3 ms·

> You're wrong that long covid evidence is anecdotal. Fine, convince me. Provide a citation from a study where: 1) the results are NOT gathered by survey from

by timr 5y ago

> You're wrong that long covid evidence is anecdotal. Fine, convince me. Provide a citation from a study where: 1) the results are NOT gathered by survey from a self-selected population 2) the study is pre-registered, with endpoints determined in advance 3) the study is prospective, longitudinal and randomized 4) there is a control group 5) the participants are verified to actually have Covid 6) a significant effect is shown on the pre-registered endpoints and I'll take your claims seriously. The only current study that I'm currently aware of that remotely comes close to this standard of evidence (but doesn't meet all of them) is this one: a controlled, retrospective study of 250,000 kids, which found that kids without covid had roughly equivalent long-term symptoms to kids who did have covid. https://www.medrxiv.org/content/10.1101/2021.05.05.21256649v1 https://www.medrxiv.org/content/10.1101/2021.05.05.21256649v... I bring up this study only to underscore the point: essentially all current "long covid" data is self reported and uncontrolled, and when you start to be rigorous, the claims of severity rapidly diminish. Here's a relevant editorial in Nature: "Studies of long COVID thus far—nearly all from high-income countries—have been difficult to interpret or compare, not only because of the lack of a consistent case definition but also because of systematic selection bias of cases, often without appropriate controls....Meanwhile, community-based studies have raised questions about the representativeness of cases compared with the general population, especially those using online recruitment, with many studies relying on self-reported symptoms and without laboratory-confirmed diagnosis of SARS-CoV-2 infection. Selection bias becomes even more pronounced when people are recruited through online support groups, where very high rates of often nonspecific long-term symptoms have been reported, even in children." https://www.nature.com/articles/s41591-021-01402-w https://www.nature.com/articles/s41591-021-01402-w > Patients often show reduced lung capacity of 60% Citation required. See above for standards of evidence. > The brain fog is also often much more severe than just not being able concentrate a bit, people can't do simple calculations, can't follow conversations, can't read more than a page in a book, immediately forget things they just talked about. Citation required. See above for standards of evidence. > Many people become completely unable to perform their usual tasks. Citation required. See above for standards of evidence.

dragontamer 5y ago
> Fine, convince me. Why do you need a control group? We know that typical people don't just randomly get coughs on a daily basis. Anyway: https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2776560 https://jamanetwork.com/journals/jamanetworkopen/fullarticle... A pretty small study (250-ish people) but plenty of evidence for long-haul COVID19, since ~20+% of them got long-haul COVID19. ----------- Random, normally functioning adults don't have breathing issues under normal circumstances. There's just no need for a control group in that kind of study. The only question is how long do symptoms last for COVID19 patients. There's a bit of self-selection since these individuals decided to get a COVID19 test to begin with. But otherwise, the % of long-haul COVID19 is high enough that we know its a real thing. For you to be suggesting that long-haul COVID19 is entirely made up is just... ignoring the evidence we have available. You're making perfect the enemy of good: studies don't need to be "perfect" in every way to provide evidence. ----------- Besides, you're playing "judo" with the argument. If you want to convince ME that long-haul COVID19 is fake news, YOU are the one who has to pull up the study indicating evidence in your favor. I'm perfectly fine leaving this argument where it is and walking away. I don't _need_ to convince you of anything to be satisfied.
- timr 5y ago
  > Why do you need a control group? We know that typical people don't just randomly get coughs on a daily basis. Because this is science, not a spread in the New York Times. And if you read the link to the paper I cited, you'd see that "typical people" do randomly get these "symptoms" on a regular basis (I'm coughing right now!) And in fact, once you control for this, you find out that some of what you thought was "long covid" was just...what happens in a normal population of people. As for your paper, it fails the criteria established: "COVID-19 symptom data were obtained at the time of acute illness or retrospectively recounted at a 30-day enrollment visit. A total of 234 participants with COVID-19 were contacted between August and November 2020 to complete a single follow-up questionnaire between 3 and 9 months after illness onset." In other words, this is self-reported data. The control group isn't actually a control group, because the people in each group knew if they had Covid, and are self-reporting their own symptoms...in a survey about Covid. It's a recipe for confirmation bias. A good study of this sort would follow N people over time, randomly selected and matched on various demographic criteria (age, weight, gender, etc.), split into known sero-positives and known sero-negatives. You'd then survey all of those people over time about a few selected health issues, without telling them you're doing a study about Covid. A gold-standard study would find a way of objectively confirming the illness, and wouldn't rely on self-reported data (e.g. medical exam with lung volume test). But even if you can't do that, you can still do a lot better than this. This is just...meaningless. It's like taking a group of people who broke their legs in the last decade, and asking them if their leg hurts today. Sure, maybe some people have "long broken leg", but your experiment is a bad way to find out. > Besides, you're playing "judo" with the argument. If you want to convince ME that long-haul COVID19 is fake news, YOU are the one who has to pull up the study indicating evidence in your favor. Nope. Again, this is science. The burden is on the person making the claim to prove the claim. If you believe that Cooties are a thing, or that head shape predicts personality, or that fusion can occur at room temperature in a can of Mr. Pibb, I am not obliged to believe you simply because you want very badly for it to be true.

by timr 5y ago

dragontamer 5y ago

> Fine, convince me. Why do you need a control group? We know that typical people don't just randomly get coughs on a daily basis. Anyway: https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2776560 https://jamanetwork.com/journals/jamanetworkopen/fullarticle... A pretty small study (250-ish people) but plenty of evidence for long-haul COVID19, since ~20+% of them got long-haul COVID19. ----------- Random, normally functioning adults don't have breathing issues under normal circumstances. There's just no need for a control group in that kind of study. The only question is how long do symptoms last for COVID19 patients. There's a bit of self-selection since these individuals decided to get a COVID19 test to begin with. But otherwise, the % of long-haul COVID19 is high enough that we know its a real thing. For you to be suggesting that long-haul COVID19 is entirely made up is just... ignoring the evidence we have available. You're making perfect the enemy of good: studies don't need to be "perfect" in every way to provide evidence. ----------- Besides, you're playing "judo" with the argument. If you want to convince ME that long-haul COVID19 is fake news, YOU are the one who has to pull up the study indicating evidence in your favor. I'm perfectly fine leaving this argument where it is and walking away. I don't _need_ to convince you of anything to be satisfied.

timr 5y ago
> Why do you need a control group? We know that typical people don't just randomly get coughs on a daily basis. Because this is science, not a spread in the New York Times. And if you read the link to the paper I cited, you'd see that "typical people" do randomly get these "symptoms" on a regular basis (I'm coughing right now!) And in fact, once you control for this, you find out that some of what you thought was "long covid" was just...what happens in a normal population of people. As for your paper, it fails the criteria established: "COVID-19 symptom data were obtained at the time of acute illness or retrospectively recounted at a 30-day enrollment visit. A total of 234 participants with COVID-19 were contacted between August and November 2020 to complete a single follow-up questionnaire between 3 and 9 months after illness onset." In other words, this is self-reported data. The control group isn't actually a control group, because the people in each group knew if they had Covid, and are self-reporting their own symptoms...in a survey about Covid. It's a recipe for confirmation bias. A good study of this sort would follow N people over time, randomly selected and matched on various demographic criteria (age, weight, gender, etc.), split into known sero-positives and known sero-negatives. You'd then survey all of those people over time about a few selected health issues, without telling them you're doing a study about Covid. A gold-standard study would find a way of objectively confirming the illness, and wouldn't rely on self-reported data (e.g. medical exam with lung volume test). But even if you can't do that, you can still do a lot better than this. This is just...meaningless. It's like taking a group of people who broke their legs in the last decade, and asking them if their leg hurts today. Sure, maybe some people have "long broken leg", but your experiment is a bad way to find out. > Besides, you're playing "judo" with the argument. If you want to convince ME that long-haul COVID19 is fake news, YOU are the one who has to pull up the study indicating evidence in your favor. Nope. Again, this is science. The burden is on the person making the claim to prove the claim. If you believe that Cooties are a thing, or that head shape predicts personality, or that fusion can occur at room temperature in a can of Mr. Pibb, I am not obliged to believe you simply because you want very badly for it to be true.