The objective of this study was to investigate how many patients who die from causes other than cancer might benefit from specialist palliative care. This was achieved by secondary analysis of data from the Regional Study of Care for the Dying, a retrospective national population-based interview survey. The investigation involved 20 self-selected English health districts, nationally representative in terms of social deprivation and most aspects of health services provision. A total of 3696 patients were randomly selected from death registrations in the last quarter of 1990; an interview concerning the patient was completed 10 months after the death by bereaved family, friends or officials. The results show that a third (243/720) of cancer patients who were admitted to hospices or had domiciliary palliative care scored at or above the median on three measures of reported symptom experience in the last year of life. That is the number of symptoms (eight or more), the number of distressing symptoms (three or more) and the number of symptoms lasting more than six months (three or more). A total of 269 out of 1605 noncancer patients (16.8%) fulfilled these criteria. On this basis, it is estimated that 71,744 people who die from nonmalignant disease in England and Wales each year may require specialist palliative care. An increase of at least 79% in caseload would, therefore, be expected if specialist palliative care services were made fully available to noncancer patients. This is a conservative estimate, as non-cancer patients were matched to only one-third of cancer patients who had specialist palliative care. It is concluded that clinicians and patient groups caring for patients with advanced nonmalignant disease must work together with specialist palliative care services and with health commissioners to develop, fund and evaluate appropriate, cost-effective services which meet patient and family needs for symptom control and psychosocial support.