Death and dementia

Objective To investigate the last phase of dementia and the causes of death ; comparing autopsy and death certificate diagnoses. Design Prospective, 11-year, longitudinal study of behavioural and psycholo gical changes in dementia, with autopsy follow-up. Setting Participants with dementia, living at home with a carer. All lived in Oxfordshire, UK. Participants Ninety-one people with dementia (Alzheimer's disease and/or va scular dementia) who died during the course of the study. Measures At four-monthly intervals, carers were interviewed about the parti cipants' behaviour using the Present Behavioural Examination. Participants were also assessed cognitively. Causes of death were established from death certificates and, where possible, from post-mortem examination. Results The main immediate cause of death recorded at autopsy was pneumonia (57%), followed by cardiovascular disease (16%) and pulmonary embolus (14% ). This agreed with the immediate cause of death on 53% of death certificat es. Pulmonary embolism and bronchopneumonia were under-reported on death ce rtificates. Dementia was mentioned on 73% of death certificates. The dement ing illness lasted for a mean of 8.5 years with 58% dying in a debilitated state associated with severe dementia. In the period before death, 35% were unable to walk, 58% were hypophagic, 73% were incontinent of urine, 21% in continent of faeces and 76% entered an institution permanently for a mean p eriod of 18 months. Conclusion Family and professional carers of people with dementia can be gi ven some information about prognosis to help them plan for the likely outco me of dementia. Copyright (C) 2001 John Wiley & Sons, Ltd.