4 ms·
If that's the case, I wish the information would include percentage of cases. 90% side effect of severe depression and 2% side effect of dry mouth is a very dif
by cabaalis 8y ago
If that's the case, I wish the information would include percentage of cases. 90% side effect of severe depression and 2% side effect of dry mouth is a very different medication than 90% dry mouth and 2% severe depression.
- jessriedel 8y agoUnfortunately, the typical patient doesn't interpret probabilities very well. Still, lists of side effects often use language like "very rarely".
- nonbel 8y agoFirst, frequency != probability. What negative outcome is expected by sharing the actual values though?
- gowld 8y agoTo the extent that probability means anything in the real world, it means frequency. The domain of the frequency is the rub.
- jessriedel 8y agoYes, and correlation doesn't imply causation except when it does. Say, a patient doesn't take an important drug because the 0.001% chance of death seems too large, or alternatively they are falsely comforted by a seemingly-small 1% chance of death.
- nonbel 8y agoI don't get it, "seems to0 large" and "seemingly-small" according to who? It seems like you are against the patient doing their own cost-benefit analysis and instead someone else needs to be doing it for them. Who and on what basis?
- jessriedel 8y agoSo, we can first infer that the patient isn't correctly taking the numbers under advisement because they use them inconsistently (e.g., they worry more about four things that are 1 in a million rather than one thing that's 1 in 10 thousand). The question of how to align the patients understanding and actions with what they would do if they had perfect knowledge and intelligence is of course fraught, but one way is to compare medical risks to other risks they understand more intuitively. It's a big topic.
- stordoff 8y agoIs that not the case in the US? Every prescription drug I take in the UK has them broken down by "Very common" (>1/10), "Common" (1/10 to 1/100), "Uncommon" (1/100 to 1/1000), "Rare" (1/1000 to 1/10000), "Very rare" (1/10000 to 1/100000), and "Frequency not known" (frequency cannot be estimated from the available data).
- dvlsg 8y agoI don't think I've ever seen that in the US. Would be useful information to know, though, for sure.
- pkaye 8y agoI see that on my prescription brochures in the US.
- joveian 8y agoIf you get a tiny print information sheet it has that information somewhere. More conveniently, drugs.com has that information in the professional section of the side effects page. The inconvenient aspect is that the number of listed side effects seem to inevitably be proportional to how long the drug has been around and how much it has been used and researched, so you also need to have a general idea of all of that if you are trying to figure out which of two drugs is likely to be safer. Also, if two (or more) drugs are commonly used together, the side effects of the combination can be listed on both drugs, even if one of the drugs is most likely primarily responsible.
- cryoshon 8y agothey break it down into "common" and then various others. the trick is that most of the ones on the tier immediately after common are actually in the common tier, but they may not have the severity necessary for a patient to report it. example: chantix(https://en.wikipedia.org/wiki/Varenicline https://en.wikipedia.org/wiki/Varenicline) patient reports ubiquitously report nightmares almost without exception, yet nightmares are listed as a rare side effect because the nightmares might not necessarily be severe enough to mention to a doctor or meet the clinical criteria for a side effect.