Problem is, by its nature this research is pretty exploratory. Most studies enroll people who are engaged and interested, and thus the population is very selected
-
Show this thread
-
For example, one of the best sources of information is the COVID Symptom Study, which uses an app to log details. By their very nature, apps exclude some people from using them https://covid.joinzoe.com/about
2 replies 0 retweets 13 likesShow this thread -
Some of my research shows this for chronic disease - lots of people stop using apps even if they are beneficial, and this pattern is non-random https://www.jmir.org/2020/9/e20283/
1 reply 1 retweet 19 likesShow this thread -
What we'd need to really understand the prevalence of
#LongCovid in populations is to enroll a large group of people - including asymptomatic cases and people who don't have COVID - and follow them over time to see what happens3 replies 4 retweets 28 likesShow this thread -
We'd also need to try and get data from pretty much everyone at every time point, because it's likely that those who drop out are different to those who don't in ways that may be all but impossible to measure
1 reply 0 retweets 17 likesShow this thread -
Unfortunately, that's really hard to do, so we are left to interpret the observational data that gives us, at best, a biased view of the situation
1 reply 0 retweets 13 likesShow this thread -
Whenever you see a proportion being proposed (i.e. 1 in 20 people have Long COVID) it's important to understand the denominator Is it everyone? Only those in one study? Who did that study enroll?
1 reply 0 retweets 19 likesShow this thread -
Important note - it's possible I've missed a large prospective study that does answer this question, I've tried to look but please do let me know if one has been published
2 replies 0 retweets 13 likesShow this thread -
Important note #2 - this bias could easily bias the results IN BOTH DIRECTIONS, meaning that there could be less or far more
#LongCovid than reported. There are reasonable arguments both ways!4 replies 1 retweet 30 likesShow this thread -
Replying to @GidMK
Don’t know if you saw this one: https://www.medrxiv.org/content/10.1101/2020.10.07.20208702v1.full.pdf … But when people ask me for a stat on the frequency in the general population this has been the reference I’ve deferred to. It’s been really hard to track down data for certain subgroups
1 reply 0 retweets 1 like
That's a useful paper, thanks! Sadly it's a pretty tiny COVID-19 sample, so I think limited usefulness, but definitely on the right track
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.