Minnesota State University, Mankato
Minnesota State University, Mankato
Cornerstone: A Collection of Scholarly
Cornerstone: A Collection of Scholarly
and Creative Works for Minnesota
and Creative Works for Minnesota
State University, Mankato
State University, Mankato
All Theses, Dissertations, and Other Capstone
Projects
Graduate Theses, Dissertations, and Other
Capstone Projects
2020
The Effects of Cognitive Training on Behavioral Functioning in
The Effects of Cognitive Training on Behavioral Functioning in
Persons with Dementia
Persons with Dementia
Abigail J. Dye
Minnesota State University, Mankato
Follow this and additional works at: https://cornerstone.lib.mnsu.edu/etds
Part of the Clinical Psychology Commons, and the Gerontology Commons
Recommended Citation
Recommended Citation
Dye, A. J. (2020). The effects of cognitive training on behavioral functioning in persons with dementia
[Master’s thesis, Minnesota State University, Mankato]. Cornerstone: A Collection of Scholarly and
Creative Works for Minnesota State University, Mankato. https://cornerstone.lib.mnsu.edu/etds/1029
This Thesis is brought to you for free and open access by the Graduate Theses, Dissertations, and Other Capstone
Projects at Cornerstone: A Collection of Scholarly and Creative Works for Minnesota State University, Mankato. It
has been accepted for inclusion in All Theses, Dissertations, and Other Capstone Projects by an authorized
administrator of Cornerstone: A Collection of Scholarly and Creative Works for Minnesota State University,
Mankato.
The Effects of Cognitive Training on Behavioral Functioning in Persons with Dementia
By
Abigail J. Dye
A Thesis Submitted in Partial Fulfillment of the
Requirements for the Degree of
Master of Arts
In
Clinical Psychology
i
April 30, 2020
The Effects of Cognitive Training on Behavioral Functioning in Persons with Dementia
Abby Dye
This Master’s thesis has been examined and approved by the following members of the
student’s committee.
__________________________________________
Chairperson
Jeffrey Buchanan, Ph.D.
__________________________________________
Committee Member
Daniel Houlihan Ph.D.
__________________________________________
Committee Member
Angelica Aguirre, Ph.D.
ii
Table of Contents
Introduction ………………………………………………………………………………………1
Methods ………………………………………………………………………………………….12
Results ……………………………………………………………………………………………20
Discussion ……………………………………………………………………………………….39
References ……………………………………………………………………………………….46
Tables
1. Participant Characteristics ………………………………………………………..……..14
2. Summary of Findings, Aggregated Data ………………………………..………………25
3. Quality of Life Scores ………………………………………………………….………..26
4. Summary of Findings, Participant One ……………………………………………….…29
5. Summary of Findings, Participant Two …………………………………………………31
6. Summary of Findings, Participant Three …………………………………..……………34
7. Summary of Findings, Participant Four ………………………………..………………..36
8. Summary of Findings, Participant Five ……………………………………..…………..38
Appendices
A. Participant Informed Consent ………………………………………………..…………55
B. Staff Informed Consent …………………………………………………………………58
C. Caregiver Interview ………………………………………………………..……………60
D. Summary of Scores by Observation Interval ……………………………………………62
E. Graphs for Observational Data, Aggregated …………………………….………………68
F. Individual Quality of Life Scores ……………………………………….………………71
iii
G. Graphs for Observational Data, Participant One ………………………………….…….75
H. Graphs for Observational Data, Participant Two ………………………………….…….78
I. Graphs for Observational Data, Participant Three ……………………………………….81
J. Graphs for Observational Data, Participant Four ……………………………….………84
K. Graphs for Observational Data, Participant Five …………………………….………….87
iv
The Effects of Cognitive Training on Behavioral Functioning in Persons with Dementia
Abigail J. Dye
A thesis submitted in partial fulfillment of the requirements for the degree of Mast of Arts in
Clinical Psychology
Minnesota State University, Mankato
Mankato, Minnesota
May, 2020
Abstract
Lack of engagement in pleasant activities and negative mood are two factors that decrease
quality of life (QoL) for older adults with moderate to severe cognitive impairment. As
enhancing QoL has become a primary treatment outcome for individuals with cognitive
impairment, investigation into the ability of nonpharmacological interventions to increase
engagement and positive mood has come to the forefront of research. Cognitive training is a
nonpharmacological intervention that utilizes manualized techniques with the primary goal of
enhancing different areas of cognitive function. Although the cognitive benefits of the programs
have been widely investigated and established, the potential benefits that cognitive training
programs may have on increasing engagement in activities and reducing negative affect have
been largely unstudied. This study investigated the effects of a cognitive training program on
engagement in activity and affect for individuals with moderate to severe cognitive impairment
through behavioral observation. An alternating treatment design was utilized to compare
engagement and affect during cognitive training program sessions and regularly scheduled
activities at a residential community for older adults. Results indicated the utility of cognitive
training programs for increasing active engagement during the program sessions while affect and
QoL remained unchanged.
1
Introduction
Despite significant efforts to the contrary, a cure for dementia has yet to be discovered.
Although efforts continue to identify a definitive cure, researchers have paid increasing attention
to creating interventions that improve quality of life (QoL). In fact, research suggests that
improving QoL should be the primary goal of treatment for individuals with this disorder rather
than focusing on cognitive outcomes (Whitehouse & Rabins, 1992; Whitehouse & George,
2008). Many different forms of nonpharmacological interventions exist for enhancing safety,
increasing independence, and improving quality of life in persons with dementia (PwD; Douglas
et al., 2004). One set of nonpharmacological interventions, called cognitive training, has received
more attention in recent years and preliminary evidence suggests promise for improving QoL for
PwD (Mate-Kole et al., 2006, Giovagnoli et al., 2017). As the number of individuals developing
dementia increases, the importance of evaluating nonpharmacological interventions to promote
their well-being also increases.
Overview of Dementia
The population of older adults in the United States is on the rise. The number of
individuals over the age of 65 is projected to increase from 63 million currently to 114 million by
2060 (United States Census Bureau, 2017). As the age of the population increases, so too does
the amount of people who experience cognitive decline as the likelihood of developing dementia
increases with age (Murman, 2015).
Broadly, dementia describes a set of symptoms indicative of cognitive and psychological
changes disrupting everyday functioning (World Health Organization, 2019). According to the
World Health Organization, 50 million people currently suffer from dementia with nearly 10
million new cases every year (World Health Organization, 2019). In the United States, an
2
estimated 5 million people have a diagnosis of Alzheimer’s disease, the most common type of
dementia (Alzheimer’s Association, 2020).
Dementia is a heterogeneous disorder involving several phenotypes with an array of
etiologies that can be partitioned into three categories: degenerative, stable, and reversible
(Rubbert et al., 2014). Degenerative dementia is the most common. Characterized by the
progressive deterioration of cognitive functioning, degenerative dementia leads to significant
functional disability (Ruppert et al., 2014). The most common form of dementia, Alzheimer’s
disease, is a degenerative dementia (Schwarz & Frolich, 2013; Ruppert et al., 2014). Stable or
slow progressive dementias include disorders such as cerebrovascular dementia and Parkinson’s
disease with dementia, and they are different from degenerative dementia in that these disorders
are characterized by stable cognitive deficits that gradually increase in severity over time
(Ruppert et al., 2014). Lastly, reversible dementia involves medical conditions of which
dementia is a symptom. These symptoms begin to abate once the cause of the disease is treated.
Examples of reversible dementias are hypothyroidism, vitamin deficiencies, and depression
(Ruppert et al., 2014). Regardless of the classification and mechanisms of the disorder, the
symptomatology is generally consistent across dementias; although some deficits are more
common in specific disorders.
Because of the overlap in symptomatology across the disorders, dementia is best
understood through the conception of stages of severity. It is generally accepted in the literature
that there are seven stages of dementia (Sclan & Reisberg, 1992; Reisberg et al., 1982) ranging
from no dementia to extremely severe. The cognitive and functional deficits prevalent at each
stage are similar across disorders.
3
In the first stage, no impairment is experienced, meaning the individual is mentally
healthy for their age. The individual experiences no objective or subjective deficits to cognitive
functioning or ability to perform instrumental of daily living (IADLs) or activities of daily living
(ADLs; Reisberg et al., 1982).
The second stage involves healthy aging with no dementia diagnosis or cognitive
impairment (Reisberg et al., 1982). To some degree, cognitive decline accompanying age is
normal. Age associated cognitive decline, as termed in the literature, constitutes non-pathological
cognitive changes that individuals experience with age (Story & Attix, 2010; Deary et al., 2009).
Typically, older adults can expect to experience a decline in critical cognitive functions like
processing speed (Salthouse, 1993), memory (Craik & Salthouse, 2008; Harada et al., 2013), and
attention (Carlson et al., 1995; Salthouse et al. 1995) that do not interrupt their ability to perform
IADLs and ADLs.
Mild cognitive impairment (MCI) is the term used to describe stage three. This term
defines a cognitive state in which the individual is not demented but is experiencing impairment
in cognitive functioning in one or more cognitive domains without the presence of disability in
IADL/ADL performance (Smith & Bondi, 2013). The National Institute of Aging-Alzheimer’s
Association task force published four diagnostic criteria for MCI to distinguish it from dementia
(Albert et al., 2011). The four criteria are: 1) Evidence of concern regarding a change in
cognition; 2) Evidence of lower performance than what is to be expected on one or more
cognitive domains; 3) Ability to independently perform IADLs and ADLs; and 4) Not meeting
the criteria for dementia. Concerns regarding change in cognition can be observed in a variety of
cognitive domains including memory, executive function, language, attention, and visuospatial
skills (Smith & Bondi, 2013). Although concern about forgetfulness is not a necessary precursor
4
for a diagnosis of MCI, it is the most common and the most indicative of an eventual conversion
to dementia (Schmidtke & Hermeneit, 2008). When MCI includes concern about memory loss, it
is referred to as amnestic MCI (Schmidtke & Hermeneit, 2008). Hallmarks of this stage include
forgetfulness, decreased ability to concentrate, insufficient work performance, and word finding
problems (Reisberg et al., 1982).
Early stage cognitive impairment, stage four, is difficult to distinguish from MCI at times
(Smith & Bondi, 2013). Limitations in the ability to perform IADLs is the main distinguishing
factor between stages three and four. For example, in stage four, it becomes more difficult to
manage finances and drive in unfamiliar places without getting lost (Reisberg et al., 1982).
Behavioral changes also occur. Across dementias, verbal aggression becomes more prominent
(Nagaratnam et al, 1998; Weiner et al., 2005). Individuals with Alzheimer’s disease may
experience less energy, become more socially withdrawn, be more apathetic and have a greater
dependence on others (Schoenberg & Scott, 2011; Weiner et al., 2005) while dementia with
Lewy bodies patients are more likely to experience visual hallucinations and depression (Borroni
et la., 2008). Generally, withdrawal from family is evident as communication becomes more
difficult and the PwD is becoming more aware of their deficits (Reisberg et al., 1982).
Neuropsychologically, memory impairment and dyssomnia are the first indicators of
Alzheimer’s disease while psychomotor slowing, impaired attention/concentration, and
constructional skills deficits are prevalent in vascular dementia (Schoenberg & Scott, 2011).
Memory loss is typically reserved for recent episodic memories, or memory for past events and
their details, while semantic memory, or memory that aids in the understanding of words, objects
and events, remains unscathed (Almkvist et al., 1998; Rogers et al, 2006). A hallmark of
Alzheimer’s disease in this stage is the inability to gain from recognition cueing (Tuokko et al.,
5
1991). Individuals in this stage will sometimes try to minimize their cognitive problems and
confabulate to cover memory deficits (Schoenberg & Scott, 2011).
The next stage, moderately severe cognitive decline, or stage five, illustrates more
predominate and readily observed cognitive and behavioral deficits. In this stage, memory loss
becomes exceptionally prominent and deficits in semantic memory are readily observable
(Reisberg et al., 1982; Rogers et al., 2006). Additionally, functional communication becomes
difficult and, in some cases, impaired (Fromm & Holland, 198). Independence in ADLs also
decreases significantly to the extent that at least some assistance is generally necessary to
complete routine tasks like bathing or dressing (Galasko et al., 2005). Behaviorally, depressive
and anxious symptoms begin to arise more prominently, especially for individuals with vascular
dementia (Schoenberg & Scott, 2011).
Stage six, severe cognitive decline, involves complete dependence on others to perform
activities of daily living (Reisberg et al., 1982). Memory deficits increase and expand to include
forgetfulness of family members and very recent events (Reisberg et al., 1982). However,
memories of early life are generally intact. Language deficits are more pronounced with
significant difficulty in verbal comprehension and participation in communication with some
dementias showing decline in non-verbal communication (Rousseaux, 2010). In Alzheimer’s
disease, confrontation naming becomes markedly impaired along with verbal fluency and
semantic fluency (Schoenberg & Scott, 2011). Behaviorally, personality and emotional changes
become more prevalent. Agitation, confusion, wandering, apathy, and emotional blunting are
common manifestations of symptoms for Alzheimer’s disease at this stage (Schoenberg & Scott,
2011).
6
The last stage, stage seven, is indicated by the inability to communicate with others and
complete assistance in activities of daily living (Hendryx-Bedalov, 2000). For individuals with
Alzheimer’s disease, psychomotor deficits appear, and individuals experience the hallmark of
global cognitive impairment for Alzheimer’s dementia: agnosia, apraxia, and aphasia (Kramer &
Duffy, 1996). Behaviorally, individuals with Alzheimer’s dementia may begin to experience
hallucinations and delusions (Schoenberg & Scott, 2011).
Dementia treatment is economically brutal. Caring for individuals with dementia has a
national cost of $305 billion a year with $244 billion worth of unpaid care provided by a family
caregiver (Alzheimer’s Association, 2020). Because of the economic and emotional burden that
accompanies caring for a loved one, many family members elect to place their loved one in
residential communities (Etters et al., 2008). In fact, about 40% of individuals residing in nursing
homes have a diagnosis of Alzheimer’s disease or another dementia (Caffrey et al., 2010). The
role of nursing home caregivers is to provide a caring and stimulating environment to their
residents in an effort to increase their QoL despite significant cognitive changes (Allen, 2011).
Quality of Life for PwD
Quality of life is best understood as the combination of three domains: social,
psychological, and physical (Brod et al., 1999). It is an evaluative term to understand the
perception of one’s life in general (Brod et al., 199). Therefore, this multidimensional concept
has different implications for individuals of different backgrounds. When asked subjectively
about what factors are most important for QoL, PwD reported improved mood, engagement in
pleasant activities, and the ability to perform ADLs (Logsdon et al., 2008). Interestingly, when
caregivers were asked the same question for their loved one, caregivers selected improved mood,
engagement in pleasant activities, and cognitive functioning as the most important (Logsdon et
7
al., 2008). The overlap in answers implicates the role of mood and engagement in pleasant
activities as important means to improve QoL.
Mood. As mentioned above, one factor that greatly influences QoL for PwD is mood.
Emotional changes, including increased depression and anxiety, are common symptoms of
dementia during all stages (Reisberg et al, 1982). PwD also experience higher rates of loneliness
and sadness than their less cognitively impaired peers (Holmen et al., 1999). Moreover, when
investigating individuals residing in nursing homes over a period to time, reductions in QoL
were predicted by increases in depression, anxiety, and cognitive deterioration (Hoe et al., 2009).
Because of the overwhelming role mood plays in QoL, intervention efforts have shifted toward
understanding how nonpharmacological interventions increase mood, and thus QoL, for PwD.
Engagement in Pleasant Activities. Engagement in purposeful and stimulating activities
was also noted as a significant domain for increasing QoL for PwD. The need to engage in
meaningful and stimulating activities is innate to all humans, related to health and survival, and
does not dissipate with age or cognitive deterioration (Wilcock, 1993). In fact, for PwD,
engagement in activities is positively correlated with pleasure and unrelated to affect and
agitated behavior (Cohen-Mansfield et al., 2012). Furthermore, meaningful participation in
activities indicates independence, the development of satisfying relationships, and the
maintenance of positive well-being (Chung., 2004).
Cognitive functioning plays a significant role in frequency and range of engagement.
Individuals with lower levels of cognitive impairment actively engage in activities for a longer
amount of time, are able to attend to the stimuli longer, and have a more positive attitude toward
stimuli compared to their more cognitively impaired peers (Cohen-Mansfield et al., 2010).
Furthermore, the activities of which PwD attend to differ depending on cognitive functioning.
8
Mildly impaired individuals spend most of their time involved in therapeutic or leisure activities
with less time spent engaging in passive activities (Chung, 2004). However, PwD exhibiting
severe cognitive deficits demonstrate higher frequencies of engagement in socially withdrawn
behaviors, negative behaviors, and self-stimulation (Chung, 2004). Furthermore, behaviors that
promote socialization and well-being are generally sparse (Chung, 2004). Taken together, the
evidence suggests that as cognitive impairment increases, engagement in activities decreases,
leaving more severely impaired individuals vulnerable to the negative effects of reduced
engagement, such as lower QoL.
Cognitive Training Programs
One non-pharmacological intervention that has been substantially studied in the literature
is cognitive training. These programs involve the standardized presentation of tasks designed to
increase the functioning of cognitive domains typically impaired by dementia such as memory,
attention, and problem solving (Bahar-Fuchs et al., 2013). The theory behind these programs is
that the effects of practice will generalize to different settings beyond the training courses
(Bahar-Fuchs et al., 2013).
The research on the effectiveness of these programs has generally been geared
demographically toward PwD experiencing mild to moderate impairment. Additionally, primary
outcome measures tend to focus on how these programs improve cognitive functioning with less
regard to behavioral changes. A wide body of literature suggests increases in global cognitive
functioning are relatively common while improvements in specific domains are less reliable
(Kallieo et al., 2017).
However, some researchers have investigated the impact that these programs have on
both cognitive and behavioral outcomes for individuals with mild to moderate impairment.
9
Giovagnoli and colleagues (2017) randomized 64 participants into three treatment groups:
cognitive training, music therapy, and neuroeducation. Neuropsychologically, the cognitive
training group showed clinical improvements for verbal initiative and episodic memory. The
music therapy and neuroeducation groups showed no changes to any cognitive measures.
Behaviorally, self-reports on anxiety, depression, and socialization indicated improvement in all
groups for trait anxiety and depression while the music therapy and neuroeducation groups
showed a significant increase of interpersonal relationships. This study suggests that for
individuals with mild to moderate impairment, a cognitive training program was able to
positively impact mood, thus potentially increase QoL.
Significantly less research has been conducted on the cognitive and behavioral outcomes
for individuals with moderate to severe cognitive impairment. A study conducted by Mate-Kole
and colleagues (2007) investigated the effects of a cognitive training program on six older adults
with a diagnosis of moderate to severe dementia. Participants were administered a variety of
neuropsychological assessments measuring overall cognitive ability along with specific cognitive
domains including attention and concentration, memory, language, spatial skills, reasoning, and
visual motor speed and tracking. Additionally, behavioral outcomes were assessed to measure
depression, quality of life, and functional activities through caregiver report. All participants
were administered the battery of tests before and after the six-week intensive training course
which used a combination of two established cognitive training programs. One program was an
interactive group training class utilizing hands-on activities focused on memory, attention,
cognitive flexibility, manual dexterity and problem solving. In conjunction with this, a
computerized training program was used to train attention, visual spatial and motor-skills,
problem solving, memory, and visual discrimination. Results indicated improvements in global
10
cognitive functioning along with short-term memory, visual motor coordination, psycho-motor
speed, cognitive flexibility, and attention. Furthermore, positive behavioral changes were
observed with improved socialization and initiation, alertness, and affect reported qualitatively
by nursing home staff. These, taken with improvements in functional activities, signify a positive
impact to QoL (Mate-Kole et al., 2007).
A similar study conducted by Buchanan and colleagues (2019) aggregated data from four
facilities who independently implemented the same cognitive training program. Twenty-three
individuals experiencing moderate impairment participated in the study. Cognitive factors
including global cognitive ability, attention, visual and verbal memory, visual spatial skills,
processing speed, executive functioning, and language along with behavioral factors
encompassing depression, QoL, agitated behavior, and daily functioning were measured prior to
and after the implementation of the cognitive training program. Results indicated improvements
in general cognitive functioning, divided attention, immediate verbal recall, immediate visual
recall, visual recognition, perceptual speed, and executive functioning. No improvements to
behavioral outcomes were found. Moreover, small declines in agitated behavior, QoL, and daily
functioning were observed. The authors report that this is consistent with previous research
suggesting that the benefits gained from an increase in cognition do not always generalize
everyday functioning (Buchanan et al., 2019).
The ambiguity and inconsistency in outcome measures regarding behavioral functioning
after participation in a cognitive training program makes it difficult to conceptualize the
behavioral effects of the program. However, it is important to approach the question of changes
to behavioral function through different methodologies to truly determine the effects. A different
11
methodology and means of data collection may better capture behavioral changes resulting from
engagement in these programs than caregiver and self-report measures would.
Purpose of the Current Study
Most studies investigating the effects of cognitive training primarily measure the effects
these programs have on cognition with a secondary emphasis on measuring important non-
cognitive constructs such as affect, QoL, or engagement. When non-cognitive constructs are
measured, they typically rely on caregiver reports, given that PwD may be unable to complete
self-report measures reliably and accurately. Although caregiver reports are valuable sources of
information, they may be biased and thereby inaccurately report the experience of the PwD. For
example, research suggests that patient/caregiver agreement is impacted by several patient
factors including functional and cognitive level, years of education, and severity of depression
(McPhail et al., 2008; Williams et al., 2006), while caregiver factors impacting patient/caregiver
agreement are caregiver age, financial situation, and valuation of life as a whole (Arons et al.,
2013).
Understanding the effects of cognitive training on measures of cognition is important but
may only capture some of the potential benefits of these programs, particularly for more severely
impaired individuals. For example, it may be unrealistic to expect that cognitive training will
greatly affect cognitive functioning in more severely impaired PwD. Alternatively, cognitive
training programs may have benefits on non-cognitive outcomes such as affect or increased
engagement in activity, which are important factors related to QoL. Therefore, it is important to
determine if cognitive training programs produce benefits on measures of behavior and affect. In
addition, it is important to determine if cognitive training programs produce greater
12
improvements on these measures of behavior and affect compared to less expensive and less
time-consuming activities typically offered in long-term care settings.
Consequently, the purpose of this study was to investigate the effect of a cognitive
training program on mood and engagement for older adults with moderate to severe cognitive
impairment. One novel aspect of this study was that it utilized direct observation methods to
measure affect and engagement during activities as they were occurring. The use of direct
observation allowed for the collection of direct samples of participant behavior during activities
to determine how participants responded to the content of the program. The methodology has the
advantage of minimizing the problems associated with retrospective caregiver reports of
participant behavior during the cognitive training classes. Another purpose of the study was to
determine if cognitive training programs produced greater engagement and positive affect
compared to activities typically offered in long-term care facilities. There were two main
hypotheses. First, it was hypothesized that the cognitive training program would elicit more
active engagement, thereby reducing passive and no engagement, compared to engagement
observed in normally scheduled activities. Second, it was hypothesized that the cognitive training
program would increase positive affect, thereby decreasing negative and neutral affect, compared
to affect observed in normally scheduled activities.
Methods
Participants
Participants were recruited from an assisted living facility in Southern Minnesota. Prior to
recruitment, facility staff familiar with the resident community were asked to identify individuals
they believed met inclusion criteria. To qualify for the study, participants had to be either
experiencing moderate to severe cognitive impairment as evidenced by a score of 77 or below on
13
the Modified Mini Mental State Examination (Teng & Chui, 1987) or have a diagnosis of
dementia established by a review of medical records by staff. Additionally, participants had to
regularly attend activities presented by the activities department. Participants were not
considered for the study if they met any of the following exclusion criteria:
1. The presence of a serious health problem, other than dementia, that could compromise
their ability to participate in the cognitive training classes.
2. The presence of a significant disabilities (e.g. blindness, deafness, a significant language
impairment) that could prevent the individual from participating.
3. The regular use of medications that could potentially affect functioning such as narcotic
analgesics. A caveat to this is that participants could participate if they were taking the
medications specifically for dementia (e.g. cholinesterase inhibitors) and met the
inclusion criteria.
4. If the individual’s level of cognitive functioning was not severe enough (i.e. above a 77
on the 3MS).
Five participants met inclusion criteria and had a family member provide informed consent to
participate in the study (See Appendix A for copy of consent form). All participants, four of
which were female, identified as Caucasian. Four of the five participants resided in the memory
care unit while one participant lived on the assisted living side of the same residential facility.
Ages of the participants ranged from 87 to 91 (M = 89.2, SD = 1.48). Participant’s 3MS score
ranged from 23 to 66 (M = 39, SD = 19.92). Two participants had a medical diagnosis of
unspecified dementia without behavioral disturbance while one had a diagnosis of unspecified
dementia with behavioral disturbance. One participant had a comorbid diagnosis of unspecified
dementia without behavioral disturbance and Alzheimer’s disease. Lastly, one participant had no
14
dementia diagnosis, but scored in the severe range for cognitive impairment on the 3MS. See
Table 1 for a summary of participant information.
Table 1
Participant Characteristics
Participant
Age
Gender
Diagnosis
3MS Score
Participant 1
90
Female
No Diagnosis
23 (severe
range)
Participant 2
87
Female
Unspecified dementia without
behavioral disturbance
42 (severe
range)
Participant 3
91
Female
Unspecified dementia with
behavioral disturbance
66 (moderate
range)
Participant 4
89
Female
Unspecified dementia without
behavioral disturbance and
Alzheimer’s Disease, Unspecified
N/A
Participant 5
89
Male
Unspecified dementia without
behavioral disturbance
25 (severe
range)
After the five participants were recruited, staff members familiar with the participants
were recruited to complete questionnaires and interviews on behalf of the participants. In order
to meet inclusion criteria, the staff member had to have worked with the resident nearly every
day for at least two months. One staff member was identified as meeting the criteria. This staff
member consented to participate in the study (see Appendix B for a copy of the consent form).
Materials
15
Cognitive Assessment. Severity of cognitive impairment was assessed using the
Modified Mini-Mental State Examination (3MS; Teng & Chui, 1987), a widely used screening
tool for dementia. This is a standardized measurement that assesses several cognitive domains
(e.g. immediate and delayed memory, executive function, language, and visual spatial ability).
Scores on the 3MS range from 0-100, with lower scores being indicative of more severe
cognitive impairment. Key psychometric properties of the 3MS have been widely established.
Internal consistency was high for individuals with dementia (α= 0.88; Tombaugh et al., 1996)
with test-retest reliability being excellent (0.91 to 0.93; Teng et al., 1990). It is also a valid test
with high sensitivity (.96) in differentiating individuals with severe dementia from their non-
affected peers (Tombaugh et al., 1996).
Engagement. In order to measure the degree to which participants were engaged in
activities, direct observation methods were used based on previous studies (e.g. Judge, Camp &
Orsulic-Jeras, 2000). Three levels of engagement were assessed. Active engagement was the
main outcome variable for engagement and was operationally defined as “any motor or verbal
behavior exhibited in response to the activity in which the client was taking part” (Judge, Camp
& Orsulic-Jeras, 2000, pg. 43). Examples of active engagement include singing-along during
music therapy or placing a marker while playing Bingo. Passive engagement was operationally
defined as, “listening and/or looking behavior exhibited in response to the activity the client was
participating in” (Judge, Camp & Orsulic-Jeras, 2000, pg. 43). Examples of passive engagement
include actively watching another person complete the activity or looking at a person talking
during a discussion. Non-engagement was operationally defined as “staring off into space or
another direction away from the activity, sleeping, or any motor and or/verbal behavior activity
in response to an activity the client was not currently participating in” (Judge, Camp & Orsulic-
16
Jeras, 2000, pg. 43). Examples of non-engagement include staring at the floor or wall and
playing with own clothing.
Affect. In order to measure affective responses to activities, direct observation methods
were used based on previous literature (e.g. Lawton, Van Haitsma, & Klapper, 1996).
Definitions of affective responses fell into two main categories: positive and negative. Positive
affect was the primary outcome variable for affect and was operationally defined as the
participant showing any overt signs of pleasure such as smiling, laughing, or nodding along.
Negative affect was operationally defined as any overt signs of displeasure such as clenched
teeth, physical aggression, furrowed brow, crying, moaning, or mouth turned down at the
corners. If the participant did not demonstrate overt signs of positive or negative affect, neutral
affect was recorded.
Quality of Life. The QUALIDEM is a 40-item questionnaire analyzing nine areas of
functioning related to quality of life for individuals with dementia (Ettema, Droes, de Lange,
Mellenbergh, & Ribbe, 2007). The nine areas assessed include: care relationship, positive affect,
negative affect, restless tense behavior, positive self-image, social relations, social isolation,
feeling at home, and having something to do. The QUALIDEM is completed by a staff member
who works closely with the resident and has been found to have adequate internal consistency (α
= 0.59 to 0.89).
Qualitative data. In an effort to understand idiographic changes in participant
engagement, affect, and quality of life, members of the assisted living home staff were asked
open-ended questions regarding participant engagement in day-to-day activities, general
communication with staff and peers, emotional expression, and general emotional state. To
ensure the care member had acute knowledge of the participant’s general functioning, only
17
caregivers who had a working relationship with the participant for at least two months were
interviewed. Refer to Appendix C for interview questions.
Research Design
An alternating treatment design with a baseline phase was employed. The alternating
treatment design utilizes rapid alteration of two conditions with a single subject where each time
a client is observed they are receiving the opposite condition (Barlow & Hayes, 1979). In this
case, the alternating conditions were the cognitive training sessions and regular activities.
Procedure
Observer training. Three research assistants were trained to collect data by the primary
researcher. Procedures for observer training were adapted from Hartman & Wood (1990). The
first step included orientation and learning the observation training guide. During this stage, all
assistants met to discuss the setting, personnel to contact at the assisted living facility, number of
participants, informed consent, and confidentiality. Assistants were also provided a manual that
outlined the study design, operational definitions, and observation schedule as well as an
example of the record form. Assistants memorized the operational definitions and observation
schedule for a quiz the following week. The second step was the first criterion check. During this
phase, assistants were quizzed on the operational definitions and observation schedule. The
assistants and primary researcher also discussed different hypothetical observations and
concluded how those situations would be recorded. Once informed consent was attained for the
participants, all assistants began the third step which involved in situ practice. Practice
observations were conducted during participant’s participation in regularly scheduled activities.
All assistants collected data in groups of two until all assistants felt comfortable with the
observational process. All observers attended between one and four practice sessions.
18
Throughout the data collection process, retraining and recalibrating sessions occurred as needed
to discuss discrepancies or questions in the data and to assess for observer drift and bias.
Baseline. Direct observation sessions were conducted over a two-week period during
regularly scheduled activities at the assisted living facility. The activities director informed the
researchers as to which regularly scheduled activities were frequented by participants.
Behavioral observations were conducted for the entirety of each activity, which typically lasted
between 15 and 60 minutes. Researchers recorded affect and engagement using a 10-second
partial interval observation schedule, which included five seconds between each interval in order
to record data. If more than one study participant attended the activity, the researcher recorded
data for the individual closest to their right-hand first, then rotated clockwise until all participants
had been observed. This procedure was repeated until the activity was completed.
During baseline, participants were administered the 3MS and a staff member was asked
to complete the QUALIDEM and the qualitative interview.
Experimental Phase. The experimental phase involved collecting data during cognitive
training classes as well as regularly attended activities. The cognitive training program was
conducted three days per week over eight weeks. In order to obtain an adequate sample of data
throughout the cognitive training class, data were scheduled to be collected during the following
cognitive training classes: 1) classes 1 or 2; 2) classes 5 or 6; 3) classes 10 or 11; 4) classes 15 or
16; 5) classes 20 or 21; and 5) classes 23 or 24. Throughout the experimental phase, observations
during regularly scheduled activities continued in the same manner as was done in baseline.
Observations occurred in conjunction with cognitive training observations, and, when possible,
occurred on the same day as cognitive training sessions. When this was not possible, regularly
scheduled activities were observed one to two days after cognitive training sessions. Observation
19
procedures used during the baseline phase were also used during the experimental phase of the
study.
Interobserver Agreement. A second observer independently collected data at 16% of
the data observation sessions. Interobserver agreement (IOA) scores were calculated on the
dependent variables of engagement and affect separately. Interval agreement was calculated by
dividing the number of agreements per interval by the number of agreements plus disagreements
and multiplying by 100. A score of 80% indicates standard agreement. The mean IOA for
engagement was 88.9% (SD = 11.12; range, 75.4% – 100%) while the mean IOA for affect was
97.0% (SD = 4.80; range, 89.9% – 100%).
Intervention
Active Mind (AM) was the cognitive training program utilized in the experimental phase
of this study. AM was developed by the New England Cognitive Center (NECC), a non-profit
company with the goal of creating and disseminating innovative brain training programs to
maximize mental functioning for older adults with and without cognitive impairment. AM was
created for individuals with moderate to severe cognitive impairment. The program consists of
24, one-hour group training sessions that occurred three times a week for eight weeks. The
manualized sessions incorporated a variety of paper and pencil activities targeting cognitive
domains that are often impaired for individuals with dementia including reaction time,
psychomotor speed, attention and concentration, memory, visual-spatial acuity and language
along with problem solving and executive functioning. The activities were created to be
challenging, enjoyable, social, and appropriate for adults. All in-session activities were
constructed in a way that requires little time for teaching, so more session time can be devoted to
engaging in activities. Each activity requires 5 to 12 minutes to complete. Additionally, all
