A helping hand for addicts

Source: naturenews
Erik Vance

Vincent Clark, of the University of New Mexico in Albuquerque, thinks he has something like a crystal ball for drug addicts. By applying traditional psychiatric evaluation and modern fMRI brain imaging to people recovering from drug addiction, he claims to be able to spot who is likely to relapse — months before the relapse actually happens.

Clark puts people recovering from cocaine and methamphetamine addiction in an fMRI machine, then asks them to play a game called 'oddball task' which is common in addiction research. Participants hit a button when they see an 'X' on a screen, but not when they see a 'T'. Mixed in are a few distracting 'C's: when these appear, they trigger activity in the posterior cingulate region of the brain in some addicts. Clark later meticulously tracks the volunteers, taking hair and urine samples, to see if they have begun using drugs again.

With more than 80% accuracy, Clark says, the test predicted who would relapse (those whose posterior cingulate did not light up) and who would stay straight (those whose posterior cingulate did) over the next six months. Combined with a simple test for a history of mania, it was 89% accurate, he says.

Clark presented the results during the annual meeting of the Organization for Human Brain Mapping in San Francisco, California, on 19 June. Nature News talked to him about how he keeps such research going.

http://www.nature.com/news/2009/090625/full/news.2009.600.html

;)

He's seeing a correlation, not a cause.

He is also skewing the results by taking hair and urine samples. A certain number of participants will not relapse simply because they are being meticulously tracked and do not want to come up positive. Sometimes a watchful eye is all people need to stay clean.

The article also states that they used participants who were in outpatient treatment "living at home and getting some kind of treatment a couple of times a week". That introduces extra variables. Social and economic status, family environment, etc play heavily into what kind of outpatient treatment is chosen.

On a side note, it seems odd that he didn't use the residential treatment centers for the recruits, where the treatment environment is at least much more controlled. He also didn't have a control group as far as I can tell.

The single most harmful part about the study is the following:
"Did the patients know what this study was about? Absolutely. We said we were looking for ways to predict who is going to relapse. They do care. These people want to be a part of the solution."

By letting them in on what the study is about, he's already invalidated the results. There is a reason that blind (and double-blind) studies are used. Once you let the participant know what they are being studied for, the results WILL be skewed, especially with behavioral studies. For example, let's say someone is accepted into a study about how often males wash their hands after using a urinal. If he is not told of the content of the study, he isn't going to change those habits. However, if he is told about the content of the study, he is more likely to have that thought run into his brain every time he uses a urinal and change his habits.

Overall, it's a completely non-scientific study and wouldn't stand a chance in a peer review. However, I do give him kudos for making the correlation, however accurate the results were. That in and of itself will lead to many more studies and a better understanding of the way the brain works.

Just be thankful this isn't Orwell's 1984 with him making the call on who needs to be re-educated :)