A rating scale to measure anxiety in dementia sufferers was developed and e
valuated in a sample of 51 inpatients and 32 day-hospital patients. Anxiety
scores were not related to sex, age, accommodation or DSM-IV diagnosis of
the type of dementia. However, both subjects with physical illnesses and su
bjects with insight into their memory problems had significantly higher anx
iety scores. The kappa values for inter-rater reliability ranged from 0.51
to I and for test-retest reliability from 0.53 to 1, which indicates modera
te to good reliability. The overall agreement on individual items ranged fr
om 82-100% (inter-rater) and 84-100% (test-retest). The professionals worki
ng in the care of the elderly and carer groups felt that the scale was comp
rehensive and all the items in the scale were important, thereby confirming
that it has good content validity. The scale significantly correlated with
other anxiety scales and also with independent ratings both by a consultan
t psychiatrist and also nursing staff, indicating good concurrent validity.
Anxiety scores were significantly higher in dementia patients who fulfille
d modified DSM-IV criteria for anxiety and clinical diagnosis of anxiety di
sorder. This showed evidence of good criterion validity. Factor analysis sh
owed five factors, including all items of the scale. Scores of 11 and above
on the scale indicated significant clinical anxiety. Overall, the scale ha
d good reliability and validity. It should be a useful clinical and researc
h instrument for assessing anxiety in dementia sufferers.