We describe the seeding events that have led to over 13,200 cases reported within 54 days of the first known introduction into the sub-Saharan African region. Most documented importation events were associated with recent travel history to Europe. As of April 21st, eight countries -- Burkina Faso, Cameroon, Côte d'Ivoire, Ghana, Guinea, Niger, Nigeria, South Africa -- had reported at least 500 confirmed cases each, while two countries had reported no cases.²⁹ Regional heterogeneity in reported case counts suggests that higher incidence in some places may be due in a large part to preparedness and current surveillance efforts, particularly around air travel passengers, and not solely due to more transmission. Relatedly, lower incidence in other countries may be due to reduced international air travel and the inability to surveil all air passengers and informal border crossings. Overall, importations into a country may take weeks or months to seed community spread at noticeable levels.³⁰ Many importations and subsequent transmission will go unnoticed without active monitoring through quarantine and testing to identify infections among travelers³¹, and risk communication directed at population subgroups most likely interacting with international travelers to encourage vigilance around symptom onset.

In terms of assessing community spread, lack of testing capacity and decisions to test upon symptom onset for quarantined individuals³² could also be misconstrued for lower incidence. More worryingly, among individuals with no recent travel history, conflation of disease symptoms may lead to lack of timely health seeking and reduced access to healthcare could prove detrimental as the disease progresses and transmission increases.³³ All of these factors could contribute to lower incidence being observed and reported. This in turn contributes to a false sense of the SSA region being less vulnerable to COVID-19 infection -- an argument that has been largely focused around age distribution alone -- when the sub-Saharan population may in fact be more vulnerable given the number of endemic, infectious and immunocompromising diseases,¹⁶ as well as increasing non-communicable conditions,¹² and poor critical care infrastructure in the region.

We found that existing preparedness indicators, namely the GHSI, was associated with the number of cases reported but not with risk of earlier introduction. The sensitivity of these indicators to capturing COVID-specific differences in countries’ vulnerability to widespread transmission warrants further investigation and analysis, accounting for the different efforts being implemented across the continent to prevent disease spread.^18,21 Our clustering analysis suggests that countries with fewer beds and healthcare workers and higher probabilities of premature death due to NCDs (thereby putting them at risk for severe COVID-19 morbidity and mortality) were less likely to report any cases during the first month post-introduction and are now reporting the highest case fatality ratios. Factors around population demographics and comorbidities uniquely relevant to COVID-19 morbidity and mortality may not be fully captured by the GHSI or other epidemic preparedness indices. The situation in SSA warrants context-specific consideration in terms of policies around timing of control efforts relative to reported case counts.

The COVID-19 situation is among the first pandemics in which high-capacity computing technology can be leveraged to quickly (and responsibly) disseminate large-scale information globally.^34,35 Information on importation events was extracted from online reporting platforms, such as situation reports, press releases from ministries of health, and dashboards, although more often, the data were being reported via social media accounts of political leaders and governmental institutions. This reflects a growing trend intended to better connect the populations they serve with updates on the situation locally and globally.³⁶

In addition to informing the general population, use of online media to share data on cases is facilitating rapid efforts at better understanding the epidemiology of the disease and its control. Notably, information from health agencies has guided mathematical modeling efforts to project case counts and assess potential impact of interventions, such as social distancing.^37-39 The robustness of modeling efforts to inform decision-making depends on the quality of data used to develop them. Especially at the start of an outbreak, it is important to distinguish between importations and community spread among all recorded cases in order to capture the growth of the epidemic accurately.

Data needs may be overwhelming local teams, as was observed with the decreasing data availability scores with increased numbers of imported cases. This reflects competing priorities and resources, not uncommon throughout the world, and could lead to delays in providing information to decision makers. The significance of such data extends beyond information on direct consequences of COVID-19 and includes indirect consequences -- for example, if data from routine surveillance and immunization activities or care-seeking behaviors become less available to capture detrimental impacts upon routine care, as has been documented during previous large-scale outbreaks.^40–42 Data on the burden of COVID-19, relative to the social and economic burden of control measures, are critical for informed decision-making, Support to data teams could enhance data collection so that modeling and other analyses can be done effectively and efficiently. Investing in this capacity could have long-term benefits to countries particularly those in SSA.