PAIN EXPRESSION IN MILD COGNITIVE IMPAIRMENT: ITS RELATION TO
FRONTAL LOBE INVOLVEMENT IN NON-VERBAL PAIN EXPRESSION

by

Tammy Klassen-Ross
B.A., University o f British Columbia, 2002
M.Sc., University o f Northern British Columbia, 2009

DISSERTATION SUBMITTED IN PARTIAL FULFILLMENT OF
THE REQUIREMENT FOR THE DEGREE OF
DOCTORATE OF PHILOSOPHY
IN
PSYCHOLOGY

UNIVERSITY OF NORTHERN BRITISH COLUMBIA
September 2014

© Tammy Klassen-Ross, 2014

UMI Number: 3663188

All rights reserved
INFORMATION TO ALL USERS
The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if material had to be removed,
a note will indicate the deletion.

Di!ss0?t&Ciori Publishing
UMI 3663188
Published by ProQuest LLC 2015. Copyright in the Dissertation held by the Author.
Microform Edition © ProQuest LLC.
All rights reserved. This work is protected against
unauthorized copying under Title 17, United States Code.

ProQuest LLC
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, Ml 48106-1346

Abstract
Background/Purpose: Pain is prevalent and often undertreated in elderly individuals suffering
from cognitive impairments. Interestingly, such individuals may actually show an increase in
facial expressions o f pain. The reasons for this are unclear, but may possibly relate to
disinhibition due to frontal lobe impairment. It was hypothesized that elderly individuals who
suffer from mild cognitive impairments (MCI) would display higher levels o f pain expression
than normal controls, particularly if they have deficits in frontal lobe functioning.
Methods: A sample o f elderly individuals were recruited from the community and labeled as
“M CI” if they scored <26 on the Montreal Cognitive Assessment Task (MoCA). A comparison
sample (control group) consisted o f individuals scoring >26 on this test. Facial expressions were
videotaped and then rated and scored for pain intensity at the time o f an influenza vaccination
and participants were subsequently questioned regarding their subjective experience o f the
painful stimulus. They also underwent neuropsychological testing o f frontal lobe functions (in
particular, executive functioning) including the Wisconsin Card Sorting Test (WCST), Stroop
Test, and verbal fluency testing. Social inhibition was assessed through a self-disclosure
interview.
Results: There were 58 participants ranging in age from 56 to 92 yrs (M=69.7; SD=8.3). O f this
group, 34 met criteria for MCI based on MoCA scores (i.e., MCI group) while 24 did not (control
group). There were no significant differences between the two groups with respect to facial
expressions o f pain, subjective reports o f pain, nor social inhibition scores. While executive
functioning was relatively impaired in the MCI group, the degree o f impairment did not correlate
with facial expressions o f pain nor with social inhibition scores.

Conclusions: Contrary to expectations, no significant differences were found between
participants identified as having MCI and normal controls in their facial expressions o f pain post­
immunization and executive functioning scores did not significantly predict observed facial
expressions o f pain in this sample. If such differences do indeed exist, they may be too mild at
this stage o f cognitive impairment to be demonstrated with such a small sample size. Further
investigations, utilizing a larger sample size and/or participants with a more moderate degree o f
cognitive impairment may be warranted.

Acknowledgement
I would like to express my deepest appreciation and thanks to my supervisor Professor Ken
Prkachin, who continually conveys an enthusiasm for pain research. Dr. Prkachin provided
support and encouragement when things didn’t always go to plan. Without his guidance and
persistent help this dissertation would not have been possible. I would also like to thank my
committee members, Professor Glenda Prkachin, Dr. Candida Graham, and Dr. Jacqui Peterson,
for their insight and comments throughout the dissertation process.
In addition, a thank-you to Dr. Ken Craig, who introduced me to pain research in vulnerable
populations. Without him, I would not have gone down this path. Finally, I would like to thank
my husband, Chris Ross, who encouraged me the most to pursue my dream.

TABLE OF CONTENTS
Abstract............................................................................................................................................................ i
Acknowledgements...................................................................................................................................... iii
List o f Tables.................................................................................................................................................vi
List o f Appendices......................................................................................................................................vii
Introduction.....................................................................................................................................................1
Literature Review .......................................................................................................................................... 2
Mild Cognitive Impairment............................................................................................................ 2
Prevalence o f M ild Cognitive Im pairm ent...................................................................................4
Defining Pain.....................................................................................................................................4
The Effects o f Aging on Pain......................................................................................................... 7
Prevalence o f Pain in the Elderly.................................................................................................. 8
Assessing Pain in the Elderly......................................................................................................... 9
Pain in Elderly Individuals with Cognitive Impairments..........................................................11
Frontal Lobe and Social Inhibition.............................................................................................. 13
Specific Hypotheses.................................................................................................................................... 18
M ethods......................................................................................................................................................... 19
Participants...................................................................................................................................... 19
Apparatus and M aterials................................................................................................................20
Procedure......................................................................................................................................... 26
Results........................................................................................................................................................... 30
Demographics................................................................................................................................. 30
Pre-Analysis....................................................................................................................................31

Specific Hypothesis Testing......................................................................................................... 32
Discussion.....................................................................................................................................................38
Limitations o f Study....................................................................................................................................42
Suggestions for Future Research...............................................................................................................45
Conclusion.....................................................................................................................................................46
References.................................................................................................................................................... 47

List o f Tables
Table 1: Demographic Characteristics o f Study Sample..................................................................... 61
Table 2: Descriptive Characteristics o f Dependent Variables.............................................................62
Table 3: Skewness and Kurtosis o f Pain Related V ariables................................................................63
Table 4: Correlations between Executive Functioning Assessments................................................ 64
Table 5: Correlations between Executive Functioning Assessments and Pain Expressions Post
Needle Insertion......................................................................................................................................... 65

vi

List o f Appendences
APPENDIX A: Informed Consent Form
APPENDIX B: Demographic and health status o f participant
APPENDIX C: McGill Pain Questionnaire
APPENDIX D: Geriatric Depression Scale
APPENDIX E: Social Inhibition Interview Protocol

Pain expression in Mild Cognitive Impairment: its relation to frontal lobe involvement in non­
verbal pain expression
Pain is prevalent and often undertreated among elderly individuals, especially among
those suffering from cognitive impairments (Gabre & Sjoquist, 2002). O f great concern is the
potential for caregivers and health care providers to misinterpret or not recognize pain signals in
a population that may have difficulty verbally expressing physical discomfort. Researchers have
made great strides in understanding pain in this population; however, a conflicting pattern has
emerged. Recent research has shown that elderly individuals with cognitive impairments are
likely to express higher levels o f pain through facial movement than cognitively intact elderly
individuals (Hadjistavropoulos, Craig, Martin, Hadjistavropoulos & McMurty, 1997; Kunz,
Scharmann, Hemmeter, Schepelmann & Lautenbacher, 2007). Such findings conflict with
research indicating that those with cognitive impairments report less pain, therefore implying that
these individual must feel less pain (Parmelee, Smith & Katz, 1993; Haasum, Fatbom,
Fratigliomi, Kareholt & Johnell, 2011). The belief that those with cognitive impairments feel less
pain may be manifested in the reduced analgesic treatment in this population (Won et al. 1999;
Morrison & Siu, 2000). Building on research that has demonstrated that elderly individuals with
cognitive impairments are more likely to express higher levels o f pain through facial movement,
two logical questions emerge: why does this population expresses higher levels o f pain through
facial expressions even though they report feeling reduced pain, and will this pattern o f increased
pain expression be present in individuals who are in the stages o f pre-dementia?
To answer these questions it is important to look at how aging and the development o f
dementia can affect brain structures, in particular the frontal lobe. The frontal lobe is generally
thought to be responsible for higher order executive functioning (Chan, Shum, Toulopoulou &

1

Chen, 2008) and it is also involved in the regulation o f social behavior (Stuss & Benson, 1984).
Contemporary research has shown that damage to the frontal lobe impairs the ability to regulate
behavior. Patients with frontal lobe damage have been observed engaging in uncontrolled and
“tasteless” social behavior such as inappropriate joking (Stuss & Benson, 1984). In addition,
individuals with frontal lobe damage have been observed greeting strangers by giving them a kiss
or a hug (Rolls, Homak, Wade & McGrath, 1994), social displays most often reserved for more
intimate acquaintances and inhibited by cognitively intact individuals. As dementia is a
mechanism that can damage the frontal lobe, it can be expected that elements o f social inhibition
would decrease in some people during the early stages o f cognitive impairment. One could
further theorize that the release o f social inhibition would affect numerous areas o f social
behaviour, such as pain expression. In fact, research has shown that the expression o f pain is
subject to social inhibition under normal circumstances (Williams, 2002). If general social
inhibitions were released, an individual would have a more arduous time hiding pain expressions
from those around them.
The purpose o f the present study was to obtain evidence o f the possibility that reduced
frontal lobe functioning causes an increase in facial expressions o f pain. It was hypothesized that
the elderly who suffer from mild cognitive impairments and show evidence o f a reduced level o f
frontal lobe functioning would display higher levels o f pain expression, when compared to
elderly individuals who do not suffer from such cognitive impairments.

Literature Review
Mild Cognitive Impairment
Cognition is defined “the mental action or process o f acquiring knowledge and
understanding through thought, experience, and the senses” (Oxford Dictionary, 2013). It is

2

anticipated that some cognitive decline will occur at older ages, but this decline is not uniform
across all cognitive tasks. It is important to be aware o f the specific stages o f cognitive decline
when an individual is moving from normal to impaired cognitive functioning.
Mild cognitive impairment (MCI) is a clinical syndrome thought to represent the
transition phase between normal cognitive functioning and dementia (Albert & Blacker, 2006).
One o f the earliest references to mild cognitive impairment described a predementia stage
(Reisberg, Ferris, de Leon, et al., 1988). Considerable research and clinical evidence supports the
argument that there is a transitional phase between normal functioning and dementia. Research
has indicated that the reduction in memory is important in the early detection o f Alzheimer’s
disease (Backman, 2008). Dysfunctions in cognitive domains other than memory may be an
early sign o f vascular or other non-Alzheimer dementias (Padovani, Di Piero, Bargoni, et al,
1995). Studies o f imaging and other biomarkers o f AD pathology demonstrate the presence o f
alterations in non-demented individuals with cognitive impairments that are intermediate
between normal individuals and those with mild AD (Albert & Blacker, 2006). Therefore, MCI
can be thought o f as the clinical intermediary between cognitive impairment and not cognitively
impaired.
MCI is currently thought to identify a spectrum o f diseases that include impairments in
both memory and non-memory cognitive domains (Petersen, 2004; Winblad, Palmer, Kivipelto,
et al., 2004). This is in contrast to the earlier diagnostic criteria for MCI in which memory
impairment was one o f the requirements for diagnosis. The current criteria for a MCI diagnosis
are as follows: cognitive complaint, decline, or impairment; objective evidence o f impairment in
cognitive domains; essentially normal functional activities; and not demented (Petersen, 2004;
Petersen, Roberts, Knopman, et al., 2009). MCI can, therefore, be classified into two subtypes:

3

amnestic and non-amnestic (Petersen, 2011). Amnestic mild cognitive impairment is clinically
significant memory impairment that does not meet the criteria for full dementia. Alternatively,
non-amnestic mild cognitive impairment is characterized by a subtle decline in cognitive
functions not related to memory, but affecting attention, use o f language or visuospatial skills
(Petersen, 2011). According to Peterson et al. (2009), this classification by subtype relates
directly to the underlying etiology and pathology, the clinical presentation, and outcomes.
Furthermore, MCI may consist o f impairment in a single cognitive domain or impairment in
multiple cognitive domains. The number o f affected domains has vital implications for
understanding the extent o f the underlying brain pathology or disease, disease severity, and
likelihood o f progression to dementia. Information from both the MCI phenotype (amnestic MCI
vs. non-amnestic MCI) and the number o f cognitive domains affected (single vs. multiple) is
hypothesized to determine future outcomes (Petersen et al., 2009). Single-domain or multipledomain amnestic MCI is hypothesized to progress to AD if there is an underlying degenerative
etiology (Petersen et al.). The non-amnestic mild cognitive impairment subtype is less common
than the amnestic subtype and may be the forerunner to dementias not related to AD, such as
frontotemporal lobe degeneration or dementia with Lewy bodies (Molano, Boeve, Ferman, et al.,
2010 ).

Prevalence of Mild Cognitive Impairment
Research on the prevalence o f mild cognitive impairment has produced inconsistent
results, mainly because o f differing definitional criteria and assessment procedures (Bischkof,
Busse & Angermeyer, 2002). However, population-based studies indicate that rates o f mild
cognitive impairment are almost double those o f dementia (Morris, Storandt & Miller, 2001).
Similar to dementia, incidence rates o f MCI seem to increase with age, and are higher for people

4

with lower levels o f education (Bischkof, Busse & Angermeyer, 2002). In contrast to the
epidemiology o f dementia, incidence rates for mild cognitive impairment may be higher in men
than women. (Ganguli, Dodge, Chen, Belle & Dekosky, 2000). To determine the prevalence and
incidence rate o f MCI, Roberts, Geda, Knopman, Cha, Pankratz, Boeve, et al. (2008), conducted
a prospective population based study. It was found that 12.25% o f the sample (n = 2685) met
criteria for an MCI diagnosis.
According to Statistics Canada, 1 in 11 Canadians over the age o f 65 has Alzheimer’s
disease or a related dementia (Statistics Canada, 2008). It is expected that in the next five years
these numbers will grow to 1 in 6 and they will double within 25 years. Specific to British
Columbia, more than 70,000 individuals are thought to be currently living with some form o f
dementia. Research has shown that cognitive decline may reduce an individual’s ability to
effectively communicate pain to health care providers (Tsai & Means, 2005). Therefore, it is
essential to not only understand pain, but to understand how individuals who have mild cognitive
impairment may display pain.

Defining Pain
Before proceeding to the discussion o f mild cognitive impairment involvement in pain
expression, it is beneficial to understand the concept o f pain in further detail. The International
Association for the Study o f Pain (IASP) defines pain as:
An unpleasant sensory and emotional experience associated with actual or potential tissue
damage, or described in terms o f such damage. Pain is always subjective. Each individual
learns the application o f the word through experiences related to injury in early life.
(2008, np).

5

This definition is widely endorsed, as it includes criteria related to physical stimulation and
subjective experience and allows for pain in situations where tissue damage may not be apparent
(Owens, 1984).
Pain, at the most basic level, serves as a protective mechanism for the human body. Pain
perceptions signal the body to move away from tissue damage or from noxious stimuli. It is a
largely subjective and individualized perception because it requires the central nervous system to
interpret the damaging stimuli. The same level o f tissue damage may cause different levels o f
pain and pain expression in different individuals. Pain can be perceived differently from one
individual to another individual and even from one time to another time in the same individual
(Montes-Sandoval, 1999). This variability in pain perception makes consistent pain
identification difficult for the outside observer.
In some cases fulfilling the criteria for pain identification is simple, yet in other cases it is
quite complex. Identification o f potential pain through tissue damage can take place by direct
observation or by considering contextual factors such as the occurrence o f surgery, vaccinations,
or tissue damage resulting from an accident. Despite the fact that potential pain from tissue
damage is readily identifiable to the observer, the central feature o f IASP’s pain definition is
subjective distress. Identifying subjective distress is more complex in that it requires the
observer to assess pain based on physical or verbal cues (Owen, 1984).
There are distinct pain categories attended to by clinicians and researchers: (a) acute, i.e.
following an injury or a surgery, (b) chronic malignant, i.e., resulting from a progressive, often
fatal, disorder such as cancer, and (c) chronic non-malignant, i.e., a progressive disorder that
does not directly lead to death, such as arthritis (Forrest, 1995). Chronic pain is arbitrarily
defined as any pain that persists for more than one month, beyond the course o f an acute illness,

6

or endures after a reasonable time to expect that healing should have been achieved (Bonica,
1990). A major distinction between acute and nonmalignant chronic pain is that acute pain can
be viewed as a warning sign or clinical indicator o f tissue damage whereas after time, chronic
nonmalignant pain serves no clinical usefulness (Forrest, 1995). Although pain affects all ages,
its perception becomes more complex as a person ages.

The Effects of Aging on Pain
Researchers have attempted to evaluate age-related changes in pain perception, but the
results have been inconsistent. Studies have been used to assess age-related changes in pain
reactivity using controlled stimuli. Laboratory studies often use acute pain stimuli to determine
pain thresholds and tolerances in differing age groups. Experimental pain stimuli often include
exposure to heat or cold, electrical muscle stimulation, or mechanical stimulation and are
delivered only to the point o f passing the individual’s pain threshold or their tolerance for pain.
Laboratory results have been inconsistent with some researchers finding that pain thresholds do
not differ in older adults (Zheng, Gibson, Khalil, Helme & McMeeken, 2000) while other
researchers find that pain thresholds and pain tolerance increase with age and cognitive status
(Tucker, Andrew, Ogle & Davidson, 1989; Benedetti et al., 1999; Carlino et al., 2010). The
difference in findings may be explained by the variance in experimental pain stimuli used to test
pain thresholds and pain tolerance. Lautenbacher, Kunz, Strate, Nielsen & Arendt-Neilson (2005)
found that somatosensory thresholds (warmth, cold, and vibration) for non-noxious stimuli
increased with age. They also found that pressure pain thresholds decrease with age, but no age
differences were found when heat was used to induce pain.
Research has been done concerning the effects o f aging on autonomic pain responses and
spinal motor responses. Mylius, Kunz, Hennighausen, Lautenbacher & Schepelmann (2008)

7

investigated nocifensive responses to noxious electrical stimulation, and compared physiological
responses in young adults to elderly adults. They found that galvanic skin responses declined
with age, whereas no differences were found between young and elderly adults when the
nociceptive flexion reflex was stimulated. The study suggests that a decline in the sympathetic
skin response indicates that the central or peripheral efferent sympathetic functions are altered by
age.
Acute pain experiences induced in a laboratory setting are very different from acute pain
experiences in a clinical setting. The major difference between laboratory pain stimuli versus
pain stimuli in a clinical setting is that laboratory stimuli are acute, controlled, and can be
terminated as soon as the individual indicates they have reached their maximum threshold or
tolerance. Conversely, clinical pain neither stops at an individual’s threshold o f pain, often going
beyond it, nor can the stimulus be immediately terminated.
In summary, the effect o f aging on pain is still unclear, as laboratory results have shown
inconsistent findings. Caution should be used when generalizing acute laboratory pain
experiences to real life clinical pain. Pain tolerance and threshold methods used in a controlled
setting do not truly imitate or represent clinical levels o f pain or chronic pain.

Prevalence of Pain in the Elderly
The prevalence o f pain in the elderly is higher than other age populations, due to higher
rates o f painful diseases such as arthritis and cancer in the elderly (Reyes-Gibby, Aday &
Cleeland, 2002). It is estimated that the prevalence o f elderly Canadians suffering from chronic
pain is 15% - 18% (Reitsma, Tranmer, Buchanan & Vandenkerkhof, 2011). Feldt (2000)
reported that 45 to 80% o f nursing home residents experience pain on a daily basis. Ferrell and
Ferrell (1990) reported that 70% o f nursing home patients described having significant pain, and

8

one-third o f that same population suffered from consistent or chronic pain. Furthermore, Ferrell
et al. (1990) indicated that 71% o f nursing home residents reported having some pain part o f the
time. Among residents with pain, 66% reported intermittent pain and 34% indicated consistent
pain.
Pain is also prevalent in community-dwelling elderly individuals. In 1994, Mobily, Herr,
Clark and Wallace conducted a large community-based survey. They surveyed over 3,000
individuals for any type o f pain they may have experienced over the previous year. Eighty-six
percent o f the participants reported having pain. Interestingly, the study further indicated that
individuals over the age o f 85 reported fewer pain complaints than those aged 65 - 74 (Mobily et
al., 1994).
The prevalence o f pain among the elderly is thus generally high, not only in nursing
homes but in the community as well. It is important to be able to accurately assess pain in the
cognitively intact elderly, and it is even more imperative to be able to assess pain in those with
cognitive impairments. Those with cognitive impairments have a more difficult time
communicating their level o f pain, in customary ways, to those around them.

Assessing Pain in the Elderly
Assessing pain in verbal populations can be a relatively simple task, as self-report
measures can be utilized. Self-report has been termed the “gold standard” in pain assessment
(Abu-Saad, Bours, Stevens & Hamers, 1998). Even though self-reports allow for an evaluation o f
a person's subjective pain experience and are easy to gather in a methodological sense, they have
major limitations. Hadjistavropoulos, Craig & Fuchs-Lacelle (2004) point out that self-reports o f
pain may be affected by response bias, situational demands and conscious distortion. They
summarized their critique with the implication that self-reports o f pain are actually “fools’ gold".

9

They argue further that researchers are over-dependent on self-reported pain; in failing to
recognize other forms o f pain assessment, they consequently ignore populations that do not have
strong verbal communication skills (i.e. infants, autistic individuals, and the elderly with
cognitive deficits). In these populations, the reliance on self-report measures for pain assessment
has resulted in the underuse o f nonverbal expression in measuring (Barr, 1992).
The overemphasis on self-report measures and the underuse o f nonverbal pain
expressions brings the generalizability o f the IASP’s definition o f pain into question, as applied
to non-verbal populations or those with limited verbal skills. In regards to pain assessment, the
IASP’s definition becomes problematic when dealing with populations with limited abilities to
communicate effectively in a verbal manner (Anand & Craig, 1996). Hadjistavropoulos, von
Baeyer & Craig (2001) summarize the problem, stating:
It is often assumed that because the experience o f pain is a subjective state, the only
means whereby it can be tapped is through the suffering person’s verbalizations... The
current definition o f pain, which emphasizes the use o f self-description, can only be taken
to imply that states o f pain and suffering cannot be understood in nonverbal persons.
This position limits attention to the availability and usefulness o f nonverbal
expression, (p. 137)
The limitations o f this position result in difficulties when assessing pain in the elderly with verbal
difficulties due to cognitive impairments. As Feldt (2000) indicates, verbal reports from elderly
individuals with cognitive impairments may be difficult to obtain or they may be unreliable
because o f short-term memory loss or impairment in language skills.

Pain in Elderly Individuals with Cognitive Impairments

10

In 2002, an estimated 15% o f the Canadian population was over the age o f 65, with this
figure expected to rise to over 18% by the year 2025 (Martin-Matthews, 2002). With the rapid
increase in people over the age o f 65 in Canada, the number o f elderly individuals suffering from
cognitive impairments has also increased substantially. In 1994, it was estimated that over
250,000 elderly Canadians were suffering from dementia, with this number expected to increase
to 592,000 by the year 2021 (Canadian Study o f Health and Aging Working Group, 1994). The
large number o f elderly individuals with cognitive impairments suffering from pain further
highlights the importance o f understanding nonverbal signs o f pain in this population. Cognitive
impairments often progress to a point where the individual becomes non-verbal or verbal skills
are compromised. Consequently, they are unable to report pain levels through traditional means
such as self-report. For these reasons, it often falls to an outside observer (i.e., physician,
caregiver) to identify the patient’s pain and provide appropriate treatment.
Research has been conducted to determine if elderly individuals with cognitive
impairments identify and display pain in a similar manner to non-cognitively impaired
individuals. Studies have shown that individuals suffering from dementia will rate pain stimuli
similarly to healthy individuals (Kunz, Mylius, Scharmann, Schepelman & Lautenbacher, 2009).
It has also been found that facial expressions o f pain are preserved in individuals with dementia
(Kunz, Scharmann, Hemmeter, Schepelmann & Lautenbacher, 2007). An interesting outcome of
the research is the level o f facial pain displays, as individuals with cognitive impairments display
pain differently when compared to cognitively intact individuals.
Clinically, it has been found that individuals with cognitive impairments have more
intense facial displays o f pain when compared to cognitively intact elderly. Facial expressions in
those with dementia have been assessed, using facial coding, during flu injections

11

(Hadjistavropoulos, Craig, Martin, Hadjistavropoulos & McMurty, 1997), venipuncture
(Hadjistavropoulos, LaChapelle, MacLeod, Hale, O'Rouke & Craig, 1998; Porter, Malhotra,
Wolf, Morris, Miller & Smith, 1996), and exacerbated musculoskeletal pain during physical
exercise (Hadjistavropoulos, LaChapelle, MacLeod, Snider & Craig, 2000). Even though the
type o f clinical pain varied in these studies, the results are homogenous. It was consistently
found that facial expressions to noxious stimuli increased in patients with dementia
(Hadjistavropoulos, LaChapelle, MacLeod, et al., 2000; Porter et al., 1996), suggesting that these
facial expressions could be reliably used to assess levels o f pain in dementia patients who are
verbally compromised. Results from these studies also uniformly found that facial expressions in
patients with dementia significantly intensify when compared to baseline periods.
To further explore the increase in facial expression, Kunz, Scharmann, et al. (2007)
conducted an experimental study to determine if facial expressions o f pain would be more
intense in dementia patients when compared to cognitively intact individuals in a controlled
environment. Their findings confirmed previous clinical research, showing that facial responses
to noxious electrical stimulation were significantly increased in patients with dementia when
compared to healthy controls. The researchers further found that facial responses were closely
related to the intensity o f the pain stimulation. To extend these findings, Kunz, Mylius et al.
(2009), assessed multiple components o f pain in a sample o f patients with MCI. They found that
when they assessed subjective, facial, motor reflex and autonomic responses to a noxious
electrical stimulus, dementia patients were more reactive in their behavioural responses to pain
relative to intact controls. Once again, facial responses to noxious electrical stimulation were
significantly increased in patients with dementia. Thus, the available literature is consistent
across clinical and experimental studies in showing that pain reactivity is enhanced among

12

patients with dementia. The question remains, why are facial pain responses more intense in
those elderly suffering from dementia, when compared to facial pain responses o f cognitively
intact elderly individuals? The Kunz et al. study investigated the differences between facial pain
responses o f elderly people with and without cognitive impairment. Within these groups,
psychometric indicators o f frontal lobe functioning were examined as they relate to facial
responses to pain.

Frontal lobe and Social Inhibition
There is a rich history regarding the current understanding o f how the human brain
controls behaviour. A well-known contributor to this level o f understanding is John Hughlings
Jackson, who is often regarded as the father o f British neurology (Gillett & Franz, 2013). As
most physicians in the later part o f the 19th century Hughlings Jackson was faced with the need to
diagnose neurological diseases with no systematic scientific methodology. Hughlings Jackson’s
approach to clinical neurological symptoms was greatly influenced by Herbert Spencer’s
emerging evolutionary approach to the mind and brain (Franz & Gillett, 2011). Spencer
championed the dismissal o f metaphysical and supernatural explanations in science and human
knowledge and he explained the emergence o f higher mental processes - the cognitive functions
to do with thought, memory, imagination and mortality. He achieved this through the expansion
o f primitive sensorimotor ‘forbearers’, using only the resources o f biological theory (Spencer,
1885). From Spencer’s theories o f sensorimotor forbearers, Hughlings Jackson emphasized the
importance o f “the corresponding organism with the environment” (Hughlings Jackson, 1884,
pg. 705); therefore, the higher brain centers are suited to the re-representation o f the body and the
conditions to which it is responsive. This led to his hypothesis that the brain evolved with
increasingly higher levels o f re-representation o f the basic (primitive) sensorimotor

13

representations. He further proposed a three-level system that effectively encompasses a
sensorimotor machine (the whole brain), with its lower level defined as anterior spinal horns and
motor nerve nuclei, its middle level o f motor control and basal ganglia and highest level, that
which “re-re-represents the body”, consisting o f the premotor (frontal) cortex (York & Steinberg,
2006, pg. 19). Hughlings Jackson further postulated the dissolution o f evolved higher functions
was the key to clinical conditions in which nervous system damage affects an individual’s ability
to function (Hughlings Jackson, 1884). Therefore, damage to the highest level o f the
sensorimotor machine (e.g. the frontal lobe) would affect an individual’s ability to function.
In humans, the frontal lobes encompass all the tissue anterior to the central sulcus and
constitute roughly 20% o f the neocortex (Kolb & Whishaw, 2009). The frontal lobe is
responsible for higher-level executive functions, including organizing and following social
norms, and can be divided into the motor, premotor and prefrontal areas. Damage to this lobe can
cause changes in personality and result in difficulties following social norms. The most
publicized example o f personality change subsequent to frontal lobe damage is that o f Phineas
Gage. John Harlow first reported this extraordinary case in 1868. Gage was a railway
construction supervisor who survived an explosion that blasted a meter long, 3cm wide at its
widest point iron-tamping bar through his frontal cortex. According to Harlow, G age’s behaviour
changed completely after the accident. Gage had been o f average intelligence and was energetic
and persistent in executing all o f his plans. However, following the injury G age’s personality
was described by Hallows as:
The equilibrium or balance, so to speak, between his intellectual faculties and animal
propensities seem to have been destroyed. He is fitful, irreverent, indulging at times in
the grossest profanity, manifesting but little deference to his fellows, impatient o f

14

restraint or advice when it conflicts with his desires, at times pertinaciously obstinate, yet
capricious and vacillating, devising many plans o f operation, which are no sooner
arranged than they are abandoned in turn for others appearing more feasible. (Blumer &
Benson, 1975, p. 153).
If a person with damage to the frontal lobe cannot follow social norms, he or she may not inhibit
social responses, such as occurs when hiding pain. Thus, if progressive neurological
impairments have damaged the frontal lobe it would be expected that the individual would be
unable to inhibit the social pain response and would show pain levels through their facial
expressions.
Facial expressions are an elemental mechanism for communication in many species and
an important element in the management o f interpersonal relationships and social interactions
(Darwin, 1872; Ekman, 1999). The degree to which this mechanism can display emotional states
can vary greatly from individual to individual, with one extreme o f individuals being stoic and
not expressing emotions to other individuals being very expressive. Ekman and Friesen (1969)
suggest that a lack o f expressiveness reflects a form o f inhibitory control, which is regulated by
societal rules and norms. This concept suggests that the expression o f emotions would be a
default response that individuals learn to suppress following social demands. As facial
expressions play an important role in social interactions, research has investigated the neural
basis o f communication via facial expressions. Behavioural studies in patients with frontal lobe
lesions suggest an involvement o f the frontal cortex in regulating facial expressions (Blair, 2003;
Weddell, 1994). Goldin, McRae, Ramel and Gross (2008) added to this research by conducting a
functional magnetic resonance imaging study in healthy controls. Results from this study
indicated that while watching negative emotion-eliciting films, activity o f the prefrontal lobe

15

increased when individuals were instructed to suppress their facial expressions. Not only do
individuals suppress facial expressions related to negative emotions, but individuals are also
known to hide their pain through the inhibition o f their facial expressions o f pain. Kunz, Chen,
Lautenbacher, Vachon-Presseau and Rainville (2011) conducted a study to determine which
brain areas were activated when individuals inhibited their pain processes. They hypothesized
that low facial expressiveness would be accompanied by stronger prefrontal activation consistent
with an inhibitory process. Results indicated that pain expression was indeed inversely related to
frontostriatal activity, consistent with the down-regulation o f facial displays (Kunz et al., 2011).
Kunz et al. used healthy adults to demonstrate the activation o f the frontal lobe through the
suppression o f pain expressions. One could speculate that an inverse effect would occur if elderly
individuals with mild cognitive impairment were tested for the association between mild
cognitive impairment and frontal dysfunction. That is, impairment to the frontal lobe should
yield an increase in facial expressions when individuals are exposed to acute noxious stimuli, as
their ability to inhibit their facial expressions o f pain would be impaired.
The notion that individuals with frontal lobe damage would lose the ability to control
their facial expressions o f pain can further be explained in relation to social inhibition. People are
often compelled to act on impulse and inclination, but they simultaneously regulate their social
behavior according to social norms. The act o f regulation is termed social inhibition (Beer et al.,
2003). It is theorized that self-conscious emotions (e.g., shame, embarrassment) have developed
for the purpose o f regulating the impulsive approach and lack o f inhibition tendencies that could
threaten or violate social relations (Tangney, Miller, Flicker & Barlow, 1996). Therefore,
individuals who violate norms governing social behaviour due to frontal lobe damage should
show deficits in these self-conscious emotions. Research has shown that damage to the

16

orbitofrontal cortex, an area richly connected to areas associated with emotional and social
processing (Brothers, 1996), can cause disruptions in social regulation. Research has further
shown that patients with frontal lobe damage who engage in self-disclosure during interviews
disclosed unnecessarily intimate information when describing their past emotional experiences;
whereas, health controls did not disclose unnecessary intimate information (Beer, Heerey,
Keltner, Scabini & Knight, 2003). Building on this idea, the present study addressed possible
relationships between frontal lobe functions, their involvement in social inhibition, and pain
responses in aged individuals in relation to their mild cognitive impairment.
The link between level o f pain expression and disinhibitory processes was assessed
indirectly by neuropsychological tests known to be sensitive to frontal dysfunction and through a
self-disclosure interview. Numerous areas o f functioning can be so tested to determine frontal
lobe involvement. These areas o f functioning include response inhibition, verbal fluency,
nonverbal fluency, motor control, language comprehension, working memory, and planning
(Kolb & Wishaw, 2009). Furthermore, it has been established that executive functioning tests are
also useful in characterizing frontal lobe brain lesions (Kennedy, 2004; Raazani, Boone, Miller,
Lee & Sherman, 2001). Poor performance on executive function tests is thought to represent a
variety o f cognitive deficits including deficient abilities in planning, organization and initiation
(Lezak, Howieson & Loring, 2004). A number o f tests exist that are drawn on to assess
executive functioning o f the frontal lobe. To assess response inhibition, the Wisconsin Card
Sorting test (Milner, 1964) and the Stroop Test (Perrett, 1974) are commonly administered. To
examine a patient’s level o f verbal fluency, the Thurstone Word Fluency Test (Milner, 1964) may
be used while the Tower o f London (Owen, Sahakian, Hodges, Summers, Polkey & Robbins,
1995) is used to assess planning abilities. The Wisconsin Card Sorting Test, the Stroop test, and

17

the Phonemic Verbal Fluency Task were used in the present study to assess frontal lobe
functioning as they are the most frequently used executive functioning tests (Stuss & Levine,
2002). In conjunction with the executive functioning tests, a self-disclosure interview was
conducted to evaluate social disinibition in participants with mild cognitive impairment.
In addition to testing the relationship between executive functioning, social inhibition and
pain expression in individuals with mild cognitive impairment, memory for pain was also
explored. Research has indicated that individuals without cognitive deficits have good memory
for acute pain intensity and can accurately report their pain intensity at a later date (Erskine,
Morley & Pearce, 1990). Research has also shown that elderly individuals suffering from
chronic pain often complain o f memory impairments and concentration issues (Munoz & Rosa,
2005). Despite these findings, research has not adequately investigated whether elderly
individuals with cognitive impairments are able to accurately report previous acute pain
instances. Therefore, this study also examined pain memory in the elderly to determine if
individuals with and without cognitive impairments are able to accurately recall past pain
experience.
The following specific hypotheses were addressed.
1). Participants with probable Mild Cognitive Impairment (MCI) will have increased facial
expressions o f pain when compared to controls.
2). Participants with probable MCI and control participants will report similar levels o f
pain directly post needle insertion.
3). Participants with probably MCI will have a poorer memory for their pain experience
when compared to controls.

18

4). Participants with probable MCI will have impairment in selected neuropsychological
measures o f frontal lobe functioning when compared to controls.
5). There will be a negative relationship between frontal lobe functioning and facial
expressions o f pain in participants with probable MCI.
6). Control participants will have higher rates o f social inhibition when compared to
participants with probable MCI.
7). There will be a negative relationship between social inhibition and facial expressions o f
pain in participants with probably MCI.

Methods
Participants
Similar to previous studies, which have examined pain in individuals with cognitive
impairments (see Felt, Ryden & Miles, 1998; Bendetti et al., 1999; Hadjistavropoulos et al.
2000), a community sample o f elderly individuals was recruited from 4 local health clinics.
Recruitment was conducted between October 2012 and November 2013. Inclusion criteria for
participation included: (1) a score o f 25 or lower on the Montreal Cognitive Assessment
(MoCA) (target group only), (2) no medical conditions which could cause pain sensitivity such
as diabetes, arthritis, osteoarthritis, cancer and peripheral vascular disease (Forrest, 1995), (3) the
participant was at the health clinic to receive a routine seasonal flu vaccination, (4) the
participant could read and write English, (5) the participant was able to give informed consent,
and (6) the participant agreed to either a home visit or a university visit. A comparison group o f
elderly individuals was also recruited from the same health clinics. Inclusion criteria for the
comparison group only differed from the target group in that they had to have a MoCA score o f
26 or more.

19

Apparatus and Materials
A Sony HD AVCHD Handycam, model number HDR-XR100 was used to record the
participants' behavioural responses to the vaccination. All digital video recordings were
downloaded into a standard iMac computer for behavioural coding and analysis. The downloaded
video was further edited into separate participant files using iMovie, a movie-making computer
program.
Using iMovie, a 26-minute video was compiled o f all the participants’ behavioural
responses to the vaccination procedure. This video consisted o f randomly sequenced 10-second
video clips, which focused on the participants’ facial expressions prior to the injections and post­
injection. Between each participant’s 10-second clips, a 5-second blank screen was presented to
facilitate ratings. To evaluate the participant’s pre-injection and post-injection pain expressions,
five independent judges viewed the behavioural video.

Demographic questionnaire (Appendix B). Participants filled out a questionnaire
inquiring about their age, level o f education, overall health status, the presence o f any chronic
pain conditions or any diseases that could cause chronic pain, and if they took any analgesic
medication prior to their vaccination.
The questions inquiring about their age, level o f education, overall health and chronic
pain status were used to ascertain study eligibility. The question regarding analgesic medication
was used as a control for pain expressions, as analgesic medication could potentially reduce the
physical pain sensation when the needle is inserted in arm.

Neuropsychological/Executive measures.
M ontreal Cognitive Assessm ent (M oCA; Nasreddine et al., 2005). The MoCA is a brief
one-page, 30-point test used to assess cognitive abilities. Specifically, the MoCA assesses the

20

following cognitive domains: short-term memory recall, visuospatial abilities, executive
functioning, attention, language, and orientation. This measure was used to assess participant’s
level o f cognitive functioning. It is important to note that the MoCA cannot provide a clinical
diagnosis o f MCI, but rather can be used as part o f the diagnostic assessment. To make a clinical
diagnosis o f MCI the individual should be assessed using the following criteria: individual is
neither normal nor demented; shows essentially normal function in activities; and can either
present memory or no memory impairment (Petersen, 2004). For the purpose o f this study,
however, a score o f 25 or lower on the MoCA indicated probable MCI (Nasreddine et al., 2005).
The MoCA has shown acceptable psychometric properties, and results suggest that it is more
sensitive than the Mini Mental Status Exam in sensitivity and equivalence in specificity (Sweet,
Van Adel, Metcalf, Wright, Harley, Leiva, et al., 2011). The MoCA has also been shown to be a
sensitive and accurate instrument for screening individuals with behavioural-variant
frontotemporal dementia (Freitas, Simoes, Alves, Duro & Santana, 2012).
Wisconsin Card Sorting Task (W CST; H eaton, 1981). The WCST consists o f 128 cards
divided equally into three categories: forms (circles, crosses, stars and triangles), numbers (one to
four), and colours (blue, green, red, and yellow). The task required participants to match each
card from the pile, to one o f the 4 stimuli cards by sorting into one o f the three categories (forms,
numbers, or colours). Correct category sorting was achieved by finding the correct classification
principle by trial and error, and by examiner feedback. For example, initially the correct solution
was colour; when the participant determined this solution, after 10 correct, consecutive matches
the principle changed without warning, demanding a flexible shift in set. The WCST was not
timed and sorting continued until all cards were sorted, or a maximum o f six correct sorting

21

criteria was reached (Heaton, 1981). Outcome measures included number o f categories
completed, and number o f sorting errors.
The WCST was employed as a measure o f response inhibition thought to reflect frontal
lobe function. This test has been found to be an effective tool in measuring frontal lobe function
(Nagahama et al., 1996). Research has shown that patients with brain lesions to the frontal lobe
due to stroke (Struss et al., 2000; Mukhopadhyay et al., 2008), tumors (Goldstein, Obrzut, John,
Ledakis & Armstrong, 2004) and epilepsy (Giovagnoli, 2001) perform worse on the WCST when
compared to healthy controls.
Stroop Test (Trenerry et al., 1989). To further assess response inhibition thought to
reflect frontal lobe damage a variant o f the Stroop test was administered in two conditions. In
the congruent task, participants were required to identify and say aloud the colours o f 112 colour
bars (blue, green, red, tan). In the incongruent task they were shown 112 colour printed in a
different colour o f ink (e.g., the word “green” written in blue ink). Participants were required to
name the colour o f the ink in which the word was printed (e.g. if the word 'green' is written in
blue ink, the participant should say "blue"). Two outcome measures were calculated. The first
was the difference in time between the congruent and the incongruent task. The second was the
difference between the number o f errors made in the congruent task and the number o f errors in
the incongruent task.
The Stroop test measured the interference that the automatic process o f reading has on
more effortful processes. Research has shown that individuals with frontal lobe damage showed
deficits in inhibiting reading the words in the interference condition (Perrett, 1974).
Verbal flu en cy test. To determine participants’ level o f verbal fluency a phonemic
fluency task, the F-A-S task, was administered. The participants were instructed to generate as

22

many words as possible beginning with the letters “F”, “A” and “S” within a 1-minute time period
for each given letter, excluding proper nouns and the same word with a different suffix. The
following instructions were given to all participants: “I will say a letter o f the alphabet. Then, I
want you to give me as many words you can that begin with this letter, as quickly as possible.
For example, if I say B, you can say bed, big, but you cannot say proper nouns like Bob or
Brazil. Also, you can’t say the same word with a different ending like bug and bugs”.
Subsequently, participants were asked if they understood the instructions.
The F-A-S is a sensitive task that has been used to assess frontal lobe functioning (for
review see Alvarez & Emory, 2006). Vilkki and Holst (1994) used F-A-S to determine
differences in the types o f frontal lobe lesions. The F-A-S phonemic fluency task has also been
used to assess word fluency performance in post-stroke aphasia. Samo, Postman, Cho and
Norman (2005) found that the task was sensitive in identifying increases in verbal fluency
following comprehensive treatment.

Self-report: Pain assessment.
Coloured Analogue Scale (CAS; M cG rath et a l, 1996). To assess participants’
subjective pain directly following their vaccination, the CAS was administered. The CAS is a
visual analogue scale that was initially developed to provide a practical pain measure for children
and others with marginal self-report skills (McGrath et al., 1996). Pain on the CAS is rated by
moving a plastic glide along a 14.5 cm. long triangular shape that varies in width and colour,
from 1 cm width and a light pink colour at the bottom to a 3 cm width and deep red colour at the
top. The extremes o f the scale are anchored with “No Pain” at the bottom and “Most Pain” at the
top. This scale allows participants to have visual cues for scaling their pain severity, not only in
the variation in width and changes in colour, but also in the length o f the scale and the anchoring

23

words. The scale has numbers marked on the back so that the person administrating the scale can
record a number (ranging from 0-10), representing the highest level o f pain felt by the participant
during the vaccination procedure.
This visual analogue scale has been shown to be valid and reliable when used with the
elderly without cognitive deficits and with elderly individuals who have mild to moderate
cognitive deficits (Hadjistavropoulos et al., 1998).
M cG ill Pain Questionnaire (MPQ; M elzack, 1975; A ppendix C). To further assess
participants’ subjective pain perception at the time o f the vaccination and to assess their pain
memory, they were asked to complete a subset o f the McGill Pain Questionnaire. The McGill
Pain Questionnaire is a self-report pain-rating tool that can be used by most elderly persons with
normal to moderately impaired cognitive functioning (Manz, Mosier, Nusser-Gerlach, Bergstrom
& Agrawal, 2000). The McGill Pain Questionnaire consists o f three sections: verbal descriptors
o f pain, temporal aspects o f pain, and pain severity. For the purpose o f this study, only the
verbal descriptor section was used. This section consists o f 20 word groupings such as temporal,
spatial and punctate pressure. Within each grouping, a number o f pain descriptor words are
provided. These groupings are further collapsed into four categories: sensory, affective,
evaluative and miscellaneous (Melzack, 1983). The participant was instructed to circle one
descriptor word within each category that best described the pain they felt during the vaccination.
If a category did not include a descriptor word describing their pain experience, the participant
had the option to select “no words applicable” for that particular group. Each descriptor word had
a previously assigned numeric rating, and a final pain score for each o f the four categories was
calculated by adding all the ratings within each specific category. An overall error score was also
calculated for each o f the four categories. Comparing pain words selected directly post

24

immunization (Time 1) to pain words selected during the follow-up visit (Time 2) derived this
score. An error was scored if the participant chose a word at Time 1 but failed to indicate that
same pain descriptor word at Time 2, an error was also scored if a participant chose a pain
descriptor at Time 2, but they had not selected that word at Time 1.
Geriatric Depression Scale S h o rt Form (GDS; Sh eikh & Yesavage, 1986; A ppendix D).
Depression has been shown to be prevalent in elderly populations living in private households or
in institutions, with rates ranging from 1% to 16% (Djemes, 2006). Depression has also been
shown to amplify pain sensations (von K orff & Simon, 1996). To control for depression and its
potential effects on pain behaviours within the target population, it was assessed using the GDS.
The GDS short form is a self-report measure consisting o f 15 yes-no items specifically
designed to assess depression in an elderly population. This instrument has been shown to be
both reliable and valid. These strong psychometric properties have been demonstrated when the
GDS is administered to samples o f functionally impaired, cognitively intact, community dwelling
or primary care participants (Friedman, Heisel & Delavan, 2005).

Semi-Structured Interview.
Social inhibition: Self-disclosure task (Beer et a l, 2003; A ppendix E). To test social
inhibition, participants were asked to complete a face-to-face interview, which focused on levels
o f self-disclosure. Participants were presented with a set o f emotional terms. The emotional
terms included five self-conscious emotions: embarrassment, guilt, pride, self-conscious, and
shame. After each emotional term was given, the participant was asked to define the term and to
provide a narrative o f when they felt that particular emotion. The self-conscious emotional terms
were counterbalanced.

25

Each participant’s self-disclosure interview was recorded using a hand-held Sony digital
voice recorder. These recordings were later transcribed, word for word, for further coding
analysis.

Procedure
Influenza vaccination clinic. Upon arrival at the influenza vaccination clinic, a research
assistant approached potential participants and explained the study objectives and study
procedures. Participants who agreed to take part in the study were asked to read and sign the
consent form (Appendix A). To ensure patient privacy, the research assistant entered the cubical
after the vaccination procedure was explained to the participant. The vaccination procedure was
recorded with the digital camera, which captured the participants' behavioural responses to the
vaccination. For the purposes o f later behavioural coding, the research assistant recording the
procedure clicked a pen to indicate the exact moment o f needle insertion.
To determine subjective pain experience, participants were asked to make a pain rating by
drawing a line on the Coloured Analogue Pain Scale immediately following the influenza
vaccination procedure. To further assess their subjective pain experience, participants were
given a list o f pain descriptors from the McGill Pain Questionnaire and asked to identify which
pain words best described their pain experience during the injection.
Due to time and space considerations following the influenza vaccination procedure, not
all questionnaires could be completed immediately. Therefore, a research assistant made a
follow-up appointment with the participant to complete the remaining questionnaires (see followup appointment procedure below).

Follow-up appointment. After an average o f 30 days (SD = 35 days) following the
influenza vaccination, follow-up appointments for participants were completed. During the

26

follow-up appointment a trained research assistant gathered demographic information, measured
cognitive functioning, assessed memory for pain with the MPQ, and conducted a semi-structured
Social Inhibition Interview (Beer et al., 2003). Specifically, the Montreal Cognitive Assessment
(MoCA) was used to assess the overall level o f cognitive function and frontal lobe functioning
was measured using the Wisconsin Card Sorting Task, the Stroop test and the Phonemic Verbal
Fluency Task.
To test the level o f social inhibition o f participants, each participant was asked to engage
in a self-disclosure interview. The self-disclosure interview used in this study was modeled after
Beer et al. (2003). Each participant was presented with a set o f self-conscious emotional terms
and asked to define each term and provide an example o f when they felt that emotion.
To assess the memory for pain o f participants, the same list o f MPQ pain descriptors used
during the vaccination procedure was presented and participants were asked to recall their pain
experience during the influenza vaccination.
O bservational m easures. Video recordings o f participant behaviour were examined for
evidence indicative o f pain. Measures o f pain were observed looking for specific indices o f
facial expressions o f pain post needle insertion.
To provide further rigor to the assessment o f observed pain in this elderly sample,
volunteer independent judges examined participant pain, post needle insertion. The independent
raters based their pain ratings upon participants’ general behaviour post needle insertion.
In d ex o f Facial Pain Expression (IFPE; Prkachin & Rash, unpublished manuscript).
Facial activities indicative o f pain displayed by participants during the vaccination procedure
were coded using the IFPE. The IFPE is a facial coding system that has been derived from the
more complex Facial Action Coding System (FACS; Ekman & Friesen, 1978). FACS identifies

27

44 actions that the face is capable o f performing. These facial actions are defined in terms o f the
underlying musculature and the changes the muscle movement makes in the facial appearance.
Research in pain has shown that there are a limited and distinct number o f facial actions that
appear when a person is experiencing pain. The IFPE is a coding system that has been developed
to observe these limited and distinct facial actions and has been used in a number o f studies
(Prkachin, 1992; Prkachin & Solomon, 2008; Rocha, Prkachin, Beaumont, Hardy & Zumbo,
2003). Using the system, observers code four facial actions: brow lowering, orbit tightening,
levator tightening, and closing o f the eyes. If the facial action is present, it is further coded for
intensity. Intensity ratings can range from 1 (trace) to 5 (extreme). For eye closure, a code o f Y
(yes) is given a score o f 1; N (no) is given a score o f 0. The numeric scores for each action are
simply summed, yielding the IFPE score, which can range from 0 to 16 for any individual.
Behavioural coding. The time segments analyzed for IFPE coding were the 10-second
pre-immunization phase and the 10-second post-immunization phase. These two phases were
further broken down into 2-second segments. To determine the participants’ overall change in
their facial expression o f pain from the pre-immunization segment to the post-immunization
segment, the total observed scores from the pre-immunization segment were subtracted from the
post-immunization segment, producing a single pain reaction score.
Independent ratings o f pain. To further quantify the pain expressions in this study, a
judgm ent study was conducted. Judgment studies have two components, an encoder and a
decoder. The encoder is the individual who emits behaviour and the decoder is the individual
who interprets, scores, rates, or judges the behaviour. As indicated by Rosenthal (1987), the
basic judgment study ABC model is made up o f three parts: encoders’ internal state or emotion

28

(A), encoder behaviour (B), and decoder judgm ent (C). For the purpose o f this study, the
encoders were the participants receiving vaccinations and the decoders were independent judges.
Five independent decoders (3 female, 2 male) with an average age o f 34.20 years (SD =
6.30) rated participants’ pain behaviour using the same pre- and post-immunization video
segments used to obtain the IFPE pain score. Each judge rated each o f the 58 participants’ pre­
immunization segment and post-immunization segment on a single behavioural pain dimension.
The pain dimension was assessed via a 4-point Likert confidence interval rating scale ranging
from -2 (absolutely no pain) to 2 (absolutely pain). To obtain a single observed pain score for
each participant, the pre-immunization segment pain score was subtracted from the post­
immunization pain score. This type o f rating scale has been used in previous judgm ent studies
(Ambady & Rosenthal, 1993).
The effective reliability (Rosenthal, 1987) o f the decoders’ ratings on the pain dimension
was calculated. They were found to have an intraclass correlation o f r = .64, which exceeds the
recommended cutoff o f r = .60 (see Rosenthal, 1987). This indicates that the decoders’ ratings
were sufficiently homogeneous to calculate an average behavioural score for both the pre­
immunization segment and the post-immunization segment. To simplify statistical analysis, two
behavioural ratings (pre-immunization segment and post-immunization segment) were derived
for each participant by taking the average score across all 5 decoders.
Social inhibition: Self-disclosure task (see Beer et a l, 2003; A ppendix E). To test social
inhibition, participants were asked to complete a face-to-face interview, which focused on levels
o f self-disclosure. Participants were presented with a set o f emotional terms. The emotional
terms included five self-conscious emotions: embarrassment, guilt, pride, self-conscious, shame.
After each emotional term was given, the participant was asked to define the term and to provide

29

a narrative o f when they felt that particular emotion. The self-conscious emotional terms were
counterbalanced.
Coding self-disclosure task responses. To code the participants’ responses three judges,
blind to the participant’s cognitive status, reviewed the transcribed interviews. They coded the
participants’ responses for the intimacy o f their self-disclosure when defining the emotional
terms and when providing examples o f when they had felt a particular emotion. A self­
disclosure rating for each emotion was made using a 7-point Likert Scale, from 1 (not at all) to 7
(very much). Ratings for each participant were averaged across all emotions to produce an
overall index o f self-disclosure intimacy. Furthermore, ten percent o f the interviews were coded a
second time to determine inter-rater reliability. Inter rater-reliability was calculated using
Person’s r (Rosenthal, 1987). Inter-rater reliability was found to be quite high at r = .98.

Results
Demographics
O f the 74 elderly individuals who originally agreed to participate in the study, one
participant was excluded due to having a previously diagnosed chronic pain condition, one
participant’s influenza vaccination was not captured on videotape, and 14 elderly individuals did
not complete their follow-up visit. The number o f participants included in the final analysis was
58. Table 1 summarizes descriptive data about the elderly individuals who participated in the
study. O f the 58 participants in the study, 24 (41%) were male and 34 (59%) were female. The
average age o f the participants ranged between 56 and 92 years (M = 69.7 years o f age, SD = 8.3
years o f age). O f the 58 participants in the study, 54 (91.3%) had been given an influenza
vaccination previously and 15 (25.9%) o f the participants had taken some form o f pain
medication prior to their vaccination. No significant differences on facial expressions o f pain

30

were found between individuals that had taken pain medication and those that had not taken pain
medication prior to their vaccination F ( 1,55) = 0.02, ns.
To determine the number o f participants who met the criteria for MCI, scores from the
MoCA were utilized. Results indicated that o f the 58 participants, 34 (17 males and 17 females)
had a MoCA score o f 25 or lower, therefore meeting the study criteria for MCI. The remaining
24 (7 males and 17 females) who had a MoCA score o f 26 or higher were used as the comparison
group.

Pre-Analysis
Prior to conducting the statistical analysis, pre- and post- immunization expressions o f
pain (observational and judgment rated), neurological and executive functioning tests, and selfreported pain measures were examined through various SPSS analyses for accuracy o f data entry,
missing values, extreme outliers, and tests for normality.
Conducting stem and leaf plots, extreme values were found for the dependent variables:
IFPE pain rating, the average decoder behaviour ratings for the pre and post video segments,
CAS, and the McGill pain questionnaire administered at time 1 and 2. To determine if these
extreme scores affected the normal distribution o f the variables, a test for normality was
conducted.
Tests for normality indicated that the following variables did not fit a normal distribution:
the IFPE pain ratings, the decoder behaviour ratings, and the CAS (see Table 2). If a variable
varies widely from a normal distribution, standard practice (see Tabachnick & Fidell, 2001) is to
use a transformation method that produces skewness and kurtosis values as close to zero as
possible. The aforementioned sample variables were transformed using loglinear transformation

31

to bring the distributions as close as possible to normality (see Table 2). The loglineartransformed variables were used in all subsequent analyses.
To determine if the depression variable should be entered as a covariate in the following
statistical analysis, an independent samples t-test was run. With equal variance not assumed,
results indicated no significant differences in levels o f depression between the MCI group (A /=
2.2\ , S D = 2.17) and the comparison group ( M - 1.54, S D = 1.38), /(55.5) = 1.42,/? = .16, As no
differences were found, depression was not entered as a covariate in the subsequent statistical
analyses.

Specific Hypothesis Testing
Pain expression: Observed. To determine if the influenza vaccination produced pain
expression in this sample, a single sample t-test was conducted on the IFPE scores, testing the
null hypothesis that the average IFPE score did not differ from 0. Results indicated that pain
expression was present in this sample /(57) = 28.53,/? < .001, SEM= 0.03, tj2 - 0.94. To further
determine if observed pain was evident in this sample due to the influenza vaccination, a twotailed paired-samples t-test was conducted on the decoders’ ratings o f the pre-immunization
segments and the post-immunization segments. Results indicated that the decoders observed
significantly more pain during the post-immunization segment (M = 0.43, SD = 0.15) than during
the pre-immunization segment (Af = 0.25, SD = 0.16), t(51) = 7.46,/? < .001, S EM = 0.03, rj2 =
0.50.
Additionally, to demonstrate that both the IFPE and the decoder interpretation o f the pain
segments revealed similar pain behaviours a Pearson’s r correlation was conducted. Results

32

indicated a significant, albeit weak relationship between the IFPE and the decoders’ pain scores,
r = .28, p < .05 (1-tailed).
To test the first hypothesis that MCI participants would have increased facial expressions
o f pain when compared to controls, a two-way mixed model o f analysis o f variance (ANOVA)
was conducted. The between-subjects factor was participants’ cognitive impairment status (MCI
and normal). The within-subjects factor was the decoded pain expression (pre needle segment
and post needle segment). The pre-immunization and post-immunization decoded facial
expression main effect, and the pre- and post-immunization decoded expressions X cognitive
impairment status interaction were tested. The pre- and post-immunization decoded facial pain
expressions main effect was significant, F ( l , 56) = 54.45, p < .001, MSE = 0.93, partial tj2 =
0.49, with an observed power o f 100%. Participants showed more pain post needle insertion (M
= 0.43, SD - 0.15), than before the needle insertion (M= 0.25, SD = 0.16). The interaction effect
between pre- and post-immunization X participants cognitive status was not significant, F (1, 56)
= 0.22, p = 0.61, MSE = 0.01, partial tj2 - 0.01, with an observed power o f 8%. Due to the fact
that the interaction was not significant no follow-up tests were conducted.
To further test the first hypothesis that level o f MCI could predict pain expression, two
separate regressions were conducted. To evaluate the relation o f the independent variable, the
MoCA raw score, to pain measured using the observed IFPE score, a linear regression was
conducted. Results indicated that the linear regression o f the MoCA score did not significantly
predict the observed IFPE score R2 = .01, adjusted R2 - -0.01, F (l,5 6 ) = 0.31 , p = 0.58. A
second linear regression was conducted to evaluate the prediction o f the independent variable,
the MoCA score, on observed pain measured using the decoders’ pain score. The linear
regression o f the MoCA score did not significantly predict the decoders’ observed pain, R2 =

33

.001, adjusted R2 - -0.02, F (l,5 6 ) = 0.08, p = .78. These two linear regressions indicate that MCI
scores did not significantly predict observed facial expressions o f pain during the influenza
vaccination procedure.

Pain: Self-report. To test the second hypothesis that both the MCI group and the
comparison groups had similar levels o f self-reported pain directly after the vaccine was given, a
one-way ANOVA was conducted, with MCI vs. comparison group as the independent variable
and CAS score as the dependent variable. Results indicated no significant differences, F ( l , 57)
= 0.20, p = .66, MSE = 0.03, between the amount o f self-reported pain among MCI participants

(M = 0.13, SD = 0.17) and normal controls (M= 0.12, SD = 0.16). Thus, both groups reported
similar levels o f pain after the needle was given.
To test the third hypothesis that the MCI group would have a poorer memory o f their pain
experience when compared to the control group, one-way ANOVAs were conducted on each o f
the four separate MPQ categories, with MCI vs. comparison group as the independent variable
and category error score as the dependent variable. Results indicated no significant differences F
(1, 57) = 0.17, p = .69, MSE = 0.69 between the number o f memory errors in the sensory
category among MCI participants ( M - 1.47, SD = 2.16) and normal controls (M = 1.28, SD =
1.82). In regards to recall o f the affect associated words, results indicated no significant
differences F (1, 57) = 1.31,/? = .26, MSE = 0.60 between the number o f errors made among
MCI participants (M = 0.21, SD = 0.88) and normal controls (M= 0.00, SD = 0.00). Further,
results indicated no significant differences F( I , 57) = 0.11,p = .68, MSE = 0.02 on recall errors
for evaluative associated words between MCI participants (M= 0.12, SD = 0.33) and normal
controls (M = 0.09, SD = 0.28). Finally, in regards to the number o f recall errors committed in the
miscellaneous pain word category, results indicated no significant differences F ( l , 57) = 0.22, p

34

= .65, MSE = 0.10 on the number o f errors committed between Time 1 and Time 2 between MCI
participants (M= 0.29, SD = 0.80) and normal controls (A /= 0.21, SD = 0.51). Thus, both groups
had similar rates o f recall error when asked to choose words that would best describe the pain
they felt during the immunization.
To further explore memory for pain, a linear regression was conducted to determine if
scores o f cognitive impairment could predict the number o f recall errors made between Time 1
and Time 2. Results indicated that cognitive impairment scores could not significantly predict
overall recall errors, R2 = .01, adjusted R2 = -0.02, F (l,5 6 ) = 0.11, p = .74.

Executive functioning. Prior to testing the fourth hypothesis that MCI participants had
impairments in frontal lobe activities in comparison to the control participants, correlations were
conducted between the executive functioning assessments (see Table 3). This was done to
evaluate whether the Wisconsin Card Sorting task, the Stroop test, and the Verbal Fluency task
were measuring a common feature within this sample. Specifically, Pearson’s r correlations
indicated a significant negative correlation between the Stroop test and the number o f completed
Wisconsin Card Sorting task categories, r = -0.26,/? < .05. A significant negative correlation was
also found between the Stroop test and the Verbal Fluency task, r = -0.41,/? <. 01. Finally, a
positive correlation was found between the number o f categories completed on the Wisconsin
Card Sorting Task and the Verbal Fluency task, r = 0.29, p < .01.
To test the fourth hypothesis that MCI participants had impairments in tests sensitive to
frontal lobe functioning in comparison to the control participants, a one-way ANOVA was
conducted on the executive functioning assessments. A significant difference was found on the
Verbal Fluency Task, F ( l , 57) = 14.41,/? < .001, MSE = 138.10, with MCI participants
performing worse on this task (M - 26.65, SD = 12.43), when compared to the normal controls

35

(M= 38.54, SD = 10.70). A significant difference was also found on the number o f successfully
completed Wisconsin Card Sorting Task categories, F ( l , 56) = 9.46, p < .01, MSE = 1.92, with
MCI participants completing fewer Wisconsin Card Sorting Task categories (M= 1.94, SD =
1.24), when compared to the normal controls (A/= 3.08, SD = 1.56). A moderate difference was
found on the time it took to complete the incongruent Stroop task, F ( l , 52) = 3.59,p = .06, MSE
= 33,909.33, with MCI participants taking longer to complete the task ( M = 117.50, SD =
123.89), when compared to normal controls (M= 66.50, SD = 27.03).

Executive functioning and pain expression. To test the fifth hypothesis that a negative
relationship would exist between measures o f frontal lobe functioning and facial expressions o f
pain in MCI participants, Pearson r correlations were conducted. Results indicated no significant
relationships between measures reflecting frontal lobe functioning and facial expressions o f pain
post needle insertion (see Table 4).
To determine if levels o f executive functioning could predict observed pain expressions
two separate linear regressions were conducted. To evaluate the prediction o f the combined
independent variables, number o f Stroop errors, number o f WCST errors and VFT on the
observed pain measured using the observed IFPE score a linear regression was conducted.
Results indicated that the linear regression o f the executive functioning measures did not
significantly predict the observed IFPE score R2 = .05, adjusted R2 = -0.01, F(3,50) = 0.88, p =
0.46. A second linear regression was conducted to evaluate the prediction o f the independent
variable, number o f Stroop errors, number o f WCST errors and VFT on observed pain measured
using the decoders’ pain score. The linear regression o f the executive functioning measures did
not significantly predict the decoders’ observed pain scores, R2 = .03, adjusted R2 = -0.03,
F(3,53) = 0.54, p = .66. These two linear regressions indicate that executive functioning scores

36

did not significantly predict observed facial expressions o f pain during the influenza vaccination
procedure.

Social inhibition. Prior to testing the sixth and final hypotheses, correlations were
conducted to determine the relationship between the social inhibition interview and the executive
functioning tests. As previously described, if a person has reduced frontal lobe functioning they
may also exhibit reduced social inhibition. Pearson’s r correlation indicated no significant
relationship between the WCST (number o f completed categories and overall errors), Stroop
Task (time to complete task and overall errors), VFT and social inhibition.
To test the sixth hypothesis that comparison group participants would have higher rates o f
social inhibition when compared to MCI participants, a one-way ANOVA was conducted.
Results indicated no significant differences between the amount o f social inhibition between
MCI participants and normal controls F ( l , 55) = 0.08,/? = .78, MSE = 1.58.
To further determine if rates o f social inhibition could predict pain expression, two
separate linear regressions were conducted. The first linear regression was run to determine if the
independent variable scores o f social inhibition could predict the observed IFPE score. Results
indicated that scores o f social inhibition could not significantly predict observed IFPE scores, R2
= .01, adjusted R2 = -0.01, F (l,5 6 ) = 0.56,/? = .46. A second linear regression was conducted to
determine social inhibition scores could predict decoders’ observed pain scores. Results indicated
that social inhibition did not significantly predict decoders’ observed pain scores, R2 = .00,
adjusted R2 = -0.02, F (l,5 6 ) = 0.01,/? = .95.
To test the final hypothesis that a negative relationship would exist between social
inhibition and facial expressions o f pain in MCI participants, a Pearson’s r correlation was

37

conducted. Results indicated no significant relationships between levels o f social inhibition and
facial expressions o f pain, post needle insertion, in MCI participants.

Discussion
The purpose o f this study was to obtain evidence that mild cognitive impairment,
presumably reflecting reduced frontal lobe functioning, is associated with an increase in facial
expressions o f pain in the elderly. Research has shown that individuals diagnosed with dementia
often display more intense facial expressions o f pain when compared to cognitively intact elderly
individuals (Hadjistavropoulos, LaChapelle, MacLeod, et al., 2000; Porter et al., 1996). Yet,
research has not provided clear understanding o f why cognitively impaired individuals display
higher facial expressions o f pain.
One theory is that individuals who have cognitive impairment will have a harder time
inhibiting their facial expressions o f pain because the brain areas responsible for emotional
inhibition have diminished functioning. Kunz et al (2011) conducted a study to determine which
brain areas were activated when healthy individuals inhibited their pain processes. They showed
that pain expression was inversely related to frontostriatal activity, consistent with the downregulation o f facial displays (Kunz et al., 2011). Therefore, it is reasonable to speculate that,
cognitive impairment, potentially reflecting impairment to the frontal lobe, should yield an
increase in facial expressions o f pain when individuals are exposed to acute noxious stimuli, as
their ability to inhibit their facial expressions o f pain would be impaired.
This study was designed to test these ideas in a community sample, making use o f a
commonly employed and ethical procedure as a source o f noxious stimulation.

Pain Expression: Observed

38

Elderly individuals in the present study displayed increased facial activity in response to
the influenza injection, when compared to facial activity during the harmless events preceding
the needle puncturing the skin. The increase in overall facial activity was associated with an
increase in a number o f discrete facial actions including brow lowering, orbit tightening, levator
tightening, and closing o f the eyes. This facial pain profile is consistent with the literature on
pain in elderly individuals (Hadjistavropoulos, LaChapelle, MacLeod, et al., 2000; Porter et al.,
1996). Furthermore, the observation o f pain expression after needle insertion in this sample was
confirmed by the decoders’ pain ratings. Thus indicating, the flu injection pain stimuli chosen for
this study did meet preliminary, but low pain expectations for a noxious stimuli in this
population.
Contrary to expectations, no significant differences were found between participants
identified as having MCI and normal controls in their facial expressions o f pain post­
immunization. Furthermore, tested executive functioning scores did not significantly predict
observed facial expressions o f pain in this sample. These two findings are inconsistent with
previous research findings (Kunz, Scharmann, Hemmeter, Schepelmann & Lautenbacher, 2007),
which showed an increase in facial expressions o f pain in cognitively impaired individuals when
compared to normal controls. The inconsistency between the two study results can readily be
explained, as Kunz et al. observed facial expression o f pain in individuals who had a previous
diagnosis o f moderate cognitive impairment. Whereas, the sample used for this study used
participants with undiagnosed mild cognitive impairment. Therefore, the differences in facial
expressions o f pain could be attributed to level o f cognitive impairment. That is, one could
speculate that the more severe the level o f cognitive impairment the more intense the facial
expressions o f pain would become.

39

Pain Expression: Self-Report
As expected, elderly individuals, regardless o f cognitive status, self-reported similar
levels o f pain post needle insertion. This finding is in contrast with previous research indicating
that cognitively impaired individuals are more likely to report less pain overall, compared to
normal control elderly individuals (Horgas, Elliot, & Marsiske 2009; Parmelee, Smith & Katz,
1993)
Interestingly, no cognitive differences were found in the present study in participants’
memory o f their pain experiences after the needle insertion. More specifically, MCI participants
and normal control participants had similar levels o f memory for their pain experience at followup. This finding was particularly surprising, as it was expected that participants with self­
indicated MCI symptoms would have more difficulty remembering how they had previously
described their pain experience. However, an interesting finding emerged that participants,
regardless o f cognitive status, were more likely to describe their post-immunization pain as being
sensory in nature than describing their pain as being affective, evaluative or miscellaneous in
nature. It can be concluded that within this sample, participants were more focused on the
sensation o f the needle insertion.

Executive Functioning
Consistent with expectations arising from the fourth hypothesis, MCI participants showed
significant impairment in executive functioning tasks when compared to normal controls.
Specifically, MCI participants generated fewer words on the Verbal Fluency Task, completed
fewer categories on the Wisconsin Sorting Task, and they took longer to complete the Stroop
task. This result is contrary to current research reporting that individuals with MCI normally do
not display significant deficits in executive functioning (Petersen, 2011), but rather displays

40

evidence o f memory impairment, preservation o f general cognitive and functional abilities and
absence o f a dementia diagnosis (Morris, Storandt, Miller, McKeel, Price, Rubin, et al., 2001).
However, these results are congruent to research, which has looked at individuals who have been
diagnosed with non-amnestic MCI. Non-amnestic MCI subtype has been implicated in the
development o f frontotemporal lobe degeneration (Petersen, 2011). The chosen executive
functioning tests may have been sensitive enough to pick up minor deficits in executive
functioning. These results could further indicate that a proportion o f the study sample is
displaying non-amnestic MCI. This would indicate, that in time, greater facial expressions o f
pain would be observed, as social inhibitions would be released due to frontal lobe impairment.
However, the best way to determine this would be to follow the MCI participants over a long
period o f time to observe if they developed frontotemporal lobe degeneration, vascular
degeneration or Parkinson’s dementia/Lewy body dementia, and re-observe their facial
expressions o f pain when exposed to a common noxious stimulus.

Executive functioning and pain expression. Focusing on the hypothesized negative
relationship between executive functioning and pain expression, results indicated that within this
sample no relationship existed between level o f executive functioning and pain expressions post
needle insertion. Furthermore, executive functioning did not predict pain expression in this
sample.

Social Inhibition
Contrary to the hypothesis that normal control participants would have higher levels o f
social inhibition in comparison to MCI participants, no differences were found between the two
groups in the amount o f personal information disclosed during the social inhibition interview.
Furthermore, levels o f social inhibition did not significantly predict levels o f pain expression post

41

needle insertion. This result is not surprising due to the lack o f significant differences in the level
o f disclosure between the MCI and normal control groups.
Finally, contrary to the hypothesis that a negative relationship would be found between
social inhibition and pain expression in MCI participants, results showed that no significant
relationship existed.

Limitations of Study
Research conducted with an elderly and potentially cognitively impaired population can
be quite complex. Many ethical considerations had to be taken into account during the study
design and study implementation. The first ethical consideration taken into account was the pain
stimuli used to elicit a pain response in the study sample, and we wanted to avoid subjecting a
vulnerable population to unnecessary pain. As per the study parameters, we did not determine
level o f cognitive impairment until the follow-up visit, and there was ethical concern that some
individuals may inadvertently participate in the study that could not give proper informed
consent due to their level o f cognitive capabilities. Therefore, it was decided to use a fairly
common medical procedure, influenza immunization, as the pain stimuli in this study. The
second ethical consideration taken into account was level o f cognitive impairment. Due to the
progressive nature o f cognitive impairment and its tendency to reduce an individual’s ability to
give informed consent or follow direction, it was decided to test the study hypothesis with
individuals who had mild cognitive impairment. Research has demonstrated that elderly
individuals with mild cognitive impairment are able to give informed consent (Jefferson, Lambe,
Moser, Byerly, Ozonoff & Karlawish, 2008) and can therefore follow study protocol.

42

The findings o f this study should be considered with some important limitations. The first
limitation that needs to be addressed is the vaccination procedure. First, the effectiveness o f
needle insertion in eliciting a pain response must be considered. Prior research has successfully
demonstrated that influenza immunizations (Hadjistavropoulos, Craig, Martin, Hadjistavropoulos
& McMurty, 1997) and venipuncture (Hadjistavropoulos, LaChapelle, MacLeod, Hale, O'Rouke
& Craig, 1998; Porter, Malhotra, Wolf, Morris, M iller & Smith, 1996) can induce facial pain
responses. Furthermore, pain responses were found in this sample, although they were not
strong. During initial data analysis a floor effect was observed in the IFPE scores; that is, pain
was not observed in a number o f the participants. Anecdotally, many participants indicated that
they did not even feel the needle puncture their skin. This may have been due to the size o f the
needle. Influenza injections are often given using a 30-gauge needle (0.16 mm); this is quite a
small needle in comparison to an intravenous 22-gauge needle (0.41 mm) or a blood draw 16gauge needle (1.19 mm). One could conclude that the influenza vaccination may not have been
sufficiently noxious to elicit a wide range o f pain expressions within this sample. It should also
be pointed out, the Hadjistavropoulos, et al. study, which used immunizations as the pain stimuli,
found differences in facial expressions o f pain between individuals who were cognitively
impaired and normal controls. However, there was an important difference between the
aforementioned study and the current study, and that was level o f cognitive impairment.
Hadjistavropoulos, et al. only retained participants who had “substantial cognitive disability” (pg.
73). These considerations may explain why differences in pain expressions were not seen
between the MCI participants and the normal control participants in this sample.
Secondly, while conducting the study in a clinical setting enhanced its real-world validity,
it resulted in an inability to control potential interactions o f the participants’ reactions. While all

43

the participants were receiving an influenza vaccination, the persons administering the
vaccination differed from site to site. For example, nurses administered the vaccine at two o f the
clinics, whereas a nurse practitioner and two resident doctors administered the vaccination at the
other two sites. It is possible that the majority o f the individuals administering the injections
used a gentler method resulting in lack o f variability o f subsequent pain experiences.
A final source o f potential variance in the vaccination procedure was the clinic settings.
Participants were recruited from four separate influenza vaccination clinics. At two clinics,
individuals were given their vaccination in separate closed-door rooms. In contrast, in the other
two clinics vaccinations were given in a large open room with no privacy between vaccination
stations. Ferrell, Ferrell, Ahn and Tran (1994), demonstrated that distraction is an effective pain
management technique in the elderly. The open room set-up could have offered distractions for
the participants while the needle was being inserted into their arm. Therefore, reducing the level
o f observed facial expressions o f pain.
Another limitation that needs to be addressed is the cognitive status o f the participants.
Due to the nature o f cognitive impairment and the desire to ensure we had a sample that could
freely give consent, fully understand the study procedure and give self-report ratings, we retained
individuals with mild cognitive impairment. As previously stated, mild cognitive impairment is
seen as a midway point between normal cognitive functioning and noticeably impaired cognitive
functioning. Mild cognitively impaired groups may have been too similar in frontal lobe
functioning to the normal control group; thus, levels o f frontal lobe functioning may have been
unable to predict pain expressions. If this study was to be replicated, individuals with moderate
cognitive impairment should be considered. It should be noted that a few o f our participants
likely did have a more moderate degree o f cognitive impairment (e.g., a MOCA score as low as

44

9/30); however low participant numbers did not allow for a statistical comparison between these
participants and those with higher MOCA scores within the MCI group.
The final limitation that needs to be considered is statistical power. As reported,
statistical power was quite low for many o f the null results 6% -l 1%. This lack o f power could
be due to a number o f factors. The first factor to be considered is sample size. O f the 74
originally recruited participants, 24 did not complete their follow-up visit; therefore, a full data
set was not obtained and they were removed from the study. When possible, reasons for why the
individual did not want to complete the follow-up were obtained. M any indicated they were too
busy, had family emergencies, or in one instance the individual was concerned that the cognitive
assessments may lead to the loss o f independence. The second factor that could have contributed
to a lack o f power was the pain stimuli. As indicated in the results section a floor effect was seen
in the facial expressions o f pain. A more noxious clinical pain stimuli, such as blood draws or
the insertion o f an intravenous needle, may yield a larger variance in pain expression; therefore
increasing power. The final factor that may influence power is cognitive status. As previously
indicated, the target group was categorized as having mild cognitive impairment. A wider range
o f cognitive impairment levels would increase the variance therefore increasing power.

Suggestions for Future Research
This study raised issues that should be researched further. An important follow up study
would be using the same research parameters put forth in this study in combination with a more
painful clinical stimulus and a higher level o f cognitive impairment. For example, instead o f an
influenza vaccination, one could observe the facial expressions o f elderly individuals as they
undergo rehabilitation for joint replacement. In using a more painful stimulus, a wider range o f
pain expressions may be obtained.

45

Another area o f research that could be examined is that o f more severe cognitive
impairment and its effects on facial expression o f pain through the release o f social inhibitions.
As pointed out earlier, mild cognitive impairment can be classified into two subtypes: amnestic
and non-amnestic. One could test the hypothesis that individuals diagnosed with non-amnestic
cognitive impairment show higher levels o f pain expression when compared to amnestic MCI or
normal controls.

Conclusion
In conclusion, elderly individuals are a vulnerable population who undergo numerous
painful medical procedures, regardless o f cognitive status. In addition to frequent medical
treatments that may be painful, decline in cognitive abilities is common in this population. We
learn from a young age to hide certain emotions, such as pain expressions. However, as we age,
this form o f social inhibition may be released due to frontal lobe impairment caused by dementia.
Therefore, these individuals may display higher levels o f pain then individuals who are
cognitively intact. In this study it was found that mild cognitive impairment did not influence
facial expressions o f pain or levels o f social inhibition. Frontal lobe involvement should be not
discounted in the explanation o f increased facial expressions o f pain in cognitively impaired
individuals. Further research should be conducted to determine the level o f involvement the
frontal lobe may play in the expression o f pain.
This study has raised some interesting research questions that need to be studied further
to get a better understanding o f how pain is manifested in this vulnerable population. It is
important to find ways to interpret how pain is experienced, especially during common medical
practices, in order to promote optimal quality o f life and well being o f the elderly. As care of

46

elderly moves to the forefront o f modem healthcare, it is o f great benefit to fully understand the
patient experience o f the elderly, particularly the perception and expression o f pain.

47

References
Abu-Saad, H., Bours, G., Stevens, B. & Hamers, J. (1998). Assessment o f pain in the neonate.
Seminars in Perinatology, 22, 402-416. doi: 10.1016/S0146-0005(98)80056-6
Albert, M. & Blacker, D. (2006). Mild Cognitive Impairment and Dementia. Annual Review o f
Clinical Psychology,2, 379-388. doi: 10.1146/annurev.clinpsy.l.102803.144039.
Alvarez, J. & Emory, E. (2006). Executive functioning and the frontal lobes: A meta-analytic
review .Neuropsychology Review, 16, 17-42. doi:10.1007/sl 1065-006-9002-x
Ambady, N. & Rosenthal, R. (1993). H alf a minute: predicting teacher evaluations from thin
slices o f nonverbal behavior and physical attractiveness. Journal o f Personality and
social Psychology, 64, 431-441. doi: 10.1037/0022-3514.64.3.431
Anand, K. & Craig, K. (1996). New perspectives on the definition o f pain. Pain, 67, 3-6.
doi: 10.1016/0304-3959(96)03135-1
Backman, L. (2008). Memory and cognition in preclinical dementia: What we know and what we
do not know. Canadian Journal o f Psychiatry, 53, 354-360.
Barr, R. (1992). Is this infant in pain? Caveats from the clinical setting. A P S Journal, 1, 187-190.
doi: 10.1016/1058-9139(92)90008-Z.
Beer, J., Heerey, E., Keltner, D., Scabini, D. & Knight, R. (2003). The regulatory function o f
self-conscious emotion: Insight from patients with orbitofrontal damage. Journal o f
Personality and Social Psychology, 85, 594-604. doi: 10.1037/0022-3514.85.4.594.
Benedetti, F., Vighetti, S., Ricco, C., Lagna, E., Bergamasco, B., Pinessi, L. & Rainero, I. (1999).
Pain threshold and tolerance in Alzheimer’s disease. Pain, 80, 377-382. doi:
10.1016/S0304-3959(98)00228-0.

48

Bischkopt, J., Busse, A. & Angermeyer, M. (2002). Mild cognitive impairment - a review o f
prevalence, incidence and outcomes according to current approaches. Acta Psychiatrica
Scandinavica, 106, 403-414.
Blair, R. (2003). Facial expressions, their communicatory functions and neuro-cognitive
substrates. Philosophical Transactions o f the Royal Society B Biological Sciences, 358,
561-572. doi: 10.1098/rstb.2002.1220.
Blumer, D. & Benson, D. (1975). Personality changes with frontal and temporal lobe lesions. In
D.E. Benson and D. Blumer , Eds. Psychiatric Aspects o f Neurological Disease. New York:
Grune & Stratton.
Bonica, J. (1990). The Management o f Pain (2nd ed.). Philadelphia: Lea & Febiger.
Canadian Study o f Health and Aging W orking Group (1994). Canadian Study o f Health and
Aging: Study methods and prevalence o f dementia. Canadian Medical Association
Journal, 150, 899-913.
Brothers, L. (1996). Brain mechanisms o f social cognition. Journal o f Psychopharmacology, 10,
2-8. doi: 10.1177/026988119601000102.
Carlino, E., Benedetti, F., Rainero, I., Asteggiano, G., Cappa, G., Tarenzi, L., Vighetti, S. &
Polio, A. (2010). Pain perception and tolerance in patients with frontotemporal dementia.
Pain, 151, 783-789. doi: 10.1016/j.pain.2010.09.013.
Chan, R., Shum, D., Toulopoulou, T. & Chen, E. (2008). Assessment o f executive functions:
Review o f instruments and identification o f critical issues. Archives o f Clinical
Neuropsychology, 23, 201-216. doi: 10.1016/j.acn.2007.08.010.
Darwin, C. (1872). The Expression o f the Emotions in Man and Animals. Chicago, Illinois:
University o f Chicago.

49

Djeraes, J. (2006). Prevalence and predictors o f depression in populations o f elderly: A review.
Acta Psychiatrica Scandinavica, 113, 372-387. doi: 10.111 l/j.l600-0447.2006.00770.x
Ekman, P. (1999). Facial expressions. In: The Handbook o f Cognition and Emotion (Eds. T.
Dalgleish and T. Power), p. 45-60. Sessex, UK: Wiley.
Ekman, P. & Friesen, W. (1969). The repertoire o f nonverbal behavior: categories, origins,
usage, and coding. Semiotica, 1, 49-98.
Ekman, P. & Friesen, W. (1978). Facial Action Coding System: A technique fo r the
measurement o f facial movement. Palo Alto, California: Consulting Psychologists Press.
Erskine, A., Morley, S. & Pearce, S. (1990). Memory for pain: a review. Pain, 41, 255-265. doi:
10.1016/0304-3959(90)90002-U
Felt, K. (2000). Improving assessment and treatment o f pain in cognitively impaired nursing
home residents. Annals o f Long-Term Care, 8, 36-42.
Ferrell, B. & Ferrell, B. (1990). Easing the Pain. Geriatric Nursing, 175-178. doi:
10.1016/SO 197-4572(05)80337-4
Ferrell, B., Ferrell, B., Ahn, C. & Tran, K. (1994). Pain management for elderly patients with
cancer at home. Cancer, 74, 2139-2146.
Ferrell, B., Ferrell, B. & Osterweil, D. (1990). Pain in the nursing home. Journal o f the
American Geriatrics Society, 38, 409-414.
Folstein, M., Folstein, S. & McHugh, P. (1975). Mini-mental state: a practical method for
grading the cognitive state o f patients for the clinician. Journal o f Psychiatric Research,
12, 189-198.
Forrest, J. (1995). Assessment o f acute and chronic pain in older adults. Journal o f
Gerontological Nursing, 21, 15-20.

50

Franz, E. & Gillett, G. (2011). John Hughlings Jackson’s evolutionary neurology: a unifying
framework for cognitive neuroscience. Brain, 134, 3114-3120.
Friedman, B., Heisel, M. & Delavan, R. (2005). Psychometric properties o f the 15-item geriatric
depression scale in functionally impaired, cognitively intact, community dwelling
primary care patients. Journal o f the American Geriatric Society, 53, 1570-1576.
Freitas, S., Simoes, M., Alves, L., Duro, D. & Santana, I. (2012). Montreal Cognitive
Assessment (MoCA): Validation study for Frontotemporal Dementa. Journal o f Geriatric
Psychiatry and Neurology, 25, 146-154. doi: 10.1177/0891988712455235.
Gabre, P., Sjoquist, K. (2002). Experience and assessment o f pain in individuals with cognitive
impairments. Special Care in Dentistry, 22, 174-180. doi: 10.1111/j. 17544505.2002.tb00267.x
Ganguli, M., Dodge, H., Chen, P., Belle, S. & Dekosky, S. (2000). Ten-year incidence o f
dementia in a rural elderly US community population: the Movies Project. Neurology, 54,
1109-1114.
Gillett, G. & Franz, E. (2013). John Hughlings Jackson: bridging theory and clinical
observation. The Lancet, 381, 528-529.
Giovagnoli, A. (2001). Relation o f sorting impairment to hippocampal damage in temporal lobe
epilepsy. Neuropsychologia, 39, 140-150. doi: 10.1016/S0028-3932(00)00104-4
Goldin, P., McRae, K., Ramel, W. & Goss, J. (2008). The neural bases o f emotion regulation:
reappraisal and suppression o f negative emotion. Biological Psychiatry, 63, 577-586. doi:
10.1016/j.biopsych.2007.05.031
Goldstein, B., Obrzut, J., John, C., Ledakis, G. & Armstrong, C. (2004). The impact o f frontal
and non-frontal brain tumor lesions on Wisconsin Card Sorting Test performance. Brain

51

and Cognition, 54, 110-116. doi: 10.1016/S0278-2626(03)00269-0
Hadjistavropoulos, T., Craig, K. & Fuchs-Lacelle, S. (2004). Social influences and the
communication o f pain. In T. Hadjistavropoulos & K. Craig (Eds.). Pain: Psychological
perspectives (pp. 87-112). Mahwah, New Jersey: Lawrence Erlbaum Associates.
Hadjistavropoulos, T., Craig, K., Martin, N., Hadjistavropoulos, H. & McMurty, B. (1997).
Toward research outcome measure o f pain in frail elderly in chronic care. Pain Clinic, 10,
71-79.
Hadjistavropoulos, T., LaChapelle, D., MacLeod, F., Hale, C., O ’Rourke, N. & Craig, K. (1998).
Cognitive functioning and pain reactions in hospitalized elders. Pain Research and
Management, 3, 145-151.
Hadjistavropoulos, T., LaChapelle, D., MacLeod, F., Snider, B. & Craig, K. (2000). Measuring
movement-exacerbated pain in cognitively impaired frail elders. Clinical Journal o f Pain,
16, 54-63. doi: 10.1097/00002508-200003000-00009
Hadjistavropoulos, T., von Baeyer, C. & Craig, K. (2001). Pain assessments in persons with
limited abilities to communicate. In D. Turk & R. Melzack (Eds.). Handbook o f Pain
Assessment, 2nd Ed., (pp. 134-149). New York: Guilford.
Haasum, Y., Fastborn, J., Fratiglioni, L., Karenholt, I., Johnell, K. (2011). Pain treatment in
elderly persons with and without dementia: a population based study o f institutionalized
and home-dwelling elderly. Drugs Aging, 28, 283-293. doi: 10.2165/11587040-

000000000- 00000.
Heaton, R. (1981). Wisconsin Card Sorting Test. Odessa, Florida: Psychological Assessment
Resources.

52

Horgas, A., Elliot, A. & Marsiske, M. (2008). Pain Assessment in persons with dementia:
relationship between self-report and behavioural observation. Journal o f the American
Geriatrics Society, 57, 126-132. doi: 10.1111/j. 1532-5415.2008.02071.x
Hughlings Jackson, J. (1884). Croonian lectures on the evolution and dissolution o f the nervous
system. Lancet, 26, 555-558; 649-652; 739-744.
International Association for the Study o f Pain (IASP) Task force on Taxonomy (2008). IASP
Pain Termonology [online]. Available: http://www.iasp-pain.org/terms-p.html.
Jefferson, A., Lambe, S, Moser, D., Byerly, L. Ozonoff, A. & Karlawish, J. (208). Decisional
capacity for research participation in individuals with mild cognitive impairment. Journal
o f the American Geriatrics Society, 56, 1236-1243. doi: 10.1111/j. 15325415.2008.01752.x
Kennedy, G. (2004). The assessment o f executive dysfunction: Importance for diagnosis and
prognosis. Primary Psychiatry, 11, 19-20.
Kolb, B. & Whishaw, I. (2009). Fundamentals o f Human Neuropsychology: 6th Edition. New
York, New York: Worth Publishers.
Kunz, M., Chen, J-I., Lautenbacher, S., Vachon-Presseau, E. & Rainville, P. (2011). Cerebral
regulation o f facial expressions o f pain. The Journal o f Neuroscience, 31, 8730-8738. doi:
10.1523/ JNEUROSCI. 0217-11. 2011
Kunz, M., Scharmann, S., Hemmeter, U., Schepelmann, K. & Lautenbacher, S. (2007). The
facial expression o f pain in patients with dementia. Pain, 133, 221-228. doi:
10.1016/j .pain.2007.09.007

53

Lautenbacher, S., Kunz, M., Strate, P., Nielsen, J. & Arendt-Nielsen, L. (2005). Age effects on
pain thresholds, temporal summation and spatial summation o f heat and pressure pain.
Pain, 115, 410-418. doi: 10.1016/j.pain.2005.03.025
Lezak, M., Howieson, D. & Loring, D. (2004). Neuropsychological Assessment. New York:
Oxford University Press.
Manz, B., Mosier, R., Nusser-Gerlach, M., Bergstrom, N. & Agrawal, S. (2000). Pain assessment
in the cognitively impaired and unimpaired elderly. Pain Management Nursing, 1,
106-115. doi: 10.1053/jpmn.2000.19332
Martin-Matthews, A. (2002). The Health Transition Fund-Seniors Health. Health Canada.
Available at www.hc-sc.gc.ca.
McGrath, P., Seifert, S., Speechley, K., Booth, J., Sitt, L. & Gibson, M. (1996). A new analogue
scale for assessing children’s pain. Pain, 64, 435-443.
doi: 10.1016/0304-3959(95)00171-9
Melzack, R. (1975). The McGill Pain Questionnaire: Major properties and scoring methods.
Pain, 1, 277-299. doi: 10.1016/0304-3959(75)90044-5
Melzack, R. (1983). The McGill Pain Questionnaire. In R. Melzack, Ed. Pain Measurement and
Assessment. New York: Raven Press.
Milner, B. (1964). Some effects o f frontal lobectomy in man. In J.M. Warren and K. Akert, Eds.
The Frontal Granular Cortex and Behavior. New York: McGraw Hill.
Molano, J., Boeve, B., Ferman, T., et al. (2010). Mild cognitive impairment associated with
limbic and neocortical Lewy body disease: a clinicopathological study. Brain, 133, 540556.

54

Montes-Sandoval, L. (1999). An analysis o f the concept o f pain. Journal o f Advanced Nursing,
29, 935-941. doi: 10.1046/j. 1365-2648.1999.00971 x
Morris, J., Storandt, M. & Miller, J., McKeel, D., Price, J., Rubin, E. & Berg, L. (2001). Mild
cognitive impairment represents early-stage Alzheimer disease. Archives o f Neurology,
58, 397-405.
Morrison, R. & Siu, A. (2000). Surviving in end-stage dementia following acute illness. The
Journal o f the American Medical Association, 284, 47-52. doi: 10.1001/jama.284.1.47.
Mukhopadhyay, P., Dutt, A., Kumar Das, S., Basu, A., Hazra, A., Dhibar, T., et al. (2008).
Identification o f neuroanatomical substrates o f set-shifting ability: Evidence from patients
with focal brain lesions. Progress in Brain Research, 168, 95-104. doi: 10.1016/S00796123(07)68008-X
Munoz, M. & Esteve, R. (2005). Reports o f memory functioning by patients with chronic pain.
Clinical Journal o f Pain, 21, 287-291.
Mylius, V., Kunz, M., Henninghausen, E., Lautenbacher, S. & Schepelmann, K. (2008). Effects
o f aging on spinal motor and autonomic pain responses. Neuroscience Letters, 446,
129-132. doi: 10.1016/j.neulet.2008.09.048.
Nagahama, Y., Fukuyama, H., Yamauchi, H., Matsuzaki, S., Konishi, J., Shibasaki, H. &
Kimura, J. (1996). Cerebral activation during performance o f a card sorting task. Brain,
119, 1667-1675. doi: 10.1093/brain/l 19.5.1667
Nesreddine, Z., Phillips, N., Bedirian, V., Charbonneau, S., Whitehead, V., Collin, 1., et al.
(2005). The Montreal Cognitive Assessment (MoCA): A brief screening tool for Mild
Cognitive Impairment. Journal o f the American Geriatric Society, 53, 695-699.
Owen, A., Sahakian, B., Hodges, J., Summers, R., Polkey, C. & Robbins, T. (1995). Dopamine-

55

dependent fronto-striatal planning deficits in early Parkinson’s disease. Neuropsychology,
9, 126-140. doi: 10.1037/0894-4105.9.1.126
Owens, M. (1984). Pain in infancy: Conceptual and methodological issues. Pain, 20, 213-230.
doi: http://dx.doi.org/10.1016/0304-3959(84)90813-3
Oxford Dictionary (2013). Oxford University Press.
http://www.oxforddictionaries.com/definition/english/cognition, retrieved November 4,
2013.
Padovani, A., Di Piero, V., Bragoni M., et al. (1995). Patterns o f neuropsychological impairment
in mild dementia: A comparison between Alzheimer’s Disease and multi-infarct
dementia. Acta Neurological Scandinavica, 92, 433-442. doi: 10.1111/j. 16000404.1995.tb00477.x
Parmelee, P., Smith, B. & Katz, I. (1993). Pain complaints and cognitive status among elderly
institution residents. Journal o f the American Geriatric Society, 41, 517-522.
Petersen, R. (2004). Mild cognitive impairment as a diagnostic entity. Journal o f Internal
Medicine, 256, 183-194.
Petersen, R. (2011). Mild cognitive impairment. The New England Journal o f Medicine, 364,
2227-2234.
Prkachin, K. (1992). The consistency o f facial expressions o f pain: a comparison across
modalities. Pain, 51, 297-306. doi: 10.1016/0304-3959(92)90213-U
Prkachin, K. & Rash, J. (unpublished manuscript). Quantifying pain in humans using the Index
o f Facial Pain Expression (IFPE).

56

Prkachin, K. & Solomon, P. (2008). The structure, reliability and validity o f pain expression:
evidence from patients with shoulder pain. Pain, 139, 267-274. doi:
http://dx.doi.Org/10.1016/j.pain.2008.04.010
Razani, J., Boone, K., Miller, B., Lee, A. & Sherman, D. (2001). Neuropsychological
performance o f right- and left- frontotemporal dementia compared to Alzheimer’s
Disease. Journal o f the International Neuropsychology Society, 7, 468-480. doi:
10.1017/S 1355617701744037
Reisberg, B., Ferris, S., de Leon, M., et al. (1988). Stage specific behavioural, cognitive, and in
vivo changes in community residing sybjects with age-associated memory impairment
and primary degenerative dementia o f the Alzheimer type. Drug Development Research,
15, 101-114.
Reitsma, M., Tranmer, J., Buchanan, D. & Vandenkerkhof, E. (2011). The prevalence o f chronic
pain and pain-related interference in the Canadian population from 1994 to 2008. Chronic
Diseases and Injuries in Canada, 31.
Reyes-Gibby, C., Aday, L. & Cleeland, C. (2002). The impact o f pain on self-rated health in the
community dwelling older adults. Pain, 95, 75-82. doi: 10.1016/S0304-3959(01)00375-X
Roberts, R., Geda, Y., Knopman, D., Cha, R., Pankratz, V., Boeve, B., Ivnik, R., Tangalos, E.,
Petersen, R. & Rocca, W. (2008). The Mayo Clinic study o f aging: Design and sampling,
participation, baseline measures and sampling characteristics. Neuroepidemiology, 30,
58-69. doi: 10.1159/000115751.
Rocha, E., Prkachin, K., Beaumont, S., Hardy, C. & Zumbo, B. (2003). Pain reactivity and illness
behavior in kindergarten-aged children. Journal o f Pediatric Psychology, 28, 47-57. doi:
10.1093/jpepsy/jsl036

57

Rolls, E., Homak, J., Wade, D. & McGrath, J. (1994). Emotion-related learning in patients with
social and emotional changes associated with frontal lobe damage. Journal o f Neurology,
Neurosurgery, and Psychiatry, 57, 1518-1524. doi: 10.1136/jnnp.57.12.1518
Rosenthal, R. (1987). Judgment Studies: Design, Analysis, and Meta-Analysis. New York:
Cambridge University Press.
Samo, M., Postman, W., Cho, Y. & Norman, R. (2005). Evolution o f phonemic word fluency
performance in post-stroke aphasia. Journal o f Communication Disorders, 38, 83-107.
doi: 10.1016/j.jcomdis.2004.05.001
Sheikh, J. & Yesavage (1986). Geriatric Depression Scale (GDS): Recent evidence and
development o f a shorter version. In Clinical Gerontology: A Guide to Assessment and
Intervention (pp. 165-173). New York: Haworth Press.
Spencer, H. (1855/1889). The principles o f psychology. 3rd ed., Vol 1. New York: D. Appleton.
Statistics Canada (2008). Eldercare: What we know today.
Stuss, D. & Benson, D. (1984). Neuropsychological studies o f the frontal lobes. Psychological
Bulletin, 95, 3-28. doi: 10.1037/0033-2909.95.1.3
Stuss, D., Levine, B., Alexander, M., Hong, J., Palumbo, C., Hamer, L., et al. (2000). Wisconsin
Card Sorting Test performance in patients with focal frontal and posterior brain damage:
Effects o f lesion location and test structure on separable cognitive processes.
Neuropsychologia, 38, 388-402. doi: 10.1016/S0028-3932(99)00093-7
Stuss, D. & Levine, B. (2002). Adult clinical neuropsychology: Lessons from studies o f the
frontal lobes. Annual Review o f Psychology, 53, 401-433.
doi: 10.1146/annurev.psych.53.100901.135220

58

Sweet, L., Van Adel, M., Metcalf, V., Wright, L., Harley, A., Leiva, R. & Taler, V. (2011). The
Montreal Cogntivie Assessment (MoCA) in geriatric rehabilitation: psychometric
properties and association with rehabilitation outcomes. International Psychogeriatrics,
23, 1582-1591. doi: 10.1017/S1041610211001451
Tabachnick, B. & Fidell, L. (2001). Using Multivariate Statistics: 4th ed. Needham Heights,
Massachusetts: Allyn & Bacon.
Tangney, J., Miller, R., Flicker, L. & Barlow, D. (1996). Are shame, guilt, and embarrassment
distinct emotions? Journal o f Personality and Social Psychology, 70, 1256-1269. doi:
10.1037/0022-3514.70.6.1256
Trenerry, M., Crosson, B., Deboe, J. et al (1989). Stroop neurological screening test. Odessa,
Florida: Psychological Assessment Resources.
Tsai, P. and Means K. (2005). Osteoarthritic knee or hip pain. Possible indicators in elderly
adults with cognitive impairment. Journal o f Gerontological Nursing, 31, 39-45.
Tucker, M., Andrew, M., Ogle, S. & Davidson, J. (1989). Age associated change in pain
threshold measured by transcutaneous neuronal electrical stimulation. Age and Ageing,
18, 241-246. doi: 10.1093/ageing/18.4.241
Vilkki, J. & Holst, P. (1994). Speed and flexibility on word fluency tasks after focal brain
lesions. Neuropsychologica, 32, 1257-1262. doi: 10.1016/0028-3932(94)90107-4
von Korff, M. & Simon, G. (1996). The relationship between pain and depression. The British
Journal o f Psychiatry, 168, 101-108.
Winblad, B., Palmer, K., Kivipelto, M., et al. (2004). Mild cognitive impairment - beyond
controversies, toward a consensus: report on the International W orking Group o f Mild
Cognitive Impairment. Journal o f Internal Medicine, 256, 240-246.

59

Weddell, R. (1994). Effects o f subcortical lesion site on human emotional behavior. Brain and
Cognition, 25, 161-193. doi: 10.1006/brcg. 1994.1029
Won, A., Lapane, K., Gambassi, G., Bemabei, R., Mor, V. & Lipsitz (1999). Correlates and
management o f nonmalignaent pain in the nursing home. SAGE Study Group.
Assessment o f Geriatric drug use via Epidemiology. The Journal o f the American
Geriatric Society, 47, 936-942. doi: 10807/37445
York, G. & Steinberg, D (2006). An introduction to the life and work o f John Hughlings Jackson
with a catalogue raisonne o f his writings. Medical History Supplement, 26. London: The
Wellcome Trust Centre for the History o f Medicine, UCLA.
Zheng, Z., Gibson, S., Khalil, Z., Helme, R. & McMeeken, J. (2000). Age-related differences in
the time course o f capsaicin-induced hyperalgesia. Pain, 85, 51-58. doi: 10.1016/S03043959(99)00247-X

60

Table 1

Demographic characteristics o f sample

Diagnosed Illness

Normal
(n = 24)
19(79% )

Probable MCI
(n = 34)
24 (71%)

Chronic Pain

8 (33%)

11 (32%)

Age

A /=68.1 (SD = 8.19)

M = 70.8 (SD = 8.42)

Years o f Education

M - 13.9 (SD = 3.05)

M = 12.1 (SD = 2.92)

MoCA Raw Score

M - 27.5 (SD = 1.21)

M - 22.0 (SD = 3.83)

(26 - 30)

(9 -2 5 )

Demographic

(Range)

61

Table 2

Descriptive characteristics o f dependent variables
Controls (n-24)
Mean (SD)

Probable MCI (n=34)
Mean (SD)

Variable
Geriatric Depression Scale

1.54 (1.38)

2.21 (2.17)

Verbal Fluency Task

38.54 (10.70)

26.65 (12.43)

Stroop Errors

66.50 (27.03)

111.5 (123.89)

Wisconsin Card Sorting Task: Categories

3.08 (1.56)

1.94 (7.25)

Wisconsin Card Sorting Task: Error Number

18.67 (9.54)

22.58 (10.42)

Pre MPQ: Sensory

1.42 (2.65)

1.32 (2.10)

Pre MPQ: Affect

0.00(0.00)

0.03 (0.17)

Pre MPQ: Evaluative

0.04 (0.20)

0.00 (0.00)

Pre MPQ: Misc.

0.21 (0.59)

0.09 (0.57)

Post MPQ: Sensory

1.21 (2.32)

1.35 (3.03)

Post MPQ: Affect

0.00(0.00)

0.03 (0.77)

Post MPQ: Evaluative

0.04 (0.20)

0.26 (0.96)

Post MPQ: Misc.

0.54 (1.79)

0.47 (1.33)

Social Inhibition Interview

22.13 (4.76)

21.79 (4.28)

62

Table 3

Skewness and Kurtosis o f Pain Related Variables
Variable
IFPE Pain Score

Raw Data
Skewness
Kurtosis
1.23 (SE = .32) 1.15(S£=.62)

Loglinear Transformation
Skewness
Kurtosis
-0.27 (SE =.32) 2.35 (SE-.62)

Average decoders rating
pre-needle

1.19 (S£ =.32)

1.67 (SE=.62)

0.24 (SE =.32)

-0.47 (SE= 62)

Average decoders ratings
post-needle

0.29 (S E = 3 2 )

-.24 (SE=. 62)

-0.61 (SE =.32)

0.42 (SE=. 62)

CAS

2.50 (SE =.32)

8.77 (SE= 62)

1.19 (SE =.32)

0.81 (S£=.62)

Note. SE = Standard Error. IFPE = Index o f Facial Expression o f pain. CAS = Coloured Analogue Scale.

63

Table 4

Correlations between Executive Functioning Assessments

WCST
number o f
completed
categories
WCST
number o f
errors
Stroop length
o f time

WCST
number o f
completed
categories

WCST
number of
errors

Stroop length
o f time

Stroop
number o f
errors

VFT

1

-0.11**

-0.14

-0.26*

0.29*

1

0.06

0.18

-0.09

_

1

0.03

-0.41**

1

-0.19

_

Stroop
number o f
errors
Note. WCST = Wisconsin Card Sorting Task. VFT = Verbal Fluency Task.
** p < .01 (one-tailed)
* p < .05 (one-tailed)

64

Table 5

Correlations between Executive Functioning Assessments and Pain Expressions Post Needle
Insertion
Executive Functioning
Assessment

Pain Variable
IFPE

WCST Completed
Categories
WCST Number o f Errors

-0.06

Decoder rating post
needle insertion
0.14

0.18

-0.21

Stroop length o f time

-0.08

-0.01

Stroop number o f errors

0.03

-0.23

VFT

-0.03

-0.08

Note. WCST = Wisconsin Card Sorting Task. VFT = Verbal Fluency Task

65

APPEN D IX A: Inform ed Consent Form

Informed Consent Form
Principal Investigator: Dr. Kenneth Prkachin
Co-Investigator: Tammy Klassen-Ross, MSc, PhD Candidate
Purpose: As individuals age they can experience many different painful conditions.
Some of these elderly individuals also develop cognitive impairments. Cognitive
impairments can make it difficult for som e people to express pain. We are interested in
discovering how cognitive impairments can affect how elderly individuals express pain to
those around them. We are asking you to participate in this research because you fit
into our target age group (over 55) and you are here to receive an influenza injection.
The research is being conducted a s a PhD dissertation project at UNBC. Your doctor is
offering this research opportunity to patients and participation is voluntary.
Study Procedure: We will be asking your permission to record your influenza injection
on video. We will also ask you to complete a brief test of cognitive functioning. After the
injection, we will ask you to tell us how the inoculation felt. We will also be asking you to
participate in a follow-up visit within the next 2 weeks. During the follow-up visit we will
ask you complete som e short tests of cognitive activities and som e mood
questionnaires. The follow-up visit should take around 1 hour to complete. If you decide
to withdraw from the study all information gathered to that point will be destroyed.
Risks: There are no known risks to you for taking part in this study.
Benefits: Taking part in this study will not affect your care. It will help us understand
how elderly people with and without cognitive impairments react to painful situations.
This knowledge could contribute to improved care in the future.
Confidentiality: Any information from this research study will be kept strictly
confidential, only authorized personnel will have access. The results of your cognitive
functioning tests will be shared with your doctor. All documents and video recordings
will be identified by code number and kept in a locked filing cabinet and no nam es will
be used. Only expert project staff will view the video recordings. We will retain copies of
the video for our files indefinitely, but only code numbers will be used to identify you.
Contact: I understand that if I have any questions or desire further information with
respect to this study I should contact Dr. Prkachin at 250-960-6633 or Tammy KlassenRoss at 250-960-6446. If I have any concerns about my treatment a s a research
subject, I may contact the Research Ethics Board at the University of Northern British
Columbia, 250-960-6735 or by e-mail: reb@ unbc.ca. I may also contact

66

I have received a copy of this consent form for my own records (circle) Yes/No.
I consent for my health practitioner to receive a copy of my cognitive test results (circle)
Yes/No.
I understand that the video recordings may be used for future research, but no nam es
will be on the recordings (circle) Yes/No.
I agree to participate in a follow-up visit either at my home or at the University of
Northern British Columbia (circle) Yes/No.
I would like a copy of the research results (circle) Yes/No
If yes, please write down your e-mail or mailing address and a copy of the study results
will be provided to you upon the completion of the project.

Participant Signature

Date

Participant Name (Please print)

W itness Signature

Date

67

A PPEN D IX B: D em ographic and health status o f participant

PARTICIPANT QUESTIONNAIRE
Today’s Date:_______

Participant Number_________

What is your date o f b irth ?__________________
What is your highest level o f education?_______________________
Do you have any pre-existing medical conditions?
If yes, please explain:

Do you suffer from Chronic Pain?

Y/N

If yes, please explain:

Did you take any pain medication today? Y/N
If yes, what did you take:

68

Y/N

A PPEN D IX C: M cGill Pain Q uestionnaire

McGill Pain Questionnaire
Participant Number____________

Today’s Date__________________

Please Check O n e:_______PostIn je c tio n _______ 15-min. Post In je c tio n ________Follow-up
Please select one word in each box, or select Not Applicable if no words match your pain:
1.

Flickering
Quivering
Pulsing
Throbbing
Beating
Pounding
Not Applicable

6.

Tugging
Pulling
Wrenching
Not Applicable

11.

Tiring
Exhausting
Not Applicable

16. Annoying
Troublesome
Miserable
Intense
Unbearable
Not Applicable

2.

Jumping
Flashing
Shooting
Not Applicable

7.

Hot
Burning
Scalding
Searing
Not Applicable

12.

Sickening
Suffocating
Not Applicable

17.

3.

Prickling
Boring
Drilling
Stabbing
Lancinating
Not Applicable

8.

Tingling
Itchy
Smarting
Stinging
Not Applicable

13.

Fearful
Frightening
Terrifying
Not Applicable

18.

Tight
Numb
Drawing
Squeezing
Tearing
Not Applicable

4.

Sharp
Cutting
Lacerating
Not Applicable

9.

Dull
Sore
Hurting
Aching
Heavy
Not Applicable

14.

Punishing
Grueling
Cruel
Vicious
Killing
Not Applicable

19.

Cool
Cold
Freezing
Not Applicable

5.

10.

15.

20.

Pinching
Pressing
Gnawing
Cramping
Crushing
Not Applicable

Tender
Taut
Rasping
Splitting
Not Applicable

Wretched
Blinding
Not Applicable

69

Spreading
Radiating
Penetrating
Piercing
Not Applicable

Nagging
Nauseating
Agonizing
Dreadful
Torturing
Not Applicable

A PPEN D IX D: G eriatric D epression Scale

Geriatric Depression Scale (short form)
Instructions:

Circle the answer that best describes how you felt
over the past w eek.

1.

Are you basically satisfied with your life?

yes

no

2.

Have you dropped many o f your activities and
interests?

yes

no

3.

D o you feel that your life is empty?

yes

no

4.

Do you often get bored?

yes

no

5.

Are you in good spirits most o f the time?

yes

no

6.

Are you afraid that something bad is going to
happen to you?

yes

no

Do you feel happy most o f the time?

yes

no

yes

no

yes

no

yes

no

11. D o you think it is wonderful to be alive now?

yes

no

12. Do you feel worthless the way you are now?

yes

no

13. Do you feel full o f energy?

yes

no

14. Do you feel that your situation is hopeless?

yes

no

yes

no

7.

8

. Do you often feel helpless?

9.

Do you prefer to stay at home, rather than going
out and doing things?

10. Do you feel that you have more problems with
memory than most?

15. Do you think that most people are better off
than you are?

Total S c o ts

1

Tod*

m»y b* o o p M without pormlMion

70

A PPEN D IX E: Social Inhibition Interview Protocol

Social Inhibition Interview:
Explain to the participant that you will be naming 11 different emotions. For each emotion ask
them to define the emotion and give a short story o f when they felt that emotion.

Embarrassment
Guilt
Pride
Self-Conscious
Shame
Script:
For each emotion say the following:

Please tell me w hat__________means to you.
Can you tell me a short story when you felt__________ ?
Thank-you, are you ready for the next emotion.
When they are ready for the next word repeat the above script.

71