Dear Journos,
I hope I am not teaching grandma to suck eggs, but for anyone trying to form a view point on kids and covid may I recommend the following. 🧵
I hope I am not teaching grandma to suck eggs, but for anyone trying to form a view point on kids and covid may I recommend the following. 🧵
To assess the strength of evidence:
👶🏼 Look for asymptomatic testing in kids. This is important as children usually have no symptoms at the time of infection.
👶🏼 Check to see what other mitigation might be affecting results, like economic shut downs, and in school mitigation.
👶🏼 Look for asymptomatic testing in kids. This is important as children usually have no symptoms at the time of infection.
👶🏼 Check to see what other mitigation might be affecting results, like economic shut downs, and in school mitigation.
👶🏼 check to see if the situation the kids were in, in the study, are the same as the situation you are about to face as a society. Look for mask use, ventilation and filtration use, and think... is this what we are doing? Can I expect the same results.
👶🏼 Beware of studies that take the results of lots of other studies and squeeze them together. Usually none of the parameters between different sites are the same, and also some of the measurements (rapid tests vs lab PCR for example).
👶🏼 also check to ensure there is not difficulty accessing testing for kids and people in general. If the queues to test are really long and difficult, expect a slight (or big) underestimate.
👶🏼Check demographics, age etc. does the population in the study apply to the class in question ( although it may or may not affect the underlying mechanics of transmission, it is better to use a “well matched” population).
And also “How to read a paper” by Trish Greenhalgh are excellent short reads and how I learned to read a paper. 
👶🏼 lastly, please look at the study outcomes measured. There are deaths, education, mental health, family hardship and chronic disease outcomes to investigate. Amongst others I have probably missed out.
Have a list. Is the paper answering your questions?
Have a list. Is the paper answering your questions?
I hope that this thread will make it easier to deal with experts, because just as it is wise to read the abstract last when reading a paper, so you avoid bias, it is wise to be thoroughly prepped before an expert interview, so you know when bias creeps in.
👶🏼Addendum, on the topic of studies that group other studies together, or “reviews”, always make sure you check what *wasn’t* included, and why.
If you were to run a search on this topic, what search terms would you chose? Did important stuff get left out?
If you were to run a search on this topic, what search terms would you chose? Did important stuff get left out?
• • •
Missing some Tweet in this thread? You can try to
force a refresh
