Dev.SN ♥ developers
Sections
Dev.SN
from the will-somebody-think-of-the-children? dept.
GungnirSniper writes:
"The Journal of the American Medical Association, Pediatrics, has found that use of acetaminophen (paracetamol) during pregnancy is associated with a higher risk for HKDs [hyperkinetic disorders] and ADHD-like behaviors in children. More than half of all mothers in the study reported acetaminophen use while pregnant.
The LA Times has a longer and lighter story about the study which reminds us 'that unchecked fevers have been associated with a number of poor health outcomes in babies, including lowered IQs.'
Led by neuropsychologist Miriam Cooper of the University of Cardiff in Wales, the group wrote that without more details on how acetaminophen might lay the foundations for later ADHD, and when and in whom it is most likely to boost risk, the current findings should be interpreted cautiously and should not change practice.
For pregnant women, the study underscores that, even when a medication is billed as safe, the safest route is to take it as rarely as possible and at the lowest effective dose, said UCLA obstetrician Dr. Daniel Kahn, a maternal-fetal health specialist who was not involved in the study.
Damned if you do, damned if you don't?"
(Score: 5, Insightful) by Open4D on Wednesday February 26 2014, @10:50AM
What do you mean by too broadly defined?
They claim to have established an association between "maternal acetaminophen use during pregnancy" and "HKDs and ADHD-like behaviors in children" - with a stated level of statistical confidence. Are you disputing that? If yes, please elaborate. If not, please read on ...
.
We all know that correlation != causation. But if it's not the case that acetaminophen causes ADHD then it would have to be either:
.
An example of #1 might be that certain problems during pregnancy cause mothers to be more likely to take acetaminophen and cause the child to be more likely to have ADHD. This possibility could perhaps be investigated by seeing whether any acetaminophen alternatives were also correlated with ADHD. I don't know whether they did anything like that. But the abstract does say, "results did not appear to be confounded by maternal inflammation, infection during pregnancy, the mother's mental health problems, or other potential confounders we evaluated."
I would guess that the possibility of a non-causal systematic link between acetaminophen and ADHD is the reason the researchers said that without more details "the current findings should be interpreted cautiously and should not change practice."
.
One "insightful" reply to your post talks about milk, make-up, and TV. I hope you don't agree with that poster's implied argument. Unless you geniunely think they may be part of the mechanism of a systematic link between acetaminophen and ADHD that the experts hadn't considered.
.
Disclaimer: I am just a programmer. The above just derives from what I learnt during secondary education (15 years ago). Corrections are welcomed.
(Score: 1) by useless on Wednesday February 26 2014, @11:52AM
No, I don't think milk causes ADHD, and I'm pretty sure the poster of that comment doesn't either. ;)
My main problem isn't in the study so much (the DNBC* was a great program), but the uneducated/sensationalized reaction to it.
This finding was from pouring through massive amounts of generalized data, finding a *potential* correlation, and saying "Hey, this might mean something, we should study this". This is a preliminary finding, and in the conclusion they say as much. But that hasn't stopped people from running with the headline as fact.
In programming terms, this is just the idea that leads to the business plan. Then comes the spec, prototyping, programming/debugging, then release. People are going from the idea to release, without all the lengthy work in between.
Thanks for the reply, and hope this helps you get where I'm coming from.
* Overview on the DNBC here if you're interested. It's lead to some great research:
http://sjp.sagepub.com/content/29/4/300.abstract [sagepub.com]
(Score: 2) by Open4D on Wednesday February 26 2014, @12:15PM
Agreed, but the poster seems to be implying that there is some kind of flaw in the study, so that any old random factor such as milk or make-up could be relevant.
Whereas I think the results are far more definitive than that. Either acetaminophen causes ADHD, or there is some other systematic link between the two that the researchers haven't been able to account for.
.
The poster picked up [dev.soylentnews.org] on this sentence from the abstract, "More than half of all mothers in the study reported acetaminophen use while pregnant". Possibly he or she got confused and assumed that this fact itself was the evidence that the researchers were presenting.
(Score: 2, Insightful) by useless on Wednesday February 26 2014, @12:47PM
Ah, ok, I think I see where you may be getting my point confused. Correct me if I'm wrong:
This study was not an actual clinical trial to test the effects of acetaminophen during pregnancy, or a trial to find the potential causes of ADHD. It was more of a research project on general pregnancy data, and they found an interesting, potential link. Check out the overview of the DNBC to see where their research data came from, and the wide array of follow-up studies it has lead to.
So there is not really a "flaw" in the report per se, it's that this report did not come from a focused study. Meaning there are no definitive results to base an factual conclusion on.
Cheers
(Score: 2, Interesting) by sbgen on Wednesday February 26 2014, @02:08PM
You are right this was not a clinical study, it was epidemiological inquiry. However, the results are definitive and the factual conclusion is that there is enough data in here to formulate a hypothesis to start detailed basic science research. Not everything in medical science need to be clinical study to be of valid concern to public health. I agree that sensationalizing to the general public is not good but then again I read only the actual paper, not the general press.
Warning: Not a computer expert, but got to use it. Yes, my kind does exist.
(Score: 1) by useless on Wednesday February 26 2014, @02:32PM
Exactly my point. It's a start, not the end.
(Score: 2) by Open4D on Wednesday February 26 2014, @02:19PM
The abstract gives numerical results in the form of "hazard ratios" (presumably this [wikipedia.org]) and CIs (which I assume stands for Confidence Intervals). Aren't those definitive enough? If an actual clinical trial to test the effects of acetaminophen during pregnancy had been done instead, and had come up with the same results & confidence intervals, should that have been considered more definitive?
Notwithstanding the criticisms of epidemiology [oxfordjournals.org], shouldn't all the relevant issues have been factored in to the confidence intervals? So the mere fact that this is yet another result from the Danish National Birth Cohort shouldn't be seen as lessening the validity of the result - should it?
(Score: 2, Informative) by sbgen on Wednesday February 26 2014, @12:32PM
You are right about the correlation/causation and I admire you for providing the two (properly numbered!) alternative possibilities. The study is based on the telephone interview of mothers about acetaminophen usage during pregnancy and the data available on their children (at 7 years age) about ADHD etc. The data they have used therefore can only provide correlation. While the statistical treatment of the analysis seems to be rigorous there is not functional analysis to definitively say acetaminophen usage resulted in the later life outcome for the children. The most important aspect is that now there is a definitive study correlating one to the other and the matter is really of public importance. I am sure actual functional analysis to show the biology (or to disprove causality) will follow (provided some one funds that research). The article also quotes previous studies showing that acetaminophen can cross placenta barrier, can be an endocrine disrupt or etc. ADHD/Autism/... seem to have some causes in very early part of life, perhaps prenatally. There is stuff to be worried about the effect of acetaminophen but solid data is necessary to proceed further.
Warning: Not a computer expert, but got to use it. Yes, my kind does exist.