felt very strongly that it was important not to just decide that a particular treatment is good but rather systematically test whether certain treatments might be more effective than others.

We decided to test exposure therapy — in which the survivors confront his or her feelings about the event and relive it in therapy — against traditional kinds of counseling that people might get.

Advertisement

We offered both supportive counseling and exposure therapy, and we randomized people into one of these treatment options for four weeks. After four weeks, if they weren't better, we offered people who got supportive counseling the opportunity to get the other therapy. And we collected information about their symptom improvement and their stress hormones.

We don't have the results of our study yet, but our impression is that people will do better with any therapy. It's just that certain kinds of treatment may accelerate the rate of recovery or allow people to get better with fewer sessions.

A few months from now, I think we'll have some definitive information.

There is sometimes a negative attitude toward research in the aftermath of a disaster, a feeling that it may be exploitative at a time like this.

My own view, though, is that it is far more exploitative to act without evidence that the interventions are therapeutic.