338 E
VIDENCE
-B
ASED
H
EALTH
P
RACTICE
intermediate stages of AD (Drachman, O’Donnell, Lew, & Swearer, 1990). The
relationship between disease severity and prognosis should be taken into ac-
count before discussing the differences in the rates of disease progression.
When using “survival” as the prognostic parameter for the population 75
years and older, older age, male sex, low education, comorbidity, and func-
tional disability were associated with shorter survival (Aguero-Torres,
Fratiglioni, Guo, Viitanen, & Winblad, 1998).
A population-based prospective study of persons 55 years and older was con-
ducted, and high age and low cognitive performance were the strongest predic-
tors of institutionalization and death (Ruitenberg et al., 2001). The relationship
between the presence of psychotic symptoms and dependence in both basic and
instrumental ADL were investigated in patients with AD who were referred to a
memory clinic. Dependence in ADL was associated with psychotic symptoms as
well as cognitive decline and depression, but the possible causal relationship be-
tween psychotic symptoms and dependence in ADL needs to be further investi-
gated (Tran, Bedard, Molloy, Dubois, & Lever, 2003). On the other hand, the
annual rates of change in the cognitive and functional scales have been calcu-
lated in AD patients and investigated in relation to the clinical predictors
(Swanwick, Coen, Coakley, & Lawlor, 1998). Although age and duration of symp-
toms at entry were not predictive of the rate of decline partly because of the large
variability in the progression rate, male sex was associated with a faster decline
in cognition.
Risk Factors of Mortality The number of remaining years of life and the risk fac-
tors for mortality of demented patients are the most important factors in terms of
social and economic issues. Larger numbers of older adults will require more so-
cial services and care, with an attendant increase in medical costs.
Estimates of the median survival from the clinical onset of dementia vary
from 5 to 9.3 years. However, previous studies may have underestimated the
deleterious effects of dementia because patients with rapidly progressing de-
mentia died before commencement of the studies (length bias). Recently, the
median survival from the onset of dementia symptoms, with adjustment for
length bias, was estimated in the Canadian Study of Health and Aging (Wolfson
et al., 2001). From a random sample of 10,263 subjects 65 years and older, 821
subjects with dementia, including 648 patients with AD, were followed up for 5
years. The estimated median survival was 3.3 years after the adjustment for the
length bias, even though the mean age at sampling was 83.8 years (Kawas &
Brookmeyer, 2001). In the Personnes Agees Quid prospective population-based
cohort study in France, 2,923 elderly were followed up for 8 years, and 281 per-
sons were diagnosed as having incident dementia. The median survival time of
the demented patients was 4.5 years (Helmer, Joly, Letenneur, Commenges,
& Dartigues, 2001). These studies thus showed shorter survival than in the
earlier studies.
A number of surveys were recently conducted to study the survival and risk
factors associated with dementia. An 8-year follow-up study showed that deaths
from cerebrovascular diseases and respiratory disease were more frequent in pa-
tients with dementia than in persons without dementia (Helmer et al., 2001).
With regard to late-stage AD, a 5-year follow-up study in a long-term care facility
showed that better physical health and the presence of delusions were associated