Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
They failed to find a statistically significant signal.
15 patients exhibited suicidal behavior in the paroxetine group out of 46 patients vs 21/45 in the placebo.
This is modest and not statistically significant. So they made up a bogus category, minor (<5 previous suicide attempts) offenders vs major offenders (>= 5 suicide attempts) and re-ran their analysis. And in this subgroup, they found a statistically significant signal. But in reality, data scientists for BigPharma cut the data a million ways until they found their subgroup that showed efficacy. This is an example of data mining. When your primary endpoint turns up nothing, keep digging until you find something.
That's the big con in this paper.
The other cons are "suicidal behavior" itself, which is subjective. We don't know how generous or if they applied this criteria evenly. You cannot run a double-blind RCT with an active drug vs placebo, this another fiction (I pointed it out at the very beginning of the thread).
Then there was the curious little bit where they randomized more men to the placebo, and more of the placebo were on benzos. Of course, men are more likely to commit suicide, and self treating yourself with high doses of benzos may be classified as suicidal behavior!
Most docs are terrible actually. There are good docs of course, but the average Doc doesn't stay up to date on research, they learned whatever they learned in medical school 20 years ago and now they punch in and punch out and go enjoy their evenings doing something else. That's of course fine.
To be honest, most patients would do a better job treating themselves and just using simple google searching than relying on their doc. There are great docs, of course for which this doesn't hold.
But I only found them affiliated with university research centers and/or on a concierge basis.
Right, I should have done my own cancer surgery. Or just died from a non-Covid viral infection that put me in the hospital for a week. I've been very happy with my doctor. I've been a patient for over 20 years. She knows me, knows what I've had and what my issues are. My ENT is amazing as well.
There are a lot of good doctors in Houston. Bakersfield, not so much. My doctor there went to med school in Monserrat. And did his residency at Kern Medical Center.
Please don't make such sweeping generalizations like "most doctors are terrible, actually". It's simply not true, and an insult to doctors.
Yes that sounds good in theory and may have been practice in your practice but is not true for many Americans who take drugs then take drugs to deal with the side effects of those drugs.
Some of that is patients lying to doctors, and not telling them what medications they are taking. Or, they have the beginnings of dementia and just forget to tell the new doctor what they are taking.
Vioxx is not a case of "oops, we got a drug wrong, let's correct." Certainly, I'm ok with that. Mistakes happen.
Vioxx was a case of Merck pretending to be stupid, and Docs just were really really stupid.
The drug was approved in 1999 but pulled in 2004 - just 5 years, after which it killed millions of people. Not to mention it was in a class of drugs called NSAID, so basically drugs that are supposed to make you more comfortable (reduce pain in arthritis sufferers), not kill you lol.
Diagnosis in the mental health field is part art and part science. Then one ends up having to experiment with different drugs to see what works. The process is not as scientific as many of us would like it to be.
Yep. We just don't know enough about the brain (and nervous system in general) yet. My prediction is that disorders like what we now call "depression" will probably be found someday to be part of a variety of ailments caused by different factors, benefitting from different treatments - that being depressed is symptom of some illnesses, not an illness in and of itself, if that makes sense. (Compare to what physicians used to call "dropsy" - nowadays our goal would be to find out what is causing the edema, but even if the doctors of the time HAD been able to differentiate between heart failure, kidney problems, liver problems, etc., they couldn't have done much to support or repair those organs, because they didn't have effective medications or surgical interventions yet.) Our understanding of how a combination of electrical impulses and neurotransmitter variations translates to our experience of emotions and moods and thoughts is so poor, we're currently left attempting to do palliative care for most mental illnesses, as opposed to correcting root causes.
Almost any treatment has side effects - the key is finding the treatment that offers the greatest alleviation of suffering while having the fewest drawbacks. And since we don't have cures for most mental illnesses, it's doubly important to find something the person can live with for the long run. In most cases that requires some experimentation to find what is most effective for the individual.
The bottom line is that doctors are doing their best with the currently available tools and information, barring the occasional quack, and in the face of obstacles such as insurance considerations.
They failed to find a statistically significant signal.
15 patients exhibited suicidal behavior in the paroxetine group out of 46 patients vs 21/45 in the placebo.
This is modest and not statistically significant. So they made up a bogus category, minor (<5 previous suicide attempts) offenders vs major offenders (>= 5 suicide attempts) and re-ran their analysis. And in this subgroup, they found a statistically significant signal. But in reality, data scientists for BigPharma cut the data a million ways until they found their subgroup that showed efficacy. This is an example of data mining. When your primary endpoint turns up nothing, keep digging until you find something.
That's the big con in this paper.
The other cons are "suicidal behavior" itself, which is subjective. We don't know how generous or if they applied this criteria evenly. You cannot run a double-blind RCT with an active drug vs placebo, this another fiction (I pointed it out at the very beginning of the thread).
Then there was the curious little bit where they randomized more men to the placebo, and more of the placebo were on benzos. Of course, men are more likely to commit suicide, and self treating yourself with high doses of benzos may be classified as suicidal behavior!
Please register to post and access all features of our very popular forum. It is free and quick. Over $68,000 in prizes has already been given out to active posters on our forum. Additional giveaways are planned.
Detailed information about all U.S. cities, counties, and zip codes on our site: City-data.com.
Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
