Select Country

Hotspot Purpose, Measures, and Scoring Criteria

Purpose

This dashboard was designed for internal use by PERC partners to identify countries that have growing or widespread outbreaks of COVID-19. This dashboard was intentionally designed to be highly sensitive at the cost of specificity, meaning that countries may be assigned high hotspot levels at time when transmission in the country is low. PERC partners use this dashboard to identify countries where a deeper dive into that country’s data and outbreak status is warranted and where technical support around COVID-19 may be beneficial. Hotspot levels are assigned based on the numbers of cases and deaths each country reports, and hotspot levels may be assigned too low if a country is not testing, identifying, and reporting their COVID-19 cases.

What is each measure?

Test-Case Ratio Adjusted Incidence (Cases/1 million population/week)

This indicator estimates the incidence rate for each country if they were conducting enough testing to achieve a test-case ratio of ≥20 (equivalent to 5% test positivity). Incidence is linearly scaled upward until a test-case ratio of 20 is achieved. For example, a country with a test-case ratio of 10 would have their incidence doubled to account for its low testing rate. For countries with a test-case ratio ≥20, no adjustment is made.

If a country reported no testing data over the prior 2 weeks, or if it reported more cases than tests over the prior 2 weeks, the warning level for this indicator is automatically increased one level above the level corresponding to that country’s unadjusted incidence rate. For example, if a country has an unadjusted incidence rate of 50 cases/1M/week that corresponds to warning level 2, but that country has not reported testing data, that country would be assigned warning level 3 for this indicator.

Incidence is calculated per week, and an average for the most recent two weeks is used to calculate the test-case ratio.

Equations:

If test per case ratio ≥20:

\[{[New {cases \over week} *1\ Mil.] \over population}\]

If test per case ratio <20 and ≥1:

\[ \frac{[New {cases \over week} *1\ Mil.]}{population} * \frac{[20\ tests\ per\ case\ (Target)]}{[Tests\ per\ case\ (2\ week\ avg.)]} \]

If test per case ratio <1 or no tests reported in prior two weeks, and then automatically increase warning level by 1:

\[ \frac{[New {cases \over week} *1\ Mil.]} {population} \]

Trend in Incidence Rate

This indicator calculates the percent change in the number of new cases across weeks, to determine if the incidence of COVID-19 is trending upwards or downwards.

Equations:

Increase for the latest week:

\[ \frac{[New\ cases\ most\ recent\ week]-[New\ cases\ 1\ week\ prior]} {[New\ cases\ 1\ week\ prior]} \]

Increase from latest week to two weeks prior:

\[ \frac{[New\ cases\ most\ recent\ week]-[New\ cases\ 2\ weeks\ prior]} {[New\ cases\ 2\ weeks\ prior]} \]

Increase in two consecutive weeks:

\[ \frac{[New\ cases\ most\ recent\ week]-[New\ cases\ 1\ week\ prior]} {[New\ cases\ 1\ week\ prior]} \]

AND

\[ \frac{[New\ cases\ 1\ week\ prior]-[New\ cases\ 2\ weeks\ prior]} {[New\ cases\ 2\ weeks\ prior]} \]

Deaths Rate (deaths/10 million population/week)

This indicator calculates how many COVID-19 deaths a country has reported over the prior week adjusted per capita. If the number of deaths reported is 0 for the prior week and a country reported 0 cases for the prior week, it is assumed that no data was reported for the country.

Equation:

\[ \frac{[New\ deaths/week]}{population} * 10,000,000 \]

Trend in Death Rate

This indicator calculates the percent change in the number of new COVID-19 deaths across weeks, to determine if the death rate from COVID-19 is trending upwards or downwards. If the number of deaths reported is 0 for the prior week and a country reported 0 cases for the prior week, it is assumed that no data was reported for the country.

Equations:

Increase for the latest week:

\[ \frac{[New\ deaths\ most\ recent\ week]-[New\ deaths\ 1\ week\ prior]} {[New\ deaths\ 1\ week\ prior]} \]

Increase from latest week to two weeks prior:

\[ \frac{[New\ deaths\ most\ recent\ week]-[New\ deaths\ 2\ weeks\ prior]} {[New\ deaths\ 2\ weeks\ prior]} \]

Increase in two consecutive weeks:

\[ \frac{[New\ deaths\ most\ recent\ week]-[New\ deaths\ 1\ week\ prior]} {[New\ deaths\ 1\ week\ prior]} \]

AND

\[ \frac{[New\ deaths\ 1\ week\ prior]-[New\ deaths\ 2\ weeks\ prior]} {[New\ deaths\ 2\ weeks\ prior]} \]

How is each measure scored?

Indicator Test-case ratio adjusted incidence Level 1: Lower Risk <35 cases/1M/week Level 2: Medium Risk 35 to <150 cases/1M/week Level 3: High Risk 150 to <1000 cases/1M/week Level 4: Very High Risk ≥1000 cases/1M/week
Indicator Trend in incidence rate Level 1: Lower Risk <20% increase in the latest week AND <30% increase from the latest week to two weeks prior OR <75 cases over the prior 3 weeks Level 2: Medium Risk >20% increase in the latest week OR an increase from 0 to ≥1 case in the latest week OR >30% increase from the latest week to two weeks prior Level 3: High Risk >40% increase in the last week OR >15% increase in two consecutive weeks Level 4: Very High Risk ≥25% increase in two consecutive weeks
Indicator Death Rate Level 1: Lower Risk <10 deaths/10M/week Level 2: Medium Risk 10 to <30 deaths/10M/week Level 3: High Risk 30 to <100 deaths/10M/week Level 4: Very High Risk ≥100 deaths/10M/week
Indicator Trend in death rate Level 1: Lower Risk <20% increase in the latest week AND <30% increase from latest week to two weeks prior OR no deaths reported in the last week Level 2: Medium Risk >20% increase in the latest week OR an increase from 0 to ≥1 death in the latest week OR >30% increase from the latest week to two weeks prior Level 3: High Risk >15% increase in two consecutive weeks Level 4: Very High Risk ≥25% increase in two consecutive weeks

How is the overall hotspot warning level determined?

The overall hotspot warning level of a county is determined by whichever of the four indicators is at the highest warning level. This is done to ensure that the system identifies countries as soon as data suggests that there may be a growing outbreak. This system was not designed to identify which countries have the worst outbreak or highest incidence. Rather, the objective is to highlight countries that should receive attention either due to an increasing or widespread outbreak.

What is our data source?

Africa CDC receives cases, deaths and testing data from each African Union Member State. The data is updated twice a day, at 9am and 5pm EAT. It is not uncommon for countries to report data several days late, and totals for past days will be updated as additional data is reported to Africa CDC. The most recent date that each country reported a case to Africa CDC is indicated on the “Most recent date data available” column on the front page of the dashboard. Additionally, some countries may report cases and deaths, but not testing data, and some countries report testing data in periodic batches (eg. weekly) rather than providing daily statics.

Information on other key measures

Trend measures (New Case, Death and Test Trends)

All trend measures are calculated according to the formula:

\[ \frac{[Total\ most\ recent\ week]-[Total\ 1\ week\ prior]}{[Total\ 1\ week\ prior]} * 100 \]

New Case Trend (2 week):

New Death Trend (2 week):

New Test Trend (2 week):

Warning Level Trend

This measure indicates if the warning level has changed since the prior week.

Most Recent Epi Data

This measure indicates the most recent date that each country reported either a case or a death count greater than 0. Because the dashboard is updated weekly, dates beyond the end of the current week are not displayed.

For example, if the end of the current week being shown on the dashboard is Feb 15th:

Most Recent Testing Data

This measure indicates the most recent date that each country reported a COVID daily new tests count greater than 0. Because the dashboard is updated weekly, dates beyond the end of the current week are not displayed.

For example, if the end of the current week being shown on the dashboard is Feb 15th:

Severe Crisis

This measure identifies countries that have an ongoing crisis that may impact both COVID programming and the quality of reported COVID data. The hotspot warning level of countries with a severe crisis should be interpreted cautiously. A lack ability to conduct COVID activities or report high quality COVID data may cause this warning system to underestimate the hotspot warning level.

This metric is derived from ACAPS INFORM Severity Index. This index is updated and released approximately one per month, and ACAPS provides a country level score that is designed to provide the overall crisis severity for each country facing at least one ongoing crisis. Any country with an INFORM Severity Index Score ≥4.0 is indicated as having a Severe Crisis on the dashboard.

© 2021 Africa CDC