries the highest burden of infectious diseases globally. Sub-Saharan Africa has the highest global prevalence of the immune-modulating pathogen HIV (75) and parasitic helminths like Schistosoma mansoni (76). Ongoing chronic infections with these pathogens suppress the immune system and may influence both the efficacy and effectiveness of vaccines in all populations in endemic regions. Whilst HIV directly infects and depletes CD4 + T-cells (77), S. mansoni indirectly modulates the immune system by expressing and secreting immunoregulatory proteins which suppress Th1 immune pathways and promote Th2 responses (78). Vaccines, particularly those directed towards eliciting cellular immune responses, are therefore unlikely to induce the desired protective immune response in those infected with these parasites. A number of studies have in fact reported attenuated responses to various vaccines in individuals infected with HIV (79) and in those repeatedly exposed to helminth infections (80).

Africa bears the greatest burden of malnutrition, particularly in children under 5. Cases of malnutrition continue to rise and present significant public health challenges for these nations (48), including impaired immune responses to vaccines. Specifically, deficiencies in micronutrients such as calcium, potassium, folic acid, iron, zinc, selenium, and vitamins A, E, K, and D have previously been associated with attenuated pneumococcal vaccine-induced immunity (81). More recently, studies have also found that COVID-19 vaccinations induce weaker antibody responses in those who are underweight and malnourished (82).

As the prevalence of malnutrition is relatively higher in Africa compared to all other continents, and the efficacy of multiple vaccines, such as those against pneumococcus, rotavirus, polio, cholera amongst others, have been found to be low in malnourished individuals (83), the overall effectiveness of vaccines may be reduced in Africa.

Anaemia, a common comorbidity resulting from malnutrition and iron deficiency, is of particular importance for the African population. It has been shown to limit immune responses to many vaccines including those against pneumococcus, diphtheria, pertussis, and measles (84). In Africa, the prevalence of anaemia is greatest in children under 5 years of age, hence vaccine effectiveness may be greatly reduced in this vulnerable population.

With increasing urbanisation leading to more sedentary lifestyles and greater consumption of ultra-processed foods, the burden of non-communicable diseases continues to pose a major global health challenge, especially in many developing countries in Africa (85). Examples of such comorbidities include chronic liver disease, diabetes, hypertension, and coronary heart disease. Even though treatments for many of these conditions are available and can limit complications, healthcare is suboptimal in Africa and many face barriers of access due to healthcare costs (86). Because these diseases are known to weaken the immune system, they may also affect the level of protection induced by vaccines, especially when disease progression is not halted by effective treatment. Advancements in vaccine technology have however enhanced vaccine-induced immunity for patients living with these comorbidities (87). mRNA and viral vector COVID-19 vaccines validated this argument as they were found to be equally safe and effective in those with or without comorbidities (88).

The human microbiome, particularly the gut microbiome, has increasingly been shown to play a role in modulating the immune system (89). This implies that differences in microbiome composition between individuals and populations could drive varied responses to vaccines. Although the vast majority of studies characterising the human microbiome primarily exclude individuals from the Global South (90), there is some recent evidence suggesting that the microbiome of the African population is more diverse, and is composed of different proportions of varied taxa (91). The diversity and composition of the microbiome also varies across regions within Africa (92).

Given the gut microbiome’s direct role in the absorption and processing of oral vaccines, most experimental studies on mouse models and human population-based intervention studies have examined the gut microbiome’s impact on the efficacy and effectiveness of vaccines administered orally. These studies have shown significant differences in responses to oral vaccines across different geographical locations such as Ghana, Pakistan, and Bangladesh that can be attributed to diverse gut flora (93).

Responses to vaccines administered through alternative injectable routes have also been shown to be influenced by relative compositions of the gut microbiome (94–96). However, these studies only show associations, further studies are needed to determine causality and elucidate potential mechanisms by which the host microbiome can impact immune responses to vaccines.

Effectiveness of vaccines has been observed to be poorer in low- and middle-income countries and in rural regions (87, 97). Vaccine effectiveness calculations in these regions may however be skewed by limited coverage (98) which may be attributed to multiple factors, including affordability and access to vaccines (87). Since governments have limited budgets to acquire vaccines for their states, vaccination programmes may primarily be targeted towards high-risk, vulnerable populations who are more likely to experience severe disease. This would inadvertently reduce vaccine effectiveness estimates. Healthcare infrastructure in these areas is also often fragmented which may negatively impact vaccination programmes by further reducing coverage (99). Lack of appropriate storage facilities in remote rural regions further compromises vaccine potency and efficacy.

Literacy levels, which tend to be globally the lowest in sub-Saharan Africa (Our World in Data), also influence vaccine uptake. Individuals educated to a higher level and on the importance of vaccine uptake are more likely to accept and be willing to participate in vaccination programmes (100, 101).

Variation in immune-related genes within and between populations can influence efficacy of vaccines across different populations. For example, heterogeneity in HLA types and toll-like receptors (TLRs) have been found to induce varied responses to different types of vaccines (102–104). This is because HLA molecules play a key role in directing immune responses by presenting epitopes to T-cells and inducing T-cell activation, whilst TLRs recognise foreign antigens and stimulate initial responses by the innate immune system. As different HLA types and TLR polymorphisms differentially recognise the same epitopes, vaccine constructs may be more or less immunogenic across different populations.

A study found additional polymorphisms in other genes beyond those encoding for HLA types which influence antibody responses to COVID-19 vaccines. These include TP53, ABO, APOE, ACE2, and CRP-related genes (105).

Studies investigating dynamics of vaccine-induced immunity across different ethnic groups and ancestry have also found that both cellular and humoral immune responses subsequent to measles and COVID-19 vaccines vary significantly by ethnicity (106, 107). This heterogeneity has been supported by a study showing significant differences in pre-vaccination gene expression profiles of myeloid and B-cell specific genes between African Americans and European descendants (108).

Pre-existing immunity acts as a double-edged sword, as it can both positively and negatively influence the efficacy and effectiveness of vaccines. Factors such as timings between infection and vaccination, and antigenic divergence can influence whether prior immunity is beneficial or detrimental. Table 2A summarises studies on various vaccines against different pathogens which have shown pre-existing immunity to either protect against reinfection, or to augment vaccine effectiveness; whilst Table 2B summarises key studies which have observed and reported negative outcomes of pre-existing immunity on vaccine efficacy and effectiveness.

Pre-existing immunity can enhance vaccine effectiveness, as vaccination can further boost pre-existing protection. This benefit is pronounced with longer time intervals between exposures (109, 110). However, due to a higher level of baseline protection present in populations with pre-existing heterologous immunity, the relative gain from vaccination is lower compared to naïve populations. This thus results in relatively lower vaccine effectiveness in populations with prior protective immunity (111). This is accentuated in scenarios where pre-existing immunity provides even greater protection than vaccinations. For example, natural cholera infection has been found to induce long-lasting protection that significantly reduces the severity of disease in subsequent infections, and is more protective than vaccination (112).

However, pre-existing cross-reactive protection can also compromise the efficacy of vaccines. Immune interference, which can be caused by immune imprinting from previous infections and vaccinations, may limit the extent to which one’s immune system can be primed by new vaccinations to protect against emerging pathogens (14). Immune imprinting refers to a phenomenon whereby immune memory from initial exposures to similar immunogens is preferentially reactivated, even though it may not be the most appropriate for subsequent interactions with similar but unrelated immunogens. Hence, vaccines with antigens that are very similar to those that individuals have previously been exposed to, are likely to fall victim to immune interference (113, 114). As highlighted in Table 2B, this phenomenon has previously been observed and reported for vaccines against different pathogens, including recent COVID-19 vaccines.