Understanding Pandemic Statistics
October 4, 2020 Leave a comment
Real examples help to better understand how to interpret the statistics.
False Positive Paradox: A particular medical test for a disease is 96% accurate. If one has the disease, the test comes back ‘Yes’ 96% of the time, and if one does not have the disease, the test comes back ‘Yes’ 4% of the time.
If 100 of 10000 tested patients have the disease, what is the probability that the person with the diagnosis ‘Yes’ has the disease?
_____________________________________________________
Effective Average Infection Ratio: R is the effective average infection ratio for a disease, also known as the reproduction number. It is the average number of secondary infections caused by one person. (Infections caused by the secondary infections – which would be tertiary infections – are not counted). Consider 50 infected people. Suppose 49 spread the infection to nobody, but one person spreads the infection to 60 people.
What is the R-value?
_____________________________________________________
Leslie Green asks: How would you propose to deal with the COVID-19 pandemic, given information current at the time of writing (11 October 2020).
A. Severe lockdown for 4 weeks
B. Ignore it and carry on as normal
C. Partial lockdown and wait for a vaccine
D. Some other idea
_____________________________________________________
The Pfizer COVID-19 vaccine-candidate interim results from 8 Nov 2020 showed 38,955 participants in a placebo controlled double-blind trial.
94 participants became “evaluable”, which we presume to mean they showed COVID-19 symptoms. The analysis presented was that the vaccine efficacy rate was above 90%. What is the maximum number of (genuinely) vaccinated people who showed COVID-19 symptoms?
Variability Analysis on COVID-19 Interim Trial Data by Leslie Green
_____________________________________________________
Is there any rational justification for being wary of vaccines?
_____________________________________________________
In the UK, December 2020, up to 2 million university students were potentially going home for Xmas during the middle of the COVID-19 pandemic. A new fast COVID-19 test had been developed to spot the causative virus, SARS-CoV-2. These lateral flow tests had the characteristics shown in the image. At the time, around 1% of the population had the virus within the community.
The scientific advice, given on prime-time news channels, was that a pair of negative tests meant it was safe to go home, as a negative test meant a 99.75% chance of not having the virus.
Was this true, and good advice?
_____________________________________________________
