AI Could Help Patients With Chronic Pain Avoid Opioids

Cognitive behavioral therapy is an effective alternative to opioid painkillers for managing chronic pain. But getting patients to complete those programs is challenging, especially because psychotherapy often requires multiple sessions and mental health specialists are scarce.

At three months, patients’ pain intensity and pain interference were just as good with the AI-supported program, and at six months, substantially more patients in the AI-supported group had clinically important improvements in their outcomes, Piette said.

Eighty-two percent of patients in the AI-CBT group completed all 10 weeks of treatment, compared to 57% of patients who were offered 10 weeks of telephone counseling by a therapist.

“Despite receiving more weeks of treatment, the AI-supported program used less than half the therapist time, meaning that we could double the number of patients who can be treated with the same number of clinicians,” Piette said.

“This finding could have a dramatic impact on how we think about delivering psychotherapies for people with pain.”

https://neurosciencenews.com/pain-ai-21240/

 

Oh, fabulous. "Here's your app, now sod off."

Clearly they've never actually experienced chronic pain, or realised that many people with it are already very skilled at deploying management strategies.

 

I also know of one case where they turned to alternatives, including alcohol and street drugs. If someone in chronic pain needs strong pain relieving medication its not helpful to dismiss that, including in the guise of a CBT app.

 

Perhaps AI could be an improvement over the average CBT technician or practitioner.

But more germaine is the development of new drugs for chronic pain. Ones that are at least 60% effective and don't cause constipation or upset the addiction community.

 

If this quackery is so effective, why is it so 'challenging' to get patients to complete the programs?

I imagine that chronic pain patients would be incredibly enthusiastic about persisting with a treatment that works.

 

I forget any specifics, but I am sure in a comparison between mental health professionals and a software programme responding to key words and using a number of stock phrases, people found the software more understanding and helpful than real people on the other end of a keyboard.

 

When you make up false excuses for why your thing is awful and doesn't work, then pretend those excuses aren't just cheap excuses to avoid facing reality. How is it they put it with us? Fear avoidance? Yeah, it's all projection with these people. They are avoiding reality because they don't like it. Irony overload.

To be clear: the AI claims in this are a scam. At best this is a decision tree, it's hard enough to train neural networks using reliable data, it's nonsense to pretend this is using AI. I've seen electric toothbrushes claiming to use AI, the claims are just as valid here. It's not possible to train a neural network without accurate feedback. This is also the case with humans, it's not possible to learn from experience unless experience provides reliable and precise feedback. Did it hit the target? How close? It has to be precise. This here is all wishy-washy stuff, it has none of that.

What it truly says is that you can replace a CBT therapist with a programmed decision tree and it makes no difference. Because a useless thing is just as useless if it's automated in a scripted program. You can even slap an "AI-powered" label if you want to, it makes no difference at all.

 

Possibly people find an app which can't help, but provides platitudes, more 'friendly' than a person, who could help, but chooses not to, possibly in a very unfriendly manner e.g. blames the patient for still being in pain despite being told not to be when they've deliberately, and maliciously, obstructed any chance of useful treatment because of their moralistic (or something else unuseful) dogma?

 

This is the paper: Patient-Centered Pain Care Using Artificial Intelligence and Mobile Health Tools, A Randomized Comparative Effectiveness Trial.

Not sure what they added the redundant words between randomized and trial.

 

Or it could be an expert system. The old expert systems worked using pattern matching, so if there is a comment made that it has a rule for it replies with that rule's response. So its like a decision tree with loops, and where it can jump to other branches if the conversation changes.

Expert systems rely on the experts having some genuine expertise. This is an area where experience does not equal expertise.

In learning AI programming the first major assignment in my degree was to build an Eliza-like program that tries to have an "intelligent" conversation. Its just rule matching. The better the rules, the better the response, but it does not learn once deployed and can be repetitive.

 

CBT is pretty simplistic. Either you are rationalizing, generalizing, catastrophizing, doing black or white thinking, etc. etc.

Aren't there about 10-12 principles of CBT?

If AI can parse your responses (which I am pretty sure it can), then spitting out the correct CBT principle to 'correct' one's thinking is easy.

I am not adverse to seeing AI in action; I think CBT is just a very limited approach to complex problems.

 

Just sitting here realizing the silliness of noninferiority trials when comparing two useless things, both of which will give you the same score since they're equally useless. Of course given the poverty of assessment, those scores will not actually be zero, because it's not a calibrated measure relating to anything real. But they sure will be equal, even though they're useless, somehow rating non-zero.

Every time it will find that the "novel" treatment being compared is noninferior, it literally can't be otherwise. Like comparing one person's wealth, owning nothing, with another person who owns nothing. The other person is not poorer than the first one, somehow in that logic that should make them not poor, if the first person's poverty is somehow denied.

I think that the other CBT apps also used noninferiority. They also had the same results: equally useless, aka noninferior. You can't do less than nothing, so you always "win".

 

So AI is better than CBT alone. Not that much of a win!

“We’re very excited about the results of this study, because we were able to demonstrate that we can achieve pain outcomes that are at least as good as standard cognitive behavioral therapy programs, and maybe even better. And we did that with less than half the therapist time as guideline-recommended approaches.”