In science & esp medicine, risk is a tricky thing to contend with - besides of course being a mathematical one - and involves many considerations in order to have a more fulsome understanding.
In this post @LizMarnik and I attempt to explain these concepts. 1/9
In simple terms, risk is the probability of an event occurring. In statistical terms, the risk of an event occurring is simply defined as its probability. 2/9
Take a fictitious disease - AhCrap. Risk of dying due to this is 2/million & using GoodStuff tx can reduce risk to 1/million.
While relative risk reduction is 50%, only one less person out of 1M w AhCrap are saved by GoodStuff, making absolute risk reduction fairly small. 3/9
Because the risk of dying decreased from two in a million to one in a million, the relative risk decreased by 50%.
However, the absolute risk only changed from 2/1,000,000 to 1/1,000 000, that is, a difference of 0.00005%. 4/9
A few other things to consider, such as:
what the likelihood of the risk occurring is and what the risk-benefit analysis including downstream impacts are.
The examples in this slide demonstrate these points to help put overall risk into context. 5/9
Changing gears, let’s talk about cumulative risk, which is the combined risk from multiple exposures or behaviors over time.
This requires several pieces of information, such as time period, how many exposures, risk assoc w each & how other behaviors impact the likelihood. 6/9
Take this example: risk of TerribleOutcome after the first time one has BadInfection is 2 in 100 or 2%. Risk after each subsequent infection is 1 in 100 or 1%.
Cumulative risk in this example of person having TerribleOutcome over lifetime, i.e. a 15 year period is 3.95% 7/9
So as we can see in this situation, while the individual risk of a subsequent Badinfection may infact be lower than the first infection, the cumulative risk of developing BadOutcome is higher with each additional infection. 8/9
Understanding these can help you make informed decisions.
Putting into context how likely risk is, what impact of the risk is, what overall risk-benefit analysis is & downstream impacts are all important considerations when assessing & evaluating risks + decisions.
9/9
• • •
Missing some Tweet in this thread? You can try to
force a refresh
A week ago, a new paper was published evaluating the safety of Pfizer, Moderna & AstraZeneca vaccines. Unfortunately, the media’s response with sensational headlines has been both irresponsible & misleading.
Here is a deep dive I did with @LizMarnik & @niniandthebrain
Samples were collected via a prospective observational study that enrolled individuals with prior SARS-CoV-2 infection with and without long COVID symptoms 3/7
Two new Canadian studies demonstrating the efficacy of masks.
These studies show that mandatory masking, esp during periods of high respiratory viruses in the community, can result in reductions in absenteeism, prevent infections and deaths & reduce economic burden. 🧵 1/9
First study by @NishaOttawa , @DrKevinSchwartz , @KevinAnBrown et al.
This study examined whether school-based mandatory masking policies were associated with reduced educational disruptions due to COVID-19 and other respiratory viruses 2/9
A total of 77 640 students from 205 schools were included:
166 were elementary schools, of which 60 (36.1%) had mandatory masking policies
39 were secondary schools, of which 12 (30.8%) had mandatory masking policies 3/9
A thing that invariably shocks a lot of people is when I say I don’t trust most science-related books or documentaries.
So, I want to explain why, and situate it within one of my all-time fav things: evidence. 🧵 /1
What constitutes good evidence?
First, all data is not created equally.
When it comes to publications, most people, including academics, tend to read the abstract and perhaps skim quickly through the data. /2
But that doesn’t tell you how robust or valid the study and results are, ie, whether the design, conduction and methodology are able to successfully prevent systematic errors or bias. /3
New study comparing myocarditis due to Covid-19 vaccines vs infections.
There was a markedly higher risk of heart failure and death for myocarditis associated with Covid-19 infections compared to Covid-19 vaccination. 1/4
Subanalysis of age 12-39 & no comorbidities showed Covid-19 infection related myocarditis was assoc with 5.8x higher risk of heart failure or death vs vaccine related myocarditis.
Confirms Covid-19 related myocarditis far less severe with vaccinations than with infections. 2/4
Beware those who compare the numbers of the events - we do not have the denominator for any of the groups, so this study cannot be used to discern incidence or rate. Those with no scientific background will use this data nefariously. 3/4
Part 1 of the series on Covid-19 vaccinations and pregnancy, fertility and breastfeeding.
This one focuses specifically on the updated data we have in pregnancy - from the impact of infections to safety and benefits of the vaccine. 🧵1/11
Risk of severe outcomes due to Covid-19 infection in pregnancy continue to be high for the pregnant person.
Studies show these risks are highest for those unvaccinated. 2/11
There are also considerable risks to the fetus/baby due to Covid-19 infections in pregnancy.
Studies show that risk of stillbirth and neonatal mortality highest in those unvaccinated. 3/11