Oops! A Very Embarrassing Story

Cites a study which concludes that psychotherapy doesn't work very well at all. The study 'Who--Or What--Can Do Psychotherapy?' by Andrew Christensen and Neil Jacobson in 'Psychological Science,' January 1994; The contention that therapy delivered by nonprofessionals is just as effective as that performed by professionals.

By Virginia Rutter, published March 1, 1994 - last reviewed on June 9, 2016

A major study shows that psychotherapy doesn't work very well at all. Butbefore you jump to simple conclusions, consider this: When it comes to matters of the mind, drug therapy isn't any more effective.

Does psychotherapy work? Does professional training actually make therapists more effective? These are some of the questions that researchers are currently debatin ghotly across e-mail networks, at conferences, and, to a lesser extent, in psychotherapy journals.

The matter is scarcely insignificant. Some 16 million people a year use mental-health services such as psychotherapy. And an estimated 24 million more need help, though many of them get it outside the mental-health system.

Now, two heavyweight psychologists have completed a thorough review of the literature. Their findings are eye-opening--though you won't find the mental-health establishment calling a press conference.

The two psychologists report that years of experience, professional education, or lawful credentials do not determine the success of psychotherapy. Never mind that millions of dollars are spent each year on studies comparing the approaches of experienced therapists. Never mind that the more experienced, more educated therapists charge more money for their services. The outcome of therapy is not enhanced by training, education, or years of experience. It may not even matter whether there is a live therapist present!

This is the startling conclusion of Andrew Christensen, Ph.D., and Neil Jacobson, Ph.D. Their study, entitled "Who--Or What--Can Do Psychotherapy?" appeared in the January issue of Psychological Science, a publication put out by the politically brave American Psychological Society.

Christensen, professor of psychology at University of California at Los Angeles, and Jacobson, professor of psychology at the University of Washington, contend that no one has made much of an effort to look at therapy delivered by nonprofessionals, despite the fact that it proves just as effective, or more effective, than therapy performed by psychiatrists, psychologists, social workers, and family therapists.

Nor has anyone rushed to fund the study of inexpensive yet promising alternatives such as self-help and support groups led by nonprofessionals, to say nothing of self-administered treatments via self-help books or interactive computer programs--even though the American Psychiatric Association estimates that 15 million people in the United States participate in self-help groups.

"Most studies compare different types of professional treatments," Christensen notes. "It is discouraging, because the most common finding is that there is no difference among treatments. We suggest that more money be allocated to studying nonprofessional treatments--where the results are promising and cost less to achieve."

Comparing professional and nonprofessional treatments tells us more about what actually helps people get better, Christensen insists. He and Jacobson reviewed studies done since 1979 that asked: Are professionally trained therapists more effective than paraprofessional helpers without professional degrees but often with specific training? One study they reviewed showed no differences in the rates of psychological improvement when professional therapists, averaging 23 years of experience--were compared with liberal arts college professors having no experience or training. Both were "treating" disturbed college students.

The duo also discussed a 1979 review of 42 studies that compared professional and paraprofessional therapists. Only one component of the study demonstrated superiority of professionals; in 12, paraprofessionals actually helped people more. The remaining 29 found no differences.

Over the years, the data from the troublesome 1979 review have been reanalyzed using more stringent standards; each time the results have come back stronger for paraprofessionals. One study concluded: "Clients who seek help from paraprofessionals are more likely to achieve resolution of their problem than those who consult professionals. Hmmm.

"These are provocative findings for the psychotherapy community," note Christensen and Jacobson. "It is hard to imagine a study comparing trained and untrained surgeons, or trained and untrained electricians, for that matter. Dead patients in the first instance or dead trainees in the second could be the unfortunate outcome."

The difference between surgeons and psychotherapists has a lot to do with the difference between surgery and therapy. Across the board, psychotherapy researchers agree that a positive therapist/client relationship is the most important feature of successful treatment.

But get this--the Christensen and Jacobson report questions even the highly touted need for a therapist/client relationship. Self-administered treatments worked just as well as those delivered by live therapists, in certain cases. Computerized treatments have been shown to work for obesity, phobias, and depression.

Observes Christensen: "With most professions it is very clear there is a specific skill involved, but in psychotherapy it is not clear that the skills of the therapist are any more helpful than the skills of people with life experience in dealing with a problem."

Despite the solidity of Christensen and Jacobson's findings, a long line of scientists take issue with their conclusions. Psychotherapy researcher Kenneth Howard, professor of psychology at Northwestern University, argues that "it is not possible to test treatment in a lab model." In other words, you can't study psychotherapy in controlled clinical trials the way medical research does. "You or I would never volunteer for a study involving nonprofessionals, so why would anyone else?" Better are naturalistic studies of cases from everyday practice--like one of his own. It shows, he says, that experienced therapists progress more than inexperienced therapists in the first three sessions. But the inexperienced ones catch up by about the sixth session.

Naturalistic research does not necessarily support Howard's optimism for psychotherapy. In a review of child psychotherapy research, John Weisz, professor of psychology at UCLA, found that child psychotherapy actually works only in lab settings. When child psychotherapy is studied in "real life" it shows no benefits.

Whether professionals do better than paraprofessionals is irrelevant, Howard insists. He worries that Americans' overall health will be affected by the Clinton health care reform plan, which may limit access to psychotherapy by limiting coverage.

States Christensen, "If you are concerned about access to care, then a crucial question is who can provide the care. If only a small group of people can, then access is going to be limited automatically. But if a much larger group can provide care, then the benefits are much greater."

Wait a minute, says Larry E. Beutler, Ph.D., editor of the journal of Consulting and Clinical Psychology, the leading journal of psychotherapy research. Most psychotherapy research, including the studies cited by Christensen and Jacobson, look at only half the question.

The full answer will come only from asking who is getting the psychotherapy and just what kind of therapy are they getting, Beutler argues. What's needed are "matching" studies that figure out which treatments work for which clients. Then services can be tailored to people based on the characteristics of clients, not of experts.

Jacobson's concern goes beyond any difference training or matching makes. As a practicing therapist (as is Christensen), he asks whether behavior changes even temporarily. The National Mental Health Association, a mental health lobby, recently reported that mental health professionals have an 80 percent success rate in treating depression.

Jacobson points out that some studies declaring "success" measure how many patients return to a healthy state--that's called clinical significance. Other studies measure only whether improvement has occurred--called statistical significance. At the end of treatment a severely anxious person maybe improved--a statistical difference--but still a long way from a functional state--clinical significance.

When he applies the clinical-significance standard, psychotherapy shows disappointingly low success rates--and so does drug therapy--for treating marital distress, agoraphobia, and children's disorders, among others. "All the treatments are pretty weak. Some patients get better without treatment," he explains.

A new study of depression treatment by the National Institute of Mental Health puts the success rate--for drugs or psychotherapy--at 19 to 30 percent. "My mother wonders what I get paid for if this is the best I can do," quips Jacobson.

So how can 40 million Americans get effective help without breaking the public or private bank? The answer lies in examining what makes people get better--not what psychotherapy can do to help. it is indisputable that solving mental health problems--with psychotherapy, computer programs, or help cleaning your house--offers many benefits. It saves employers money by reducing sick days, improves physical health, reduces doctor visits, and reduces violence and abuse.

Sometimes psychotherapy does the trick; sometimes less costly approaches work. It depends on the person and the problem.

However psychotherapy works--whether it heals, or even treats, what ails you--people want it and like it. (There's research that says so.) On the average, psychotherapy offers more benefits than no therapy--it just may be that much simpler interventions yield similar results.

CARTOON