Hacker News new | past | comments | ask | show | jobs | submit login

Are there any documents laying out the specific issues found in the papers? It sounds like there's a lot wrong with them but I'm curious how much worse than your median paper they actually are. There are a lot of politically motivated studies that could be argued to have unsupported assumptions and present data in a misleading way that have not been retracted.

Here's the retraction itself: https://journals.sagepub.com/doi/10.1177/23333928231216699




There is some additional detail in the reporting from Ars Technica[0]:

> The study looked at all emergency department visits, not only visits related to abortion. This could capture medical care beyond abortion-related conditions, because people on Medicaid often lack primary care and resort to going to emergency departments for routine care. When the researchers tried to narrow down the visits to just those related to abortion, they included medical codes that were not related to abortion, such as codes for ectopic pregnancy, and they didn't capture the seriousness of the condition that prompted the visit. Medication abortions can cause bleeding, and women can go to the emergency department if they don't know what amount of bleeding is normal. The study also counted multiple visits from the same individual patient as multiple visits, likely inflating the numbers. Last, the study did not put the data in context of emergency department use by Medicaid beneficiaries in general over the time period.

[0] https://arstechnica.com/science/2024/02/anti-abortion-groups...


It's not only about fake science, don't forget the authors and a reviewer lied about their affiliations and conflicts of interest. When a Journal Editor and a publisher realize that they were deceived by plain lies, I would expect them to react as strongly as they did.


Of course, but that's all pretty clear cut (at least in this example) and well spelled out in the retraction.


There are some specifics about which parts of the paper had problems, but I can’t find the original paper to see how those parts look (failed to find at libgen, doesn’t seem to be open-access from what I can tell)




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: