Sunday, April 24, 2011

Guitar calls Lidcombe critics ridiculous

Barry Guitar launched a frontal attack on those who are doubting Lidcombe in this radio interview on NPR. The journalist Jane Lindholm asks Guitar why some doubt the evidence base of Lidcombe due to small sample siz. He says that "This is the most ridiculous critique you can imagine". Check at 7:30 minutes. He refers to 100 studies showing efficacy with gold standards. He says that critics don't want to see 100 studies but 1000s studies. He also says that other treatments have no evidence at all that they work.

He falls into the common traps:

(a) equating perceived success in an outcome trial with success of the method itself. (it could well be that other treatments are as effective, and success is not due to Lidcombe specific issues.)



(b) He did not mention the lack of good long-term follow-up. The only follow-up with small sample size did not show results above natural recovery, even if I assume that the considerable drop out did not relapse.

(c) He tells us a wonderful story of how he "cured" a boy, without telling us that it could very well have been natural recovery.

He also attacks "this experimental psychologist Pete Howell who feels that it is inhuman or heartless to treat children with Lidcombe" for his article in Nature. When in fact, Pete Howell only wrote:
More controversially, some researchers believe they can induce fluency in people, children in particular, using verbal operant procedures, similar to the reward and punishment techniques used to train dogs and other animals.
He only said that (a) the method is controversial, (b) conditioning is what you also do to train dogs. So what's wrong with his statement. He does not say that it is ethically wrong to train kids because that is how you train dog!! Barry Guitar should read the article!

I was always suspicious about Barry Guitar's credentials as a scientist. Especially after hearing his "naive" talk on his work on temperament. (Check out Per Alm's deconstruction of his temperament work. Sorry don't have a reference right now) For me, Guitar is a dedicated therapist, and I would recommend him to others. But he is a mediocre scientist in the mantle of "a respected authority" and an academic professorship draining resources and young brains from good scientists.

13 comments:

Satyendra said...

Interesting. Yes, I can see- being a good therapist (been one) and a good scientist are two different things..
Why cant people say- "I dont know. Or I am in doubt."?

Anonymous said...

It does look like Lidcombe has the most evidence for the treatment of preschool kids who stammer. If Lidcombe is not you #1 choice, what is your #1 choice?

Therapists have to make decision right now, today!!!

How about you show some respect to Dr. Guitar?

Professor Guitar is a good clinician and knows more about clinical research than you, wouldn't you agree?

Finally, focus on the arguments, the temperament work is irrelevant.

Tom Weidig said...

@Sachin: "I don't know. Or I am in doubt." Because he thinks he knows and is therefore not in doubt!


@Anonym: Can you please use your real name?

>> It does look like Lidcombe has the most evidence for the treatment of preschool kids who stammer.

Either the evidence is sufficient or it is not. There is no "more or less evidence". For me, it's not sufficient, due to absence of follow-up data with big sample and a comparison with other treatments.

>> If Lidcombe is not you #1 choice, what is your #1 choice?

This question is not relevant what I believe. I am only arguing against their being an evidence base for support of Lidcombe-specific treatment.

>> Therapists have to make decision right now, today!!!

I completely agree, but I am not a therapist so I don't have to make this decision. I take the stand-off view, and ask: is the evidence enough to say it's effective AND better than others. My answer: No.

>> How about you show some respect to Dr. Guitar?

No respect for anyone in an intellectual debate. I focus on arguments and I say what I have on my mind.

>> Professor Guitar is a good clinician and knows more about clinical research than you, wouldn't you agree?

Clinician: yes. Clinician research: given his low sample sizes and other strange setup that I saw. I seriously doubt it. But he is no worse than most others.

>> Finally, focus on the arguments, the temperament work is irrelevant.

No, they are linked. If he is blind to certain methodological traps in one piece of research, he is in others, too.

Ora said...

Tom - Can you explain your point "(a) equating perceived success in an outcome trial with success of the method itself"? I don't understand your point.

I think that this may be the Per Alm article you're referring to - http://www.mendeley.com/research/stuttering-in-adults-the-acoustic-startle-response-temperamental-traits-and-biological-factors/

Oro said...

This is a war now. You have attacked the Lidcombe and the people for many years, now they are just fighting back.

You are attacking and taking away Guitar's paychecks.

http://www.montrealfluency.com/continuing-education/faqs/

Vanwannabe? said...

Tom -

Barry is one of the SFA's go to guys. If the queen decides to enter the battle VS you, you will quickly come to your "expert" status demise.

The SFA and its professionals totally endorse the Lidcombe method. Guitar and Chemla preach it during their sermons on the professional mount.

It is,as Barry so eloquently shared, the "cure".

Just give it up, Tom. Money rules, you'll drool.

Tom Weidig said...

@Ora: Tom - Can you explain your point "(a) equating perceived success in an outcome trial with success of the method itself"? I don't understand your point.

Let us assume that treating children with Lidcombe shows clear long-term benefits with respect to an un-treated sample. We can then still not say that Lidcombe-specific technique have caused the success. For example, very different treatment approaches might get a similar success rate. It could well be that ANY treatment setting will improve fluency. Franken et al are currently running a trial with 50 kids on Lidcombe and 50 kids on demands & capacity. Both could get similar results. What then? Is the success due to Lidcomb and D&C specific technique, due to generic treatment methods, or even due to the same measurement bias??

Tom Weidig said...

@Oro: This is a war now. You have attacked the Lidcombe

I have NOT attacked Lidcombe as a treatment, but the evidence basis of Lidcombe.

Satyendra said...

"What then? Is the success due to Lidcomb and D&C specific technique, due to generic treatment methods, or even due to the same measurement bias??"

Hmm- I though I understood research- but TOM you have truly extended my horizon!! I am afraid many people out there just want to hear "good news"- any "good news"- they may not be much concerned about the "facts out there".. Anyway, this debate, has certainly educated me- a stammerer, who often counsels other PWS and their families..
Keep up the good work and objectivity..

Anonymous said...

Introducing measurement bias into this is a great idea, but I see two problems.

1. Because Libcombe presents positive results, it means it's just as good as any other therapy so the treatment is viable.

2. The title "Libcombe never fails, only parents do" is an attack on Libcombe. "Therapist never fails, only parents do" would be a better title, because she would blame the parents without Libcombe probably. I'm convinced that it's not the therapy as much as it is the therapy in terms of experiences and outcomes. I also don't think you would have created that title if any other therapy was used.

Tom Weidig said...

@Anonym:

1a. I have never said that Lidcombe is worse than other treatments.

1b. in my view Lidcombe has not shown long-term success rate above natural recovery, where success means fluency as if naturally recovered.

2. I agree that "therapist never fails, only parents do" is also a valid statement. However, Lidcombe very explicitly claims great successes ("getting rid of stuttering"), and therefore the therapist's conclusion must be that the parents are at fault.

I do not see how your statements conflict with measurement bias.

Anonymous said...

> Tom Weidig said...
> 1b. in my view Lidcombe has not shown
> long-term success rate above natural
> recovery, where success means fluency
> as if naturally recovered.

Yes, Tom, they have shown that. Saying "in my view" doesn't allow you to neglect the scientific facts. If you criticize all Lidcombe data, you should also interpret the recovery data with an equally critical eye. What you will see if you can be objective, is that recovery (fluency as if naturally recovered) is both much larger and much faster for Lidcombe than for spontaneous recovery.

> 2. I agree that "therapist never
> fails, only parents do" is also a
> valid statement. However, Lidcombe
> very explicitly claims great
> successes ("getting rid of
> stuttering"), and therefore the
> therapist's conclusion must be that
> the parents are at fault.

That is completely false logic to think/state that "therefore the therapist's conclusion must be ..." Please show me where Onslow, Packman, Shenker, or Guitar have actually said what you are implying they say. They have not done so. They always openly recognize that Lidcombe does not work in ALL casses and they never state that the parents are at fault when this occurs.

Anonymous said...

I read the below non stuttering related article. Reminded me of Lidcombe.

http://www.cnn.com/2011/US/06/07/sissy.boy.experiment/index.html