JMIR Publications

ABSTRACT

Background:

Since the novel coronavirus emerged in late 2019, the scientific and public health community around the world have sought to better understand, surveil, treat, and prevent the disease, COVID-19. In sub-Saharan Africa (SSA), many countries responded aggressively and decisively with lockdown measures and border closures. Such actions may have helped prevent large outbreaks throughout much of the region, though there is substantial variation in caseloads and mortality between nations. Additionally, the health system infrastructure remains a concern throughout much of sub-Saharan Africa, and the lockdown measures threaten to increase poverty and food insecurity for the subcontinent’s poorest residents. The lack of sufficient testing, asymptomatic infections, and poor reporting practices in many countries limit our understanding of the virus’s impact, creating a need for better and more accurate surveillance metrics that account for under-reporting and data contamination.

Objective:

The goal of this study is to improve infectious disease surveillance by complementing standardized metrics with new and decomposable surveillance metrics of COVID-19 that overcome data limitations and contamination inherent in public health surveillance systems. In addition to prevalence of observed daily and cumulative testing, testing positivity rates, morbidity and mortality, we derive COVID transmission in terms of: 1) speed, 2) acceleration or deceleration, 3) change in acceleration or deceleration (jerk), and 4) 7-day transmission rate lag which explains where and how rapidly COVID is transmitting, and quantifies shifts in the rate of acceleration or deceleration in order to inform policies to mitigate and prevent COVID and food insecurity in sub-Saharan Africa.

Methods:

We extracted 60 days of COVID data from public health registries and employed an empirical difference equation to measure daily case numbers in 47 sub-Saharan countries as a function of the prior number of cases, the level of testing, and weekly shift variables based on a dynamic panel model that was estimated using the generalized method of moments (GMM) approach by implementing the Arellano-Bond estimator in R.

Results:

Kenya, Ghana, Nigeria, Ethiopia and South Africa have the most observed cases of COVID and the Seychelles, Eritrea, Mauritius, Comoros, and Burundi have the fewest. In contrast, the speed, acceleration, jerk, and 7-Day Lag indicate rates of COVID transmissions differ from observed cases. In September 2020, Cape Verde, Namibia, Eswatini, and South Africa had highest speed of COVID transmissions at 13.1, 7.1, 3.6, and 3 infections per 100,0000; Zimbabwe has an acceleration rate of transmission while Zambia has the largest rate of deceleration this week compared to last week referred to as a jerk. Finally, the 7 Day Lag or persistence rate indicates the number of cases on Sept 15, 2020 that are a function of new infections from September 8, 2020 decreased in South Africa from 216.7 to 173.2 and Ethiopia from 136.7 to 106.3 per 100,000. The statistical approach is validated based on the regression results; they determine recent changes in the pattern of infection; and during the weeks of September 1-8 and September 9-15 there were substantial country differences in the evolution of the SSA pandemic. This change represents a decrease in the transmission model R value for that week, and is consistent with a de-escalation in the pandemic for Sub-Saharan African continent in general.

Conclusions:

1) Standard surveillance metrics such as daily observed new COVID cases or deaths are necessary but insufficient to mitigate and prevent COVID transmission. Public Health leaders also need to know where COVID transmission rates are accelerating or decelerating, whether those rates increase or decrease over short time frames because the pandemic can quickly escalate, and how many cases today are a function of new infections 7 days ago; and 2) Even though SSA is home to some of the poorest countries on the globe, development and population size are not necessarily predictive of COVID transmission, meaning higher income countries like the USA, can learn from African countries on how best to implement mitigation and prevention efforts.