Until recently, few pharmaceutical companies pre-registered antidepressant trials, which allowed them to selectively publish. This means that antidepressants may be nowhere near as effective as commonly believed, and so doctors are overweighting the benefits compared to the costs of side effects. .
“While only half of these trials had formally significant effectiveness, published reports almost ubiquitously claimed significant results. "Negative" trials were either left unpublished or were distorted to present "positive" results.”
If this is true, then many people who currently take antidepressants may actually be better, not worse, without them, since at least they’d avoid the side effects.
Further, the “chemical imbalance” theory that is propagated through paid advertising has no applicability to a doctor prescribing an antidepressant, since the doctor has no way of measuring and monitoring that imbalance should one exist.
It’s become increasingly clear that genetics play a huge role in which psychiatric drugs are effective and which ones are not. Genetic testing for psych med effectiveness should become standard practice given how in some people, SSRIs (often the first line of defense in psych meds) can cause extended periods of psychosis.
Funny thing about psychosis — when you’re psychotic, your erratic and extreme behavior doesn’t feel abnormal. You don’t feel like anything has changed. This happened to me on Prozac — the most commonly prescribed antidepressant — and it took me 3 years to put my life back together after all the damage I did to my personal relationships during that period.
The anxiety I was dealing with was crippling, but it didn’t dismantle my life the way the side effects of that drug did. The treatment was absolutely worse than the disease. If they had done a genetic test first, they would have seen a red flag and avoided SSRIs all together.
This testing exists today, but it’s only used in patients who show strong resistance to pharmacological intervention. It should be standard.
I feel like we are so close to wirelessly monitoring all brain waves (next 50 years) we will probably be able to, in real time, recognize and diagnose abnormal brain patterns. I think it would be promising if we get this sensor tech up and then perhaps deploy nanotech to interfere with maladaptive brain patterns. The ethics of such a device is considerably sticky.
paraphrased: "wireless non-contact mind reading tech may be cheap and mass producible within the next 50 years."
the only rate limiting step with his method of decoding thoughts is that it takes time under a high energy device to pair associations in the brain.
Think of how much screen time we currently use. If we could pair a wireless 'headset' like headphones while looking at pictures on a screen it would be possible to build a brain bank of children->tenenagers. Currently we just need a lot of processing power and reserved bandwidth on the internet to do some cool stuff.
In my honest opinion as a neuroscientist, we are not that close. We have only an obtuse understanding of how the brain works. Most 'mind reading' demonstrations, though cool, occur within highly constrained circumstances and training and test data, and even then perform just a bit over chance. For context, we barely achieve good accuracy classifying ASD vs typical development, ADHD vs typical development using neuroimaging, much less the much harder problem of classifying and decoding the contents of our thoughts. Slightly more impressive results have been achieved in studies collecting data from actual brain implants found in epilepsy patients, but even then, I do not see this as something that can be a reliable technology in a long long time.
I appreciate someone in the field taking the time to respond. I think we have the ability to scale our knowledge with machine learning and mass monitoring. Again i recognize that 50 years is extremely optimistic. Ethics, money, and politics never agree.
The poster before you is presumably talking about physical limitations of the current measurement methods which leads to very high noise. I doubt machine learning can overcome that.
1. Our understanding of the brain is poor, and I think that will be necessary to have good machine decoding. In principal, a machine learning algorithm does not need to know how the brain works to make an accurate decoding prediction, but I suspect understanding will be needed to help construct and constrain machine learning models learn to deal with the myriad of mental states, intentions, emotions, etc that can occur. While we can train classifiers better than chance to identify whether one or another item is observed, remembered, etc, these training are quite constrained to a type of information or class. I imagine those classifiers are hopeless when you start performing another activity or mental state, and the machine would need to recognize the new state and apply different decoders.
Second, yes there is too much noise in most available methods (EG, fMRI, MEG) and the information is not dense enough (spatial, temporal resolution) and not specific enough to the type of activity (excitation, inhibition, chemical diffusion, etc). Electrodes have done cool things, but I suspect that even millions of wires would not provide enough information and types of information to provide a full fledged brain interface.
"Treatment is worse than the disease". This is also true in cancer treatment, where many drugs have measured effects on reducing tumor size and yet not increasing survival times...yet the side effects make the patients' quality of life significantly worse.
I don't think this is fair for several reasons. First, cancer treatment drugs are tested heavily in-vitro first, where the only significant metric is reduction of tumor mass. Clinical trials for cancer treatments are different from other drugs, as you don't generally do double-blind trials (would you want to get a placebo if you had a terminal cancer? No, probably not).
So many cancer treatments go into clinical trials with only the evidence that it shrinks tumors, and tends to kill tumors faster than it kills the rest of your body. Side effects are expected to be severe since most cancer treatments are quite literally killing any cells that divide... tumors just divide faster and thus get hit harder. The end result is cancer treatment often has very poor quality of life.
Clinical researchers _do_ perform analysis to see if survival increases relative to other treatments. But often you don't know until you've already tried it on a cohort of patients (with their consent, obviously).
There is push-back in the medical field in areas for what you suggest. For example, colonoscopies are becoming widely recognized as causing more problems than they solve. The complications they can induce tend to lead to a poorer patient outcome than missing colon cancer because you didn't screen (e.g. survivability is not affected by finding/treating colon cancer early).
I just don't think it's fair to paint cancer treatments with the same brush strokes, since dealing with terminal diseases is a totally different ball game.
Edit: I should say, there are some parts of cancer treatment which is moving towards the dont-treat opton. For example, there is growing evidence that treating prostate cancer is not necessary, since you'll probably die of something else long before prostate cancer kills you (it is very slow growing on average)
The other problem with cancer is that there are different types of cancer cells. There's the fast dividing and there's also the cancer stem cells which can lay dormant for years without dividing. This is the reason for most cancer returning. There's a promising new tech that hopefully pans out that activates these cells so standard chemo can attack them. Regarding clinical trials, they are very selective. You have to for a very specific criteria and your cancer needs to be measurable by a CT scanner typically to prove efficacy. Add to that the typical cancer patient is either a child or an older person and you're bound to have poorer results.
So... As a cancer patient myself going through chemo, I'll take the chemo over the cancer any day if the week. While some chemo treatments suck, it typically depends on what stage you're at, if you're curable, your comorbidities, and your personal genetics. I can say without any doubt whatsoever I'm alive right now because of chemo. Reducing the tumor allowed me to eat, which has kept me from wasting away into death. Survival times tend to increase with treatment, however quality of life of that survival time is reduced drastically for some people which I believe is what your point is not accurately reflecting, but again it's different for every single patient. I'm young, otherwise healthy, and I'm tolerating the chemo far better than expected with a far better response than average. I'm halfway through average expected survival time and if I live 2 more years I'll be drastically beating the odds. Right now that looks possible for me and that's because of the cancer treatment. There's also been a groundswell of new standard treatments, premeds, and novel treatments as better understanding of the disease in the past 10 years is just starting to bear fruit. If you look at survival rates of some common cancers in the United States many of them are downright not terrible odds of surviving through unless you catch them very late in their development.
only slightly beyond that, 23andme doesn't cover all of the relevant SNPs but it does cover the ones that allow you to infer that you are less likely to respond to p-glycoprotein based drugs like citalopram and paroxetine. You have to export your 23andme raw data and use something like promethease to reprocess the data.
Isn't this mostly from companies like GeneSight? Considering that they relied on 5-HTTLPR and HTR2A, which both had enormous edifices of completely useless (and now debunked) scholarship behind them, I wouldn't bet on those tests' effectiveness. (For reference, GeneSight's test relied on 7 genes. The five not mentioned deal with liver enzymes)
That is a very odd way of looking at the issue. The scientific literature certainly has a bias in favor of publishing positive results rather than failures. However, the scientific literature in no way affects FDA/regulatory approval. Drug/treatment candidates that fail regulatory approval are not on the market.
As far as negative results being "distorted" to present "positive results" is concerned, yes it is quite common to take a patient population that overall fails to show statistically significant benefit and see if there is a subset that does in fact show a statistically significant benefit. One could call that p-hunting, but as long as it is followed up and confirmed, it doesn't seem like a nefarious thing to do.
> the “chemical imbalance” theory that is propagated through paid advertising
I've never seen a pharmaceutical ad that used that term. The only thing I recall from the ads is them distinguishing clinical depression from feeling blue.
ok, I concede that one. They used the term "imbalance of chemicals". To be fair, though, that was a "vintage commercial" and undated. And everything else is not disputed: how to recognize the symptoms of depression. However, I don't fault Big Pharma for not figuring out how to explain the neurochemistry of SSRI's in a 30 second TV ad.
That's quite a statement. I mean, sure, the entirety of why selectively inhibiting serotonin reuptake might affect psychological outcomes may be a bit fuzzy at the end, but there is a pretty solid theory here isn't there? The drugs work by selectively inhibiting serotonin reuptake. This leads to more serotonin signaling. Sure it gets a little fuzzy from there to how this mechanism affects mood disorders with various competing theories, but "unknown" is quite a stretch here.
the mechanism that many drugs use, such as nitrous oxide aren't fully known. the wikipedia page for SSRI doesn't seem to look like marketing. a drug doesn't have to be fully understood in order for it to be effective.
“While only half of these trials had formally significant effectiveness, published reports almost ubiquitously claimed significant results. "Negative" trials were either left unpublished or were distorted to present "positive" results.”
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2412901/
If this is true, then many people who currently take antidepressants may actually be better, not worse, without them, since at least they’d avoid the side effects.
Further, the “chemical imbalance” theory that is propagated through paid advertising has no applicability to a doctor prescribing an antidepressant, since the doctor has no way of measuring and monitoring that imbalance should one exist.