3 ms·
A read through of the methodology makes it sounds like they took solid precautions to prevent biases that could taint the results- As an MD, I'm impressed by th
by drcode 7y ago
A read through of the methodology makes it sounds like they took solid precautions to prevent biases that could taint the results- As an MD, I'm impressed by the design of this study, and the promising results.
This may be what more studies in the future will look like: Bundle multiple interventions together to help create statistically stronger results, at the cost of less clarity around each individual treatment.
EDIT: Someone else on HN noticed there possibly was no placebo for the control group- This worries me. Also, there's no indication in the methodology that the "field team" was blinded as to treatment groups :(
- vanderZwan 7y ago> This may be what more studies in the future will look like: Bundle multiple interventions together to help create statistically stronger results, at the cost of less clarity around each individual treatment I'm probably stating the obvious, but it may also be that multiple interventions together have a larger net effect than the interventions separately would suggest. I recently read an article by a physician arguing for many of his patients, treating ADHD with a combination of a stimulant and a non-stimulant resulted in bigger benefits with less side-effects, due to being able to have a lower dosage for each individual medication[0]. [0] https://www.additudemag.com/adhd-medication-combination-therapy-treatment/ https://www.additudemag.com/adhd-medication-combination-ther...
- drcode 7y agoI agree that synergy is an important element, but to me a study like this could make sense even without synergy: To me, it seems better to have 4x the statistical data on 4 drugs at once than just running the individual studies, given how long a study like this takes.
- Madmallard 7y agoIt bothers me that doctors are still out of the loop on chronic risky side effects of medication that are now becoming more prominent. Statins are quite bad for you and increase risk of death from all causes.
- drcode 7y agoI agree with you mostly, but if we can establish with enough certainty that the benefits of something like this are far greater than the costs, it seems unwise to let "perfect" treatment be the enemy of "good" treatment.
- Madmallard 7y agoI don't know if our research is good enough at all to adequately make the risk versus benefit distinction in risk factor but not imminent danger related medicinal decision making. I don't think the integrity of the system and the rate of technological advancement are allowing it to happen. Companies conduct dozens of trials with new drugs and toss out ones that don't surpass placebo and publish the one study that by chance did. I'm sure the thing that happened with Vioxx is happening more covertly on a much larger scale now. Most smart people are just avoiding doctors unless something obviously bad pops up. It's kind of tragic.
- drcode 7y agoYeah, unfortunately those are all good points...
- Madmallard 7y agohttps://youtu.be/qP07oyFTRXc?t=345 https://youtu.be/qP07oyFTRXc?t=345 Here's an excerpt from a video about double speak and it brings up statins. It actually infuriates me that doctors try to bully people into medications that don't actually provide them a net benefit because the doctors themselves have been duped.