Tylenol fallout

[Unknownparadox]

Oh, look -- another one who did not get the pre-conceived answer he decided was the right one.

We know that tactic.

I know the right answer. Autism is great, being born in the wrong body is great. Don't change a thing.

[Encia22]

Enough!!

 

Do not reply to this post.

[Ignorant]

No causal relationship has been shown between acetaminophen use during pregnancy and a higher risk of disorders like autism.

No one has gone into the science of it all yet. I'll try to explain why the recent meta-analysis that Trump's statements are based on is misleading, and why, as of right now, we have no reason to believe acetaminophen use during pregnancy causes autism, and in fact we have every reason to believe it doesn't.

First, we need to understand correlation vs causation, confounding, and sibling controls. This may be asking a lot, but since everyone, including people who are clearly not scientifically trained, wants to have an opinion about this, surely they'd also be willing to spend time understanding these basic concepts. Regardless, if you're ever talking to a pregnant woman about this, please just tell them to defer to their doctor. It's an absolute certainty that you do not know better than their doctor.

Correlation vs causation & confounding
People who buy ice cream have a higher chance of drowning in the next 24 hours than people who don't buy ice cream. This is a true statement, yet it's not a causal relationship. In summer, more people buy ice cream. They also go swimming more often, and drown more often. The relationship between ice cream and drowning is a correlation. The real cause is hot weather. Hot weather is a confounding variable; a hidden variable that we missed and caused both more ice cream buying and more drowning.

The first claim, "People who buy ice cream have a higher chance of drowning in the next 24 hours than people who don't buy ice cream", is therefore misleading. It's not directly saying there's a causal relationship, but it's intentionally phrased in such a way that the reader will think it's a causal relationship. The confounding variable is not mentioned.

Sibling controls
In scientific studies, to "control for X" is to change our study so that it removes X as a possible confounding variable. So if we wanted to control for hot weather in our ice cream study, we would make sure that people are getting ice cream randomly assigned throughout the year, and then observing if they drown. If we still notice that people drown more often when they've had ice cream, we can more confidently say ice cream causes drowning.

When it comes to studies about the effects things have on children, it's often unethical to do this. For example, ethically we can't randomly assign acetaminophen to mothers (and force them to take it) during pregnancy. But there is a great way to control for many possible confounding variables in studies like these: Sibling controls. How does that work? If, for example, increased risk for disorders was found in children exposed to acetaminophen and no increased risk was found in their siblings, that means acetaminophen is very likely to be the cause. But if increased risk was found both in children exposed to acetaminophen and in their siblings who weren't exposed, acetaminophen was not the cause; It's a correlation. The real cause, then, is a third hidden variable.

In the meta-analysis study I linked above, they point out:
 

Numerous well-designed studies have indicated that pregnant mothers exposed to acetaminophen have children diagnosed with neurodevelopmental disorders (NDDs), including autism spectrum disorder (ASD) and attention-deficit/hyperactivity disorder (ADHD), at higher rates than children of pregnant mothers who were not exposed to acetaminophen.

Remember correlation vs causation? They make no statement here about acetaminophen actually causing these things. They just note the relationship. As I mentioned before, this is misleading, because the reader will assume that it's a causal relationship when it isn't. They've done exactly what I did when I stated "People who buy ice cream have a higher chance of drowning in the next 24 hours than people who don't buy ice cream".

So how did they get to this conclusion? They've taken a number of studies about the effects of acetaminophen use during pregnancy, and noted that many of them found a relationship between acetaminophen use and neurodevelopmental disorders in children. Okay. That is true. But is it a causal relationship? This is unclear here, and we need to explore other studies to find out.

It turns out we're in luck. There's a huge study about this (2.5m participants) which used sibling controls to control for confounding variables. Their results should tell us whether or not the relationship that other studies found is causal. In the key points section, they are immediately very clear:
 

Question  Does acetaminophen use during pregnancy increase children’s risk of neurodevelopmental disorders?

Findings  In this population-based study, models without sibling controls identified marginally increased risks of autism and attention-deficit/hyperactivity disorder (ADHD) associated with acetaminophen use during pregnancy. However, analyses of matched full sibling pairs found no evidence of increased risk of autism (hazard ratio, 0.98), ADHD (hazard ratio, 0.98), or intellectual disability (hazard ratio, 1.01) associated with acetaminophen use.

Meaning  Acetaminophen use during pregnancy was not associated with children’s risk of autism, ADHD, or intellectual disability in sibling control analyses. This suggests that associations observed in other models may have been attributable to confounding.

Sibling controls ruled out confounding variables, and they found that the relationship was no longer there. That suggests confounding and it invalidates the results of the non-sibling-controlled part of the study. It also suggests that the increased risk found by other studies without sibling controls, which includes by extension the meta-analysis from earlier, is likely due to a third, confounding variable. [1]

Therefore, we have no reason to believe the relationship between acetaminophen use during pregnancy and neurodevelopmental disorders in children is causal. We have every reason to believe it's just a correlation, and the real underlying cause is something else.

What irks me about the whole thing is that the people who did the meta-analysis as well as Trump's advisors should know all this. It's pretty basic stuff. If I credited Trump with any intelligence I would even expect him to know it. In any case, the scientists do, and his advisors should. If the authors of the study in particular were serious scientists, they would not have been a part of it. They were probably paid a lot of money for this, because it's really quite embarrassing for them. I hope they never hear the end of it.

I saw some people argue "why not avoid acetaminophen?". Well, of course you should never take medicine for no reason, but sometimes you actually need it. High fevers in the mother during pregnancy have been shown to damage the baby, and the best way to lower a fever during a pregnancy is acetaminophen. It's the best way precisely because acetaminophen has so few side effects compared to other fever reducers, and is safe to take as long as you stick to the recommended dose. It's easy to imagine Trump supporter mothers with fevers are now less likely to take the medicine they need, and could actually end up damaging their baby. I hope they listen to their doctor over their cult leader.

Follow your doctor's advice.

[1] One example of this could be that pregnant women who are more likely to use acetaminophen are also more likely to use other, harmful medicines or even things like alcohol. I actually wouldn't be surprised if higher use of OTC medicines in general correlates with substance abuse or other behaviors that are harmful to the fetus. Anyway, we can speculate, but we don't know what this hidden variable is. We just know it exists.

[MichSwampbuck]

@Ignorant

An excellent and detailed explanation that was easy (for me at least) to comprehend. You have a nice approach, like a kind teacher who really wants their students to understand.

There are some that use that information like a club to beat people with in what appears to be a show of superior strength and authority. That attitude only makes me resist their message no matter how factual it may be.

Your presentation has got me thinking that the connection that is implied may really be shinola. Also, if a fever is that dangerous for a pregnant woman's unborn, even if there is really is some risk of autism, being safest for use it is better than severe damage or death when left untreated.

Alternately, I must consider the that Tylenol can't be the only safe treatment for fever or to bring down a temperature during pregnancy. Plus, I won't forget that big pharma wants an uninterrupted revenue stream. I also wonder how dangerous a fever is for the unborn and how often this is a problem during pregnancy statistically. Another good question is, "How often is it prescribed during pregnancy?" Is it given out like candy to pregnant women?

[quintessentone]

No causal relationship has been shown between acetaminophen use during pregnancy and a higher risk of disorders like autism.

No one has gone into the science of it all yet. I'll try to explain why the recent meta-analysis that Trump's statements are based on is misleading, and why, as of right now, we have no reason to believe acetaminophen use during pregnancy causes autism, and in fact we have every reason to believe it doesn't.

First, we need to understand correlation vs causation, confounding, and sibling controls. This may be asking a lot, but since everyone, including people who are clearly not scientifically trained, wants to have an opinion about this, surely they'd also be willing to spend time understanding these basic concepts. Regardless, if you're ever talking to a pregnant woman about this, please just tell them to defer to their doctor. It's an absolute certainty that you do not know better than their doctor.

Correlation vs causation & confounding
People who buy ice cream have a higher chance of drowning in the next 24 hours than people who don't buy ice cream. This is a true statement, yet it's not a causal relationship. In summer, more people buy ice cream. They also go swimming more often, and drown more often. The relationship between ice cream and drowning is a correlation. The real cause is hot weather. Hot weather is a confounding variable; a hidden variable that we missed and caused both more ice cream buying and more drowning.

The first claim, "People who buy ice cream have a higher chance of drowning in the next 24 hours than people who don't buy ice cream", is therefore misleading. It's not directly saying there's a causal relationship, but it's intentionally phrased in such a way that the reader will think it's a causal relationship. The confounding variable is not mentioned.

Sibling controls
In scientific studies, to "control for X" is to change our study so that it removes X as a possible confounding variable. So if we wanted to control for hot weather in our ice cream study, we would make sure that people are getting ice cream randomly assigned throughout the year, and then observing if they drown. If we still notice that people drown more often when they've had ice cream, we can more confidently say ice cream causes drowning.

When it comes to studies about the effects things have on children, it's often unethical to do this. For example, ethically we can't randomly assign acetaminophen to mothers (and force them to take it) during pregnancy. But there is a great way to control for many possible confounding variables in studies like these: Sibling controls. How does that work? If, for example, increased risk for disorders was found in children exposed to acetaminophen and no increased risk was found in their siblings, that means acetaminophen is very likely to be the cause. But if increased risk was found both in children exposed to acetaminophen and in their siblings who weren't exposed, acetaminophen was not the cause; It's a correlation. The real cause, then, is a third hidden variable.

In the meta-analysis study I linked above, they point out:
 

Remember correlation vs causation? They make no statement here about acetaminophen actually causing these things. They just note the relationship. As I mentioned before, this is misleading, because the reader will assume that it's a causal relationship when it isn't. They've done exactly what I did when I stated "People who buy ice cream have a higher chance of drowning in the next 24 hours than people who don't buy ice cream".

So how did they get to this conclusion? They've taken a number of studies about the effects of acetaminophen use during pregnancy, and noted that many of them found a relationship between acetaminophen use and neurodevelopmental disorders in children. Okay. That is true. But is it a causal relationship? This is unclear here, and we need to explore other studies to find out.

It turns out we're in luck. There's a huge study about this (2.5m participants) which used sibling controls to control for confounding variables. Their results should tell us whether or not the relationship that other studies found is causal. In the key points section, they are immediately very clear:
 

Sibling controls ruled out confounding variables, and they found that the relationship was no longer there. That suggests confounding and it invalidates the results of the non-sibling-controlled part of the study. It also suggests that the increased risk found by other studies without sibling controls, which includes by extension the meta-analysis from earlier, is likely due to a third, confounding variable. [1]

Therefore, we have no reason to believe the relationship between acetaminophen use during pregnancy and neurodevelopmental disorders in children is causal. We have every reason to believe it's just a correlation, and the real underlying cause is something else.

What irks me about the whole thing is that the people who did the meta-analysis as well as Trump's advisors should know all this. It's pretty basic stuff. If I credited Trump with any intelligence I would even expect him to know it. In any case, the scientists do, and his advisors should. If the authors of the study in particular were serious scientists, they would not have been a part of it. They were probably paid a lot of money for this, because it's really quite embarrassing for them. I hope they never hear the end of it.

I saw some people argue "why not avoid acetaminophen?". Well, of course you should never take medicine for no reason, but sometimes you actually need it. High fevers in the mother during pregnancy have been shown to damage the baby, and the best way to lower a fever during a pregnancy is acetaminophen. It's the best way precisely because acetaminophen has so few side effects compared to other fever reducers, and is safe to take as long as you stick to the recommended dose. It's easy to imagine Trump supporter mothers with fevers are now less likely to take the medicine they need, and could actually end up damaging their baby. I hope they listen to their doctor over their cult leader.

Follow your doctor's advice.

[1] One example of this could be that pregnant women who are more likely to use acetaminophen are also more likely to use other, harmful medicines or even things like alcohol. I actually wouldn't be surprised if higher use of OTC medicines in general correlates with substance abuse or other behaviors that are harmful to the fetus. Anyway, we can speculate, but we don't know what this hidden variable is. We just know it exists.

That's the messages I've been trying to get across all along on this thread, but there are those here that only see the right/left and have Trump Devotion Syndrome, so it's impossible for the message to be received.

[MichSwampbuck]

That's the messages I've been trying to get across all along on this thread, but there are those here that only see the right/left and have Trump Devotion Syndrome, so it's impossible for the message to be received.

Ignorant sets a great example.  Good job and my thanks.

I'd like to see more people here respond like that instead of tossing in a jab or two while making their remarks sound condescending.

[BeyondKnowledge]

removed

And yet you are the first to bring up politics again. 

I do not agree completely with the thinking that Tylenol has not been drought into question during pregnancy. I do believe the recommendation should have been consult your doctor before taking it while pregnant. You are failing to consider that most people think it is safe simply because it is an over the counter drug and has no specific warnings about pregnancy on the package. 

The medical investigationn should be if it is safe for use while pregnant before recommending without consulting a doctor. It is better to be safe in this casre. At least people would be more aware there are questions about its use under certain circumstances.

You are still ignoring the manufacturers recommendation that everyone needs to consult there doctor before use when pregnant.

[quintessentone]

And yet you are the first to bring up politics again. 

I do not agree completely with the thinking that Tylenol has not been drought into question during pregnancy. I do believe the recommendation should have been consult your doctor before taking it while pregnant. You are failing to consider that most people think it is safe simply because it is an over the counter drug and has no specific warnings about pregnancy on the package. 

The medical investigationn should be if it is safe for use while pregnant before recommending without consulting a doctor. It is better to be safe in this casre. At least people would be more aware there are questions about its use under certain circumstances.

You are still ignoring the manufacturers recommendation that everyone needs to consult there doctor before use when pregnant.

Politics and the left/right divide here needs to be brought up and discussed if at all possible. I will always call out right/left wing BS when I see it, so hopefully we can all recognize it and begin civil and fair dialogues.

I am the first one to object to Trump's 'do not take it' and post here that everyone should be getting their doctor's advice. Pregnant women with a high fever should be taking it or something else to lower their fevers because it may harm their unborn child. Trump IMO is irresponsible with his messaging now as he was with the Covid pandemic.

Good, so you also agree with me that Trump's message should have been as Vance stated, 'Consult your doctor'.

[BeyondKnowledge]

Politics and the left/right divide here needs to be brought up and discussed if at all possible. I will always call out right/left wing BS when I see it, so hopefully we can all recognize it and begin civil and fair dialogues.

I am the first one to object to Trump's 'do not take it' and post here that everyone should be getting their doctor's advice. Pregnant women with a high fever should be taking it or something else to lower their fevers because it may harm their unborn child. Trump IMO is irresponsible with his messaging now as he was with the Covid pandemic.

Good, so you also agree with me that Trump's message should have been as Vance stated, 'Consult your doctor'.

I don't understand why labels like right, left, Democrat, republican, liberal, conservative, have so much control and apparent meaning for some. The extreamists are all that really matter. All sides are nuts in the extreams.

Have you considered that Trump said that to make a bigger impact. To get more people to notice. Follow the manufacturers recommendation is hardly news worthy.

[quintessentone]

I don't understand why labels like right, left, Democrat, republican, liberal, conservative, have so much control and apparent meaning for some. The extreamists are all that really matter. All sides are nuts in the extreams.

Have you considered that Trump said that to make a bigger impact. To get more people to notice. Follow the manufacturers recommendation is hardly news worthy.

It's not just me stepping forward denouncing Trump's messaging, as I posted it was also JD Vance, GOP Leader Thune, where they didn't explain it away, they said the opposite of what he said. There are far more dangerous over the counter drugs out there that pregnant women should be wary of, but he did not mention any of them...make it make sense.

It almost appears that the political division spans across each of all the political spectrums because I rarely see civil and open discussion from the Americans here, except from a few seemingly Independent or Centrists, or people which change their party depending upon which issue they find more of a priority. Or from the members who just don't care or who have given up.

I believe in free speech, women's rights, freedom of expression, peaceful protest, I believe capitalism only serves the rich, and I am anti-war. Doesn't that make me straddling both sides of the fence?