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
Theses, Dissertations, and Other Capstone
Projects
2019
The Effects of Elderspeak on the Mood of Older Adults with
The Effects of Elderspeak on the Mood of Older Adults with
Dementia: A Preliminary Report
Dementia: A Preliminary Report
Kenia Torres-Soto
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
Torres-Soto, K. (2019). The effects of elderspeak on the mood of older adults with dementia: A
preliminary report [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/931/
This Thesis is brought to you for free and open access by the 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 Elderspeak on the Mood of Older Adults with Dementia: A Preliminary Report
by
Kenia Torres-Soto
A Thesis Submitted in Partial Fulfillment of the Requirements for the Degree of
Master of Arts
In
Clinical Psychology
Minnesota State University, Mankato
Mankato, Minnesota
May 14th, 2019
May 14th, 2019
The Effects of Elderspeak on the Mood of Older Adults with Dementia
Kenia Torres-Soto
This Master’s Thesis has been examined and approved by the following members of the student’s
committee.
_________________________________________
Jeffrey Buchanan, Ph.D., Chairperson
_________________________________________
Daniel Houlihan, Ph.D., Committee Member
_________________________________________
Kristen Anderson, Ph.D., Committee Member
i
Abstract
Barriers to effective and harmonious communication between caregivers and those affected with dementia
are common and inevitable. An example of a common barrier is aphasia, which is a prevalent
communication deficit associated with dementia. The social environment may further hinder harmonious
communication through the use of well-intended, but ineffective speech patterns. Elderspeak (ES), which
is infantilizing speech directed at older adults, is one such speech pattern that is commonly used in long-
term care facilities (LTC) and is related to negative outcomes for older adults receiving it. Older adults
with mild to moderate cognitive impairment who were residents at a LTC facility were exposed to two
videos depicting a typical interaction between a nursing staff at a LTC facility and a resident. Prior to
using the videos with older adults, they were validated for accuracy by nursing staff at a LTC facility. The
videos depicted a “neutral” interaction (N-ES) and one that used elderspeak. Self-reports of mood were
collected for older adults before and after each video. Behavioral observation of affect was collected
while the older adults watched the videos. Qualitative interviews investigating preferences and opinions
were administered after each video. Contrary to existing literature, results indicated that the participants in
this study had similar emotional and behavioral responses to both videos. Further research is necessary in
order to more fully determine what contextual variables affect how individuals with dementia respond to
elderspeak.
ii
Table of Contents
Introduction ……………………………………………………………………………………………………………..1
Method ……………………………………………………………………………………………………………………8
Results ……………………………………………………………………………………………………………………13
Discussion ………………………………………………………………………………………………………………19
References ………………………………………………………………………………………………………………25
Tables
1. Sample Characteristics………………………………………………………………………………………9
2. Mood Avergaes…………………………………………………………………………………….16
3. Affect Behavioral Observation…..………………………………………………………………17
4. Phase II Interview…………………………………………………………………………………………………..18
Figures
1. The Communication Predicament of Aging Model……………………………………………………..5
2. Positive and negative change scores compared between video conditions……………………..16
Appendices
A. Video Scripts………………………………………………………………………………………………………..28
B. Phase I Interview…………………………………………………………………………………………………..34
C. Mood Measure……………………………………………………………………………………………..36
D. Phase II Interview………………………………………………………………………………………….37
E. Affect Recording Instrument………………………………………………………………………………..38
F. Phase I Informed Consent Form………………………………………………………………………………39
G. Phase II Participant Informed Consent Form…………………………………………………………….41
H. Phase II Guardian Informed Consent Form………………………………………………43
1
The Effects of Elderspeak on the Mood of Older Adults with Dementia: A Preliminary Report
According to demographic census data from 2010, the number of older adults (i.e., age 65 and
older) in the U.S. is projected to increase by 92.3% between 2016 and 2060 (US Census Bureau, 2018).
Furthermore, the number of individuals 85 years and older is projected to increase by roughly 200%, and
by over 600% for centenarians within that timeframe. By 2030 one in five Americans is projected to be an
older adult, with that year being the first in which older adults outnumber children.
The demographic changes projected to take place by 2030 will mark a societal shift for the U.S.,
with one key feature being a change from a youth-dependent to predominantly elderly-dependent society.
In contrast to the youth dependency ratio projections for the upcoming decades, the old-age dependency
ratio is estimated to increase considerably (US Census Bureau, 2018). For 2020, there may be three-and-
a-half working-age adults for every older adult qualifying for retirement, decreasing to two-and-a-half
working-age adults to one older adult for 2060. These changes, which will be new to the U.S. compared
to other countries (e.g., Japan, Canada, and European countries), suggest that more time and resources
may need to be allocated to the aging population (US Census Bureau, 2018).
According to the World Alzheimer’s Report by the International Association for Alzheimer’s
Disease (2018), there are over 50 million individuals with a dementia diagnosis worldwide. This number
is estimated to increase to 152 million by 2050. This prediction is congruent with the estimated increase
in the older adult demographic. Hence, a proportion of the required resources allocated to this population
may include long-term care services. Although many challenges exist between older adults with dementia
and professional caregivers in long-term care settings and one prominent challenge is enhancing effective
communication between caregivers and older adults.
The Importance of Communication
2
Dementia inevitably leads to progressive declines in various abilities (e.g., attention, memory,
motor skills), which can obstruct harmonious interactions between affected individuals and their
caretakers. In particular, dementia causes declines in expressive and receptive language skills, which
makes it difficult to comprehend what others are saying and impedes effective communication of needs
and preferences. Due to the persistent retrograde nature of dementia, it is important to consider what
impact the social environment may have on the behavior and quality of life of the affected. Although
“language” and “communication” are large constructs that can include a wide array of specific skills,
existing research has uncovered a number of communication variables that directly influence the quality
of interactions between caregivers and persons with dementia. For example, Christianson et al. (2007)
examined what type of commands issued by caregivers maximized the likelihood of compliance when
working with older adults with dementia. Results indicated that alpha commands (i.e., straightforward,
concise, and feasible) accounted for more compliance than beta commands (i.e., ambiguous, interrupted,
and complicated). Compared to alpha commands, beta commands resulted in a lower frequency of
compliance and a higher frequency of noncompliance and forced compliance. Commands that were stated
directly, that repeated or clarified a previous command, and that were repeated exactly were found to
produce better compliance. Questions that required a motoric response were found to produce relatively
poor compliance. Collaborative commands (e.g., “let’s brush our teeth”) were found to result in equal
rates of compliance and noncompliance and high rates of forced compliance. This finding is consistent
with previous literature indicating that this communication style may be related to resistiveness to care
(Williams et al., 2008). Overall, this study has implications for training caregivers to issue more specific
and effective instructions.
Gentry and Fisher (2007) investigated whether responses of conversation partners contributed to
excess verbal deficits in individuals with Alzheimer’s disease (AD) by indirectly punishing verbal
behavior. There is evidence that indirect repairs used by caregivers when patients make mistakes
reinforce patient efforts to communicate verbally while direct repairs may serve to punish these efforts. In
3
an indirect repair, the listener paraphrases the speech of the person with AD. In a direct repair response,
the listener interjects with corrective feedback. Use of indirect repairs was associated with more words
spoken, longer speech duration, fewer topic changes, and fewer incomplete interactions compared to
direct repairs.
Caretakers may present additional barriers to communication, compounding those already
existing due to abilities deficits. For example, caretakers are often faced with the need to balance heavy
workloads and provide humanistic care. This struggle can lead to care that is largely task-driven
(Williams, 2006). Personal stressors and cultural differences may also provide barriers to harmonious care
(Williams, 2006). Picture an elderly nursing home resident, with the characteristic verbal manifestations
of dementia, interacting with a nursing staff during activities of daily living (ADL). This may look like
two individuals struggling to understand one another during a task (e.g., grooming) under time
constraints. One can presume that frustration will develop in both parties. It is then easy to see why it may
be tempting for a caretaker to avoid frustration on both sides by simply completing the task him/herself.
However, engaging residents in these ADLs has been empirically shown to be related to longevity in the
affected (Lichtenstein, Federspiel, & Schaffner, 1985; Ryan, Hummert, & Boich, 1995). At the same
time, resident frustration during these activities may lead aggression, noncompliance, negative mood,
social isolation, idleness, and dependence.
It is also important to consider how resident-caretaker communication may unintentionally affect
dependency. The literature suggests that LTC staff communication typically reinforces dependent
behavior (Baltes & Wahl, 1996). This evidence suggests that dependence behaviors may be indirectly
reinforced by caregiver attention, while those who appear to be independent and strong receive less
attention. In other words, over time a pattern of dependence may emerge as a function of obtaining
attention. Dependence-support communication may also indirectly establish a perception of imbalance of
power and project a message of control over rather than that of rehabilitation and independence
(Lanceley, 1985). Fortunately, there is evidence that training staff to avoid dependence-support
4
communication reduces resident depression and problem behaviors, and improves staff retention (Ryan et
al., 1995).
Elderspeak. An example of a dependence-support communication style that is relevant to elderly
individuals, but particularly to those dwelling in LTC facilities, is elderspeak. Elderspeak is a simplified
form of speech that resembles “baby talk” and is typically directed towards older adults (Williams, 2006).
This communication style is characterized by exaggerated intonation, simplistic vocabulary and grammar,
elevated pitch and volume, inappropriate use of diminutives, and use of collective versus first person
pronouns, among others (Ryan et al., 1995; Williams, 2006). Although the use of elderspeak may reflect
caretakers’ attempts to communicate effectively and provide affection to the individuals receiving it, the
communication predicament of aging model attempts to explain how it may be failing to do that and
contributing to negative outcomes.
The communication predicament of aging model (CPAM), introduced by Ryan et al. (1995), was
developed from the communication accommodation theory. The communication accommodation theory
suggests that people adjust their communication style according to who they are speaking in an attempt to
achieve effective communication. The CPAM goes further by modifying this model for older adults,
suggesting that this tendency may result from erroneous assumptions of dependence and/or incompetence
in this population (see Figure 1). In line with negative feedback models, the CPAM suggests that such
communication modification may reinforce dependence behaviors and constrain opportunities for
satisfying communication in the elderly. This may negatively affect self-esteem and psychological well-
being of older adults (Ryan et al., 1995).
As seen in Figure 1, individuals encountering elderly persons first recognize old-age cues (e.g.,
using a walker, gray hair, posture, setting of interaction, social role). While it is possible for individuals to
hold positive and negative stereotypes for older adults, this model specifically refers to the eliciting of
negative stereotypes. Thus, the recognition of old-age cues elicits negative stereotypes associated with
5
older adults, such as dependency, incompetence, and cognitive decline, which can lead the conversant to
modify their speech according the stereotypes. According to the CPAM, these erroneous assumptions are
linked to age-dependent overaccommodations of patronizing communication.
Figure 1. The Communication Predicament of Aging Model (Ryan et al., 1995)
While this communication style is clearly destructive at face value, the implications of its use are
profound. The literature surrounding elderspeak revolves around the paradox that while caretakers may
use it to convey care and warmth, this patronizing communication style projects messages of
incompetence, helplessness, and dependence that can negatively affect the self-esteem, personal identity,
and psychological well-being of those receiving it (Williams, 2006). This may be particularly true for
older adults dwelling in LTC facilities whose self-esteem may already be threatened by the setting and
6
context of their social interactions. A brief review of this literature on the negative effects of elderspeak is
provided.
Caporael (1981) investigated the prevalence of an intense form of elderspeak, labeled secondary-
baby-talk, in a nursing home setting and found that it was used considerably (up to 20% of the time).
Speech recordings of secondary-baby talk were content-filtered and judged to be characteristically
identical to primary baby-talk that is used with children. Moreover, Caporael (1981) found that nurse
ratings of residents’ cognitive abilities did not predict whether secondary baby-talk was used.
There is also evidence suggesting that elderspeak may be a precursor to resistance to care (e.g.,
pushing, swearing), which is common in older adults with dementia (Cunningham & Williams, 2007).
Williams et al. (2008) assessed interactions between nursing home residents and staff during ADLs to
evaluate the effects of elderspeak. The results indicated that elderspeak led to an increased probability of
resistance to care compared to normal speech.
Elderspeak has also been judged to be condescending, disrespectful, and unwelcomed by
individuals receiving it. La Tourette and Meeks (2000) compared community-dwelling older adults and
nursing home residents regarding their perceptions of elderspeak and neutral speech. Both nursing home
residents and community residents rated the nurse more favorably and the resident more satisfied in a
vignette depicting neutral speech compared to a vignette depicting elderspeak. Community residents also
viewed the resident as less competent in the elderspeak vignette.
Other studies have investigated contextual factors that increase the likelihood that elderspeak is
used. According to Lombardi et al. (2014) elderspeak may be considered more acceptable in institutional
settings such as hospitals or nursing homes. For example, ES was found to be more acceptable for
residents older than 70 and those with mild to severe memory problems or are disoriented. In contrast, ES
was found to be less acceptable for use with residents who appear to be angry compared to residents who
appear happy or sad. In regard to the perceived acceptability based on relationship to the resident, results
7
indicated that the use of ES is more appropriate when nurses had regular interaction with residents,
compared to infrequent or no interactions. The use of ES was also found to be less appropriate when other
residents or family members were present during nurse-resident interaction. Finally, there was a
significant difference in perceived appropriateness of the use of ES regarding task-type, such that it was
rated more appropriately when the task was hands-on (e.g., grooming) compared to a hands-off
interaction.
Balsis and Carpenter (2005) investigated the effects of the speaker’s age and relationship to older
adults on participant perceptions of those using and receiving elderspeak in dialogue vignettes. Results
indicated that individuals using and receiving elderspeak were viewed negatively regardless of the age of
the speaker or their relationship to the recipient. The speaker was evaluated as having a worse demeanor
in the elderspeak than in the neutral speech vignette. Individuals receiving elderspeak were viewed as
having a worse mood and decreased ability.
Although there is evidence suggesting that elderspeak is an undesirable and possibly ineffective
form of communication, there is also evidence for some beneficial aspects to using elderspeak. Kemper
and Harden (1999) investigated the effectiveness of using elderspeak by having older adults watch a
video that described a route that was traced on a map. The results indicated an improvement in
performance and reports that the instructions were easier to follow when the speaker reduced the
grammatical complexity (e.g., minimized the number of subordinate and embedded clauses) and used
semantic elaboration (i.e. repeated and expanded upon what was said). These findings suggest that some
features of elderspeak may improve older adults’ ability to follow and comprehend directions. Reduced
grammatical complexity, frequent repetition, and slower rate may also be beneficial when communicating
with individuals with dementia (Williams, 2006).
Mood
8
Guzaman-Valez et al. (2014) investigated whether feelings could persist in persons with AD,
even after their declarative memory for what caused the feelings had faded. Participants experienced two
emotion-induction conditions in which they watched film clips intended to induce feelings of sadness or
happiness. Real-time emotion ratings were collected at baseline and at three post-induction time points
(immediately after, 10-15 minutes after, and 20-30 minutes after) and a test of declarative memory was
administered shortly after each condition. Results indicated that individuals with AD can experience
prolonged states of emotion that persist well beyond their memory for the events that originally elicited
the emotion. The preserved emotional life evident in persons with AD has important implications for their
management and care and highlights the need for caretakers to foster positive emotional experiences.
Purpose of the Study
Relatively little research has been devoted to understanding how individuals with dementia
respond to elderspeak, particularly those individuals with more severe cognitive impairment. Therefore,
the purpose of this study is to expand on the existing literature regarding the effects of elderspeak on the
mood of older adults with moderate to severe dementia. More specifically, the purpose of the study is to
determine: 1) how elderspeak affects mood in persons with dementia when compared to speech that does
not include elements of elderspeak and, 2) if individuals with dementia have a preference for elderspeak
or communication that does not include elderspeak.
It is predicted that participants will exhibit more negative than positive mood and affect in the
elderspeak compared to the neutral speech video condition. It is also predicted that negative mood and
affect will increase, and positive mood and affect will decrease between pre and post elderspeak-video
exposure. Finally, it is predicted that the neutral speech video condition will not be related to significant
mood and affect changes between pre and post video exposure.
Methods
9
Participants & Setting
Phase I. Participants were seven nursing staff members at a LTC facility located in the Midwest
that serves retired nuns. Participants were recruited by asking administrators to identify direct care staff
who would be willing to participate in the study.
Phase II. Participants included individuals who had mild to moderate cognitive impairment and
resided in an assisted living facility. Participants were recruited by asking facility administrators or direct
care staff to identify residents that likely had cognitive impairment, and would enjoy interacting with
research staff, and who had sufficient verbal abilities to provide responses to questions. Refer to Table 1
for participant characteristics.
In total, 17 individuals were screened for participation in the study, seven were included, seven
were screened but did not qualify for inclusion, and three were withdrawn. In order to estimate the
severity of cognitive impairment, the Brief Interview for Mental Status (BIMS) was administered
(Chodosh, et al., 2008). The BIMS appraises cognitive function using temporal orientation and recall
items. The memory portion consists of immediate and delayed (after a period of approximately one-two
minutes after presentation) recall of three items (i.e., sock, blue, bed). For delayed-recall, individuals who
were unable to recall the items on their own were given category cues (i.e., “a color”). The orientation
portion included items inquiring about the current year, month, and day of the week. Scoring the BIMS
involves assigning whole number scores to each accurate item, which are then added to yield a sum score
(0-15). Participants scoring below 13, which indicates at least moderate cognitive impairment, were
included in this study. The mean BIMS score for those included (n=7) was 8.71 (SD= 2.98), with scores
ranging from 3 to 12 (Table 1).
Table 1
Sample Characteristics
10
Variable
Mean (SD)
Mode
N (%)
Age
Memory Unit
Diagnosis:
None
Unspecified
dementia (UD)
Mild cognitive
impairment
Mild
intellectual
disability
AD
Transient
ischaemic
attack
Total: 85.35 (8.81)
Included: 89.43 (8.58)
Total: No (88.20%)
Included: No (85.70%)
Total: None (29.40%)
Included: UD (42.90%)
5 (29.40%)
4 (23.5%)
2 (11.80%)
1 (5.90%)
1 (5.90%)
1 (5.90%)
BIMS
Total: 10.53 (4.02)
Included: 8.71 (2.98)
*Includes all screened participants
Materials and Instruments
Videos. Videos used in all phases and conditions of the study were scripted by students and
faculty with background in clinical psychology and nursing. Scripts were based on the professional
experience of nursing students who worked in nursing homes with elderly individuals with dementia.
Two videos were created that were similar in terms of content and length (~ 6 min in length) and differed
in terms of communication style used. The video scenarios presented an older adult, with some cognitive
and physical impairment interacting with a nursing staff in a long-term care facility. The interaction
11
depicted a student actor with experience as a CNA and an older adult actor as a nursing staff getting a
resident ready for lunch after a nap. Specifically, the scenario involves the nurse 1) arousing the resident
from a nap, 2) assisting the resident out of bed, 3) helping the resident put on socks, shoes, and a sweater,
4) assisting the resident to a sink and mirror to groom, and 4) assisting the resident to the door to so she
can go to lunch.
The first video (i.e., N-ES) demonstrated an interaction with “typical” or “neutral”
communication (e.g., neutral tone and volume, the use of titles and last names). In addition, in this video,
the nursing assistant provided the resident with opportunities to make choices (e.g., about clothes to wear,
whether she wanted to get out of bed).
The second video (i.e., ES) depicted an interaction where the nursing assistant used
communication that included elements of elderspeak (e.g., elevated pitch, terms of endearment, childish
vocabulary). In addition, the resident was provided limited choices in this video. Refer to Appendix A for
transcripts of the videos.
Staff interview. A nine-item interview (i.e., Phase I Interview) was used for Phase I of the study
to determine the accuracy and relevance of the scenarios depicted in the two videos. Sample interview
questions include: Have you witnessed this communication style in a work setting? Do you think older
adults living in long-term care settings will find the interactions in the video realistic or relatable? Refer
to Appendix B for the full interview.
Mood measure. For Phase II of the study, a seven-item mood measure was administered to
participants before and after they viewed each video to assess current mood (Guzman-Velez et al., 2014).
Participants rated how they “feel right now, at the present moment” by rating their experience on seven
emotions (sadness, anger, disgust, fear, anxiety, peacefulness, and happiness) on a nine-point scale that
ranges in intensity from “none” (0) to “extremely” (8). Refer to Appendix C for the full mood
instrument.
12
A self-created, 17-item interview (i.e., Phase II Interview) containing closed and open-ended
questions was used to gather qualitative data to supplement the quantitative data gathered from the mood
measure. Sample interview questions include: How much did you enjoy the video? [“not at all” (0) to
“extremely” (8)] What did you like about the video? What did you dislike about the video? Refer to
Appendix D for the full interview.
The Affect Rating Scale (ARS), an affect measure developed from the Philadelphia Geriatric
Center Positive and Negative Affect Rating Scales, was used as a model for behavioral observation
(Lawton, Van Haitsma, & Klapper, 1996). The ARS consists of six types of affects, three positive
(pleasure, interest, and contentment) and three negative examples (anger, anxiety/fear, and sadness).
Affect types are described in terms of facial expressions, vocalizations, and body movements, among
other behaviors. The affect descriptions from the ARS were combined two create two binary affect types
(positive and negative). These affect descriptions were used by observers, who recorded the incidence of
each affect type on a 15-second partial-interval recording system. Observers recorded whether a positive
and/or negative affect was observed at any time during each 15-second interval for the duration of the
videos (approximately six minutes in length). Additionally, observers recorded baseline affect for each
participant before the introduction of each video (approximately two minutes in length). Refer to
Appendix E for the affect recording instrument. Interobserver agreement (IOA) data was collected for at
least half of the sessions, with the primary investigator serving as the second observer. Interval-by-
interval analysis on 10 sessions revealed an agreement index of 95.4%.
Design
A quasi-experimental design was used for both phases, with individuals being exposed to one
repeated-measures condition in phase I and two repeated-measures conditions in phase II (2 x 2 repeated-
measures factorial design).
13
Phase I. This phase was conducted in one session per participant. Here, all individuals were
exposed to both videos in one session. The presentation order of the video conditions was
counterbalanced across participants.
Phase II. This phase was conducted in two sessions per participant, with the individual sessions
comprising a video condition. All participants experienced both conditions, the presentation order of
which was counterbalanced across participants.
Procedure
Phase I. This phase of the study required a sample of nursing staff members to view and provide
opinions about both videos to ensure they represent typical staff-resident interactions during activities of
daily living. Once consent was received, each video was presented to the individual nursing staff followed
by the administration of the Phase I Interview, twice per session. Each session took approximately 30
minutes to complete.
Phase II. Once consent was received from a family member or legal guardian, but prior to
beginning the experimental procedures, assent was obtained from the participant. If assent was given,
phase II of the experiment began.
All data collection took place in participants’ personal rooms, with the exception of one session,
which took place in a secluded common area of the nursing home. The trained researcher first
administered the BIMS to potential participants, whose inclusion was dependent on the score they
obtained (i.e., included if score was ≤ 13).
Prior to watching each video, participants were administered the mood measure. During this time,
a research assistant collected baseline affect data from behavioral observation using the binary ARS
descriptions (~two minutes in length). The videos were displayed on a 15-inch laptop computer for all
participants. During the presentation of the individual videos, a research assistant also collected affect
14
behavioral observation data (for the length of the videos), while the primary researcher occasionally
collected inter-observer agreement data during this time (~30% of the time). The research assistant was
introduced as an assistant and was instructed to reside in a discrete location that allowed an unobstructed
view of the participants. Following the presentation of the individual videos, participants were
administered the mood measure for a second time and the Phase II Interview.
A break of at least one day was implemented following the completion of the first experimental
condition. The second experimental condition included the same procedures as the first condition, with
the only differences being that participants were not re-administered the BIMS and that they watched the
remaining video. Each session took approximately 30 minutes to complete.
Results
Phase I
Interview. Qualitative analysis of responses to the staff interview indicated that both videos
represent realistic interactions that occur between residents and staff members during personal care tasks.
Refer to Appendix B for the list of questions asked of staff.
Staff were asked for their opinion on the verbal interactions, the nurses, and the older adult shown
in the videos. All participants indicated disliking the interaction and the nurse’s style in the ES video,
while the opposite was observed in the N-ES video. With the most common dislikes for the ES video
being the use of terms of endearment, the lack of choices provided, and the task-focused style of the
nurse. A commonly reported like of the N-ES video was the resident-focused style of communication
used by the nurse. For the ES video, the majority of participants reported that the resident appeared to be
frustrated and resistant due to having the ability to express wants and needs but feeling unheard. For the
N-ES video, the majority indicated that the resident appeared to be independent.
15
Next, participants were probed regarding their use of the interaction styles depicted in the videos
and if they witnessed these communication styles in day-to-day work. All participants indicated having
observed both communication styles in the past, with the observation of the ES style ranging from a few
times to daily, and all indicating daily observation of the N-ES style. In comparison, all but one
participant reported they did not use the ES style, while all reported using the N-ES style. The majority of
participants indicated that they do not use terms of endearment, a task-focused style, or deny residents the
ability choices. Comparable to the interaction style shown in the N-ES video, the majority indicated they
provide choices and foster independence in their day-to-day work.
Participants were asked for opinions regarding the validity of the setting, interactions, and
behaviors displayed by the older individual in the videos. All participants indicated that the setting,
behaviors displayed by the older adult, and both interaction styles depicted in the videos were realistic.
With regard to the interaction between the nursing assistant and resident, the majority of participants
reported that the use of the ES style was especially realistic during busy work times. The general
consensus regarding the older adult’s behavior was that it is typical of a resident with dementia. The
majority of individuals reported that the setting should not have included a kitchen.
Participants were asked to select the video that displayed a better example of how nursing staff
should interact with residents who have a memory impairment. All participants indicated that the N-ES
video was most appropriate, with the majority indicating that the communication style shown in this
video promoted residents’ rights and dignity. Participants were asked whether they believe older adults
living in long-term care settings would find the interaction in the videos realistic or relatable. The
majority (n=5) of individuals reported that the ES video would be relatable, compared to all participants
reporting this for the N-ES video.
Phase II
16
Mood. A series of repeated-measures analyses of variance (ANOVA) were conducted to test
whether there was a difference in negative and positive mood between video conditions. The mood
measure was dichotomized into a positive and negative mood score. To adjust for the difference in
number of items comprising the two mood types, an average score was created for the two mood types.
Change scores were created from pre to post video-exposure for both positive and negative mood. As
seen in Figure 2, these change scores were compared between video conditions. The ANOVA for
positive mood change between pre and post-video measurements determined that there was no significant
difference between the ES (M=.43, SD=.73) and N-ES (M=-.50, SD=1.04) (F (1,6) = 4.16, p=.09,
ηp2=.409). Refer to Table 2 for means and standard deviations for the average scores. The ANOVA for
negative mood change between pre and post-video measurements determined that there was no significant
difference between ES (M=.23, SD=.45) and N-ES (M=.30, SD=.83) (F (1,6) = .02, p= .90, ηp2=.003).
Figure 2. Positive and negative change scores compared between video conditions
To test whether there was a difference in post-video mood measurements between video
conditions, a series of repeated-measures analyses of variance (ANOVA) with a Greenhouse-Geisser
correction were conducted on positive and negative average scores. The ANOVA for positive mood post-
video scores between video conditions determined that there was no significant difference between the ES
17
(M=5.64, SD=1.55) and N-ES (M=4.58, SD=.80) (F (1,6) = 2.80, p=15, ηp2=.318). The ANOVA for
negative mood scores between video conditions determined that there was no significant difference
between the ES (M=.17, SD=.30) and N-ES (M=.40, SD=.62; F (1,6) = .50, p=.51, ηp2=.075).
Table 2
Mood Averages
Variable
ES
N-ES
Positive
Mood
Mean
(SD)
Positive
Mood ∆
Negative
Mood
Mean (SD)
Negative
Mood ∆
Positive
Mood
Mean (SD)
Positive
Mood ∆
Negative
Mood
Mean (SD)
Negative
Mood ∆
Pre
6.07
(1.54)
.43
.40 (.53)
.23
4.07 (.84)
-.50
.70 (1.01)
.30
Post
5.64
(1.55)
.17 (.30)
4.58 (80)
.40 (.62)
Affect. Observations of baseline mood and mood expressed while watching the videos differed in
terms of length. To account for these differences in length, averages of positive and negative affect
observed were created and compared. As seen in Table 3, descriptive statistics indicate there was not
much affect observed in general or much difference between conditions for both positive and negative
affect. Visual analysis also shows that there was somewhat more affect observed during baseline than
while videos were being watched.
Table 3
Affect Behavioral Observation
Variable
ES
N-ES
Baseline
During
Baseline
During
Mean % of
interval for
positive
affect
15.23%
6.04%
10.71%
8.24%
Mean % of
interval for
9.72%
7.14%
14.30%
2.20%
18
negative
affect
Interview. During the interview participants were asked a series of close and open-ended
questions, which were analyzed visually and with descriptive statistics. Participants were asked how
much they enjoyed the videos on a 9-point scale ranging from “not at all” (0) to “a great deal” (8), and
results indicated that they marginally enjoyed the ES (M=5.43) video more than the N-ES (M=4.90) video
(refer to Table 4). All residents reported something specific they enjoyed from the ES video, with some
examples being: the nurse was helpful, the nurse showed concern, and the nurse was caring. Only two
individuals reported not enjoying the way the nurse handled the tasks, while the majority reported not
disliking anything in the ES video. All but one participant reported something specific they enjoyed about
the N-ES video, with some examples being: the nurse is helpful, the nurse is good at her job, the style of
care. All participants indicated there was nothing they disliked about the interaction in the N-ES video
(one participant did not like the décor).
Participants were asked for their opinion on how the resident depicted in the videos felt. For the
ES video, some participants indicated that (n=3) resident felt negatively (e.g., helpless, frustrated,
unheard), while others (n=4) reported the resident felt positively (e.g., okay, independent, satisfied). For
the N-ES video, the majority (n=6) reported that the resident felt positively (e.g., good, happy, above
average). As seen in Table 4, participants responded identically for both video conditions on specific
questions regarding how the resident presumably felt, with the exception being that the majority felt the
resident in the ES video did not feel happy. In general, participants felt that the resident felt positively
(i.e., respected, cared for, not irritated/annoyed, not sad).
When asked for their opinions about the nurses in the videos, all participants reacted positively
(e.g., ES: she did a good job, okay, patient and kind; N-ES: she did a good job, helpful, professional) to
both videos. As seen in Table 4, participants responded identically for both video conditions regarding
19
specific questions about the nurse. In general, participants felt that the nurse did a good job (i.e.,
respectful, caring, good at her job, not controlling).
Table 4
Phase II Interview, questions # 1, 4b, 6, & 7
Variable
ES
N-ES
M (SD)
Mode
M (SD)
Mode
How much did you
enjoy the video?
5.43 (1.9)
4.90 (1.70)
Resident:
Respected
Yes
Yes
Cared for
Yes
Yes
Irritated/annoyed
No
No
Sad
No
No
Happy
No
Yes
Nurse:
Respectful
Yes
Yes
Caring
Yes
Yes
Good at her job
Yes
Yes
Controlling
No
No
If you were to need
assistance with tasks
such as combing your
hair or getting dressed,
would you want this
nurse to assist you?
(Yes/No)
Yes
Yes
Discussion
Contrary to previous research, the results of this study did not support the hypotheses that
individuals with dementia would exhibit differences in mood, affect, or preferences between neutral and
20
elderspeak communication (Cunningham and Williams, 2007; Williams et al., 2008). In other words,
elderspeak exposure did not appear to significantly affect participants’ current mood in a positive or
negative way. The results of this study were inconsistent with the findings of Ryan et al. (1991) and La
Tourette and Meeks (2000) who found that participants who listened to scenarios depicting elderspeak
had less favorable impressions of the speaker (i.e., nurse). In the current study, participants rated the
nurses equally as being good at their job, respectful, caring, and not controlling in both video conditions.
This is also true for perceptions of individuals receiving ES, who were rated as feeling equally respected,
cared for, not irritated/annoyed, and not sad for both video conditions.
The results of this study were not consistent with the findings of Guzman-Velez et al. (2014), in
which individuals with AD exhibited prolonged periods of negative mood following the viewing of
mood-induction videos. Emotion measures in this study found minimal emotional expression and self-
reported changes in mood related to video exposure. However, there is evidence that older adults with
dementia express more facial expressions of pain than non-cognitively impaired controls (Kunz et al.,
2007), so presumably their ability to outwardly express emotions should be similar that of individuals
without cognitive impairment.
Future Directions and Limitations
Several limitations associated with this study regard the sample obtained. Notably, statistical
power in establishing relationships was limited by the small sample size. Because the sample obtained
was limited by homogeneity of the stage of dementia of the residents (mild to moderate), future research
with larger samples should examine the effect of dementia stage on how ES is perceived. An additional
limitation regarding the sample obtained for this study is that it was primarily Caucasian. It is possible
that older adults of other ethnic backgrounds or from other geographic regions of the United States may
respond differently to the use of ES. Future research should include more ethnically diverse samples to
determine how individuals from other ethnic groups are affected by the use of ES.
