There are many clinical reports on psychotherapy with depressed patients, and an enormous effort has been devoted to psychotherapy research (10, 23, 24, 27, 28). However, there is little quantitative description of the content of psychotherapy—what patients discuss during sessions and how much time is spent reflecting on problems. This situation is partially understandable given the technical difficulties in obtaining psychotherapy process data. Content analysis of transcribed tapes, process records, video-tapes and the coding of words have been among the methods used (7, 9, 12, 14). These methods are inclusive and precise but impracticable in lengthy psychotherapy or with large samples.