How can the simple act of forgetting become the impetus for a psychiatrist to develop his own series of self-administered psychotherapy computer programs? According to John Greist, M.D., it began when he forgot to ask his patients important information during the interview process. He also became aware that the very way he formed the question would lead to different responses from the patient, depending on the person he was speaking with or their frame of mind.
How can the simple act of forgetting become the impetus for a psychiatrist to develop his own series of self-administered psychotherapy computer programs?
According to John Greist, M.D., distinguished senior scientist for the Dean Foundation in Madison, Wis., it began when he forgot to ask his patients important information during the interview process. He also became aware that the very way he formed the question would lead to different responses from the patient, depending on the person he was speaking with or their frame of mind.Then in 1965, while completing an internship, he witnessed another intern doing a computer interview of a patient.
"It immediately struck me that this addressed both of the issues that I was experiencing as problems," he recalled. "The computer could be programmed to ask questions it never forgot, and further it could be carefully programmed and ultimately refined so that the stimulus would be whatever you wanted it to be."
His original thought, of designing programs for physicians to gather information such as medical histories and physical exams, lost some steam when he found that most doctors won't use a computer program unless it saves them time.
"The reality is if a computer program means that a doctor gets home late for dinner, he or she will not use it," he said.
So since 1967, Greist's emphasis has been on interacting directly with patients, gathering information from patients and giving information to them in the form of computer interviews.
Writing the Programs
In order to write the programs directly for the patients, he began asking them about symptoms that would lead to diagnoses-symptoms within a diagnosis that would permit him to assess the severity of their disorder and to measure change over time. He did this for various disorders-mood, anxiety, thought, substance, dementing.
"Right from the beginning, it was clear to us that as we gathered information from patients, we had to make sure they understood the question and that they understood the responses they were making," Greist explained.
Also, in the process of teaching them how to use the computer interview, Greist realized he might be able to teach them other things, such as health information. Better yet, he could even instruct them in self-help, where that was appropriate.
He continued working along those lines, and in the mid-1980s, his colleague, Paulette Selmi, Ph.D., developed a cognitive behavior therapy treatment program for depression. It comprised six sessions, and Selmi compared it with a human therapist giving six sessions of cognitive therapy.In a study published in the January 1990 issue of American Journal of Psychiatry Selmi randomly assigned patients either to computer-administered cognitive-behavioral treatment, to therapist-administered cognitive-behavioral treatment or to a waiting-list control condition. Unlike the usual waiting-list control groups, however, these patients would be able to get treatment if they began to deteriorate.
When that study was completed Selmi found that the computer program and the human therapist were very effective, but the waiting list control condition produced no improvement, either at the end of the six weeks or in a two-month follow-up.
That was the first treatment program Greist and his cohorts studied systematically. Since then, they've been developing and studying two other treatment programs-one for obsessive-compulsive disorder and the other for people with mild-to-moderate depression.
"In the studies we've done thus far on our behavior therapy program," Greist noted, "we get improvement on the Yale-Brown Obsessive-Compulsive Scale that is at least as great as you get with medication, but is substantially less than the very best behavior therapists can provide. The problem is, less than 10% of the people can get to those very best behavior therapists. There are some patients who do extremely well using this computer program."
Greist hopes that these computer programs, when properly studied and validated, will be able to fill part of the enormous gap that includes patients who can't get treatment, can't get good treatment or can't afford it.
Future Role of Computers
"We think that there's a very large role for computer interviewing in mental health applications, and way beyond into all of the aspects of medicine." Greist said. "Computers assist physicians in understanding what is happening with patients as we treat them. It's just not feasible, not practical to do that with human interviewers.
He continued, "The computer, in one form or another [whether patients fill out forms that are optically scanned, generating a report, or whether the data are collected directly from the patient by a computer, or the telephone in an interactive voice response mode...seems a remarkably helpful way to solve a problem that we simply cannot solve without them."
Acknowledging some physicians' hesitancy in using self-administered programs, Greist was empathetic. However, he said that it is important for the clinician to view these as tools that can enhance their clinical practice. "They're not intended to substitute for or replace a clinician, but they will permit us to do a better job helping our patients, and that's what we all want to do in medicine."