Popular CT Scans Could Account for 5% of All Cancer Cases a Year | Hacker News Reader

Popular CT Scans Could Account for 5% of All Cancer Cases a Year

ucsf.edu

26 pointshsuduebc29 comments

Discussed a few days ago at https://news.ycombinator.com/item?id=43692390

If popular CT scans do account for 5% of all cancer cases a year, don't they do that regardless of the time interval?

Thank you for highlighting the risks of overusing CT scans - especially for children. The comparison to known carcinogens like alcohol and excess weight really puts it in perspective.

I wish we could get such adverse effects stats for all other medical treatments from common medicines, to psych drugs, to other types of diagnostics (e.g. it is known that MRI causes cancer, too), all the way to vaccinations (especially the newer ones), ordered by the sizes of treated populations, descending. Guess which number is larger in the US, all homicides or deaths due to medical errors? The latter is 10x of the former, it's the third most prevalent cause of death, after heart disease and cancer. And these stats could blow the stats out even worse. It is not a coincidence that we at the same time over-medicate and don't live very long.

Well we know for certain that ionizing radiation causes cancer, whereas the link between pharmaceuticals and cancer is often based on unproven or speculative theories.

Well, let's see how "speculative" those theories are, shall we? I'm having a hard time believing that no one gets turbo cancer from e.g. 6.6 _billion_ prescriptions (month-equivalent), and 262 billion doses of prescription medication issued in the US per year. Shit's _way_ out of hand.

I didn't mean pharmaceuticals can't pose risks. Your concern about the scale of prescriptions is valid, especially given that the U.S. prescribes about 65% more medications per capita than the EU, which suggests genuine overprescribing issues. However, jumping straight from high prescription numbers to "turbo cancer" feels premature without concrete evidence therefore it is a speculation. Many of these prescriptions genuinely improve or even save lives, especially for chronic and psychiatric conditions.

On the other hand, history gives us plenty of cautionary examples: leaded gasoline, asbestos insulation, pesticides or tobacco were all once considered safe until they very clearly weren't. So I hope your assumption is wrong.

I feel like you're making the opposite case. Given such large N, it would easier to detect positive correlations to aggressive cancers, no?

To detect you have to actually collect the data in a robust way and then analyze it. Some areas of study seem to be off-limits either for political reasons, or because Big Pharma won't fund them, or both. The other day I told o3 to do deep research on the prevalence of SSRI use among mass shooters. It did compile a report which basically said the data is inconclusive because the current laws prevent conclusive data from being assembled in the first place. I frankly don't see why a mass shooter deserves HIPAA protection. Let's look into it.

Reply on news.ycombinator.com