Donepezil, rivastigmine, galantamine and memantine for ...

GBS prepared by the study authors for a previous
study. Scores range from 0 to 78 and the higher
the score the greater the degree of the deficit. In
this trial, the mean change (± SEM) from baseline
was statistically significantly lower (better) in the
5 mg donepezil group than the placebo group:
–0.72 ± 0.53 donepezil, 1.84 ± 0.69 placebo,
p < 0.05.
PGA
One unpublished study 55 reports the Patient
Global Assessment scale (PGA).
[Commercial/academic confidential information
removed]
Summary: on measures of global outcomes, treatment
with donepezil generally leads to more favourable results
when compared with treatment with placebo.
Functional outcomes
Eight included trials report data on activities of
daily living (ADL) scales; each reporting on
different scales.
Unified ADL
Rogers and colleagues 53 report on the Unified
Activities of Daily Living (ADL) assessment scale
where a negative response relates to improvement.
This trial was a four-arm comparison of three
daily doses of donepezil (1, 3 and 5 mg) and
placebo. 53 Full details can be seen in Appendix 7;
however, for consistency the discussion that
follows will concentrate on the 5 mg donepezil
group versus placebo. At 12 weeks of follow-up the
mean (adjusted) change from baseline favoured
the 5 mg donepezil group compared with the
placebo group, although this difference does not
reach statistical significance (–3.1 versus 1.5,
donepezil 5 mg/day versus placebo, respectively,
p = 0.068).
PDS
Winblad and colleagues 47 report on the PDS where
a positive score indicates clinical improvement.
This trial had an intervention dose of donepezil of
5 mg/day for 28 days followed by 10 mg/day until
study completion and will be treated as having a
10 mg/day dose. Mean change from baseline for
overall activities of daily living at 52 weeks of
follow-up was –10.8 in the 10 mg donepezil group
and –15.3 in the placebo group. This difference
(4.5 points) was statistically significant at p < 0.05.
Full details of changes in individual items can be
found in Appendix 7. Individual items on the PDS
that also reached statistical significance from
placebo were self-care (p < 0.05), memory
(p < 0.01), and telephone (p < 0.01), although it
© Queen’s Printer and Controller of HMSO 2006. All rights reserved.
Health Technology Assessment 2006; Vol. 10: No. 1
is unclear whether corrections for multiple
comparisons were made to the level for statistical
significance.
IADL
In a subsequent publication of additional outcomes
from the Winblad 47 trial, data on the proportion
deteriorating on the Instrumental ADL (IADL)
scale was also reported. 56 The study reports that
overall, statistically significantly fewer participants
in the donepezil group (10 mg) deteriorated in
individual IADL items at week 52 compared with
placebo, p < 0.05. The overall proportions are not
presented, but the proportion deteriorating on
individual items can be found in Appendix 7.
ADFACS
One study assessed function with the Alzheimer’s
Disease Functional Assessment and Change Scale
(ADFACS), which they adapted for the study (see
Appendix 7). 46 This scale has a range of 0–54,
where lower scores correspond to better function.
The adjusted mean change from baseline to the
54-week endpoint was 2.4 in the donepezil and
3.85 in the placebo group. This difference (1.45
points) was statistically significant, showing better
function in the donepezil-treated group (p < 0.001).
IDDD
A modified Interview for Deterioration in Daily
Living in Dementia (IDDD) scale was used in one
trial (see Appendix 7). 50 On this scale, a value of
68 was the boundary between clinical
improvement or decline, where a value 68
clinical decline. In this 24-week study, mean
change (± SEM) in baseline scores on the IDDD
were 70.4 ± 0.4 in the 5 mg/day donepezil group,
69.4 ± 0.4 in the 10 mg/day donepezil group and
71.1 ± 0.4 in the placebo groups. The difference
between the 10 mg/day donepezil group and
placebo was statistically significant (p < 0.01).
CMCS
One 24-week study included the Caregiver-rated
Modified Crichton Scale (CMCS), which is a
modified Crichton Geriatric Rating Scale
(CGRS). 44 On this scale, a minus change score
relates to clinical improvement. The mean change
± SEM from baseline was 1.03 ± 0.66 in the
donepezil-treated group (5 mg/day) compared with
3.45 ± 0.71 in the placebo-treated group. This
shows that both groups deteriorated but the degree
of deterioration was less in the donepezil group
than the placebo group. The difference between
the groups was statistically significant (p = 0.01).
Full details can be found in Appendix 7.
31