There is no such thing as luck; only adequate and inadequate preparation to cope with the statistical universe
– Robert Anson Heinlein
In case we have not noticed, it is good to remind ourselves that we are living in a statistical universe. Number crunching is the most favourite pastime of the politicians, advertisers and spin-doctors (experts) in the name of sharing information.
The number of people tested positive and number of deaths by Covid-19 are being checked and absorbed by almost every citizen of the country these days. As usual, politicians in power would try their best to paint only the full-half of the glass while those in the opposition paint the empty-half. If one doesn’t have the ability to analyse the numbers thrown at him then one is at the mercy of his political bias in deciding what to believe.
Numbers and people’s beliefs
Citizens of the world are being bombarded by statistics every day and we make our decisions depending on them. Often, these statistics are just numbers attached to claims and a majority of the public, with the level of numeracy they have achieved, tend to believe what the numbers show. It doesn’t normally occur to them to find out how those numbers got there and why a particular person or an organisation is trying to share those numbers with them. Once during an election campaign a politician said, “Today we have over 500,000 people unemployed in this country and if I am elected I will bring the unemployment down to 3 percent”. This statement has no meaning to a person who does not understand the numbers, percentages and how unemployment figures are calculated.
But 500,000 feels like a big number and three sounds like a very small number compared to that and it may have been good enough for some people to vote for him, especially if they were looking for employment during that time. It turned out that the 500,000 out of the working-age population of the country was holding the unemployment at three percent already even when that statement was made and the candidate didn’t have to do anything to bring it to that level. He just had to make sure that he didn’t do anything to increase it. As we know, he would worry about it only if he had any intention of keeping his promises after getting elected anyway.
There are several vaccines claiming all kinds of percentages of effectiveness in fighting the virus. When the new vaccines came out, their makers and the Governments of respective countries touted their efficacy with some impressive numbers — 95% for Pfizer, 94% for Moderna, 67% for Johnsons and Johnsons and 80% for AstraZenaca. Normal procedure would have been to test these vaccines individually and one against the other in a sequence of clinical trials conducted under the supervision of the regulatory authorities such as Food and Drug Administration (FDA) of the US and Medicines and Health Products Regulatory Agency (MHRA) in the UK. But due to urgent requirements the manufacturers were asked to conduct their own testing and report the results. Though the studies were acceptable by the authorities each manufacturer used their own sampling and methodologies and calculations. Therefore, comparing those efficacy rates would be meaningless.
Vaccinations and its efficiency
A vaccine with 95% efficacy rate does not mean that one’s chance of getting the virus is only 5% after getting vaccinated. Your chance of getting Covid is not 100% to start with anyway. Vaccine efficacy is a ‘relative risk reduction’. It’s a ratio comparing the risk of infection in people who got vaccinated versus people who did not (the control group). One could also calculate what is called the absolute risk reduction. That is simply the difference in risk for someone in the treatment group versus someone in the control group. For example, if there are 100 people who don’t get a vaccine, and 10 of them caught the disease then, the baseline risk of getting it is 10%. Also, if 100 other people got the vaccine, and only one of those got sick. There the risk is actually 1%. The absolute risk reduction (ARR) is then just 9% (10% minus 1%). But the relative risk reduction (RRR) is 90%, that is the percentage representation of that reduction of 9% divided by the baseline risk of 10%.
One of the main reasons why absolute risk reduction is not shown is because of the ‘wow; effect of the numbers. It is more attractive to say 90% effective than saying that the absolute risk reduction is 9%. Further studies have shown that even with trials on tens of thousands of people, the absolute risk reductions in Covid-19 is just around 1.2% for Moderna and 0.84% for Pfizer.
Who is fighting Corona better
Since people have heard that countries such as New Zealand, Korea and Taiwan have managed the pandemic much more successfully than others, they often are used as the bench-marks for success against Covid while India and Brazil are mentioned for comparisons at the other end. Recently there was a news item with such a comparison about Sri Lanka’s Covid numbers which said, “Sri Lanka’s fatality rate as a percentage of the total number of Covid-19 patients remained as low as 0.61 whereas it was as high as 1.78 percent in the United States, 1.09 percent in India and 0.99 percent in New Zealand”, supporting the claim that Sri Lanka is handling the pandemic better than New Zealand. Though the percentages are correct, drawing the conclusion that Sri Lanka is handling the pandemic better than New Zealand from them would be a grave mistake. The total population of New Zealand is almost five million and total infected number is 2650 and the number of deaths is 26. If Sri Lanka had controlled the pandemic as effectively as New Zealand then for our population of 21 million we should not have had more than 12,000 total infections and the death toll should have been less than 110.
But the total infections in Sri Lanka, (by Tuesday, May 11, 2021), is 129,000 (more than 10 x 12000) and the total number of deaths is 827 (about 7.5 x 110). At the moment the numbers are increasing in Sri Lanka while they are decreasing in New Zealand. If one wants to compare the performance of Sri Lanka to that of New Zealand, one should look at all such numbers, their trends and future predictions, all together before one jump to any conclusion.
This article also throws a bunch of numbers at you and the readers can verify them using the websites of WHO (World Health Organization) and AMA (American Medical Association) or any other site they trust. It was Ernest Rutherford, a physicist originally from New Zealand said: “If your experiment needs statistics, you ought to have done a better experiment.”
Here’s to statistics, the cause of, and solution to, most of the problems in the world.
The writer has served in higher education sector as an academic over twenty years in the USA and fourteen years in Sri Lanka and he can be contacted at [email protected]