Insights from mathematical modeling tell the story of what could be wide regional differentials in population-normalized COVID-19 testing rates.
As of January 19, 2021, there were more than 96.6 million confirmed COVID-19 cases globally, more than 3.3 million or 3% of those cases in Africa. The recent global case fatality rate has been more or less constant at 2.1% with a recovery rate of 72%. These figures are a noticeable contrast from Africa’s average case fatality rate of 2.4% and recovery rate of 83%. For a continent with 17% of the global population, these statistics apparently tell of a region that has been disproportionately spared the worst statistics. Deeper insight from mathematical modeling, however, tells a different story of what could be wide regional differentials in population-normalized COVID-19 testing rates. Kenya qualifies as Africa’s suitable representative for testing this hypothesis.
The Kenyan COVID-19 Case
Lately, Kenya has stayed in the list of top-ten countries in Africa with the highest number of COVID-19 cases. On January 19, 2020, Kenya had 99,308 confirmed COVID-19 cases, 15.2% of them being active cases following a case fatality rate of 1.7% and a recovery rate of 83%. Of the active cases in Kenya, 0.2% were either serious or critical. Kenya’s total cases and active cases made up 3% of Africa’s total cases and active cases, respectively. This outcome compared well, proportionately, with the population of Kenya and Africa. The serious or critical cases in Kenya made up 1% of Africa’s serious or critical cases. Kenya’s curve of positivity rates has been on a steadily declining trend since December 2020, from 10% to below 2.5% as shown in the graph.
The model projections remained within 10% margin of error, which is within the boundary recommended for most policy and planning purposes.
Point of Curiosity about Kenya’s COVID-19 Curve
A genuine point of curiosity must be about the effect of the festive season and school reopening on the COVID-19 transmission trends and what has become of the expected surge in cases following these two triggers. As shown in the graph, the actual curve of COVID-19 cases in Kenya has stayed below the red lines simulating a surge in cases in this modeling series. The model simulation estimated that Kenya’s current COVID-19 curve would theoretically peak at 103 188 cases on January 14, 2021. Between January 1 and January 14, the simulations of a curve peaking on January 14 remained within 4.6–5.5% above the actual cases. The country confirmed 89 771 total cases on January 14, 2021 (4.6% difference). The model projections remained within a 10% margin of error, which is within the boundary recommended for most policy and planning purposes. The COVID-19 trend observed in Kenya has, however, already invalidated the possibility of having the previously simulated peak of 99 000 cases on April 11, 2021 (simulated under this IBD model series).
Explaining Kenya’s COVID-19 Curve and Map: Testing Rates and Coverage
The recent progression of Kenya’s normalized testing rate, measured in tests/million people/day, has not been as encouraging as it used to be in July, October, and November. Over a period exceeding a month, the testing index has only increased marginally from 65.0 on December 12, 2020, to 66.6 on January 19, 2021. Going by the best progression rate observed earlier in Kenya, a daily increment of 0.5 in the index, the normalized testing rate expected on January 19 would be 84 tests/million people/day. Uganda has witnessed a similarly slow pace, from 55 to 57 over the same period. Ethiopia and Egypt represent African cases where the index has reduced by up to 2–4 points over the same study period. Outside Africa, a similar recession in testing rates over the study period has been evident in China, Brazil, India, and Peru. The better African examples of an increase in the normalized testing index over the study period, covering December 12, 2020 to January 19, 2021, are South Africa (+57 to reach 403), Morocco (+39 to reach 434), and Rwanda (+11 to reach 197).
In a wide contrast, the increase on the same measure in the normalized testing index for Israel, the USA, and leading European countries has been remarkable. Over the same study period, the US testing index has increased from 2023 to 2391 tests/million people/day, the UK from 2185 to 2700, Spain from 1626 to 1817, Germany from 1132 to 1207, Canada from 978 to 1175, and Israel from 2059 to 3174. The rising COVID-19 cases in these countries have informed tight containment measures and enforcement of rules. The recent surge in COVID-19 numbers cannot be divorced from the observed upward trend in these testing rates. The European countries that were recently ranking lower in the list of leading COVID-19 cases have, consequently, risen to the top again. Despite the observed recession in their normalized COVID-19 testing rates, India and Brazil have still remained top in the global list.
Spatial intelligence is a key aspect of the visual evidence needed to calibrate containment responses with active public participation. The spatial stratification of a sampling exercise has a strong role to play in revealing the actual geographical spread of confirmed COVID-19 cases. The status map below displays the dominance of Kenya’s confirmed COVID-19 statistics in the Nairobi metropolitan region and key urban counties: Nairobi (42.4%), Mombasa (9.0%), Kiambu (6.5%), and Nakuru (4.8%).
The realized surge in COVID-19 cases in the USA and leading countries in Europe has strong linkages with their substantially increasing daily average testing rates, normalized by population. Israel and the UK, for example, have respectively progressed on the normalized testing index by a daily increase of 29 and 14 over the study period. This progression is far ahead of the slow-paced daily incremental trend of only 1.5 in South Africa, which is currently the top example of increasing testing rates in Africa. The gradual reduction in the normalized testing rates for India and Brazil portend the likelihood of losing the grip of the true picture of their COVID-19 curves, a fact that should be revealed when their testing rates improve before attaining herd immunity.
Since the efficacy of testing is a critical source of the actionable intelligence needed to understand the actual COVID-19 curve and inform the timing of policy and strategic responses, there still remains much-uncovered ground ahead for Kenya and similar countries with slow or reducing testing rates. Enhancing the targeted penetration of the COVID-19 tests across the vast populations is a critical requirement. As normalcy bias sets in, there is likely to be a continuing decline in the COVID-19 testing rates across Africa. A few African representatives such as South Africa, Morocco, and Rwanda are showing the right way forward in sustaining a progressive trend of population-normalized testing rates. Despite vaccine optimism, the attainment of herd immunity could still take more time given the inherent challenges of logistics and large country population sizes.
Adequate and targeted testing, leveraged with spatial intelligence, would reveal a more accurate situation of transmissions in schools, the highly populated urban slums, and the rural parts where laxity is common. Awareness-raising among the majority of skeptical citizens would help turn them into active participants in the citizen science geared towards containing COVID-19, citizen science being an inclusive and cost-effective way of enhancing modern scientific research. The current threat of new variants of the novel coronavirus increases the urgency of adequate testing and relentless research to gain better insights into the local nature and dynamics of the pandemic. For now, existing knowledge warrants the continuation of strict COVID-19 containment measures with discipline, care, and cautious optimism. Kenya and many African countries, which are facing capacity limitations inadequately testing their large populations, are the main would-be beneficiaries of these lessons
Nashon, a geospatial expert, lecturer and trained policy analyst applies dynamic models to complex adaptive systems. He is a youth mentor on career development and the founder of Impact Borderless Digital.