RISK-FACTORS AMONG INCIDENT CASES OF DEMENTIA IN A REPRESENTATIVE BRITISH SAMPLE

In a 4-year follow up of 1042 elderly people (aged 65+) randomly sampled from the community, levels of dementia were assessed using a two-phase case-finding procedure (screening followed by clinical interview) among survivors, with additional information provided by death certificates, hospital case notes or postal questionnaires. Overall, 42 cases were clinically identified. In a case-control analysis matching for age, sex and premorbid cognitive status, risk estimates were obtained for nine self-reported factors: heart trouble; headaches; giddiness; falls; current smoking; past smoking; left-handedness; insomnia; and the use of sedative-hypnotic drugs to promote sleep. While none of these factors was associated with a significantly increased risk, the power of these case-control comparisons was limited by sample size.