Incidence of dementia in a rural community in Spain: the Girona cohort study.

Background: Information on dementia incidence in Spanish populations is still scarce, and there is a dearth of prospective studies.

Objective: To estimate the incidence rates of dementia, Alzheimer's disease (AD) and vascular dementia (VaD) in a population cohort aged 75 and over in a rural area in Spain.

Methods: A prospective population cohort study over a 5-year period in 8 rural villages in the province of Girona. The baseline study in 1990 identified 200 prevalent cases of dementia. The dementia-free cohort included 1,260 persons aged 75 and over. This was the sample used for the incidence study. We rescreened and selectively reexamined this group in 1995 using a two-phase procedure consisting of a screening interview at home using the MMSE. Diagnoses of dementia, AD and VaD were established using the Cambridge Examination for Mental Disorders of the Elderly for surviving participants. For deceased participants, we used the Retrospective Collateral Dementia Interview to establish a diagnosis of dementia and AD according to DSM-III-R diagnostic criteria.

Results: Information was obtained for 91% of the subjects at risk; 122 incident cases of dementia were identified. Incidence rates per 1,000 person-years at risk were 23.2 (95% CI = 19.1-27.3) for dementia, 10.8 (95% CI = 7.8-13.7) for AD and 9.5 (95% CI = 6.7-12.1) for VaD. All dementia subtypes showed an age-dependent pattern. Females had a relative risk of 1.8 (95% CI = 1.0-3.4) to develop AD. The inclusion of deceased cases with manifestations of dementia increased the rate of dementia incidence in 7.1 cases/1,000 person-years at risk.

Conclusion: Incidence rates were similar to those reported by other cohort studies. All dementia subtypes increased with age, but incidence rates did not increase exponentially in the oldest old. Females were at increased risk for AD. The inclusion of information about dementia symptoms from relatives of deceased participants was useful in order to avoid underestimation of the dementia incidence rates. Underestimation of the incidence rates was more important in those aged 75-84 years.