Year: 1987
Type: letter
Abstract: To the Editor:—We appreciate the comments of Borrie et al. on our study of age at onset and rate of progression of Alzheimer's disease.1 Measuring the rate of progression of Alzheimer's disease is problematic, and there is presently no generally accepted method of measurement. We elected to measure progression by determining change over time in the total score (the sum of all subscales) on the Blessed Dementia Scale (BDS).2 We propose that this is a valid method because the BDS is sensitive to a range of clinical manifestations of Alzheimer's disease, and because it has been demonstrated to correlate with the severity of neuropathological abnormalities observed in postmortem examinations.2 Borrie et al. suggest that the activities of daily living component of the BDS may change due to physical illnesses during the follow-up period, resulting in a change in total BDS score that may exaggerate the actual change in severity of dementia. The implication of this suggestion is that such artefactual change in BDS scores may be more common in senile-onset than in presenile-onset cases, perhaps explaining our observation of more rapid progression in senile-onset cases. This is an important concern, and in fact the ratings for some items (eg, “Inability to perform household tasks”) on the “Changes in Performance of Everyday Activities” subscale of the BDS may be influenced by physical illnesses that do not affect mental status. The instructions for the BDS, however, specify that inquiries be “directed toward changes in capacity,” and that allowances be made in scoring “for physical disabilities that would restrict activities”.2 Moreover, the majority of items are directly concerned with changes in cognitive abilities (eg, “Inability to recall recent events”), for which confounding effects of physical illnesses do not pose a problem. In our study, we recorded two subscores: one (hereafter, subscale “A”) comprising the sum of scores on the “Changes in Performance of Everyday Activities”, “Changes in Habits,” and “Changes in Personality, Interests, and Drive” subscales; the other (hereafter, subscale “B”) consisting of the “Information-Memory-Concentration Test” subscale. The mean scores at initial evaluation for the early-onset and late-onset patient subgroups were 6.9 and 6.4, respectively, for subscale A, and 17.7 and 17.6, respectively, for sub-scale B. Scores were not significantly different between groups by t-tests for either subscore of the BDS. For patients in whom follow-up testing was done, the rate of change (in score points per month) was greater in late-onset than in early-onset patients for subscale A (difference between groups 0.21 points/month; p = .03 by t-test) and subscale B (0.14 points/month), although the latter difference was not statistically significant. A repeated measures analysis of variance crossing subject group (early-onset versus late-onset) with BDS subscale (A versus B) revealed no interaction (p > .50), indicating that the difference in rate of progression between patient groups was similar whether measured by either subscale A or B. Although the difference between groups for subscale B was not statistically significant, the direction of the difference suggests more rapid deterioration in late-onset cases, which is the same conclusion that was drawn based on the total BDS score. We consider the failure to obtain statistical significance in examining subscale B to result from the decrease in statistical power that occurs in fractionating the BDS, rather than from a differential effect of other medical conditions in the two groups on subscales A and B. This interpretation is supported by the failure to find a group-by-subtest interaction using analysis of variance. A more direct approach to the question raised by Borrie et al. would be to examine presence of medical illnesses that might influence performance on ADL as covariates in analysis of change in BDS subscales. Such an analysis would be a useful refinement to our method of measuring progression of dementia. We don't consider presence of such illnesses to be a substantial confounding factor in determining BDS scores, however, for the reasons presented above. We don't, therefore, expect that such an analysis would alter our conclusions regarding age at onset and rate of progression of Alzheimer's disease.
Language: English
Subfield: Psychiatry and Mental health
Field: Medicine
Domain: Health Sciences
Open Access status: bronze
OpenAlex ID: W2030384137