A new projection, jointly conducted by the Directorate General of Health Services (DGHS) and Institute of Health Economics (IHE), has estimated that Bangladesh is set to hit a peak with almost 300,000 Covid-19 cases in the second half of July.
However, they said the new projection, which was made using data available at Institute of Epidemiology, Disease Control and Research (IEDCR), could differ provided adequate measures in prevention, testing, isolation and treatment are taken.
The new prediction also contradicts a previous prediction that stated that the country will enter its peak in Covid-19 infection in the first two week of the current month, June 30 to July 15.
When contacted, projection modeler IHE Associate Professor Shafiun Nahin Shimul said the projections were made using three models – SIR, time series and data analysis model.
SIR is one of the simplest mathematical models of disease spread that splits the population into three basic categories, according to disease status where “S” is the number of susceptible individuals, “I’ is the number of infected individuals and “R” is the number of recovered individuals.
Time series models are based on statistical methodology for time-indexed data and for forecasting analyzing the real world time series data of confirmed and recovered Covid-19 cases. Data analysis model projects the peak using real time data.
“The previous projection was made on June 20. SIR (also used by IEDCR) projected that we are already in the peak while time series projected that the peak would be in mid July,” Prof Shimul said.
“But, we have new data and while working on it, the time series model projected that the country will hit peak in late July with 4,500 to 5,000 cases per day.
"The time series model earlier predicted that if the testing number was increased, we would have some 140,000 to 160,000 confirmed cases in the country, that was projected to be under 100,000 cases by SIR. Now, the time series model projects that Bangladesh will have some 300,000 confirmed within this month,” he added.
According to DGHS, there were already 153,277 confirmed cases in Bangladesh.
The data analysis model is now projecting that the country will be hit by the peak in early August with some 6,000 to 6,500 cases per day.
The modeller said: “From Bangladesh's experiences, the time series model has been more effective in making good predictions.”
Talking to Dhaka Tribune on Tuesday, Dr Abu Jamil Faisel, a member of the public health expert team for the eight divisions under DGHS, said as per the earlier projection made on June 20, the peak period for Covid-19 in Bangladesh was set from June 30 to July 15.
“As per this prediction, there would be a month of plateau and there will be high cases till August and then it will start to flatten. The country might start to go back to normalcy from September,” he said.
But, the situation can be altered if the factors considered at the model varied, said Dr Faisel, who is also the former president of Public Health Association of Bangladesh.
IEDCR data not representing real scenario
IHE Professor Syed Abdul Hamid, who is also a member of the team, said the projections being made using data available from IEDCR, which however does not represent the real scenario.
“Projection with such data will never be accurate and there remains a chance that the country will enter the peak day later than that of projection,” he added
And for this, preventive measures should strictly be implemented and test numbers should be increased to 20,000 to 25,000 per day.
Besides, the novel coronavirus is showing uniqueness characteristics every day. Many people are getting re-infected by the virus, although it is a mild infection. As there is no data on how many people are contracting the infection more than once, the projection may change, the professor said.
Prof Shimul said: “Unless we can increase our testing capacity, we will not be able to detect as much as 6,500 confirmed cases per day. Moreover, infection may even soar during Eid and in that case, data analysis results are very likely to be inaccurate.”
He termed the decision to fix fees for Covid 19 tests as discouraging for the poor and said it might create class discrimination in terms of receiving results.
“Now they have to pay and also to stand in long queues to avail testing facilities. It will be a double punishment for the poor,” he added.
He also said that although the government has been craving for a lower percentage of confirmed cases in comparison to the number of tests, only a small number of countries in the world have witnessed such a higher number of confirmed cases like Bangladesh.
The average confirmation rate is only 7% throughout the world, the researcher added.
The health authority could establish a procedure, such as the one adopted during the dengue outbreak, to make people understand that the government wants them to test as much as possible, the project modeler opined.
