Peter Espeut | Examining published COVID-19 data
The government has – belatedly – admitted that Jamaica is now in the fifth wave of the COVID-19 pandemic. Two days ago (May 18) Minister of Health Christopher Tufton announced that the wave began on April 20 – about a month before – and five days...
The government has – belatedly – admitted that Jamaica is now in the fifth wave of the COVID-19 pandemic. Two days ago (May 18) Minister of Health Christopher Tufton announced that the wave began on April 20 – about a month before – and five days after the ending of the mask mandate and the nightly curfews.
The published data does not support this hypothesis. Like many others, I track the published data daily, and on 20 April 2022, 26 confirmed cases were reported, down from 35 the day before (and 31 the day before that, and 45 the day before that). Going by the published statistics, April 20 cannot be the start of the fifth wave. April 20 was a low point.
On April 27, the number of confirmed cases jumped to 115, and except for three days when testing was relatively low, the number of confirmed positives has remained in the hundreds ever since.
April 27 is twelve days after the ending of the mask mandate and the nightly curfews. In my opinion, the fifth wave began about two weeks after the ending of the mask mandate and the nightly curfews by the Government, just before the Easter holiday weekend, to facilitate the entertainment industry. The boost in the livelihoods which occurred that Easter holiday weekend was a superspreader which increased the morbidity within the Jamaican population. The policy of removing restrictions and living with COVID-19 means that some must get sick – and some die – so that others can make money. The policies you choose will depend on which you love more: people or money (as the song by A.J. ‘Boots’ Brown explores).
As a trained statistician, I have been amused (if not alarmed) at how the data has been reported and used. The COVID-19 tests carried out on any particular day are referred to as “samples”, which is the correct natural science experimental language; and then a positivity rate is published. What does that positivity rate mean?
NUMBERS WILL BE SMALL
When the number of people tested on any given day is small, the number of positives will be small; when the number tested is larger, the number of positives will be larger. With different numbers being tested each day, comparing the number of positives reported each day will not give a true idea of the progress or regress of the pandemic; this is why the positivity rate is used.
Is the positivity rate really a snapshot of the advance or decline in the COVID-19 pandemic in Jamaica on that day? That would only be true if the persons tested on that day (and every day) were a representative sample of the Jamaican population (here the word “sample” is used in the social scientific sense).
How are persons chosen for COVID-19 testing? What sampling methodology has been in use? (Again, I am using social science language).
Someone experiences flu-like symptoms and decides to get tested; if the majority of blood samples taken fall into this category, the positivity rate may be expected to be higher than in the population as a whole. Athletes on the way to an overseas track meet require a positive test to travel; or anyone travelling for any reason must be tested; if the majority of blood samples taken fall into this category, the positivity rate may be expected to be lower than in the population as a whole.
These blood samples taken for testing may not in any sense be considered a representative sample or a scientific sample of the Jamaican population at any time; we (disparagingly) call this an “availability sample”. The positivity rate thus determined refers only to those tested that day, and not to Jamaica as a whole. Since the Government has not invested in scientific sampling, at no time during this pandemic have we known the true picture of COVID-19 morbidity in Jamaica. While setting public health policy, the Government has been flying blind.
But it is the only data we have, and we blithely compare the daily positivity rates as if they were scientific samples, and use it to set public health policy. The Government has not been forthright in explaining to the public the meaning of the statistics they publish.
As of last Tuesday, the total number of reported positive tests since March 2020 was 133,250 out of 1,045,36 tests performed. No way can we conclude from this that only 133,250 persons have been infected with COVID-19 in the last two years. Some of these may be the same people, tested more than once; many of these are tourists. Some who tested negative at one point have subsequently become infected. Many more than two million Jamaicans have never been tested.
Some experts estimate that for every positive result, 20 other persons are infected but not tested. This means that it is likely that a majority of Jamaicans have been infected so far (many, of course, would have been asymptomatic, and would be completely unaware of it). We really can only guess, since no scientific data has been collected.
The Government has been half-hearted with its mantra “balancing lives and livelihoods”. There has never been such a balance; the money wins every time. “It’s the economy, stoopid!”
I predict that in the decades to come, many theses will be written on the Government’s policy choices, by ethicists and students of public administration. Choices to call a general election, to promote parties in Negril, to be soft on enforcing and prosecuting those breaching the regulations, including public servants, and the timing of ending the mask mandate and nightly curfews, will all come in for examination and analysis.
Pandemic public policy decisions – like those governing road traffic behaviour and addressing violent crime – have life-and-death implications for Jamaicans; more than 3,000 of us have died so far from complications from COVID-19. Responsible governments must follow the science to make properly informed decisions.
Peter Espeut is a sociologist and development scientist. Send feedback to email@example.com