[ad_1]
BATTLE VS COVID-19. A police officer uses a thermal scanner to check a Filipina’s body temperature upon entering Laguindingan Airport in Laguindingan, Misamis Oriental. Region 10 Tourism Department Photo
MANILA, Philippines – Is the Philippines already “flattening the curve” in its fight against the new coronavirus?
While the Department of Health (DOH) and some doctors have made this claim, Associate Professor of Mathematics at Manila University of Athenaeum, Felix Muga II, said the data does not support this.
At the Cure COVID virtual press conference on Friday, May 8, Muga said 3 things show that the curve has not yet started to flatten out. (Watch the press conference here).
First, the epidemic curve or the number of new daily cases reported by DOH continues to advance upward.
Presentation of Felix Muga II during the Cure COVID press conference
Presentation of Felix Muga II during the Cure COVID press conference
Graphing the 5-day moving average of new cases per day also shows a line that moves up, rather than down.
“We can see that the red line starts low on March 7, but it shoots up on April 3, it goes down and then it goes back up. So it doesn’t go down, it doesn’t flatten. You can see that the end of the red line is still at up gear, “he said in Filipino.
Study the government charts closely. The same goes for the line graph of the total number of coronavirus cases reported by DOH every day. Because there are always new cases every day, this line continues to rise.
However, using the same data on a logarithmic scale graph shows the flattening of the line, a possible reason why some claim that the Philippines has flattened its curve.
Total cases visualized using linear scale of the Felix Muga II presentation
Total cases visualized using the logarithmic scale of the Felix Muga II presentation
A logarithmic scale displays numerical data over a very wide range of values, with the largest numbers in data hundreds or even thousands of times larger than the small numbers.
Therefore, a line going upward on a linear scale may appear to be flattened on a logarithmic scale.
“What DOH uses for its total cases is a logarithmic curve. They used a logarithmic scale on the Y axis, but if you use a linear scale, the line really goes up,” Muga said.
He advised people to closely study the type of charts used by the authorities to present the coronavirus data.
Play number still above 1. Another indication that the curve has not yet flattened is the number of replication of the virus in the Philippines.
This “R” number is the number of people that a COVID-19 positive person can infect.
In order to say that the virus is not spreading, the breeding number must be below 1. Muga says the country’s R number “is hovering above 1.”
But the data shows a slower rate of duplication of cases, Muga said.
From the presentation of Felix Muga II
Before April 7, the total number of active cases doubled every 5 days. After April 7, this decreased to a 42-day doubling rate.
DOH has reported a recently reached test positivity rate of 10% as the basis for its “curve flattening” claim.
The positivity rate is the percentage of people who test positive out of the total number of people tested. The World Health Organization has told countries to look for a positivity rate of 10% or less.
The Philippines has 10,463 confirmed cases of coronavirus and 696 deaths as of Friday, May 8. It has the third-highest number of cases in Southeast Asia, after Singapore, which has more than 21,000 cases, and Indonesia with 13,000 cases.
But the number of 10,463 is expected to increase, not only because of really new cases, but also because of the “backlog” of DOH cases, Muga said.
DOH reported a total of 13,655 people who tested positive on May 7. The difference of approximately 3,000 between this number and the total of confirmed cases (10,463) has yet to be validated, said the undersecretary of Health, María Rosario Vergeire.
But the Philippines continues to have low test capacity, averaging about 5,000 tests a day for its population of more than 100 million. – Rappler.com
[ad_2]
