Journal of Alzheimer’s Disease 25 (2011) 679–694 679

DOI 10.3233/JAD-2011-100999
IOS Press

Effects of a Newly Developed Cognitive

Intervention in Amnestic Mild Cognitive
Impairment and Mild Alzheimer’s Disease:
A Pilot Study
Verena C. Buscherta,∗ , Uwe Friesec , Stefan J. Teipeld , Philine Schneidera , Wibke Merenskya ,
Dan Rujescua , Hans-Jürgen Möllera , Harald Hampele and Katharina Buergera,f
a Department of Psychiatry Ludwig-Maximilians University, Munich, Germany
b Department of Psychiatry, University Rostock, Rostock, Germany
c Institute of Psychology, University of Osnabrueck, Osnabrueck, Germany
d DZNE, German Center for Neurodegenerative Disorders, Rostock, Germany
e Department of Psychiatry, Psychosomatic Medicine & Psychotherapy, Goethe University,

Frankfurt/Main, Germany
f Institute for Stroke and Dementia Research, Klinikum Groβhadern, Ludwig-Maximilians

University Munich, Munich, Germany

Accepted 11 February 2011

Abstract. Recent studies have shown that patients with Alzheimer’s disease (AD) and its possible prodromal stage mild cog-
nitive impairment (MCI) benefit from cognitive interventions. Few studies so far have used an active control condition and
determined effects in different stages of disease. We evaluated a newly developed 6-month group-based multicomponent cogni-
tive intervention in a randomized controlled pilot study on subjects with amnestic mild cognitive impairment (aMCI) and mild
AD patients. Forty-three subjects with aMCI and mild AD were recruited. Primary outcome measures were change in global
cognitive function as determined by the Alzheimer’s Disease Assessment Scale-cognitive subscale (ADAS-cog) and the Mini
Mental Status Examination (MMSE). Secondary outcomes were specific cognitive and psychopathological ratings. Thirty-nine
patients were randomized to intervention groups (IGs: 12 aMCI, 8 AD) and active control groups (CGs: 12 aMCI, 7 AD). At
the end of the study, we found significant improvements in the IGMCI compared to the CGMCI in the ADAS-cog (p = 0.02) and
for the secondary endpoint Montgomery Asberg Depression Rating Scale (MADRS) (p < 0.01) Effects on the MMSE score
showed a non-significant trend (p = 0.07). In AD patients, we found no significant effect of intervention on the primary outcome
measures. In conclusion, these results suggest that participation in a 6-month cognitive intervention can improve cognitive and
non-cognitive functions in aMCI subjects. In contrast, AD patients showed no significant benefit from intervention. The findings
in this small sample support the use of the intervention in larger scales studies with extended follow-up period to determine
long-term effects.

Keywords: Alzheimer’s disease, cognitive intervention, cognitive stimulation, cognitive training, mild cognitive impairment,
stage-specific

∗ Correspondence to: Verena Buschert, Dementia Research Sec-

INTRODUCTION
tion and Memory Clinic, Alzheimer Memorial Center and
Geriatric Psychiatry Branch, Department of Psychiatry Ludwig-
The NICE-guidelines (The National Institute for
Maximilians University, Nussbaumstrasse 7, D-80336 Munich,
Germany. Tel.: +49 89 5160 5862; Fax: +49 89 5160 5865; E-mail: Health and Clinical Excellence, http://www.nice.org.
verena.buschert@med.uni-muenchen.de. uk) as well professional societies in the field of

ISSN 1387-2877/11/$27.50 © 2011 – IOS Press and the authors. All rights reserved
680 V.C. Buschert et al. / Cognitive Intervention in MCI and AD
neurology and psychiatry (e.g., the German S3 guide-
lines on dementia, http://www.dgn.org/images/stories/
dgn/pdf/s3 leitlinie demenzen.pdf) recommend par-
ticipation in non-pharmacologic cognitive stimulating
programs for patients with mild-to-moderate AD. Pre-
vious studies in healthy subjects showed cognitive
enhancement and a reduced risk of future cognitive
decline after participating in cognitive training [1]. A
recent meta-analysis covering more than 29,000 indi-
viduals found an odds ratio (OR) of 0.54 for dementia
in individuals with a high compared to a low “behav-
ioral brain reserve”, indicating a risk reduction of 46%
[2, 3]. Mentally stimulating activities were among the
most robust factors, even after excluding effects of
education, age, occupation, and other potential con-
founds. This effect was sustained over an average
longitudinal follow-up of 7 years [2]. Because “behav-
ioral brain reserve”, also referred to as “cognitive
reserve” [4], seems to modulate the clinical expres-
sion of AD pathology, direct modulation of behavioral
brain reserve using cognitive interventions might help
to delay the onset and progression of dementia in
AD. Furthermore, cognition-based interventions have
been found to attenuate cognitive decline in amnes-
tic mild cognitive impairment (aMCI) subjects [5–7],
which is a transitional stage between normal cogni-
tive aging and AD [8], and in mild-to-moderate stages
of dementia [9–14]. These studies and meta-analyses
on cognitive interventions in AD [15, 16] provided
first evidence of efficacy showing (a) improved global
cognitive functioning, (b) improved abilities of daily
living, (c) reduced behavioral disturbances, and (d)
positive effects on the quality of life in MCI subjects
and mild AD patients [17]. The findings are promising
but challenges remain for translation to daily routine
because most studies employed relatively small sam-
ples and did not focus on specific stages of disease
[11, 14, 18, 19]. A further limitation is that only a few
studies used active control conditions [14, 18, 19] and
the cognitive interventions differed between studies
so that a comparison of treatment effects is challeng-
ing. Therefore, it is desirable to clarify the nature of
related but distinct forms of cognitive interventions and
their target population. A first distinction can be made
between cognitive stimulation and cognitive training
programs. Cognitive stimulation comprises engage-
ment of the participant in a range of activities aimed at
general enhancement of cognitive and social function-
ing in a non-specific manner [20]. Potential benefits
of this approach have been demonstrated primarily in
people with advanced dementia [21]. Positive effects
were also found in patients with mild-to-moderate
AD [15], where cognitive stimulation was even more
effective than training of specific cognitive functions
[21–23]. In contrast, cognitive training involves guided
practice of a set of standard tasks designed to target
specific cognitive functions such as attention, memory,
and problem-solving as well as teaching and practic-
ing mnemonic strategies and skills (e.g., restorative
and compensatory mnemonic techniques) [7, 24]. This
approach has been shown to be high effective in people
with MCI [25]. However, classification of the type of
intervention according to these definitions is often dif-
ficult because many studies employ a blend of different
intervention strategies that include aspects of cognitive
training and cognitive stimulation [20].
Clinical, neuropsychological, neuroimaging, and
neuropathological data indicate that the progression
of AD neuropathology through the brain replicates the
ontogenetic brain maturation in reverse order [26]. This
theory of retrogenesis provides the theoretical under-
pinnings of the global deterioration assessment scale
[27], which predicts a stage specific sequence of clin-
ical and functional decline in the course of AD. Given
the specific clinical decline sequence in AD patients,
we hypothesize that to obtain best intervention effec-
tivity, one needs to match cognitive interventions with
the cognitive and global functioning level of the AD
patient [28]. Otherwise, the intervention is at risk of
overstraining or underchallenging the participants and
may lead to frustration and adverse effects [11, 29].
Therefore, we developed a stage-specific multicom-
ponent cognitive intervention for subjects with aMCI
and mild AD. According to the concept of retrogenesis,
the more impaired the participants are, the less use-
ful is training of (specific) cognitive functions, and the
more meaningful and probably successful are cognitive
stimulation and activation of everyday functions and
activities. We studied the effects of this newly devel-
oped intervention program in subjects with aMCI, who
are at risk to develop AD [30], and patients fulfilling
criteria for mild stages of clinically probable AD [31,
32] in a 6-month randomized controlled trial.
MATERIALS AND METHODS
Participants
A total of 43 subjects (27 aMCI, 16 mild AD)
were recruited at the Dementia Research Section and
Memory Clinic of the Alzheimer Memorial Cen-
ter and Geriatric Psychiatry Branch, Department of
Psychiatry, Ludwig-Maximilians University, Munich.
 V.C. Buschert et al. / Cognitive Intervention in MCI and AD
A comprehensive clinical and neuropsychological
assessment was conducted and global functioning was
assessed by the Global Deterioration Scale (GDS) [27].
Subjects were included if
• aMCI subjects met Petersen’s criteria for amnestic
single domain MCI or amnestic multiple domain
MCI [30] (GDS 3), AD subjects met the DSM-
IV/NINCDS-ADRDA criteria for probable mild
AD [31, 32] (GDS 4),
• aMCI subjects scored at least 23 points, AD
subjects at least 19 on the Mini Mental State
Examination (MMSE) [33],
• subjects were at least 50 years old,
• subjects were able to hear and see well enough to
participate in the group activities,
• subjects did not have any major physical illness,
other mental disorder (e.g., major depression) or
disability which could affect participation,
• subjects were, if treated, on stable medication
of antidementia, antidepressant or antipsychotic
drugs for at least three months prior to the start of
the study,
• subjects did not participate in cognitive interven-
tion at least one year prior to the start of the study,
• subjects provided written, informed consent.
With regard to Petersen’s MCI-criterion of normal
generalized cognition, aMCI subjects with a MMSE
score lower than 25 points were only included if
their mild cognitive impairments did not interfere sig-
nificantly with daily activities or social functioning
verified by a caregiver. The study was performed in
accordance with the Declaration of Helsinki and the
Ethics Committee of the Medical Faculty of Ludwig-
Maximilians University approved the study. Patients
participating in the study gave written informed
consent. The study was officially registered before
first subject joined study (www.clinical.trial.gov; ID:
NCT00544856).
Design
Of the 43 screened subjects, four dropped out prior
to the randomization procedure (1 concurrent illness, 3
lacking motivation to participate in the intervention).
39 persons (24 with aMCI, 15 with AD) were ran-
domly assigned to cognitive intervention groups (IGs:
12 aMCI, 8 AD), receiving cognitive intervention, and
active control groups (CGs: aMCI: 12, AD: 7), in which
participants met monthly and received paper-pencil
exercises for self-study. To achieve matched treatment
groups (IG, CG) a blocked randomization procedure
681
was used. First of all, subjects were pooled in pairs with
respect to age, gender, education, and ApoE genotype
(in decreasing degree of priority). A study-independent
person then randomly assigned pairs to the intervention
or control arm by using a computerized random num-
ber generator. To ensure optimal conditions of work,
we considered group sizes of 10–12 aMCI patients
and 8–10 mild AD patients to be reasonable upper
limits.
There was one baseline test conducted between 1 to
4 weeks before the start of the intervention, and one test
administered between 1 to 4 weeks following the end
of the intervention. The tests were conducted in a fixed
order using a standardized case report form (CRF).
The raters who conducted the testing sessions (ini-
tials: WM, Daniela Otto, see acknowledgement) were
blinded to treatment and different from the instruc-
tor who administered the interventions (initials: VCB).
Participants of the control condition crossed over to the
intervention condition after 6 months at the end of the
study.
Intervention condition
The objective of this project was to conceptualize
and evaluate a multicomponent cognitive intervention
to preserve, reactivate, and promote cognitive and non-
cognitive functions in consideration of stage-specific
needs and capabilities of subjects with aMCI and
patients with mild AD. The program builds on the
theory of cognitive reserve [4] and the theory of ret-
rogenesis as basis for intervention task selection [26],
and employs already established methods and inter-
vention principles (e.g., the face-name association [34]
or the errorless learning approach [35]). The inter-
vention program for subjects with aMCI focused on
cognitive training of specific domains such as mem-
ory function but also included features of cognitive
stimulation, whereas that for mild AD patients empha-
sized cognitive stimulation and less cognitive training.
Table 1 gives an overview of specific features of the
stage-specific cognitive intervention and the control
condition.
We provided information on meta-cognition (e.g.,
about information processing), age-associated cogni-
tive decline and cognitive changes with AD to enhance
motivation of participants [7, 36, 37]. Additionally,
caregivers were encouraged (contacted by letter) to
take an active interest in their patient’s participation
in the study, e.g., by talking about the contents of the
weekly sessions, assisting in the exercises for self-
study, and applying specific techniques in everyday
682 V.C. Buschert et al. / Cognitive Intervention in MCI and AD

Table 1
Overview of specific features of cognitive intervention and control condition
Target entity
Intervention group (IG) Control group (CG)
aMCI mild AD aMCI mild AD
Treatment Stage-specific cognitive intervention, weekly sessions Paper-pencil exercises for self-study, monthly meetings
Focus Cognitive training/ Cognitive stimulation/ Specific cognitive function: sustained attention
Cognitive stimulation Cognitive training
Expected effects Generalized effects in global cognitive function No generalized effects in global cognitive function
Format Group (max. 12 persons) Group (max. 10 persons) Self-study; meetings in group
Duration 6 month, 20 sessions à 120 min 6 month, 6 meetings à 60 min
aMCI = amnestic mild cognitive impairment; AD = mild AD; IG = intervention group; CG = control group.

situations. Non-cognitive functions such as mood and
quality of life are deemed to be crucial for cognitive
performance [7, 25, 38, 39]. To support cognitive and
additionally non-cognitive functions, we implemented
components such as reminiscence, psychomotor and
recreational tasks and conducted the programs in
groups (aMCI: max. 12 participants, AD: max. 10 par-
ticipants) to encourage social interaction [7, 11, 14].
According to the core symptom of AD, however, the
focus lied primarily on cognitive domains. The pro-
gram was designed to involve a wide range of cognitive
functions and was expected to exert generalized effects
on global cognitive functioning.
To detail the process through which the multicom-
ponent intervention leads to a particular outcome, and
to determine a measurement strategy that enables the
researchers to assess key components of this process,
we adapted and applied a methodology for describ-
ing and decomposing multidimensional psychosocial
and behavioral interventions [40]. The aim was to
characterize the intervention across a common set of
dimensions based on the proximal goals and activities
of the intervention. Fig. 1 shows goals, activities and
measurement instruments of the program according to
Czaja et al. [40].
Prior to the start of the study the cognitive inter-
vention for subjects with aMCI and patients with
mild AD was conceptualized and detailed as a man-
ual with reproducible detailed protocols. The program
particularly aims at global cognitive functioning, but
additionally addresses non-cognitive domains (e.g.,
mood) that are supposed to be also impaired in
AD and aMCI. Each program consists of 20 units,
applied weekly for 120 minutes. Units alternate weekly
between theory based topics such as:
• specific cognitive functions following a model of
information processing (e.g., attention, percep-
tion, and memory),
• teaching and practicing restorative and compen-
satory mnemonic techniques (e.g., visual imagery
[41], face-name association [34], use of external
memory aids such as calendar, notes and prompts
[42] and
• activation of everyday life activities in house and
garden and intellectual and physical leisure activ-
ities,
and topics of common interest such as season, music,
and food. Sessions with topics of common interest
addressed the natural process of reminiscence but had
an additional focus on the present situation. Mul-
tisensory stimulation was introduced when possible
[11]. Approved memory strategies and principles such
as face-name association and the errorless learning
approach were consistently employed. According to
the theory of retrogenesis, units are oriented to the stage
of disease and differ concerning difficulty and com-
plexity as well as number and duration of trials and
exercises. Differences between the two intervention
programs (aMCI/AD) correspond in fact to a quanti-
tative rather than to a qualitative distinction. However,
according to different cognitive and functional sta-
tus in MCI and AD, some topics are only covered in
the MCI intervention, others only in the AD interven-
tion. Table 2 details similarities and differences of both
interventions with regard to content, target groups, and
stage-specific topics of units. Note that the design of
our study is incomplete. A complete design would have
included AD patients participating in the interventions
designed for MCI patients and vice versa. In addition to
ethical concerns, we think that participants being sys-
tematically overstrained in one group and participants
being not challenged in the other would have distorted
the results.
Participants who were present at more than half of
the interventions (minimum 11 sessions), were judged
as compliant and included in the study.
 V.C. Buschert et al. / Cognitive Intervention in MCI and AD 683

Stage-specific cognitive intervention

for subjects with aMCI and patients with mild AD

Secondary goal 1 Secondary goal 2

Primary goal
Improvement and stabilisation of specific Improvement and stabilisation of specific
Improvement and stabilisation of global
cognitive functions ((higher) attentional non-cognitive functions (mood, quality of life)
cognitive functioning
and executive functions, memory)

Activities* Activities* Activities*

exercises concerning specific
cognitive functions, e.g. memory,
attentional and executive functions
teaching and practising of restaurative
and compensatory mnemonic techniques,
e.g. visual imagery, face-name association
use of approved strategies/principles,
e.g. errorless learning
activation of allday activities, e.g. leisure
discussions
psycho-motor/ recreational exercises
detailed exercises for self-study at
home
exercises and tasks concerning (short-
term) memory
exercises and tasks concerning
(higher) attention, problem-solving
teaching and practising of restaurative
and compensatory mnemonic techniques,
e.g. visual imagery, face-name association
use of approved strategies/principles,
e.g. errorless learning
detailed exercises for self-study at
home
roundtable and discussions
exercises to stimulate social interaction,
e.g. remote memory, reminiscence,
language, imagination & creativity
psycho-motor/ recreational exercises
education about age-associated/
pathological changes of brain functions
(meta-cognition)
oral tasks performed at home

Measuring instruments Measuring instruments Measuring instruments

ADAS-cog TMT B MADRS
MMSE RBANS story memory, story recall QoL-AD

*According to the theory of Retrogenesis activities focus on the stage of disease and differ regarding content, difficulty and complexity as well as number and duration of
administered trials and exercises
ADAS-cog = Alzheimer‘s Disease Assessment Scale – cognitive subscale; MMSE = Mini Mental State Examination; RBANS = Repeatable Battery f or the Assessment of
Neuropsychological Status; TMT = Trail Making Test; MADRS = Montgomery and Asberg Depression Rating Scale; QoL-AD = Quality of Life Alzheimer‘ s Disease Scale;

Fig. 1. Goals, activities and measuring instruments of the cognitive intervention.

Table 2
Description and differences of intervention according to functional domains and target entities
Functional domain Target group
aMCI Mild AD
Cognition
Cognitive training (memory
functions)*
Cognitive stimulation**
Meta-Cognition
Activation of all day activities
Mood
(Social-) Behavior
(Psycho-) Motor
Different stage-specific topics Information processing Memory: Everyday life: nutrition
Reminding written information Everyday life: Body & Health
aMCI = amnestic mild cognitive impairment; mild AD = mild Alzheimer’s Disease; *Cognitive training: teaching and
practicing theoretically motivated strategies and skills in order to optimize cognitive functioning; **Cognitive stimulation:
engagement in a range of activities and discussions aimed at general enhancement of cognitive and social functioning;
strong; middle; weak.
684 V.C. Buschert et al. / Cognitive Intervention in MCI and AD
In brief, a typical session included
• a welcome with a short roundtable repeating the
contents of the previous session (10 minutes),
• a warm-up-exercise (10 minutes),
• oral and/or paper-pencil exercises alone, with a
partner or within a small group (40 minutes) inter-
mitted by a small break (15 minutes),
• a psychomotor (e.g., playing with balloons)
and/or a recreational exercise with music (10 min-
utes),
• exercises targeting primarily social interaction
and mood (25 minutes) and
• a conclusion and good-bye at which participants
received both oral (incentive-cards) and paper-
pencil exercises for self-study (10 minutes).
Control condition
To control for effects of an intervention per se,
such as beneficial effects of social interactions, we
devised an active control condition. Participants of
the control group met monthly (6 times in total) and
received paper-pencil exercises for self-study focus-
ing primarily on isolated cognitive functions such
as sustained attention, which is thought to be rela-
tively preserved at least in early stages of AD [43]
(see Table 1). We assumed that these kinds of tasks
would be of no or only little benefit for improving
global cognitive performance in aMCI and AD [44,
45]. As with the intervention condition, paper-pencil
exercises of the control condition were oriented to the
patient group (aMCI, AD) and differed concerning dif-
ficulty, complexity, and number of exercises. In the
course of monthly group-meetings, solutions of the
previous at-home exercises were provided. Comple-
tion of paper-pencil exercises for self-study was not
monitored in order to minimize personal interaction
with the control subjects. The intervention and con-
trol conditions were conducted by the same instructor.
Participants who took part in at least half of the con-
trol sessions (minimum 3 sessions) were judged as
compliant and included in the study.
Assessment measures
Primary outcome measures
Main outcome measure was change in cognitive
function determined by the MMSE [33], a brief,
widely used test of cognitive functions (increasing
scores indicate improvement), and the total score of
the ADAS-cog, version B (pre-testing) and C (post-
testing) [46], which is a more sensitive scale measuring
cognitive function and including more items that
assess short-term memory (decreasing scores indicate
improvement). The ADAS-cog is frequently used in
AD trials as the principal cognitive measure. Both
raters (WM, DO) who administered the testing at base-
line and end-of-study were blinded to the treatment
condition. WM conducted all tests and scales with the
patients, and DO interviewed the caregivers.
Secondary outcome measures
Secondary outcome measures were change in spe-
cific cognitive and non-cognitive functions. Immediate
and delayed memory (story memory and story recall)
from the Repeatable Battery for the Assessment
of Neuropsychological Status (RBANS), version A
(pre-testing) and B (post-testing) [47] covered spe-
cific memory functions (increasing scores indicate
improvement). To determine intervention effects on
higher attentional and additionally executive function,
Trail Making Test (TMT), Part A and B, commonly
used measurements of attentional functions [48], were
administered (decreasing scores indicate improve-
ment). As a non-cognitive domain, mood was assessed
by the Montgomery Asberg Depression Rating Scale
(MADRS) [49], a ten-item diagnostic questionnaire,
which is sensitive to the changes caused by antidepres-
sants and other forms of treatment (decreasing scores
indicate improvement). Assuming an overlap of cogni-
tive symptoms between “cognitive-inhibition” factors
of the MADRS, such as concentration difficulties [50],
and cognitive impairments due to clinical and pre-
clinical AD [51] without diagnosis of depression, we
excluded item 6 (“concentrations difficulties”) of the
MADRS from analysis [52]. Quality of life of partici-
pants was assessed by the Quality of Life-Alzheimer’s
Disease (QoL-AD) scale [53] (increasing scores indi-
cate improvement).
Analysis
Data were entered into the Statistical Package for
the Social Sciences (SPSS), version 15.0 for Win-
dows. We conducted a series of analyses of covariance
(ANCOVA) with treatment (intervention group, IG,
versus control group, CG), and progression (baseline
versus end-of-study) as independent variables. Primary
and secondary outcome variables served as depen-
dent measures. Furthermore, we entered educational
level (years of schooling) and patients’ age as covari-
ates into the analysis. We report effect sizes as partial
eta squared (η2 ) values. When significant interaction
effects between the treatment factor and the progres-
 V.C. Buschert et al. / Cognitive Intervention in MCI and AD
sion factor indicated that the change of the outcome
measure over time was different for IG and CG, we
used post-hoc t-tests to determine differences between
baseline and end-of-study. Unless otherwise noticed,
we report statistical effects associated with p < 0.05 as
significant, and additionally report effects with p < 0.1
on an exploratory basis.
RESULTS
Feasibility and acceptance of program and study
The randomized controlled trial was conducted from
September 2007 to February 2008. Participants of the
control group attended the cognitive intervention after
the end of the study. The average attendance of the
cognitive intervention was 16.7 (83.5%) out of 20
sessions. Participants with aMCI attended 17.1 times
(85.5%, range of scores: 14–20). For organizational
reasons, two sessions of the AD intervention had to be
administered in one week. Because most of the par-
ticipants missed the non-standard scheduled session,
participants with AD attended only 16.3 times out of
19 sessions (85.8%, range of scores: 13–19). One par-
ticipant of the aMCI-IG group dropped out of the study
because of intercurrent illness and one was excluded
from the study on the basis of not reaching our com-
pliance criteria, which corresponds to a 5.1% drop-out
rate.
Fig. 2. Profile of the trial and attrition. aMCI = amnestic mild cognitive impairment; AD = Alzheimer’s Disease; IG = intervention group; CG
= control group.
685
The average attendance of the control condition (6
meetings) was 5.6 (aMCI: 5.5, AD: 5.7). None of the
participants of the CGs were excluded from the study.
Sample characteristics
Forty-three persons passed the screening. One AD
patient dropped out due to concurrent physical illness,
and three aMCI-subjects withdrew consent because
they found participation in the study too demanding.
Thirty-nine subjects were randomized into interven-
tion groups and control groups. Two participants of
the aMCI-IG dropped out of the study because of inter-
current illness and lack of compliance. A total of 37
subjects finally completed the study and were assessed
at the end of the study. Fig. 2 shows the profile of the
trial and attrition.
On average, participants were 73.1 years (SD = 7.7)
old. There was a tendency for AD patients to be
older than aMCI subjects, 75.9 years (SD = 8.1) ver-
sus 71.2 years (SD = 7.7), as revealed by a 2-factorial
syndrome (aMCI versus AD) by treatment (IG versus
CG) analysis of variance (ANOVA) (main effect syn-
drome: F1,33 = 3.08, p = 0.09). Neither the main effect
of treatment nor the interaction effect were signifi-
cant (all p-values > 0.1). An analogue ANOVA showed
that educational level did not differ between interven-
tion and control groups (all p-values > 0.1). Sample
and groups were matched for gender (20 female,
19 male). All AD patients and one subject of the
686 V.C. Buschert et al. / Cognitive Intervention in MCI and AD

Table 3
Demographic and clinical characteristics of participants at baseline assessment (n = 39)
Variable aMCI mild AD
IG CG Total IG CG Total
Number 12 12 24 8 7 15
Age 71.8 (8.6) 70.7 (5.7) 71.2 (7.0) 77.3 (7.6) 74.2 (9.0) 75.9 (8.1)
min/max 53.2–86.5 62.4–79.3 53.2–86.5 66.9–88.6 57.5–83.7 57.5–88.6
Gender f/m 6/6 6/6 12/12 4/4 4/3 8/7
Education 12.3 (3.6) 13.3 (2.2) 12.9 (2.9) 11.8 (3.2) 13.6 (6.5) 12.6 (4.9)
min/max 8–18.5 11–16.5 8–18.5 8–17 4–21.5 4–21.5
Antidementiva 1 none 1 8 7 15
MMSE 28.1 (1.5) 26.8 (1.5) 27.4 (1.6) 24.5 (1.6) 25.3 (1.5) 24.9 (1.6)
min/max 25–30 24–29 24–30 22–27 23–27 22–27
ADAS-cog 8.7 (2.9) 9.8 (4.3) 9.3 (3.7) 12.1 (5.3) 16.4 (4.8) 14.1 (5.4)
min/max 3–13 5–19 3–19 6–21 13–23 6–23
aMCI = amnestic mild cognitive impairment; AD = Alzheimer’s Disease; IG = intervention group; CG = control
group; MMSE = Mini Mental Status Examination; ADAS-cog = Alzheimer’s Disease Assessment Scale; standard
deviation in ( ); min/max = ranges of scores.

aMCI-IG were on stable medication of antidemen-
tia drugs (Acetylcholinesteraseinhibitors (AChE-Is)
and/or Memantine). Table 3 shows demographic and
clinical characteristics as well as initial MMSE and
ADAS-cog scores of all participants at baseline assess-
ment. For aMCI patients, the difference in MMSE
scores between IG and CG at baseline was associ-
ated with t22 = 2.6, p = 0.02; ADAS-cog scores did not
differ (t22 < 1). For AD patients in IG and CG, MMSE-
scores and ADAS-cog did not differ at baseline (all
p-values > 0.1). Three aMCI subjects with MMSE-
sores lower than 25 points (MMSE 23 = 0, MMSE
24 = 2, MMSE 25 = 1) were included as the cognitive
impairments did not interfere with activities of daily
living and social functioning explicitly verified by a
caregiver.
TMT-B (F1,18 = 3.5, p = 0.08, η2 = 0.16). Moreover,
main effects of treatment were found for MMSE
(F1,18 = 8.5, p < 0.01, η2 = 0.23), RBANS-story mem-
ory (F1,18 = 12.5, p < 0.01, η2 = 0.41), and RBANS-
story recall (F1,18 = 9.9, p < 0.01, η2 = 0.36).
For the AD patients, the interactions between treat-
ment and progression showed a tendency towards sig-
nificance for TMT-B (F1,11 = 5.1, p = 0.05, η2 = 0.32)
and MADRS (F1,11 = 3.4, p = 0.09, η2 = 0.24) (see
Table 4 b for details). For RBANS-story recall
the covariate education was close to significance
(F1,11 = 4.1, p = 0.07, η2 = 0.27). The ANCOVA with
TMT-B showed tendencies for an interaction between
progression and education (F1,11 = 5.1, p = 0.05,
η2 = 0.32) and for a main effect of age (F1,11 = 4.6,
p = 0.06, η2 = 0.29).

Effects of treatment Post-hoc tests

ANCOVAS were conducted for AD and aMCI
separately with treatment (IG versus CG), and pro-
gression (baseline versus end-of-study) as independent
variables; participants’ age and educational level
were treated as covariates. Dependent measures were
MMSE, ADAS-cog, RBANS story memory and story
recall, TMT-A and B, MADRS (without item 6 “con-
centration difficulties), and QoL-AD.
For the aMCI subjects, we found significant
interactions between treatment and progression
for ADAS-cog (F1,18 = 6.2, p = 0.02, η2 = 0.26)
and MADRS (F1,18 = 8.8, p < 0.01, η2 = 0.33) (see
Table 4 a for details). Furthermore, the interac-
tion was associated with a p-value of p < 0.1 for
MMSE (F1,18 = 3.8, p = 0.07, η2 = 0.17), RBANS-
story memory (F1,18 = 3.4, p = 0.08, η2 = 0.16), and
We used two-sided pairwise t-tests to additionally
evaluate differences between baseline and end-of-
study. Owing to the explorative nature of these tests,
we did not correct the significance level for multiple
comparisons. For aMCI subjects in the IG, there was a
significant improvement for TMT-B (t9 = 3.0, p = 0.01)
and MADRS (t9 = 4.0, p < 0.01). aMCI subjects in the
CG showed significant decline in ADAS-cog (t11 = 2.8,
p = 0.02) and MMSE (t11 = 3.1, p = 0.01) (see Table 4 a
for details). In the AD groups, only MADRS scores
of the IG improved significantly between baseline and
end of the study (t7 = 2.6, p = 0.03) (see 4 b for details).
Number needed to treat
A number needed to treat (NNT) analysis was
conducted for an assessment of the effectiveness of
the intervention in aMCI. It revealed, that 5 MCI-
 V.C. Buschert et al. / Cognitive Intervention in MCI and AD 687

Table 4a
Change from baseline in subgroups of aMCI (n = 22 in measures of efficacy regarding MMSE, ADAS-cog., RBANS story memory and recall,
TMT-A and B, MADRS, and QoL-AD); standard deviation in ( ). Reported are all effects associated with p < 0.1
Outcome Point IGMCI n = 10 CGMCI n = 12 ANCOVA with **η2
ADAS-cog Baseline 8.7 (2.9) 9.8 (4.3) interaction treatment by progression
(F1,18 = 6.2, p = 0.02, η2 = 0.26)
End-of-study 7.3 (3.1) 11.7 (5.6)
t-test* n.s. t(11) = 2.8, p = .02
MMSE Baseline 28.1 (1.5) 26.8 (1.5) interaction treatment by progression
(F1,18 = 3.8, p = 0.07, η2 = 0.17) main
effect treatment (F1,18 = 8.5, p < 0.01,
η2 = 0.32)
End-of-study 28.2 (1.2) 26.0 (1.3)
t-test* n.s. t(11) = 3.1, p = .01
RBANS – Baseline 15.4 (2.8) 13.1 (2, 8) interaction treatment by progression
story memory (F1,18 = 3.4, p = 0.08, η2 = 0.16) main
effect treatment (F1,18 = 12.5, p < 0.01,
η2 = 0.41)
End-of-study 16.4 (3.2) 11.7 (4.2)
t-test* n.s. n.s.
RBANS – Baseline 7.5 (2.4) 4.7 (2.9) main effect treatment (F1,18 = 9.9, p < 0.01,
story recall η2 = 0.36)
End-of-study 7.5 (3.2) 3.4 (3.0)
t-test* n.s. n.s.
TMT-A Baseline 46.1 (27.2) 50.7 (19.8) n.s.
End-of-study 36.5 (9.7) 40.9 (15.1)
t-test* n.s. n.s.
TMT-B Baseline 130. 4 (46.1) 131.4 (34.0) interaction treatment by progression
(F1,18 = 3.5, p = 0.08, η2 = 0.16)
End-of-study 102.0 (28.4) 140.1 (45.0)
t-test* t(9) = 3.0, p = 0.01 n.s.
MADRS Baseline 3.6 (2.6) 3.7 (5.9) interaction treatment by progression
(F1,18 = 8.8, p < 0.01, η2 = 0.33)
End-of-study 0.7 (1.3) 3.8 (6.1)
t-test* t(9) = 4.0, p < 0.01 n.s.
QoL-AD Baseline 35.3 (3.8) 36.6 (5.0) n.s.
End-of-study 35.7 (5.1) 34.7 (5.5)
t-test* n.s. n.s.
ADAS-cog = Alzheimer’s Disease Assessment Scale; MMSE = Mini Mental State Examination; RBANS = Repeatable Battery for the Assessment
of Neuropsychological Status; TMT = Trail Making Test; MADRS = Montgomery Asberg Depression Rating Scale (without item 6 “concentration
difficulties”); QoL-AD = Quality of Life-Alzheimer’s Disease; IG = intervention group; CG = control group; ns = not significant; aMCI =
amnestic mild cognitive impairment; AD = mild AD; ANCOVA = univariate analysis of covariance; columns indicate significant effects of 2 × 2
ANCOVAs (treatment × progression; see text for details).; *t-Test results refer to post-hoc comparisons testing for the effect of progression
(baseline vs. end-of-study) within subgroups (IG, CG).; **η2 = partial eta squared.

participants need to be treated for one to benefit
compared to controls when calculating a decrease of
4 or more points on the ADAS-cog total score as an
improvement (lower scores indicating improvement)
and an increase of 3 points or over as adverse [11]).
DISCUSSION
In our randomized controlled trial, we evaluated
the effects of a multicomponent stage-specific cogni-
tive intervention for subjects with aMCI and patients
with mild AD compared to an active control condi-
tion. We found improvement in global cognitive status,
and specific cognitive and non-cognitive functions.
Only aMCI subjects showed significant effects of inter-
vention in the primary cognitive outcome variables,
whereas effects in AD patients were limited to sec-
ondary endpoints.
Feasibility and acceptance of program and study
The cognitive intervention program and the study-
design were feasible and well accepted by the
participants. We achieved a very high level of partici-
pation in both study arms. Completion of the at-home
exercises of the control condition was not monitored,
but correct solutions of the previous at-home exercises
were provided. Furthermore, the aMCI and mild AD
688 V.C. Buschert et al. / Cognitive Intervention in MCI and AD

Table 4b
Change from baseline in subgroups of AD (n = 15 in measures of efficacy regarding MMSE, ADAS-cog., RBANS story memory and recall,
TMT-A and B, MADRS, and QoL-AD); standard deviation in ( ). Reported are all effects associated with p < 0.1
Outcome Point IGAD n = 8 CGAD n = 7 ANCOVAs with**η2
ADAS-cog Baseline 12.1 (5.3) 16.4 (4.8) n.s.
End-of-study 11.4 (6.0) 16.4 (4.9)
t-Test* n.s. n.s.
MMSE Baseline 24.5 (1.6) 25.3 (1.5) n.s.
End-of-study 25.0 (2.7) 24.4 (2.4)
t-Test* n.s. n.s.
RBANS – Baseline 12.4 (4.3) 12.4 (3.9) n.s.
story memory
End-of-study 9.5 (2.9) 10.9 (3.3)
t-Test* n.s. n.s.
RBANS – Baseline 1.9 (2.5) 3.4 (3.6) main effect education (F1,11 = 4.1, p = 0.07,
story recall η2 = 0.27)
End-of-study 2.0 (2.7) 2.1 (2.3)
t-Test* n.s. n.s.
TMT-A Baseline 46.1 (27.2) 50.7 (19.8) n.s.
End-of-study 36.5 (9.7) 40.9 (15.1)
t-Test* n.s. n.s.
TMT-B Baseline 206.9 (42.7) 192.1 (52.1) interaction treatment by progression
(F1,11 = 5.1, p = 0.05, η2 = 0.32) interaction
progression by education (F1,11 = 4.3,
p = 0.06, η2 = 0.28) main effect age
(F1,11 = 4.6, p = 0.06, η2 = 0.29)
End-of-study 178.4 (39.0) 207.9 (43.7)
t-Test* n.s. n.s.
MADRS Baseline 3.1 (4.1) 4.3 (4.0) interaction treatment by progression
(F1,11 = 3.4, p = 0.09, η2 = 0.24)
End-of-study 1.6 (3.4) 4.7 (5.0)
t-Test* t(7) = 2.6, p = 0.03 n.s.
QoL-AD Baseline 39.1 (7.4) 33.5 (4.0) n.s.
End-of-study 38.7 (7.6) 32.6 (3.8)
t-Test* n.s. n.s.
ADAS-cog = Alzheimer’s Disease Assessment Scale; MMSE = Mini Mental State Examination; RBANS = Repeatable Battery for the Assessment
of Neuropsychological Status; TMT = Trail Making Test; MADRS = Montgomery Asberg Depression Rating Scale (without item 6 “concentration
difficulties”); QoL-AD = Quality of Life-Alzheimer’s Disease; IG = intervention group; CG = control group; ns = not significant; aMCI =
amnestic mild cognitive impairment; AD = mild AD; ANCOVA = univariate analysis of covariance; columns indicate significant effects of 2 × 2
ANCOVAs (treatment × progression; see text for details).; *t-Test results refer to post-hoc comparisons testing for the effect of progression
(baseline vs. end-of-study) within subgroups (IG, CG).; **η2 = partial eta squared.

participants were highly compliant and the drop-out
rate was relatively low even though the intervention
was quite demanding. In this context, however, it is to
note that sometimes signs of fatigue could be observed
in the course of the 120-minutes sessions, especially
in participants with mild AD. Shortening the length
of weekly sessions is probably advisable. Finally, we
found that the administration of the program by the
same instructor for the whole period proved to be prac-
tical. Thus, the cognitive intervention appears to be a
clinically applicable and feasible intervention program
for amnestic MCI and mild AD.
Generalized effects on global cognitive status
Effects of intervention on the primary outcome
variables were detected only in subjects with aMCI.
The pre-post treatment comparison was significant for
ADAS-cog, one of the primary outcome-measures, and
showed a tendency toward significance in MMSE, the
other primary outcome variable. Change in ADAS-cog
in aMCI indicates improvement insofar as participants
of the CGMCI deteriorated significantly with time,
whereas participants of the IGMCI showed a tendency
to improve. This raises the question about whether
there were simply a greater percentage of aMCI sub-
jects with prodromal AD in the control group who may
have simply deteriorated over the 6 month follow-up.
In our opinion, this seems to be unlikely because inclu-
sion of all aMCI-subjects followed stringent diagnosis
criteria (see inclusion criteria above), and stratifica-
tion was carried out according to major characteristics
such as age, gender, education, and ApoE genotype
(see randomization procedure, above). Furthermore,
 V.C. Buschert et al. / Cognitive Intervention in MCI and AD
all MCI-participants showed a similar distribution of
baseline FDG hypometabolism in AD-typical brain
regions [54]. This strongly suggests that MCI patients
in both groups were comparable with respect to
preclinical AD stage. Moreover, the NNT-analysis
indicates effectiveness of the cognitive intervention.
An improvement defined as a 4-point change on
ADAS-cog occurred only in the IGMCI , whereas cog-
nitive decline was observed in both IGMCI and CGMCI .
Many regulatory authorities recognize a four-point
change on the ADAS-cog at 6 months as indicat-
ing a clinically important difference—a proposal that
has impacted how clinical trials are interpreted [55].
Considering the small sample size, results should be
interpreted with caution. The intervention did not train
to any item of the ADAS-cog so that findings are prob-
ably not related to practice effects. The positive effects
found here suggest that participating in the interven-
tion induced generalized cognitive benefits in MCI
subjects. However, there are a number of concerns
regarding generalizability of intervention effects to
everyday functions and whether a few points improve-
ment on a particular test is clinically meaningful
[55].
The AD patients showed no significant effects of
intervention on the primary outcome variables. It has
been previously hypothesized that brain reserve may be
better preserved in mildly impaired subjects at risk for
developing dementia than in already demented patients
[56]. It is thought that the capability to enhance or
compensate impaired cognitive functions declines in
the course of the disease [4]. Therefore, aMCI and
AD participants may benefit differently from cognitive
interventions. The effect sizes in the AD patients of our
study using an active control condition were similar
to those in a previous study on 201 patients in mild-
to-moderate stages of AD with a care-as usual control
condition [11] (Cohen’s d 0.37 for MMSE and 0.38 for
ADAS-cog in [11], and 0.74 and 0.20 in our study). Our
findings together with these previous findings suggest
a moderate effect of intervention in AD that requires a
sample size of at least 26 subjects per group to detect
an effect of intervention with 80% power at p < 0.05.
Future studies need to employ larger numbers of sub-
jects with an active control condition to finally resolve
this question.
Effects on specific cognitive functions
Amnestic MCI subjects in the IG demonstrated a
tendency toward significant benefits in higher atten-
tional functions as assessed by TMT-B. TMT-B is
689
considered to be a test of higher level cognitive skills
such as mental flexibility within the executive func-
tion domain [57, 58]. With the exception of the second
intervention-session, divided attention as required in
the TMT-B was not practiced in the intervention. We
assume the improvement in TMT-B might reflect a
generalized effect of the intervention, which addresses
a high level of attentional performance and linked
executive functioning due to the group setting (e.g.,
following conversations involving more than one other
person) and a number of distinct exercises (e.g., putting
something in order, finding similarities and differ-
ences, abstract thinking, and reasoning).
Results of TMT-B in AD patients indicate a tendency
toward a significant interaction between group and pro-
gression, which gives weak evidence for an impact
of the intervention on attentional functions. A study
of cognitive rehabilitation with a comparable sam-
ple size of 6 mild AD-patients in the treatment group
(AChE-I combined with cognitive rehabilitation) com-
pared to 7 mild AD-patients in the control-group (only
AChE-I) with weekly sessions of 90 minutes for 5
month also reported a tendency towards significant
improvement for participants of the intervention group
in TMT-B [13]. The statistical power of the present
study might not have been sufficient to detect a modest
change.
In both groups, intervention had no significant effect
on specific memory performance compared to the
active control condition. Recent studies have shown
significant improvements in memory tests in subjects
with aMCI and mild AD patients through cognitive
intervention, however, these effects were detected if
specific material or tasks of the intervention were used
in the subsequent evaluation [7, 59]. Therefore, these
effects seem to be specific to the memory material that
was tested in the particular context. Our findings sug-
gest that transfer of utilized mnemonic techniques does
not succeed if transferred to tasks and situations other
than directly trained during the intervention [36, 60].
This lack of behavioral change could be related to a
failure to train participants how to apply established
strategies to specific memory problems of daily life
[36]. It has been suggested that behavioral change is
more likely to be achieved when the participants of a
cognitive intervention have acquired a positive analy-
sis of costs and benefits associated with the behavior in
everyday memory problems [36]. Within this context
it is important to identify those internal and exter-
nal memory strategies which are most effective for
accomplishing everyday memory tasks, and which are
directly related to functional independence in AD. In
690 V.C. Buschert et al. / Cognitive Intervention in MCI and AD
future studies to investigate the issues of successful
transfer of mnemonic techniques they may require
from the participants intensive practice of adapting the
appropriate techniques to daily life.
Effects on non-cognitive functions
Significant improvement in mood was found in
aMCI-subjects in the IG arm. In patients with mild
AD we observed a tendency towards statistical sig-
nificance. Participants of the CGMCI and the CGAD
demonstrated no change over the 6 months. Analysis
of MADRS-items of the IGMCI showed that most often
complaints with “concentration difficulties” (item 6)
were reported (7 of 9 participants, mean 2.3, range
0-4), which improved considerably in the course of
the intervention (5 of 9 participants, mean 1.4, range
0–3). Cognition-related symptoms measured with a
depression rating scale may appear to exist in the
context of mild cognitive impairment or prodromal
AD. Assuming an overlap of cognitive symptoms
between depression and prodromal AD [51], we there-
fore excluded “concentration difficulties” from the
analysis of depressive symptoms [61]. All partici-
pants of the present IGMCI scored below a critical
cut-off for mild depression (13 to 21 points [49]) or
even a cut-off for remission (<9 and <10 [62]) on
the MADRS including scores of concentration diffi-
culties at baseline (mean: 5.4 (SD 2.4), range: 1–8).
Remaining depression scores were low in all groups
(see table 4 a), thus, improvements in depression scores
appear not to be clinically significant with regard to
depression. However, improvements in mood with-
out manifest symptoms of depression might suggest
enhancement of self-esteem and well-being [7, 63],
which is thought to be beneficial for cognitive perfor-
mance [25, 63, 64]. A previous study has shown that
participation of AD patients in a cognitive intervention
program had a positive impact on global mood [39].
Given that a considerable effect of the intervention
was on depression scores, it raises the question as to
whether the improvements seen on higher attentional
functions (see table 4 a) may arise from improved com-
plex attention related to depression. In our opinion,
it would be somewhat surprising that improvement
in mood in absence of depression should affect com-
plex attentional functions to this extent. Moreover, one
might expect that positive effects on mood should not
only be reflected by improvements in higher atten-
tional functions but also in speed of processing or
sustained attention [65]. Since there were no sig-
nificant changes in sustained attention assessed with
TMT-A (see table 4 a), a similar effect cannot be deter-
mined in the present study. Thus, we assume that
improvements in TMT-B are rather related to direct
intervention effects on cognition and less mediated by
mood. However, since mood-loaded factors such as
motivation are considered to have an impact on cogni-
tive functioning, we cannot determine to what extent
improvements in non-cognitive depressive symptoms
might have contributed to improvements in higher
attentional functions. Studies with long-term observa-
tions with greater sample sizes probably will give better
insight into interaction between cognitive intervention,
attentional functions, and mood.
Although participants of both IGs (aMCI and AD)
benefit from the intervention concerning mood, which
is thought to be crucial for self-reported well-being,
there were no significant intervention effects in qual-
ity of life in both subgroups. In contrast, a study by
Spector and colleagues with 115 mild and mild-to-
moderate patients with AD demonstrated significant
improvement in the QoL-AD in the treatment group
relative to the control group [11, 66]. An interpreta-
tion of these findings is difficult because the control
condition in the Spector study was merely the continu-
ation of “usual activities”. Positive influence of social
interactions due to the trainer was not considered or
controlled for. Hence, it remains unclear, to what extent
the measured benefits on quality of life were due to the
Cognitive Stimulation Therapy in the Spector trial. In
contrast, in our study we applied an active control con-
dition, in which the participants were in contact with
the same instructor of the intervention condition.
In general, intervention-induced effects on quality
of life in AD due to participation in a cognition-
based intervention are difficult to achieve owing to
the complexity of the theoretical construct of qual-
ity of life. Cognitive interventions focus primarily on
cognitive domains, and, if applicable, on additionally
non-cognitive functions such as mood, but they have
few or no direct emphasis on several items tested
in the QoL-AD such as physical health, living sit-
uation, family, marriage, friends, free time activities
or financial status [67]. Administration of a question-
naire concerning satisfaction with one’s own memory
ability such as the MMQ-Contentment, a part of the
Multifactorial Memory Questionnaire (MMQ) [68],
might probably be more successful for the assessment
of self-reported benefits. Since interventions target-
ing memory functioning in AD have been criticized
for negatively affecting well-being [69] it is impor-
tant to note that we observed no adverse effects of the
intervention on measures of quality of life.
 V.C. Buschert et al. / Cognitive Intervention in MCI and AD
Summary
In summary, findings of our 6-month randomized
controlled pilot study suggest positive effects of a
cognitive intervention in subjects with aMCI. For the
AD-patient group we did not find positive intervention
effects on the primary outcome variables ADAS-cog
and MMSE and only a tendency for improvement on
secondary outcome measures such as higher atten-
tional functions and mood. With regard to lack of
improvement in verbal episodic memory but signif-
icant improvements in higher attentional functions,
improvements in global cognitive status seem to rely
primarily on attentional performance, though interpre-
tation of findings should be made with due care with
regard to limited statistical power. Recent evidence in
the research on cognitive deficits in AD suggests that
after an initial amnesic stage in AD, attention is the
first non-memory domain to be affected [43]. It appears
that divided attention is particularly vulnerable while
sustained attention is relatively preserved in the early
stages as mentioned above. Improvement of attentional
functions due to practice and experience may lead
to improved performance in other cognitive domains.
Studies examining the neurobiological aspects of prac-
tice and experience on cognitive tasks indicate changes
of neural networks underlying higher cognitive func-
tions, such as working memory, in healthy humans
[70]. Petersen and colleagues [71] suggest that a novel
task requires high cognitive effort during initial perfor-
mance for successful performance and training leads
to a reduction of cognitive demand. The decrease in
cognitive demand is due to automation of the atten-
tional and effortful components of the task because the
participant becomes an expert in the task with train-
ing [71]. Thus attentional components can be reduced
and the participant can focus on the specific cognitive
demands of the task. A similar mechanism may also be
activated through cognitive interventions in cognitive
impaired subjects.
Basically, the intervention shows evidence of effi-
cacy in cognitive and non cognitive-domains in
subjects with aMCI. However, there are a number of
concerns regarding generalized intervention effects to
everyday functions and long-term, disease modifying
effects. Thus, future studies with extended follow-
up periods should focus at global clinical status as a
primary outcome measure. Additionally, to better esti-
mate whether clinically important goals are met during
cognitive intervention, e.g., Goal Attainment Scaling
[72], might be an appropriate measure which could be
employed in future studies [73].
691
Concerning impaired memory functions in AD,
future development of our stage-specific cognitive
intervention should focus strongly on teaching and
practicing mnemonic strategies which are directly
related to functional independence in AD (e.g., keep-
ing appointments or remembering to take medications
on time), and which are important for social interac-
tions (e.g., learning new names or a telephone number)
[36]. In doing so, the program should emphasize cogni-
tive and social resources of the participants providing
opportunities for (re-) activating largely intact cogni-
tive domains. Within this context the Goal Attainment
Scaling might be an appropriate measure to be adminis-
tered in future studies [73]. Finally, sessions, especially
for participants with mild AD, should probably run
weekly for less than 120 minutes to minimize partic-
ipant fatigue (see also section Discussion, Feasibility
and acceptance of program and study).
LIMITATIONS
There are limitations in our study. First, the study
included a relatively small number of participants,
which limited the statistical power to detect statistically
significant effects especially for the mild AD patient
group. In general one has to note that the MCI con-
cept remains controversial, with no generally approved
classification system and concerns about reliability and
validity [74]. For future studies it may be advanta-
geous to define clinical risk groups not only through
neuropsychological status, but also on the basis of bio-
logical markers [75]. Finally, evidence of long-term
maintenance of benefits in aMCI and mild AD due to
intervention is crucial because of the progressive nature
of the disease. A follow-up study is in preparation in
order to detect indications of long-term-efficacy of the
cognitive intervention focusing on conversion to AD in
MCI-subjects and further deterioration in AD-patients.
CONCLUSION
Our results suggest improvements in cognition and
non-cognitive performance in subjects with aMCI
having attended our 6-month stage-specific cognitive
intervention, compared to baseline and to an active con-
trol condition. Naturally, according to a pilot study, our
results should be viewed as promising findings which
need to be replicated in larger studies. Given that sub-
jects with aMCI are at high risk of developing AD,
and given the lack of disease-modifying pharmacolog-
ical treatment presently available for this population,
692 V.C. Buschert et al. / Cognitive Intervention in MCI and AD
these results may have important clinical implications
for further therapeutic approaches.
In mild AD patients, trends for a statistically sig-
nificant effect of the intervention arm on the outcome
variables were found, indicating the necessity to further
evaluate and specify under which circumstances cog-
nitive functions can improve in these patients. These
findings in a small sample of subjects are promising
and support the use of the intervention in larger scale
studies to confirm the effect in aMCI and AD subjects.
Long-term-studies with larger samples are required
to determine the extent to which cognitive interven-
tion can affect aMCI-subjects’ conversion to AD and
progressive deterioration in AD.
ACKNOWLEDGMENTS
The authors thank Arun Bokde, PhD, for critical
review of the manuscript. Moreover, the authors thank
Daniela Otto for administration of tests and scales.
Authors’ disclosures available online (http://www.j-
alz.com/disclosures/view.php?id=771).
REFERENCES
[1] Valenzuela M, Sachdev P (2009) Can cognitive exercise pre-
vent the onset of dementia? Systematic review of randomized
clinical trials with longitudinal follow-up. Am J Geriatr Psy-
chiatry 17, 179-187.
[2] Valenzuela MJ, Sachdev P (2006) Brain reserve and dementia:
a systematic review. Psychol Med 36, 441-454.
[3] Valenzuela MJ, Sachdev P (2006) Brain reserve and cognitive
decline: a non-parametric systematic review. Psychol Med 36,
1065-1073.
[4] Stern Y (2006) Cognitive reserve and Alzheimer disease.
Alzheimer Dis Assoc Disord 20, 112-117.
[5] Cipriani G, Bianchetti A, Trabucchi M (2006) Outcomes
of a computer-based cognitive rehabilitation program on
Alzheimer’s disease patients compared with those on patients
affected by mild cognitive impairment. Arch Gerontol Geriatr
43, 327-335.
[6] Rozzini L, Costardi D, Chilovi BV, Franzoni S, Trabucchi
M, Padovani A (2007) Efficacy of cognitive rehabilitation
in patients with mild cognitive impairment treated with
cholinesterase inhibitors. Int J Geriatr Psychiatry 22, 356-
360.
[7] Belleville S, Gilbert B, Fontaine F, Gagnon L, Menard E,
Gauthier S (2006) Improvement of episodic memory in per-
sons with mild cognitive impairment and healthy older adults:
evidence from a cognitive intervention program. Dement
Geriatr Cogn Disord 22, 486-499.
[8] Petersen RC, Smith GE, Waring SC, Ivnik RJ, Tangalos EG,
Kokmen E (1999) Mild cognitive impairment: clinical char-
acterization and outcome. Arch Neurol 56, 303-308.
[9] Loewenstein DA, Acevedo A, Czaja SJ, Duara R (2004) Cog-
nitive rehabilitation of mildly impaired Alzheimer disease
patients on cholinesterase inhibitors. Am J Geriatr Psychiatry
12, 395-402.
[10] Requena C, Maestu F, Campo P, Fernandez A, Ortiz T
(2006) Effects of cholinergic drugs and cognitive training on
dementia: 2-year follow-up. Dement Geriatr Cogn Disord 22,
339-345.
[11] Spector A, Thorgrimsen L, Woods B, Royan L, Davies S,
Butterworth M, Orrell M (2003) Efficacy of an evidence-
based cognitive stimulation therapy programme for people
with dementia: randomised controlled trial. Br J Psychiatry
183, 248-254.
[12] Requena C, Lopez Ibor MI, Maestu F, Campo P, Lopez Ibor
JJ, Ortiz T (2004) Effects of cholinergic drugs and cogni-
tive training on dementia. Dement Geriatr Cogn Disord 18,
50-54.
[13] Bottino CM, Carvalho IA, Alvarez AM, Avila R, Zukauskas
PR, Bustamante SE, Andrade FC, Hototian SR, Saffi F,
Camargo CH (2005) Cognitive rehabilitation combined with
drug treatment in Alzheimer’s disease patients: a pilot study.
Clin Rehabil 19, 861-869.
[14] Olazaran J, Muniz R, Reisberg B, Pena-Casanova J, del
Ser T, Cruz-Jentoft AJ, Serrano P, Navarro E, Garcia de
la Rocha ML, Frank A, Galiano M, Fernandez-Bullido Y,
Serra JA, Gonzalez-Salvador MT, Sevilla C (2004) Benefits
of cognitive-motor intervention in MCI and mild to moderate
Alzheimer disease. Neurology 63, 2348-2353.
[15] Sitzer DI, Twamley EW, Jeste DV (2006) Cognitive training
in Alzheimer’s disease: a meta-analysis of the literature. Acta
Psychiatr Scand 114, 75-90.
[16] Olazaran J, Reisberg B, Clare L, Cruz I, Pena-Casanova J,
Del Ser T, Woods B, Beck C, Auer S, Lai C, Spector A,
Fazio S, Bond J, Kivipelto M, Brodaty H, Rojo JM, Collins
H, Teri L, Mittelman M, Orrell M, Feldman HH, Muniz R
Nonpharmacological Therapies in Alzheimer’s Disease: A
Systematic Review of Efficacy. Dement Geriatr Cogn Disord
30, 161-178.
[17] Buschert V, Bokde AL, Hampel H (2010) Cognitive interven-
tion in Alzheimer disease. Nat Rev Neurol 6, 508-517.
[18] Davis RN, Massman PJ, Doody RS (2001) Cognitive interven-
tion in Alzheimer disease: a randomized placebo-controlled
study. Alzheimer Dis Assoc Disord 15, 1-9.
[19] Quayhagen MP, Quayhagen M, Corbeil RR, Roth PA,
Rodgers JA (1995) A dyadic remediation program for care
recipients with dementia. Nurs Res 44, 153-159.
[20] Clare L, Woods RT (2004) Cognitive training and cogni-
tive rehabilitation for people with early-stage Alzheimer’s
disease: a review. Neuropsychol Rehabil 14, 385-401.
[21] Spector A, Orrell M, Davies, S, Woods, RT (1998) Real-
ity Orientation for dementia: a review of the evidence for
its effectiveness. In The Cochrane Library Update Software,
Oxford.
[22] Farina E, Mantovani F, Fioravanti R, Pignatti R, Chiavari L,
Imbornone E, Olivotto F, Alberoni M, Mariani C, Nemni R
(2006) Evaluating two group programmes of cognitive train-
ing in mild-to-moderate AD: is there any difference between
a ‘global’ stimulation and a ‘cognitive-specific’ one? Aging
Ment Health 10, 211-218.
[23] Farina E, Fioravanti R, Chiavari L, Imbornone E, Alberoni M,
Pomati S, Pinardi G, Pignatti R, Mariani C (2002) Comparing
two programs of cognitive training in Alzheimer’s disease: a
pilot study. Acta Neurol Scand 105, 365-371.
[24] Clare L, Woods T, Cook EM, Orell M, Spector A (2006)
Cognitive Rehabilitation and cognitive training for early-
stage Alzheimer’s disease and vascular dementia (Review).
Cochrane Collab, 1-31.
[25] Belleville S (2008) Cognitive training for persons with mild
cognitive impairment. Int Psychogeriatr 20, 57-66.
 V.C. Buschert et al. / Cognitive Intervention in MCI and AD
[26] Reisberg B, Franssen EH, Hasan SM, Monteiro I, Boksay
I, Souren LE, Kenowsky S, Auer SR, Elahi S, Kluger A
(1999) Retrogenesis: clinical, physiologic, and pathologic
mechanisms in brain aging, Alzheimer’s and other dementing
processes. Eur Arch Psychiatry Clin Neurosci 249 Suppl 3,
28-36.
[27] Reisberg B, Ferris SH, de Leon MJ, Crook T (1982) The
Global Deterioration Scale for assessment of primary degen-
erative dementia. Am J Psychiatry 139, 1136-1139.
[28] Reisberg B, Franssen EH, Souren LE, Auer SR, Akram I,
Kenowsky S (2002) Evidence and mechanisms of retrogen-
esis in Alzheimer’s and other dementias: management and
treatment import. Am J Alzheimers Dis Other Demen 17,
202-212.
[29] Clare L, Wilson BA, Carter G, Roth I, Hodges JR (2002)
Relearning face-name associations in early Alzheimer’s dis-
ease. Neuropsychology 16, 538-547.
[30] Petersen RC, Doody R, Kurz A, Mohs RC, Morris JC, Rabins
PV, Ritchie K, Rossor M, Thal L, Winblad B (2001) Current
concepts in mild cognitive impairment. Arch Neurol 58, 1985-
1992.
[31] McKhann G, Drachman D, Folstein M, Katzman R, Price
D, Stadlan EM (1984) Clinical diagnosis of Alzheimer’s dis-
ease: report of the NINCDS-ADRDA Work Group under
the auspices of Department of Health and Human Services
Task Force on Alzheimer’s Disease. Neurology 34, 939-
944.
[32] Saß H, Wittchen H, Zaudig M (2003) Diagnostisches und
Statistisches Manual Psychischer Störungen. DSM-IV-TR.
Deutsche Bearbeitung und Einführung, Hogrefe, Göttingen.
[33] Folstein M, Folstein S, McHugh P (1975) “Mini Mental
State”: A practical method for grading the cognitive state of
patients for the clinician. J Psychiatr Res 12, 189-198.
[34] Clare L, Wilson BA, Breen K (1999) Errorless learning of
face-name associations in early Alzheimer’s disease. Neuro-
case 5, 37-46.
[35] Wilson B, Baddeley A, Evans J, Shiel A (1994) Errorless
learning in the rehabilitation of memory impaired people.
Neuropsychol Rehabil 4, 307-326.
[36] Troyer AK, Murphy KJ, Anderson ND, Moscovitch M, Craik
FI (2008) Changing everyday memory behaviour in amnes-
tic mild cognitive impairment: a randomised controlled trial.
Neuropsychol Rehabil 18, 65-88.
[37] Kinsella GJ, Mullaly E, Rand E, Ong B, Burton C, Price
S, Phillips M, Storey E (2009) Early intervention for mild
cognitive impairment: a randomised controlled trial. J Neurol
Neurosurg Psychiatry 80, 730-736.
[38] Lachmann ME, Weaver SL, Bandura M, Elliott E, Lewkowicz
CJ (1992) Improving memory and control belief through cog-
nitive restructuring and self-generated strategies. J Gerontol
47, P293-P299.
[39] Ermini-Fünfschilling D (1997) Gedächtnistraining und
Milieutherapie bei seniler Demenz In Alzheimer-Krankheit.
Abgrenzung normalen Alterns, Epidemiologie, Aetiologie,
Pathogenese, Klinik, Behandlung, Ethik., Roesler M, Retz
W, Thome J, eds. Deutscher Studien Verlag, Weinheim, pp.
284-295.
[40] Czaja SJ, Schulz R, Lee CC, Belle SH (2003) A methodology
for describing and decomposing complex psychosocial and
behavioral interventions. Psychol Aging 18, 385-395.
[41] Breuil V, Rotrou Jd, Forette F (1994) Cognitive stimulation
of patients with dementia: preliminary results. Int J Geriatr
Psychiatry 9, 211-217.
[42] Wilson B (1999) Memory rehabilitation in brain-injured peo-
ple In Cognitive Neurorehabilitation, Stuss D, Wincour, G,
693
Robertson, IH, ed. Cambridge University Press, Cambridge,
pp. 333-346.
[43] Perry RJ, Hodges JR (1999) Attention and executive deficits
in Alzheimer’s disease. A critical review. Brain 122 (Pt 3),
383-404.
[44] Gatterer G (2007) [Efficiency of specific neuropsycholog-
ical and clinical-psychological interventions in old age]. Z
Gerontol Geriatr 40, 88-95.
[45] Gurka P, Marksteiner J (2002) [Psychosocial therapy of
patients with dementia]. Wien Med Wochenschr 152, 102-106.
[46] Rosen WG, Mohs RC, Davis KL (1984) A new rating scale
for Alzheimer’s disease. Am J Psychiatry 141, 1356-1364.
[47] Randolph C, Tierney MC, Mohr E, Chase TN (1998) The
Repeatable Battery for the Assessment of Neuropsychologi-
cal Status (RBANS): preliminary clinical validity. J Clin Exp
Neuropsychol 20, 310-319.
[48] Reitan RM (1979, 1992) Trail Making Test, Reitan Neuropsy-
chology Laboratory, Tucson.
[49] Montgomery S, Asberg M (1979) A new depression scale
designed to be sensitive to change. Br J Psychiatry 134 382-
389.
[50] Gabryelewicz T, Styczynska M, Pfeffer A, Wasiak B,
Barczak A, Luczywek E, Androsiuk W, Barcikowska M
(2004) Prevalence of major and minor depression in elderly
persons with mild cognitive impairment–MADRS factor anal-
ysis. Int J Geriatr Psychiatry 19, 1168-1172.
[51] Heun R, Kockler M, Ptok U (2003) Lifetime symptoms of
depression in Alzheimer’s disease. Eur Psychiatry 18, 63-69.
[52] Lazarus LW, Newton N, Cohler B, Lesser J, Schweon C
(1987) Frequency and presentation of depressive symptoms in
patients with primary degenerative dementia. Am J Psychiatry
144, 41-45.
[53] Logsdon R, Gibbons L, McCurry S, Teri L (1999) Quality of
life of Alzheimer’s Disease: Patient and caregivers report. J
Mental Health Aging 5, 21-32.
[54] Förster S, Buschert VC, Buchholtz H-G, Teipel SJ, Friese U,
Zach C, la Fougere C, Rominger A, Drzezga A, Hampel H,
Bartenstein P, Buerger K (2011) Effects of a 6-month cog-
nitive intervention program on brain metabolism in amnestic
mild cognitive impairment and mild Alzheimer’s disease. J
Alzheimers Dis, in press.
[55] Rockwood K, Fay S, Gorman M, Carver D, Graham JE
(2007) The clinical meaningfulness of ADAS-Cog changes
in Alzheimer’s disease patients treated with donepezil in an
open-label trial. BMC Neurol 7, 26.
[56] Stern Y (2002) What is cognitive reserve? Theory and
research application of the reserve concept. J Int Neuropsy-
chol Soc 8, 448-460.
[57] Bowie CR, Harvey PD (2006) Administration and interpreta-
tion of the Trail Making Test. Nat Protoc 1, 2277-2281.
[58] Reitan RM (1958) Validity of the trailmaking test as an indi-
cation of organic brain damage. Perceptual Motor Skills 8,
271-276.
[59] Lee SB, Park CS, Jeong JW, Choe JY, Hwang YJ, Park CA,
Park JH, Lee DY, Jhoo JH, Kim KW (2009) Effects of spaced
retrieval training (SRT) on cognitive function in Alzheimer’s
disease (AD) patients. Arch Gerontol Geriatr 49, 289-293.
[60] Glück J, Heckhausen J (2001) Kognitives Training im Alter:
Potential und Grenzen der Plastizität In Handbuch Kogni-
tives Training, Klauer KJ, ed. Hogrefe Verlag für Psychologie,
Göttingen, Bern, Toronto, Seattle, pp. 431-488.
[61] Koerts J, Leenders KL, Koning M, Portman AT, van Beilen
M (2007) Striatal dopaminergic activity (FDOPA-PET) asso-
ciated with cognitive items of a depression scale (MADRS)
in Parkinson’s disease. Eur J Neurosci 25, 3132-3136.
694 V.C. Buschert et al. / Cognitive Intervention in MCI and AD
[62] Hawley CJ, Gale TM, Sivakumaran T (2002) Defining remis-
sion by cut off score on the MADRS: selecting the optimal
value. J Affect Disord 72, 177-184.
[63] Dittmann-Kohli F (1986) [Training-induced changes in per-
formance self concept and cognitive abilities in the aged]. Z
Gerontol 19, 309-322.
[64] Platz S, Weyerer S (1990) Gedächtnistraining im Alter:
Theoretischer Hintergrund und Entwicklung eines Inter-
ventionsprogramms für Altenheimbewohner. Zeitschrift für
Gerontologie 23, 197-204.
[65] Christensen H, Griffiths K, Mackinnon A, Jacomb P (1997)
A quantitative review of cognitive deficits in depression and
Alzheimer-type dementia. J Int Neuropsychol Soc 3, 631-651.
[66] Woods B, Thorgrimsen L, Spector A, Royan L, Orrell M
(2006) Improved quality of life and cognitive stimulation
therapy in dementia. Aging Ment Health 10, 219-226.
[67] Logsdon RG, Gibbons LE, McCurry SM, Teri L (2002)
Assessing quality of life in older adults with cognitive impair-
ment. Psychosom Med 64, 510-519.
[68] Troyer AK, Rich JB (2002) Psychometric properties of a new
metamemory questionnaire for older adults. J Gerontol B
Psychol Sci Soc Sci 57, 19-27.
[69] APA (1997) Practice guideline for the treatment of patients
with Alzheimer’s disease and other dementias of late life.
American Psychiatric Association. Am J Psychiatry 154,
1-39.
[70] Kelly C, Foxe JJ, Garavan H (2006) Patterns of normal human
brain plasticity after practice and their implications for neu-
rorehabilitation. Arch Phys Med Rehabil 87, S20-S29.
[71] Petersen SE, van Mier H, Fiez JA, Raichle ME (1998) The
effects of practice on the functional anatomy of task perfor-
mance. Proc Natl Acad Sci U S A 95, 853-860.
[72] Kiresuk T, Sherman R (1968) Goal Attainment Scaling: A
general method for evaluating comprehensive community
health programs. Community Ment Health J 4, 443-453.
[73] Rockwood K, Graham JE, Fay S (2002) Goal setting and
attainment in Alzheimer’s disease patients treated with
donepezil. J Neurol Neurosurg Psychiatry 73, 500-507.
[74] Stephan BC, Matthews FE, McKeith IG, Bond J, Brayne C
(2007) Early cognitive change in the general population: how
do different definitions work? J Am Geriatr Soc 55, 1534-
1540.
[75] Dubois B, Feldman HH, Jacova C, Dekosky ST, Barberger-
Gateau P, Cummings J, Delacourte A, Galasko D, Gauthier S,
Jicha G, Meguro K, O’Brien J, Pasquier F, Robert P, Rossor M,
Salloway S, Stern Y, Visser PJ, Scheltens P (2007) Research
criteria for the diagnosis of Alzheimer’s disease: revising the
NINCDS-ADRDA criteria. Lancet Neurol 6, 734-746.