Potentially Harmful Therapies: Don't "Misunderestimate" the Problem

I was taken aback by a question I was asked a few days ago. This surprising question emerged in the course of my testimony as an expert witness in a juvenile court matter that focused on Holding Therapy, the physically-intrusive “complementary and alternative” treatment for childhood mental health problems. My interlocutor, the attorney representing one of the parties involved in the matter, first asked an ordinary question: was I familiar with a journal article from 2007 that discussed potentially harmful treatments? I asked whether he mean the article by Scott Lilienfeld (the man who has been called the Ralph Nader of psychology, but we can count on him not to be a third-party candidate for president). Yes, that was the one (Lilienfeld, S.O. [2007]. Psychological treatments that cause harm. Perspectives on Psychological Science, 2, 53-70). Lilienfeld had listed Holding Therapy as one of a group of potentially harmful treatments (PHTs).

Now came the surprising question. Well, said the attorney, what was the problem if the treatment was only potentially harmful? Lilienfeld had not said it was actually harmful; didn’t that mean he thought Holding Therapy was all right? My off-the-top-of the-head answer was that walking into the street in the face of heavy traffic is only potentially harmful to the individual who hasn’t done it yet, but it’s still inadvisable. And someone who tells you it’s potentially harmful does not mean it’s all right.

What does it mean to say that a treatment is potentially harmful? The dictionary definition says “potential” refers to something that hasn’t happened yet; it’s not “actual” (presently happening, or presently showing the effects of having happened.) “Potential” is “power”-- the power to have a certain effect. But how would we know that potential exists? What facts suggest that a treatment has the potential to do harm?

There are a couple of ways we would be likely to tag a therapy as a potentially harmful treatment or PHT. One is by making a comparison to the known outcomes of similar actions. Knowing other outcomes, or using logic, can we conclude that it is plausible that a treatment will be safe? For example, how about “compression therapy”, a CAM treatment that involves having an adult lie down on top of a child? What do we know that is relevant to the safety of this kind of treatment? We certainly know that it’s possible for a person to be crushed or asphyxiated by a heavy weight on the chest and abdomen. Crushing of this kind was even used as a method of execution in the past (see Arthur Miller’s play The Crucible for a dramatic example). Using the outcomes of similar events, and knowing that not everyone with heavy weights on the chest has died, we can legitimately label “compression therapy” as potentially harmful. It has the power or possibility to do severe harm, although it does not always do so.

But what if there is no comparable event that we can use to assess the potential for harm in a treatment? We are left with the more difficult task of watching for adverse events from the treatment and collecting that information. Given that psychotherapies are not expected to have harmful effects, one or two adverse events are enough for us to point out the potential for harm associated with a treatment. Where any such events exist (and some have occurred), we need to investigate them carefully. This is no time for what a formerly-prominent political figure used to call “misunderestimation”.

What sort of adverse events are we talking about here? The obvious ones are death or serious physical injury, but we should also pay attention to incidents like weight loss or inadequate growth. Emotional trauma is less easy to ascertain unless a concerned adult calls attention to a child’s condition, or unless an adult is able to complain of the effects of a treatment in the recent or distant past.

Adverse events can be less obvious, as well. PHTs may do harm by interfering with a child’s education, thus limiting his or her options in adulthood. They may limit social interactions and prevent the development of normal social skills. And, of course, they may use up a family’s energy and resources and prevent the use of safe and effective treatments. All these problems may emerge from the social pressures of a cult-like network in which a family becomes intensely committed to participation in a treatment that should properly be called “potentially harmful”.

Most people are still getting used to the idea that psychotherapies can have real potential for harm. But no one should assume that “potential” mean “unfounded speculation” or “fantasy”.