Summary One hundred cancer patients undergoing active treatment were interviewed to determine how they perceived their illness and how their perceptions compared with those of their attending physicians.Ninety-eight patients recognized that they had cancer and 87 correctly identified the tumour type. Sixtyfour of 67 patients with local or regional disease were aware of this, but 11 of 33 patients with metastatic disease incorrectly believed that the cancer was localized. Five of 52 patients being treated for cure thought they were being treated palliatively, and 16 of 48 patients receiving palliative treatment believed that the doctor's aim was to cure them. Forty of these 48 patients significantly overestimated the probability that the treatment would prolong their lives. (Parsons, 1951). Others have since stressed that "mutual participation" must replace paternalism as the basis of the modern doctor-patient relationship, if physicians are to succeed in combating the public's growing distrust of the profession (Haug & Sussman, 1969;Brody, 1980;Jensen, 1981). It is now generally agreed by doctors and lay people in North America that patients should participate actively in decisions about their care. This participation can only be meaningful if the patient understands his situation well enough to perform the kind of cost-benefit analysis that has traditionally been left to the doctor. Thus it is now more important than ever for patients to be well-informed about their illness and its treatment.Patients must already give their 'informed consent' for any form of medical intervention, but 'informed consent' has proved difficult to define and even more difficult to realize in practice. Information is not always effectively transmitted to the patient, and it has been suggested that complex medical information may be of little value to those who do not have the educational background necessary to interpret it (Robinson and Merav, 1976;Cassileth et al., 1980a; Mackillop & Johnson, 1986 Over a three-month period we attempted to interview every available patient, but, if a member of the healthcare team preferred that the patient not be interviewed, these wishes were respected. One hundred and sixteen patients were approached. They were told that the aims of the study were to find out how well patients understand their illness and to learn if communication between doctors and patients is effective. Written consent was obtained from each participating patient. The consent form outlined the study's objectives and explained that the interview would be videotaped. Sixteen of the 116 patients who were approached chose not to participate. Eleven felt that the interview process would be too stressful and five felt that it would be too inconvenient. The interviewer (WES) is not medically trained and was unaware of the patient's diagnosis. Interviews were videotaped to avoid notetaking during the course of the conversation, but a checklist was used to ensure that all patients were asked the same questions. The content and format o...