A methodology for collecting outcome measures for common hospital conditions

An eight-month prospective analysis of routinely collected hospital data with 24-week follow-up of mortality and morbidity was carried out at two district general hospitals in Somerset Health District, to test the feasibility of collecting useful and valid outcome measures for two common hospital conditions, strokes and fractured hips. Data were collected on 163 consecutive admissions with a primary diagnosis of stroke (83 cases) or fractured hip (80 cases). At 24 weeks, 38 patients with stroke had died (mortality ratio 46 per cent, 95 per cent C.I. = 35.3-56.7 per cent) and 11 with fractured hip had died (mortality ratio 14 per cent, 95 per cent C.I. = 6.4-21.6 per cent). Seven patients (four with fractured hip and three with stroke) died after 24 weeks and before responding to the NHP questionnaire. One stroke patient could not be traced. Nottingham Health Profiles were received from 106 patients (41 with stroke and 65 with fractured hip). Both groups of patients had problems with physical mobility and lack of energy. Patients with fractured hips were more likely to complain of pain. It is concluded that the methods used to collect outcome measures in this study are widely applicable to other conditions, but the interpretation and comparability of the results require the collection of similar data elsewhere.