Addiction and Mental Health

Learn about addiction and mental health issues

Therapeutic alliance as a predictor of outcome in treatment of cocaine dependence: A Review

with one comment

Therapeutic alliance as a predictor of outcome in treatment of cocaine dependence

by Barber, et al.

A Review by Julie Myers, PsyD, MSCP

Licensed Clinical Psychologist in San Diego

There has been an increasing interest in outcome measures in therapy.  Although there is a school of thought that technique is the primary predictor of outcome, some therapists believe that it is the therapeutic alliance that is the major predictor of outcome.  The study by Barber, et al. examines outcome measures in drug treatment.

Substance abusers are often difficult to engage in treatment, yet the therapeutic alliance has rarely been studied.  This study is the first to examine the self-reports of the therapeutic alliance as a means to predict the outcome of cocaine treatment.  Because there is a high drop-out rate for cocaine dependence (50-80 % dropout within 3 months), the research also attempts to distinguish between intent-to-treat and completer samples to see if the alliance early in treatment predicts treatment outcome.

The study used a sample size of 252; many co-occurring disorders were excluded.  Clients were randomly assigned to three different treatment techniques:  Cognitive therapy (based on Aaron Beck), dynamic therapy (support-expressive therapy), and 12-step drug counseling.  A fourth treatment group was added later in the study, which used 12-step group drug counseling only.  In addition to the targeted technique, all treatment groups were mandated to attend group drug counseling 2 times per week for initial two weeks, then once per week for four months.  Therapists in the different treatment groups were not similarly trained; drug counselors had the least training.

Both patient and therapists completed two alliance scales, the Helping Alliance (Haq-II) and the California Psychotherapy Alliance Scales (CALPAS).  In addition, patients were administered measures of functioning.  Statistical analysis of the results looked at a number of different relationships, including the prediction of outcome from alliance given symptom improvement, completer sample, and the number of sessions.

The results did not find a strong relationship between the outcome measures and the therapeutic alliance.  The alliance didn’t predict drug outcome at six months, although at one month there was stronger correlation; the therapeutic alliance was a better predictor of outcome for depression scales.  Results also showed that the therapist’s rating of alliance was less predictive than patients.  Measures were similar across completer and intent-to treat samples.  The only strongly conclusive results were that the shorter the lag time between assessment and outcome measures and the longer clients remain in treatment, the better the outcome associations.   These results do not seem to be particularly surprising to me.  The authors state that a “good therapeutic alliance with the therapist, as viewed by the patient early in treatment, is important in predicting outcome when it is embedded in a long term relationship with that therapist.”

The authors state that there are several possible reasons for the weak predictive results, including the choice of the outcome and alliance variables, the nature of the patient population and/or disorder, and the restriction of range in the measures of alliance.  Although these seem likely influences on the results of the study, I found there to be several other possible explanations.

If a study is not well designed, all of the statistical analysis in the world will be meaningless.  In my own graduate-level econometrics classes, I was taught that when designing a study, the researcher must be careful not to examine too many variables, otherwise it becomes a study that is “hunting for” statistical significance, which biases and invalidates the results.  In my opinion, this study threw too many variables into the mix, both in the design of the study and in the statistical analysis.  A far more rigorous study would have chosen one or two associations to measure, then designed the groups with stricter protocol.

In my opinion, the addition of the fourth group-counseling treatment group invalidated the results, not only because of the late introduction of the treatment, but because of the cross-over between the different treatment samples in this group counseling.  Although the researchers attempted to make the treatment samples significantly different from one another, using different therapy methods and different therapist qualifications, for therapists and counselors who do not use 12-step methods, the mandatory attendance in a 12-step group treatment adjunct to therapy may harm the therapeutic alliance.  Cognitive therapists, in particular, may have little or no belief in the 12-step method.  From my understanding, when therapists do not believe in the methods being used, outcome is compromised.

This study may be better used as a guide to setting up treatment protocols to assess therapeutic alliance than it is useful for the results of the study.  Significant changes that I would suggest are:  less lag-time between assessment of outcome and therapy, more clearly defined and independent treatment protocols, and fewer measurements of outcome.  Although the large sample size was a positive aspect, a smaller sample would not compromise the results.

– Julie Myers, PsyD, MSCP

http://www.DrJulieMyers.com

References:

Barber, J., L. Luborsky, P. Crits-Christoph, M. Thase, R. Weiss, A. Frank, L. Onken, R. Gallop (1999),  Therapeutic alliance as a predictor of outcome in treatment of cocaine dependence.  Psychotherapy Research 9(1), pp. 54-73.

Written by Julie Myers, PsyD, MSCP

July 23, 2011 at 7:32 am

One Response

Subscribe to comments with RSS.

  1. This design is spectacular! You most certainly know how to keep a reader entertained.
    Between your wit and your videos, I was almost moved to start my own blog
    (well, almost…HaHa!) Great job. I really loved what you had to say, and more
    than that, how you presented it. Too cool!

    apple macbook pro deals uk

    November 11, 2013 at 2:18 am


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: