Does Treatment Have an Effect?

The patients (18 male, 16 female, ages 16-71, mean age 37) all had been referred by their treating psychiatrists. Their cases were reviewed by the MGH Cingulotomy Assessment Committee (comprised principally of a subset of the authors) to become candidates for the procedure. The committee had to agree unanimously that the patient should undergo the procedure, and the patient had to give his or her informed consent. Prerequisites for the committee to favor treatment included patient's resistance to pharmacotherapy, psychotherapy, behavior therapy, and electro-convulsive therapy (when applicable), and the symptoms had to be severe and unremitting for at least one year. All patients had a documented DSM III-R disease that interfered significantly with normal functioning (e.g.., major affective disorder with unipolar depression (10 patients), or bipolar depression (5 patients), undiagnosible developmental disorders (1 patient), and obsessive-compulsive disorder with major depression (15 patients)). For some patients whose condition failed to improve 6 months after surgery, additional cingulotomies were performed.

The cingulotomy procedure is to insert a pair of electrodes into holes drilled in the skull just above the hairline, to a pre-determined position in the brain (guided by MRI). Radio-frequency electricity is passed through the electrodes, to raise the temperature in the part of the brain between the electrodes to 85°C (185°F) for 90 seconds, essentially cooking part of the brain between the electrodes. (This causes what is called a radio-frequency thermocoagulation lesion). The basic procedure was changed in 1993 to include a larger part of the brain.

Of the 34 subjects, 31 were available for follow-up assessment of some sort; of the 31, two had committed suicide. Follow-up assessment was done two ways: either the referring psychiatrist or the treating psychiatrist completed two questionnaires---the Clinical Global Improvement Scale, and the Current, Global, Psychiatric-Social Status Scale---or, if the patient was no longer seeing a psychiatrist, the most recent correspondence with the patient was used to assign ratings. I could not find a reference to the number of patients who were in each of these categories. On the basis of these ratings, each patient was classified as a "responder" (R), a "possible responder" (P), or a "non-responder" (N).

Fourteen of the patients who had not responded six months after the initial procedure had multiple procedures. The categorized responses were as follows:

The article breaks down the responses further by type of mental illness.

I wish to study the effect of daily consumption of small amounts of alcohol (say, two glasses of wine) on the incidence of heart disease. In many European countries, it is customary to drink wine with dinner, while in most Islamic countries, it is not. I therefore decide to compare the average incidence of heart disease for several European countries in which the average daily consumption of wine is two glasses, with the incidence of heart disease in Saudi Arabia, where the consumption of alcohol is negligible (it is a capital offense). Consider the non-drinkers to be the control group, and the drinkers to be the treatment group. Suppose it turns out that the incidence of heart disease is lower in Saudi Arabia.

It is well known that smoking cigarettes is bad for you. Many smokers try to quit, but quitting can be very difficult. Nicotine patches are a popular medication prescribed by doctors to help people quit smoking. One of the objectives of the study described here was to determine how effective a nicotine patch, combined with the National Cancer Institute program for physician advice and nurse follow-up, is in helping people "stay quit" for a year.

Those subjects were assigned at random either to receive a 22 mg nicotine patch (and physician's advice and nurse follow-up) --- the treatment group --- or to receive a placebo patch (and physician's advice and nurse follow-up) --- the control group. One hundred twenty subjects were assigned to treatment and 120 were assigned to control. Subjects wore patches for 8 weeks.

Nurse follow up was weekly, face-to-face for the first 8 weeks and by phone thereafter. Neither the subjects, the doctors, nor the nurses, knew which subjects wore active patches and which wore placebo patches.

The study classified a subject as having quit smoking if:

1. the subject reported that he or she had not smoked at all since the last visit, and
2. the level of CO in the subject's expired air was 8 ppm or less.

Subjects with missing information or who dropped out of the study for any reason were considered smokers.

Here are some characteristics of the two groups after random assignment, but before the treatment began:

Here are the results: