Uganda’s COVID-19 model can be used across Africa. – National Planning Authority.
Written by Editor on July 7, 2021
The National Planning Authority (NPA) has revealed that the model developed to make predictions of COVID-19 infection cases for every two weeks can be used by any African country.
The model has been in use since 2020 and has proved to be 97% accurate leaving only a 3% degree of error. It was published on June 28 in the proceedings of the National Academy of Sciences (PNAS) of the USA.
It was developed as a result of a research collaboration between NPA Science Department and Pennsylvania State University with funding from the National Institute of Health (NIH), USA aimed at determining the causes of neonatal mortality in Uganda.
This was revealed by the Executive Director of NPA, Dr. Joseph Muvawala during a press briefing convened at the Uganda Media Centre on Wednesday in Kampala.
Development of the model was initiated following a directive by President Yoweri Museveni four years ago that portrayed the need for the creation of a Science department at the NPA to aid planning and decision making in the country.
After this, a number of models have been developed in line with this purpose. The COVID-19 model projects infection cases a fortnight prior to the actual date.
The model projects a total of 7,214 cases for 4th to 10th of July as compared to the case for the week of 27th to 3rd July where cases were summed up to 7,259.
“In the week of June 12th to 19th, the model projected 10,144 cases and the actual reported cases were 9,926. In the week of June 20th to 26th, the model projected 10,468 new cases and the actual reported cases were 7,329,” Muvawala noted.
Muvawala further revealed that the model projects COVID-19 cases to rise and most likely reach a peak in the mid of July as was the case in June. This implies that in mid July, the COVID-19 infections will be reaching the maximum peak.
The model thus indicates that the measures of the partial lockdown that had been put in place on June 6 were not effective in terms of curbing down the spread of the virus, especially the increased transmissibility of new variants and decreased compliance with interventions.
He, against this reasoning, noted that the full lockdown measures that were put in place by the government were very necessary. He hinted that the extension of the lockdown after the 42 days will depend on the increase or decrease in the infection rates.
Muvawala urged the public to strictly observe the Standard Operating Procedures (SOPs) put by the Ministry of Health to beat down the crisis bearing in mind that the current wave is influenced by internal factors unlike in the previous wave where factors influencing the spread were mostly from outside the country.
“The public is advised to strictly adhere to the SOPs to slow the rising cases and the consequences on the health sector. There is also need for increased testing, more enforcement of the lockdown measures to curb community transmissions and increased public awareness aimed at behavioural change in the local population,” Muvawala said.
Muvawala pledged that NPA will continue giving the bi-weekly projections and trends in order to support the Ministry of Health in effective decision making.
More about the model
The modeling tools used for the projection are based on the “Endemic-epidemic modeling” (HHH4 model) propounded by Leonard held in 2005 in which spatial-temporal data of parameters that can explain the factors that affect the spread and reporting of COVID-19 from all African countries were tested, observing trends that are most consistent with the number of reported cases in different countries.
The factors incorporated in the model give insights of how parameters from these three categories influence the spread of the pandemic in Africa namely; geographic, demographic and economic factors and Human Development Index (HDI).
Government responses (a composite measure of containment and health policies introduced in the country to manage the pandemic) and Weather factors (Temperature, humidity, rainfall) were also considered in the model because most respiratory diseases are more prevalent in cold, humid conditions.