Natural history of Alzheimer's disease and other dementias : findings from a population survey

The aim of this doctoral thesis is to increase understanding of the natural history of different types of dementia in a very old population. Five studies were performed using data from the Kungsholmen Project, a population-based study on aging and dementia in Stockholm that includes subjects aged 75+. Both prevalent and incident dementia cases, as well as cross-sectional and longitudinal analyses, were used in this thesis, depending on the specific aims of each study. The main results were:

1. Memory problems were the most frequent symptoms in incipient dementia, more so in Alzheimer's disease (AD) than in vascular dementia (VaD).

2. Extrapyramidal signs (EPS) were frequent neurological findings (25% of all dementia cases and 20% of AD cases), more often among severe cases. However, longitudinal analyses did not confirm the negative prognostic value of these signs.

3. The prevalence of questionable, mild, moderate, and severe dementia in Kungsholmen, on October 1, 1987, was 1.0, 3.1, 5.5, and 2.3 per 100, respectively.

4. Fifty-five percent of the demented subjects but only 3% of the non-demented were institutionalized. Demented people living in institutions were affected mostly by moderate-severe dementia and by VaD.

5. In a very old population, dementia, independent of the presence of other chronic diseases, made the strongest contribution to both the development of long-term functional dependence and decline in function.

6. Mini-Mental State Examination (MMSE) scores decreased by 2.4 points each year in demented subjects who survived 3 years and had an initial score > 5. This cognitive decline was more rapid in AD than in dementia of other etiology. Higher initial MMSE scores and poorer functional status predicted faster cognitive decline.

7. Dementing disorders were a major risk factor for death. Even in the oldest old (85+), dementia shortens life, especially among women.

8. Fourteen per cent of all deaths could be attributed to dementia when all other factors were controlled for. Mortality risk ratios were 2.0 for AD (95% confidence interval, 1.5-2.7) and 3.4 for VaD (95% confidence interval, 2.1-5.5).

9. Shorter survival in dementia was related to male gender, advanced age, low education, comorbidity and initial poor functional status. Type, severity and duration of dementia were less important predictors.

In conclusion, our results support the hypothesis that the onset of dementia is variable. However, as memory problems are very common in early stages, when these problems present, dementia assessment is strongly recommended. EPS are not confirmed unfavorable prognostic factors. Our finding that as many demented people live at home as in institutions underscores the need for more attention by health care planners to the large requirements for domiciliary help as well as institutionalized care of demented elderly. Dementia is a long and progressive process, a major cause of functional dependence and major risk factor for death in the elderly. It is possible to identify demented subjects with a worse prognosis by using clinical and demographic data. Clinicians and health care planners should be aware of the potential usefulness of functional dependence, as measured by activities of daily living, which is a good predictor of both shorter survival and more rapid cognitive decline.