The material was obtained in a longitudinal study of general practice in a variety of organisational and physical settings. More specifically, it was designed to assess the effects of a move to a Health Centre on providers and recipients of care by means both of before and after comparisons and by comparing Health Centre practices with practices not accommodated in Health Centres. Two group practices, one with six, the other with five principals, were scheduled to move to a purpose-built Health Centre in 1973. The practices agreed to be studied over a period covering the year before the move as well as two years after the move. Additionally two practices, a partnership and a single-handed one, not planning to move to a Health Centre, agreed to participate in the study. Changes occurred in the latter practices in the course of the study: the single-handed practice became a partnership and the partnership, with the retirement of a principal, single-handed. A third comparative practice, a single-handed one, joined the sample. The patient studies were intended to contribute information on patient views and experience of the care provided by the practices, and, in the case of the group practices, the effects of the move on them. The information was obtained from medical records, from taped recordings of doctor-patient consultations, from interviews with patients in their homes and their doctors' surgeries. Additionally, data on the consultation was abstracted from the doctor's notes. The data on the providers of care were obtained mainly by intensive unstructured open-ended interviews and by observation at practice meetings and elsewhere. Variables 1. The Attenders' Enquiries The object was to obtain information about patient requirements of specific consultations with a doctor and their assessments of the consultation immediately afterwards. Patients attending their doctors' surgeries during sampled sessions were asked to complete a simple questionnaire in two parts: the first, completed before the consultation, focused on the decision to consult and on expectations of the consultation; the second, completed after seeing the doctor, asked what happened during the consultation and what they felt about it. Additionally, data on the consultation were abstracted from the doctors' notes. 2. The Home Interview Surveys In the home interviews patients were asked to comment on the service provided by their doctors, about their own state of health as they perceived it, their belief in the efficacy of remedies, their willingness to self- treat and about their age, occupation, family and household composition and proximity to kin and friends. The samples of patients selected for these interviews were adults aged sixteen or over registered with the study practices and resident in the catchment area of the Local Authority's (later Area Health Authority) Family Welfare Clinic. They were systematically chosen to include more relatively frequent users of primary care services. To do this a brief 'screening' questionnaire was sent to all adult patients of the study practices living in the catchment areas. In 1972, 6365 questionnaires were sent out of which 3638 were completed and returned by the addressee. A further 1883 were returned <i>addressee moved away</i>. There was no response of any kind from the remaining 844 patients even though two more questionnaires, mailed at fortnightly interviews, were sent to them. The responders to the screening questionnaire were categorised, albeit rather crudely, as first, those suffering from long-term chronic conditions such as diabetes, chronic bronchitis, anxiety or depression; second, those with disabling conditions such as osteoarthritis; third, the vulnerable, so described because they were heavy smokers or women in the fertile age group likely to use the services with some frequency for contraceptive advice, care in pregnancy and early motherhood; and, fourth, the rest. In 1975 those who were interviewed in 1972 who were still registered with the practices and residing in the specified catchment areas were re-interviewed. In addition those members of the original sample not previously interviewed, that is, not 'contaminated' by the interview process, as well as a sample of patients joining the practices after the first sample was drawn were interviewed, using the same method to sample new arrivals as was done for the original sample. 3. Tape-recorded Consultations In 1975, with two exceptions, the principals and trainees of the study practices with the consent of the patients agreed to tape-record a sample of their consultations, resulting in the recording of some eight hundred consultations of fifteen doctors. A random tenth were used to construct a coding scheme.