A Dark Underbelly of Therapy?

Some journal titles catch the eye more than others and so it was with a recent edition of The Psychologist, I saw, ‘When Therapy Causes Harm’ by Christian Jarrett. The very serious focus of Jarrett’s article is the oft suspected but rarely acknowledged or declared issue that some psychological therapies may result in more harm than good. According to Jarrett around 10 per cent of clients get worse after starting therapy.

Potentially Harmful Therapies
Citing research by Lilienfeld (2007), a provisional list of potentially harmful therapies has been drawn up:

• Critical incident stress debriefing
• Facilitated communication
• Recovery memory techniques
• Boot camps for conduct disorder
• Attachment therapy
• Dissociative identity disorder-oriented psychotherapy
• Grief counseling for normal bereavement
• Expressive-experiential psychotherapies

Professor Lilienfeld makes the case for a ‘dark underbelly’ of clinical practice where many practitioners underestimate the extent and magnitude of the problem.

But what exactly is a harmful therapy? In some cases it may be obvious but in others less so. The fact that a psychological therapy has the potential to harm is no different in many respects to conventional medicine. There are, as Jarrett points out, a number of methodological problems in deciding what really constitutes a harmful therapy. For example, if a client gets worse after starting therapy what is there to say this might not have happened anyway? If a client takes longer to respond to a particular therapeutic approach over one that might have been speedier has the client suffered harm because they could have recovered more quickly?

The fact that bereavement counseling is on the list provides a useful illustration of the methodological problems. It has been found that:

‘ . . . bereavement counseling’s dire reputation is thanks entirely to an unpublished student dissertation, in which it was claimed 38 per cent of bereaved clients would have fared better if, instead of receiving counseling, they had been in the no-treatment control group’. (p.11)

In his repost, Lilienfeld acknowledges the source, but makes the point that the burden of proof is with clinicians to prove that a particular therapy is effective.

Getting Therapists to Change

One of the first issues to tackle would seem to be getting some therapists to appreciate and acknowledge that some clients will deteriorate during therapy. Jarrett refers to a study by Professor Michael Lambert at Brigham Young University. Lambert asked 40 clinicians to predict which of their 500 clients would get worse during therapy. Almost none of the therapists predicted a worsening of their client’s condition. In fact of the 40 clients that did deteriorate nearly 90 per cent of these were picked up Lambert’s own predictive system and only 3 per cent were identified by the therapists.