Is Psychology a real science?

[hadespussercats]

The distinction between psychiatrists and clinical psychologists is arguably a bit more controversial, but the easiest way to explain it is that they approach similar problems from different perspectives. The problem with an issue like mental health is that the brain is an organ of the body, so mental disorders can either be a product of physical malfunctions, or they can be purely mental (or, to confuse matters, a bit of both). This means that we need professionals with degrees in medicine and a specialisation in psychiatry in order to study the brain and how drugs can help certain problems. Clinical psychologists, on the other hand, largely deal with the 'mental' causes of disorders. Generally, clinical psychologists have more tools to deal with mental disorders (e.g. cognitive behavioral therapy, systematic desensitisation, etc).

To summarise:

Psychologists - scientists who study what makes organisms tick
Clinical psychologists - apply the science of psychologists to understanding mental disorders
Psychiatrists - medical doctors who treat mental disorders as if they are physical diseases.

From a patient's point of view, in the UK at least, the distinction is certainly blurred. I've been involved with 3 separate family members who have needed (different) psychiatric/psychological treatment, and their specialists in all 3 cases used a combination of clinical psychology and pharmalogical treatment methods...which makes sense, but I'm not sure which, if either, category they fell into.

Mixing talk therapy and pharmacological therapy seems to give better results (no, I don't have links-- just reading, talking with doctors, and experience.)

I liken it to being on pain killers and going to physical therapy.

Generally, psychiatrists aren't there to talk about how you're coping with life-- they want to know how you feel physically, so they can monitor how medications are working. clinical psychologists, on the other hand, want to know how you're dealing with stress, work, family, etc.-- and, depending on the therapist, might offer up coping strategies.

Sometime, psychiatrists and psychologists collaborate on a patient's care. I think this team approach is far more effective than the dim sum a little of this, a little of that that most patients get.

[Ronja]

Can you clear up what exactly is the difference between psychologists and clinical psychologists? You said that there is a huge difference, but didn't say what it was. ( and the wiki article makes no distinction at all ! ).
And why do we need psychiatrists, psychologists and clinical psychologists?
Surely there should be a mental health qualification that covers all of it?

Why, when the brain is such a complicated organ, should there be one discipline that covers all aspects of its study?

And let's not forget neurologists and neurosurgeons - already a very small tumor in the brain can change personality and/or behavior, even radically.

[mistermack]

Why, when the brain is such a complicated organ, should there be one discipline that covers all aspects of its study?

If mental problems were very clearly defined, so that you could safely say, "this person need psychiatric help" or "psychological help" then keeping the two separate would be good. But it doesn't seem to be the case.

Psychologists in the United States have campaigned for legislation changes to enable specially trained psychologists to prescribe psychiatric medicine. New legislation in Louisiana and New Mexico has granted those who take an additional masters program in psychopharmacology permission to prescribe medications for mental and emotional disorders in co-ordination with the patient's physician. Louisiana was the second state to provide such legislation.[8]

Also, it seems that a combination of types of therapies is often the most effective.

[Mr.Samsa]

So if you test a drug to a placebo, then test CBT to the active drug (that outperforms the placebo), and find that CBT works better than the active drug, you wouldn't accept the conclusion that CBT is better than the placebo? Wouldn't the very nature of logic necessitate this conclusion? (Not that all CBT trials are done this way, but given what you're suggesting I'm not sure how you can't accept it as a valid conclusion).

I wouldn't count a placebo pill as a suitable placebo for CBT.
It doesn't mirror the mental stimulation of face to face counselling.

But the point is that CBT outperforms antidepressants, which in turn outperforms a placebo. Are you suggesting that antidepressants are better than placebo treatments, but a different kind of "placebo" (CBT) can outperform antidepressants?

Essentially, we know that CBT cannot be a placebo for two reasons: 1) the effect size is far too large, and 2) it lasts for too long (e.g. permanently, in a large number of cases).

Your next answer covered what I meant, that to test CBT, you would have to compare it with a very similar procedure that had no active ingredient.

I did start out by asking if psychology was any better than common-sense counselling, so I would personally compare it to that in a trial, rather than CBT with the salient bits removed.
Somehow, I can't see fake CBT succeeding, the sincerity and enthusiasm just wouldn't be there.

Usually they compare it to some other kind of "well-being" counselling which has been demonstrated to have no clinical effects for a certain condition. This type of counselling is, in effect, CBT without the active component. It avoids the risk you're talking about.

But it seems pretty obvious that regular one-to-one sessions could help in many ways. An incentive to keep taking the tablets could be just one. So of course you would expect CBT to out-perform just a placebo pill.

But this wouldn't explain why, for instance, CBT outperforms cognitive therapy (which is the same as CBT but without the active component).

I would expect CBT combined with antidepressants to outperform just antidepressants as well, for the same reasons.

Well of course two evidence-based treatments at once will outperform a single evidence-based treatment.

Do you understand though, that it is not CBT that supplements medication, but rather the other way around? That is, CBT is the empirically stronger treatment, and far surpasses the results of antidepressants (in both effect size, and in continuing effects).

Out of interest, what did you think of the metaanalysis I linked you to?

I'm not sure what to make of it. What does the "meta" mean in that context?

A metaanalysis looks at most rigorously conducted studies (usually large scale RCT designs with appropriate controls etc), and combines their results to see what the overall result is for the available data. This is done to avoid the problems associated with single studies, where a result might be a product of chance or some other confounding factor. Importantly, only the most methodologically sound studies are taken, with the weaken studies removed.

Only the conclusions are available free, and the sentence at the end "While limitations of the meta-analytic approach need to be considered in interpreting the results of this review, our findings are consistent with other review methodologies that also provide support for the efficacy CBT." didn't fill me with confidence.

Why not? It's standard science-talk for "Our conclusions are not absolute and subject to change given contradicting evidence". If you're concerned about the validity of metaanalyses, then you can just look at all the individual (statistically significant) studies by themselves. (You'd also need to reject a large number of medical treatments which are prescribed solely because of the results found in metaanalyses). In other words, the common misconception in the public is that RCTs are the strongest form of evidence we have in medical science - this is untrue, metaanalyses are.

But a look at hard data and methodology is the only way that you could draw your own conclusions.
Otherwise, you're limited to agreeing because they said so, and trusting their conclusions blind.

I think it's fair to question the methodology of testing.

You certainly are free to check their methodology, if you do a google scholar search for CBT then you'll find millions of articles with the conclusion that it's a valid treatment, so I'm sure a couple of them will be freely accessible.

With a drug, you have a fixed formula, and it has to jump extensive and well set-out hurdles before it can be prescribed, and it's quality doesn't depend on the person giving it to the patient.
Even then some pretty useless drugs sometimes get through.

Indeed, psychological treatments are the same.

Compare that to a therapy, and almost anything can be changed at any time. It's bound to be less defined.

I'm not sure why you think this is the case? Do you think psychological treatments are just made on the fly, and that therapists can just change any aspect they want?

As far as I know, the psychologist doesn't impact the effect of CBT at all. This is because the psychologist is irrelevant to the treatment, all they do is teach the person the rules of how to control their own thought processes. Essentially all they do is read instructions from a list. It's because of this that CBT can be, and has been, automated to the point that people can sign up to a website and receive CBT sessions from a computer program - with similar effects to those found by those receiving CBT from a psychologist in person.

OK, I should have read down to this before I wrote the above, but I'll leave it there anyway.
If that can be demonstrated to work, you have a good proof.
But I do note that most use a combination of drug and CBT, which clouds the water again.
I read the arguments given for this, but was neither convinced nor sceptical.

They only use a combination in applied settings, not during the research. The evidence still highlights the fact that CBT is more effective than antidepressants on every measure.

One would have thought that if CBT out-performs drugs in trials, it wouldn't be necessary to combine them.

Hadespussercats' analogy is a good one: antidepressants and CBT is similar to pain killers and physical therapy. The principle behind them is the same, and interestingly the arguments you have against CBT apply to physical therapy too (even more so, since it's often impossible to blind their studies).

Also, I should note, it's not necessary to combine them at all. You get better results, sometimes, if you do though. With cancer, doctors also tend to combine treatments: they excise the tumour and then they provide chemotherapy. Does the combination of the two mean that one doesn't work by itself? Or is it simply a case of medical professionals combining two empirically valid treatments to increase the chances of success in their patient?

But maybe CBT can't make the initial step as well as drugs can.
It certainly weakens the argument though.

How so?

Mr. Samsa wrote: Psychiatrists - medical doctors who treat mental disorders as if they are physical diseases.

There's no as if about it-- mental disorders are physical diseases.

I'm not sure if we're misunderstanding each other, but I can't see how your statement is accurate. By "physical diseases", I'm referring to the idea that mental disorders are a product of a brain disease, malfunction, abnormality, or some other general kind of biological misfire. This idea is obviously incorrect, which is why it was rejected from psychology and psychiatry in the 1950s.

Why, when the brain is such a complicated organ, should there be one discipline that covers all aspects of its study?

If mental problems were very clearly defined, so that you could safely say, "this person need psychiatric help" or "psychological help" then keeping the two separate would be good. But it doesn't seem to be the case.

But it is the case. Do you think mental disorders are subjectively defined or something?

Psychologists in the United States have campaigned for legislation changes to enable specially trained psychologists to prescribe psychiatric medicine. New legislation in Louisiana and New Mexico has granted those who take an additional masters program in psychopharmacology permission to prescribe medications for mental and emotional disorders in co-ordination with the patient's physician. Louisiana was the second state to provide such legislation.[8]

Also, it seems that a combination of types of therapies is often the most effective.

Not usually, it depends on the symptoms of the disorder. If, like depression, one of the symptoms is the aversion to going outside and talking to people, then it becomes difficult for CBT to work because the patient refuses to go to see the psychologist to learn CBT. Once the patient has seen the therapist for a couple of sessions and learn the basics of CBT, they are usually taken off anti-depressants because they've done their job (i.e. get them out of the house).

[hadespussercats]

Can you clear up what exactly is the difference between psychologists and clinical psychologists? You said that there is a huge difference, but didn't say what it was. ( and the wiki article makes no distinction at all ! ).
And why do we need psychiatrists, psychologists and clinical psychologists?
Surely there should be a mental health qualification that covers all of it?

Why, when the brain is such a complicated organ, should there be one discipline that covers all aspects of its study?

And let's not forget neurologists and neurosurgeons - already a very small tumor in the brain can change personality and/or behavior, even radically.

Yup. Actually, I hate the term "mental illness"-- I think "neurological illness" would be less trivializing, more indicative of actual physical ailment.

[hadespussercats]

hadespussercats wrote:

Mr. Samsa wrote: Psychiatrists - medical doctors who treat mental disorders as if they are physical diseases.

There's no as if about it-- mental disorders are physical diseases.

I'm not sure if we're misunderstanding each other, but I can't see how your statement is accurate. By "physical diseases", I'm referring to the idea that mental disorders are a product of a brain disease, malfunction, abnormality, or some other general kind of biological misfire. This idea is obviously incorrect, which is why it was rejected from psychology and psychiatry in the 1950s.

I don't see how this idea is "obviously incorrect"-- and saying that psychologists and psychiatrists rejected the idea over sixty years ago, given the extensive advances in our understanding of the brain in the time since, is hardly compelling.

[Mr.Samsa]

I don't see how this idea is "obviously incorrect"-- and saying that psychologists and psychiatrists rejected the idea over sixty years ago, given the extensive advances in our understanding of the brain in the time since, is hardly compelling.

The point is that the evidence was overwhelming back then, and no evidence has come up to change our opinion on this. It's 'obviously true' because people can have entirely healthy brains but have a debilitating mental disorder. Many problems are caused by thought patterns which are perfectly normal in a given situation, but become maladaptive when generalised to other situations. This maladaptiveness doesn't come from a brain issue, it is simply mistake in thinking. For example, suppose a kid is raised in an abusive household and learns to be anxious of people, this is a healthy behavior and not a mental disorder. But when the kid grows up, and cannot make friends or form intimate relationships because of this learnt behavior, then it becomes a mental disorder. There is nothing wrong with their brain though and it is essentially indistinguishable from someone who learnt to be gregarious, or outgoing.

[Geoff]

hadespussercats wrote:

Mr. Samsa wrote: Psychiatrists - medical doctors who treat mental disorders as if they are physical diseases.

There's no as if about it-- mental disorders are physical diseases.

I'm not sure if we're misunderstanding each other, but I can't see how your statement is accurate. By "physical diseases", I'm referring to the idea that mental disorders are a product of a brain disease, malfunction, abnormality, or some other general kind of biological misfire. This idea is obviously incorrect, which is why it was rejected from psychology and psychiatry in the 1950s.

I don't see how this idea is "obviously incorrect"-- and saying that psychologists and psychiatrists rejected the idea over sixty years ago, given the extensive advances in our understanding of the brain in the time since, is hardly compelling.

Some certainly are purely physical, but by no means all. Another, perhaps helpful, analogy would be between hardware and software problems in a computer.

[Mr.Samsa]

Some certainly are purely physical, but by no means all. Another, perhaps helpful, analogy would be between hardware and software problems in a computer.

And to further the difficulty in understanding this issue by extending your analogy, just because something is a hardware problem doesn't mean it can't be fixed by a software change. For example, (as I think I've mentioned in this thread?) autism is almost certainly a "hardware" problem, yet it's only treatment/cure is behavioral therapy (a "software" solution).

[hadespussercats]

hadespussercats wrote:

Mr. Samsa wrote: Psychiatrists - medical doctors who treat mental disorders as if they are physical diseases.

There's no as if about it-- mental disorders are physical diseases.

I'm not sure if we're misunderstanding each other, but I can't see how your statement is accurate. By "physical diseases", I'm referring to the idea that mental disorders are a product of a brain disease, malfunction, abnormality, or some other general kind of biological misfire. This idea is obviously incorrect, which is why it was rejected from psychology and psychiatry in the 1950s.

I don't see how this idea is "obviously incorrect"-- and saying that psychologists and psychiatrists rejected the idea over sixty years ago, given the extensive advances in our understanding of the brain in the time since, is hardly compelling.

Some certainly are purely physical, but by no means all. Another, perhaps helpful, analogy would be between hardware and software problems in a computer.

I never said anything about "purely physical"-- but there are certainly physical aspects to mental illness. This doesn't preclude the effectiveness of non-pharmacological interventions. There are all sorts of non-medicinal interventions that are effective for physical ailments. But if the illness isn't physical, where does it reside? Are you implying that someone can be spiritually sick?

[Mr.Samsa]

I never said anything about "purely physical"-- but there are certainly physical aspects to mental illness. This doesn't preclude the effectiveness of non-pharmacological interventions. There are all sorts of non-medicinal interventions that are effective for physical ailments. But if the illness isn't physical, where does it reside? Are you implying that someone can be spiritually sick?

I see where the confusion is now. All behaviors are the result of physical processes in the brain, but not all problem behaviors are a result of a physical problem in the brain.

[Schneibster]

I can testify to the effectiveness of cognitive therapy combined with an appropriate antidepressant; I went through four, one of which had extremely dire side effects on me. Of course, anecdote is not evidence, but it sure worked for me. I had problems from the trauma of multiple heart attacks; I won't say I'm all fixed up, but I feel a lot better (and isn't that the goal of medicine, after all?).

Most such problems are emotional in nature. And most of them come from trauma, and the responses to it, as you said above, Samsa, in your example about the abused kid who can't trust anyone. You learn by repetition, which implies a feedback loop; feedback loops are notoriously stable even in the absence of the original phenomenon that started them. See for example the phase locked loop (PLL) which is an integral component of all FM radio receivers; because the frequency is modulated, it does not exist in the signal and must be synthesized in the receiver to be compared with to determine the modulation envelope and allow the data (sound) to be retrieved. The PLL ensures that the synthesized frequency does not waver, even in the face of changing component values as the radio heats up or cools down.

[mistermack]

While you find plenty of support for CBT online, it's not unanimous by any means.

Even the validity of past trials is questioned in much the same way as I did.
And I've noticed that most patients who advocate it stress the quality of their own therapist, they claim that this is important, and most of them are also on antidepressants.
And rather worryingly, many of them have been on CBT with a therapist for many years, which rather indicates that it's the therapist, not just the therapy. ( presumably, you should know all there is to know after six months or so ).

This meta-analytical review was nowhere near so gushing. :
http://journals.cambridge.org/action/di ... id=6778264
Even though I don't know much about the subject, they are much better qualified and are making similar comments about the methodology of some studies.

Also, asserting that CBT is more effective than antidepressants may not be the good news that it sounds :
http://www.webmd.com/mental-health/news ... an-placebo

[Mr.Samsa]

While you find plenty of support for CBT online, it's not unanimous by any means.

Even the validity of past trials is questioned in much the same way as I did.

You won't find many negative reports of CBT in the literature. You will find some, of course, as with every single medical treatment. But science is not about cherry picking, it's about taking data on the whole.

And I've noticed that most patients who advocate it stress the quality of their own therapist, they claim that this is important, and most of them are also on antidepressants.
And rather worryingly, many of them have been on CBT with a therapist for many years, which rather indicates that it's the therapist, not just the therapy. ( presumably, you should know all there is to know after six months or so ).

Are you sure you know what CBT is? Because CBT is so effective, it only requires a few sessions. Typically, patients receive 5-20 sessions, which means that as an extremely difficult patient, one that can only spare an hour a week, you would only be with a therapist for an absolute maximum of 5 months. The fact that it requires so little interaction with a therapist is one of its advantages in care, as it prevents any kind of biasing interaction or relationship forming with the therapist.

This meta-analytical review was nowhere near so gushing. :
http://journals.cambridge.org/action/di ... id=6778264
Even though I don't know much about the subject, they are much better qualified and are making similar comments about the methodology of some studies.

Interestingly, you have actually demonstrated one of the major flaws with metaanalyses: "Garbage in, garbage out". One of the defining features of a metaanalysis is what it chooses for its inclusion and exclusion criteria, and as demonstrated by the formal replies to the article you've linked to, your article fails miserably. Even without being qualified in the subject, you could quickly google the studies they used as part of their metaanalysis. I'll focus on major depression for now, but I can look into schizophrenia and bipolar disorder later if you like.

Okay, so they looked at 9 studies. Some things to note straight away:

1) Some of the studies are very old (one is from 1989). Why did the authors include such old studies?
2) Most of the studies included had extremely small sample sizes. Why did the authors include such small studies?
3) Only 9 studies were identified. How did they miss all the other more recent, larger, methodologically sound studies?

This left me skeptical enough to actually read the papers they looked at. Of the 9 studies, only 4 of them looked at CBT. This is certainly worrying, considering that it's supposed to be a metaanalysis on CBT. The other 5 were looking at cognitive therapy (i.e. the control/placebo version of CBT) and, unsurprisingly, those studies had largely negative results. The 4 studies that were referenced that actually looked at CBT all found significant, and large, effects that support CBT. Interestingly, all of the studies that found negative results were from studies using patients that were hospitalised against their will.

In sum, very poorly done metaanalysis, that used ridiculous inclusion/exclusion criteria, used poorly designed studies with small sample sizes, and worst of all, included (a majority of) studies that weren't even looking at CBT.

Also, asserting that CBT is more effective than antidepressants may not be the good news that it sounds :
http://www.webmd.com/mental-health/news ... an-placebo

Antidepressants are hugely successful for people with severe depression, as your article points out, and CBT outperforms antidepressants.

[mistermack]

I don't think I cherry picked. I googled "is CBT effective?".
And the bit about patients being on CBT for many years was taken straight from comments of the patients who were on it.
It was nothing to do with what I know or don't know.
You mentioned the comments section, if you'd looked, you would read a comment of a SUPPORTER of CBT who had been with the same therapist for 10 years. And there were plenty more.

I never saw one comment of a person who only received a few sessions.
I'm only reflecting what I read.

Also, you rubbished the study, but gave no reason as to why they should deliberately try to produce errant results.
The authors were well qualified, so there is no reason to think that they are incompetent.
You seem to be hinting that they wanted a certain result, and were able to produce it.
But when I suggested that that could happen, you rubbished that.