Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Feb 1.
Published in final edited form as: Pain Manag Nurs. 2021 Aug 2;23(1):55–61. doi: 10.1016/j.pmn.2021.07.001

The Relationship Between Pain, Function, Behavioral and Psychological Symptoms of Dementia and Quality of Life

Barbara Resnick 1, Elizabeth Galik 2, Ann Kolanowski 3, Kimberly VanHaitsma 4, Marie Boltz 5, Shijun Zhu 6, Jeanette Ellis 7, Liza Behrens 8, Karen Eshraghi 9, Cynthia Renn 10, Susan G Dorsey 11
PMCID: PMC8807789  NIHMSID: NIHMS1721541  PMID: 34353739

Abstract

Purpose:

This study evaluated the association between age, gender, comorbidities, cognition, and administration of opioids with pain and the impact of all of these variables plus function, agitation, resistivenes to care and depression on quality of life among nursing home residents with severe dementia.

Design:

This was a descriptive study using baseline data from the Evidence Integration Triangle for Behavioral and Psychological Symptoms of Dementia implementation study.

Methods:

Model testing was done using structural equation modeling.

The sample included 553 residents from 55 nursing homes with a mean age of 83.88 (SD=10.44) and mean Brief Interview of Mental Status of 4.30 (SD=3.50).

Results:

There were significant associations showing those who were older, male, had fewer comorbidities, better cognition and were black were more likely to have pain. Pain, in combination with the demographic and descriptive variables, explained 32% of the variance in function, 75% of the variance in depression, 88% of the variance in agitation, 98% of the variance in resistiveness to care and 92% of the variance in quality of life. The model however did not show a good fit to the data.

Conclusions:

The model did not have a good fit with the data which likely was due to the lack of variance in outcomes. The hypothesized paths, with the exception of opioid use, were significant.

Clinical Implications:

Based on significant associations, the findings suggest the need to evaluate and treat pain, particularly among black male residents, to optimize function, mood, behavior, and quality of life.

Keywords: nursing homes, pain, quality of life, dementia, older adults

Pain is a commonly reported symptom among older adults in nursing home settings. The incidence of pain varies depending on the population evaluated and the type of measure used (observation versus subjective data). For those in nursing home settings, the verbal reports of pain tend to be high ranging from approximately 30% to 80% (Agit et al., 2018; Brennan, Greenbaum, Lemke, & Schutte, 2019; Kalinowski, Drager, Kuhnert, Kreutz, & Budnick, 2019; Ulbricht, Hunnicutt, Gambassi, Humer, & Lapane, 2091; Wróblewska, Talarska, Wróblewska, Susło, & Drobnik, 2019). Generally the incidence of pain is higher among those who are able to verbally report pain (Brennan et al., 2019; Kalinowski et al., 2019; Ulbricht et al., 2091; Wróblewska et al., 2019). For those with moderate to severe dementia verbal reporting of pain may not be reliable (Agit et al., 2018). Consistently there has been no significant relationship between verbal versus observational evaluations of pain among these individuals (Agit et al., 2018).

The sensation of pain includes four dimensions (Auvray, Myin, & Spence, 2010): (1) the sensory-discriminative dimension which addresses the location, quality and intensity of pain; (2) the affective-motivational dimension which refers to feelings of unpleasantness, distress and threat; (3) the cognitive-evaluative dimension which includes the individuals’ attitudes and beliefs about pain; and (4) the social dimension which addresses how individuals interact with peers, family and individuals in society. The sensory-discriminative dimension of pain, which depends on the lateral nociceptive pathway, is preserved in individuals with dementia even as they progress to severe cognitive impairment. In a mouse model it was noted that the 3xTg-AD mice (pre-clinical model of early and advanced stages of dementia) had an intact sensorial-discriminative threshold to a noxious cold stimulus based on latency of tail-flicking when compared to age matched non-transgenic mice or those without evidence of dementia (Baeta-Corral, Defrin, Chagi, & Gimenez-Liort, 2015). The cognitive-evaluative and the affective-motivation pain dimensions are impaired among those with severe dementia (Cole et al., 2011). Based on objective measures of pressure intensity and Quantitative Sensory Testing (Beach, Huck, Miranda, & Bozoki, 2015) there is evidence that those with cognitive impairment continue to experience pain even when they are unable to express pain.

Repeatedly it has been noted that those with more severe cognitive impairment are less likely to verbally report having pain (Agit et al., 2018; Brennan et al., 2019; Ulbricht et al., 2019) and more likely to demonstrate pain based on observational measures of pain (Agit et al., 2018). In addition, those who are older, male, or belong to a racial or ethnic minority group were less likely to verbally report having pain than those who were younger, female, and white (Brennan et al., 2019; Dahlhamer et al., 2018; Ulbricht et al., 2019). Comorbidities, particularly musculoskeletal disease, cancer or cardiovascular disease have been associated with pain among nursing home residents (van Kooten et al., 2017).

Pain has also been noted to directly influence function (Brennan et al., 2019; Coutinho, Santos, Queiros, & Silva, 2019; Jeon, Tudball, & Nelson, 2019; Kalinowski et al., 2019) and quality of life (Coutinho et al., 2019; Wróblewska et al., 2019). Because individuals with severe dementia can’t always communicate pain verbally, pain often presents as aggression, agitation, repetitive vocalizations, withdrawal, apathy, confusion, impaired or worsening of function (Kaiho, Sugawara, Sugiyama, & et al, 2017). This has negative implications as symptoms of distress may be inappropriately treated with antipsychotics versus pain medications and these individuals are generally not given behavioral interventions to manage pain (Seidel et al., 2013). The staff may assume that decreased physical activity and functional decline in individuals with dementia is due to cognitive decline rather than pain. There is, however, a reciprocal relationship between pain and function as decreased participation in functional activities can results in disuse and exacerbate pain (Ohmichi, Tashima, Osuka, et al., 2020). This occurs in nursing homes when staff underestimate the functional abilities of residents and bathe and dress residents rather than having them participate in their personal care (Galik, Resnick, Hammersla, & Brightwater, 2014; Galik, Resnick, & Pretzer-Aboff, 2009). This exacerbates pain and causes a decline in function. Consequently, the resident may not receive the appropriate range of motion exercises or physical activity needed to optimally manage his or her pain and maintain function. In addition to its association with reduced function, poorly managed pain has also been associated with delirium (Kolanowski et al., 2015).

Prior studies have demonstrated that pain and behavioral symptoms are associated with quality of life in nursing home residents (Beerens, Zwakhalen, Verbeek, & et al., 2013; Corbett, Husebo, Malcangio, & et al., 2012; Corbett et al., 2016; Kooten et al., 2017; Torvik, Kaasa, Kirkevold, & et al., 2010; van Kooten et al., 2017). Pain is expressed via behavior and together both the pain and the behavioral symptoms can negatively influence quality of life. Understanding the direct and indirect influence of pain on behavioral and psychological symptoms associated with dementia (BPSD) and quality of life will help guide the development of interventions to ultimately decrease pain and improve quality of life of residents. BPSD include aggression, agitation, depression, anxiety, apathy and hallucinations and are exhibited by up to 90% of nursing facility residents with dementia (Kales, Gitlin, Lyketsos, & Detroit Expert Panel on Assessment and Management of Neuropsychiatric Symptoms of Dementia, 2014). The purpose of this study was to test an hypothesized model which included the following hypotheses (Figure 1): (1) Age, gender, comorbidities, cognition, and administration of opioids will be directly associated with pain; (2) Cognition and pain will be directly associated with function; (3) Pain and race will be directly associated with agitation and depression; (4) pain will be directly associated with resistiveness to care; (5) pain, resistiveness to care, agitation, function, depression and opioid use will be directly associated with quality of life; (6) age, gender, comorbidities, cognition, race and opioid use will be indirectly associated with quality of life through pain; and (7) pain will be indirectly associated with quality of life though depression, resistiveness to care, agitation and function.

Figure 1 .

Figure 1 .

Open in a new tab

Full Hypothesized Model

METHODS

This study used baseline data from the Evidence Integration Triangle for Behavioral and Psychological Symptoms of Dementia (EIT-4-BPSD) implementation study. The study was approved by a University based Institutional Review Board. The parent study involved testing the use of the Evidence Integration Triangle (EIT) to help staff in nursing homes provide person centered behavioral management of BPSD while optimizing function and physical activity of the residents. A description of the parent study is provided elsewhere (Resnick, Kolanowski, Van Haitsma, Galik, Boltz, Ellis, Behrens, et al., 2018)

SAMPLE

The full baseline sample included residents from 55 nursing homes in the United States specifically from Maryland and Pennsylvania. The mean number of beds was 155 (SD = 81), the majority were for profit (62%) and the mean star rating of these facilities was 3.46 (SD = 1.26). To participate in the study the nursing homes had to: (1) agree to actively partner with the research team on an initiative to change practice; (2) have at least 100 beds or at least 50 beds if the facility had a dedicated dementia care unit; (3) identify a staff member to be an Internal Champion and work with the research team in the implementation process; and (4) be able to access email and websites via a phone, tablet, or computer.

Residents were eligible to participate if they: (1) lived in a participating nursing home at the time of recruitment; (2) were 55 years of age or older; (3) had cognitive impairment based on a score of 0-12 on the Brief Interview of Mental Status (BIMS) (Mansbach, Macea, & Clarka, 2014); (4) were noted to have demonstrated a behavioral symptom over the past month; (5) were not enrolled in Hospice; or (6) were not admitted for short-stay rehabilitation care. Eligible residents were given the Evaluation to Sign Consent (ESC) (Resnick et al., 2007) to determine if they were able to self-consent. If the resident could not correctly respond to all items on the ESC, verbal or written assent was obtained from the resident and the legally authorized representative (LAR) was approached to complete the consent process. A total of 1100 residents were evaluated for eligibility and approached and 547 were excluded. Of the excluded residents, 37 did not meet inclusion criteria and were not approached, 246 refused to participate, and the remaining 264 were situations in which assent from the resident could not be obtained, the legally authorized representative could not be reached or the individual did not meet the cognitive screening criteria or were transitioned to Hospice.

PROCEDURE

Data collection was done by research evaluators who were blinded to the details of the intervention or the randomization of sites. The assessments included chart reviews, resident observations, and input from nursing assistants. The measures included in this study have been used previously with similar older adults in long term care settings and have prior evidence of reliability and validity. Inter-rater reliability for use of the measures during this study was also done, evaluated with Intra-class Correlations (ICCs).

MEASURES

Descriptive information about residents included age, race, gender, cognitive status, and comorbidities. Comorbidities were calculated using the Cumulative Illness Rating Scale (CIRS) (Linn, Linn, & Gurel, 1968). A total score was calculated by summing the 14 categories included in the measure: heart, vascular, hematopoietic, respiratory, ears/nose/throat, upper gastrointestinal, lower gastrointestinal, liver, renal, genitourinary, musculoskeletal, neurologic, endocrine, and psychiatric disease. Cognitive status was evaluated using the BIMS (Mansbach et al., 2014). BIMS scores range from 0 to 15 with scores 13 to 15 indicative of no impairment, scores 8 to 12 indicative of moderate cognitive impairment, and scores 0 to 7 indicative of severe cognitive impairment.

Pain was evaluated using the Pain in Advanced Dementia Scale (PAINAD) (Warden, Hurley, & Volicer, 2003). The PAINAD scale is an observational measure that includes five behaviors commonly observed in individuals with dementia who have pain. The behaviors include breathing independent of vocalization, negative vocalization, facial expression, body language, and consolability. Evaluators completed the observations during periods of activity such as transferring or ambulating. The scores on the PAINAD can range from 0 to 2 for each pain behavior. A total score of 1 - 3 is indicative of mild pain, 4 - 6 is moderate pain and 7-10 is severe pain. Prior use of the PAINAD provided support for inter-rater reliability (DeWaters et al., 2008) and validity based on a correlation between the PAINAD and the numeric pain scale and the Discomfort Scale for Dementia of the Alzheimer’s Type (DeWaters et al., 2008). Inter-rater reliability in this study was .92.

The functional status of residents was based on the Barthel Index (F. Mahoney & Barthel, 1965) which is a 10-item measure focused on resident performance of basic activities of daily living (e.g., bathing, dressing). Items are weighted to account for the amount of assistance required. A score of 100 indicates complete independence. Evaluators completed the measure based on input from the nursing assistant providing care to the resident on the day of testing. Prior testing of the Barthel Index provided evidence of inter-rater reliability and construct validity with significant relationships with other measures of function(F. Mahoney & Barthel, 1965). Inter-rater reliability in this study was .90.

MEASURES TO EVALUATE EVIDENCE OF BPSD

The Cornell Scale for Depression in Dementia (CSDD) (Alexopoulos, Abrams, Young, & Shamoian, 1988) was used to evaluate depressive symptoms (Alexopoulos et al., 1988). The CSDD includes 19 common symptoms associated with depression among individuals with dementia. Prior testing supported evidence of internal consistency and construct validity (Alexopoulos et al., 1988; Barca, 2010). Inter-rater reliability in this study was .99. The 14-item Cohen-Mansfield Agitation Inventory (CMAI) (Cohen-Mansfield, 1986) was used to measure agitation. A 5-point Likert response scale is used to rate the frequency of behavioral symptoms. Prior testing of the CMAI provided support for internal consistency and inter-rater reliability and construct and convergent validity (Cohen-Mansfield, 1986). Inter-rater reliability in this study was .99.

The Resistiveness to Care Scale (E. Mahoney et al., 1999) (RTC) was used to assess the residents’ behavior during care interactions. The RTC includes 13 observed behaviors indicative of resistiveness to care such as hitting, kicking, biting, or refusing care. Evaluators noted if the behavior was present during an observed care interaction. Prior testing of the RTC scale provided support for internal consistency and criterion related validity (E. Mahoney et al., 1999). Testing of the measure with individuals with dementia provided additional support for the psychometric properties of the RTC scale (Galik, Resnick, Vigne, Holmes, & Nalls, 2017). Inter-rater reliability in this study was .95.

The Quality of Life in Late-stage Dementia Scale (QUALID) (Weiner et al., 2000) was used to evaluate quality of life among participants. The QUALID includes 11 observable behaviors that provide evidence of quality of life among individuals with dementia. Items include such things as whether or not the individual smiles, appears sad, cries, has facial expressions of discomfort, appears emotionally calm and comfortable or is irritable or aggressive. Responses are based on a Likert scale and lower scores are indicative of better quality of life. Evaluators obtained input via interviews with nursing assistants who were familiar with the residents. Psychometric testing provided support for the internal consistency, test-retest reliability and construct validity of the QUALID (Resnick, Kolanowski, Van Haitsma, Galik, Boltz, Ellis, Vigne, et al., 2018; Weiner et al., 2000). Inter-rater reliability in this study was .98.

DATA ANALYSIS

Descriptive analyses were done using SPPS version 23. Model testing was done using the AMOS statistical program (Arbuckle, 1997). Multi-collinearity was evaluated based on the variance inflation factor (VIF). The VIFs of all of the model variables were less than three indicating no significant multi-collinearity (Grewal, Cote, & Baumgartner, 2004). For all models tested, the sample covariance matrix was used as input and a maximum likelihood solution sought. The chi-square statistic divided by degrees of freedom (χ2 / df) and the Stiegers Root Mean Square Error of Approximation (RMSEA) were used to estimate model fit. The χ2 / df was used in this analysis as it is the original fit index for testing structural equation models, is derived directly from the fit function and is the basis for most other fit indices (Hu, & Bentler, 1995). Chi-square alone is affected by sample size so a relative χ2 was used, which was calculated by dividing the χ2/df. The larger the probability associated with the chi-square divided by degrees of freedom the better the fit of the model to the data. A ratio of < 3 was considered to be a good fit (Bollen, 1989). In addition to χ2 / df, a noncentrality-based measure, the RMSEA was used. The RMSEA is a population based index and consequently is insensitive to sample size. A RMSEA of < 0.10 is considered good, and <0.05 is very good (Bollen, 1989). Path significance (i.e., significance of the Lambda values) was based on the Critical Ratio (CR), which is the parameter estimate divided by an estimate of the standard error. A CR > 2 in absolute value was considered significant (Arbuckle, 1997). Significance for path estimates was set at p ≤ 0.05.

RESULTS

A total of 553 residents were included in the study. As shown in Table 1, the participants had a mean age of 83.88 (SD=10.44) and a mean score on the BIMS of 4.30 (SD=3.50) indicating that overall the participants had severe cognitive impairment. The participants had a mean of 7.10 (SD=2.16, range 2-12) comorbidities. Although the majority of the sample was white (N=419, 76%) and female (N=398, 72%) there were a relatively large percentage of males (N=155, 28%) and black participants (N=134, 24%). Overall the participants had low levels of agitation with a mean score on the CMAI of 21.02 (SD=8.36, range 14 to 54), little resistiveness to care with a mean score of .60 (SD=2.06, range 0-19), few depressive symptoms with a mean score of 4.26 (SD=4.54, range 0-28), and generally good quality of life with a mean score of 17.16 (SD=6.72, range 10-48 with higher scores indicative of worse quality of life). Lastly, the participants displayed low levels of pain behaviors with a mean score of .68 (SD=1.50, range 0-8). As shown in Table 2, 411 (7%) participants had no pain (a score of 0 on the PAINAD), 125 (23%) had mild to severe pain (a score of 1 to 8 on the PAINAD) and there was missing data on the remaining 17 (3%) participants due to death or transfer out of the facility prior to baseline assessment. Only 28 participants (5%) were receiving opioids for the treatment of pain.

Table 1.

Descriptive Statistics of Demographics and Outcome Variables

Minimum Maximum Mean Std.
Deviation
Age 56.00 105.00 83.88 10.44
Gender N %
 Male 155 28%
 Female 398 72%
Race
 Black 134 24%
 White 419 76%
Agitation 14.00 54.00 21.02 8.35
Resistiveness to Care .00 19.00 .60 2.06
Function 3.00 100.00 38.94 30.85
Pain .00 8.00 .68 1.50
Quality of Life 10.00 48.00 17.15 6.72
Depression .00 28.00 4.26 4.54
Comorbidities 2.00 12.00 7.10 2.15
Cognition 0 12 4.30 3.47
Open in a new tab

Table 2.

Frequency of Pain Scores (Potential Range= 0-10)

Pain Score N Percent
.00 411 74.1
1.00 39 7.1
2.00 23 4.2
3.00 21 3.8
4.00 14 2.5
5.00 15 2.7
6.00 9 1.6
7.00 3 .5
8.00 1 .2
Missing 17 3.3
Open in a new tab

Full model testing (Figure 1) indicated that the paths between opioid use and pain and opioid use and quality of life were nonsignificant. These paths were removed from the model and revised model testing was done. Table 3 provides the path coefficients for the revised model, all of which were significant. Age, gender, comorbidities, cognition, and race were all significantly associated with pain and explained 59% of the variance. Those who were older, male, had fewer comorbidities, better cognition and were not white were more likely to have pain. Pain and cognition were the only variables to be directly associated with function and combined with the indirect association of age, gender, comorbidities, cognition, and race, 32% of the variance in function was explained. Residents with less pain and better cognition had higher function. Pain and race were the only variables to be directly associated with agitation and depressive symptoms. Those residents who were white and had more pain had more agitation and more depressive symptoms. Combined with the indirect association with age, gender, comorbidities, cognition, and race on pain, 88% of the variance in agitation and 75% of the variance in depressive symptoms was explained. Pain was the only variable to be directly associated with resistiveness to care such that those with more pain had more resistiveness to care. Combined with the indirect association of age, gender, comorbidities, cognition, and race on resistiveness to care through pain, a total of 98% of the variance in resistiveness to care was explained.

Table 3.

Standardized Regression Weights for Hypothesized Paths for Final Model

Path Estimate p
Pain □ Age .710 .001
Pain □ Gender −.149 .044
Pain □ Comorbidities −.056 .001
Pain □ Cognition .143 .001
Pain □ Race .203 .001
Function □ Pain −.571 .001
Agitation □ Pain .947 .001
Depression □ Pain .876 .001
Depression □ Race −.058 .009
Agitation □ Race −.038 .012
Function □ Cognition .141 .001
Resistiveness to care □ Pain .988 .001
Quality of life □ Pain .492 .001
Quality of life □ Function −.041 .004
Quality of life □ Agitation .157 .001
Quality of life □ Depression .328 .001
Quality of life □ Resistiveness to care .185 .016
Open in a new tab

Race, cognition, comorbidities, gender, and age were indirectly associated with quality of life through pain, and pain was directly and indirectly associated with quality of life through function, agitation, resistiveness to care and depressive symptoms such that those with more pain, more agitation, more depressive symptoms, more resistiveness to care and lower function had worse quality of life. Together these variables explained 92% of the variance in quality of life. Although the paths in this model were all significant there was a poor fit of the model to the data with a χ2 / df of 57.91, and a RMSEA of .32. There was no improvement in fit using the revised model.

DISCUSSION

The hypothesized model proposed in this study was supported in that all the paths, with the exception of opioid use, were significant. The model did not demonstrate a good fit to the data. A good model fit means that the observed data fits a previously established theoretically or empirically based model as proposed in Figure 1. The lack of fit based on both the χ2 / df and RMSEA in this study may be because of the large sample size, the limited evidence of pain and BPSD and thus lack of variance in the outcomes, and the complexity of the model (Nevitt & Hancock, 2000). It is possible that the model would not be replicable and that other factors may influence pain, function, behavioral symptoms and quality of life. Examples of other factors to consider include the perceptions, beliefs and possible biases of health care providers regarding pain levels among older adults across race and gender differences (Wandner, Scipio, Hirsh, Torres, & Robinson, 2012) and life stressors experienced by the resident (Brennan, 2020). Retesting of the model using a more heterogeneous sample and with these additional factors considered is needed to confirm our findings and establish a model that better explains pain and impact of pain on behavioral factors and quality of life of residents.

The lack of significant association between the use of opioids and pain or quality of life among participants may be due to the low use of these drugs in this population (only 5% were receiving opioids). The low rate of opioid use in this sample of nursing home residents with severe dementia may be due to the clinical challenges of evaluating pain in a population that cannot clearly describe their pain and/or may lack recognition of pain, provider concerns about use of opioids in this population, and our exclusion of those at end of life (i.e., those on Hospice). The measurement of pain using the PAINAD also may have influenced the outcomes. Many of the items on the PAINAD are based on observable physical findings that may not actually be due to pain such as negative vocalization, facial expressions, body language and consolability. Better objective measures of pain are needed to truly establish evidence of pain in this population. For example there may be biomarkers that can be used to identify pain in older adults with cognitive impairment (Galik, Resnick, Mocci, Renn, Song, Dorsey, under review). Further, there is a major concern about opioid use in the United States as the use of opioids is 10 times more than the world average (Berterame, Erthal, Thomas, et al., 2016). The low use of opioids in this study may reflect current concerns about an opioid epidemic in this country and strong encouragement to avoid opioid initiation across all settings of care and ages of patients. There continues to be a major focus on decreasing use of opioids including use for those in nursing home settings.

In this sample those who were older, male, had fewer comorbidities and were black were more likely to demonstrate symptoms indicative of pain. Although findings are inconsistent (Brennan et al., 2019; Ulbricht et al., 2019) there have been prior studies with older adults indicating that those who are black tend to have more pain than their white counterparts (Brennan, 2020; Green & Hart-Johnson, 2010; Wandner, Hirsh, Torres, et al., 2013). Likewise prior studies have reported that males tend to have higher pain levels across all levels of cognitive status (Cowan, Beach, Atalla, et. al., 2017; Romano, Anderson, Failla, et. al., 2019; Wadner, Hirsh, Torres, et al., 2013). Pain assessments in this study were provided by caregivers which may have influenced these findings as health care providers tend to rate pain as higher in blacks versus whites (Wandner, Scipio, Hirsh, Torres, & Robinson, 2012). The differences in gender associations and pain likely include multiple biological and psychosocial processes (Bartley & Fillingim, 2013). It is possible that men with cognitive impairment, for example, are disinhibited and may be more likely to express pain rather than try and exhibit gender expected responses to pain (Bartley & Fillingim, 2013). The association between comorbidities and pain is difficult to interpret as the impact of a specific diagnosis and associated symptoms was not considered.

As noted in a prior study in nursing home residents considering the relationship between pain and behavioral symptoms (Nortonab, Allenab, Snowabc, Hardinbd, & Burgio, 2010) pain was directly associated with function and BPSD in the current study. Pain alone or in combination with the demographic and descriptive variables discussed above explained 32% of the variance in function, 75% of the variance in depression, 88% of the variance in agitation, and 95% of the variance in resistiveness to care. It should be noted that this was a correlational study so causation cannot be assumed. Despite the lack of model fit the relationships between variables in this model suggest significant associations and support good clinical care which involves the evaluation and management of pain among residents with severe cognitive impairment. The relationship between pain and function may be reciprocal, which was not tested in this model. Pain may cause a decrease in function which can exacerbate pain due to stiffness, pressure areas or contractures from immobility. It is reasonable to assume that pain, particularly when the individual is unable to understand or describe his or her pain, logically may result in aggressive behavior and resistance to care.

Likewise, as hypothesized, pain directly and indirectly through function and BPSD was associated with quality of life in this sample and explained 92% of the variance in quality of life. As expected, those with more pain, lower function, more agitation, and more depressive symptoms had lower quality of life. Prior research has shown associations between pain and quality of life (Brennan et al., 2019; Coutinho et al., 2019; Jeon et al., 2019; Kalinowski et al., 2019; Wróblewska et al., 2019). One prior study (van Kooten et al., 2017), however, using multiple regression analyses to test the association between pain and neuropsychiatric symptoms and quality of life among nursing home residents with moderate to severe impairment reported that pain was not associated with quality of life. This difference may be sample specific or due to differences in the way in which quality of life was measured. Further research, however, is needed to replicate our model and continue to explore the best ways in which to measure quality of life and the factors that are most likely to influence quality of life among those with severe cognitive impairment.

STUDY STRENGTHS AND LIMITATIONS

This study was limited in that it was a secondary data analysis and detailed information about diagnoses and associated symptoms were not included. Because the participants had severe cognitive impairment consent was done by proxies and the high rate of refusals may have biased the sample with those most likely to have behavioral problems not assenting and proxies refusing to allow participation. In addition all assessments were based on input from staff and may have been biased. There is a need to develop methods for the evaluation of pain in late-stage dementia that are reliable and valid with a focus on more objective measures such as biomarkers of physiological findings. The poor fit of the model limits the generalizability of these findings. The poor fit may have been due to limited evidence of pain and BPSD and sample size (Bollen, 1989). The study included residents from two states and 55 nursing homes and thus may not be reflective of all nursing home residents. Despite these limitations, this study included a large sample of nursing home residents with a high percentage of black residents as well as males and provided additional support to reinforce the value of evaluation and management of pain and consider the association of pain with BPSD and quality of life among nursing home residents with severe dementia.

PRACTICE POINTS.

  • Assessment and management of pain, particularly in black males, is important as pain is associated with behavioral symptoms, functional decline, and quality of life.

  • Pain is challenging to evaluate among individuals with moderate to severe dementia although observational measures such as the PAINAD can be easily used in practice.

  • Ongoing consideration of the use and value of opioids for pain management among residents with moderate to severe dementia is needed.

Acknowledgements:

This study was funded by the National Institute of Nursing Research R01 NR015982.

Contributor Information

Barbara Resnick, University of Maryland School of Nursing, 655 West Lombard Street, Baltimore MD 21218.

Elizabeth Galik, University of Maryland School of Nursing, 655 West Lombard Street, Baltimore MD 21218.

Ann Kolanowski, Pennsylvania State University, Nursing Sciences Building, University Park, PA 16802.

Kimberly VanHaitsma, Pennsylvania State University, Nursing Sciences Building, University Park, PA 16802.

Marie Boltz, Pennsylvania State University, Nursing Sciences Building, University Park, PA 16802.

Shijun Zhu, University of Maryland School of Nursing, 655 West Lombard Street, Baltimore MD 21218.

Jeanette Ellis, University of Maryland School of Nursing, 655 West Lombard Street, Baltimore MD 21218.

Liza Behrens, Pennsylvania State University, Nursing Sciences Building, University Park, PA 16802.

Karen Eshraghi, Pennsylvania State University, Nursing Sciences Building, University Park, PA 16802.

Cynthia Renn, University of Maryland School of Nursing, 655 West Lombard Street, Baltimore MD 21218.

Susan G. Dorsey, University of Maryland School of Nursing, 655 West Lombard Street, Baltimore MD 21218.

REFERENCES

  1. Agit A, Balci C, Yavuz B, Cankurtaran E, Kuyumcu M, Halil M, et al. (2018). An iceberg phenomenon in dementia: Pain. Journal of Geriatric Psychiatry and Neurology, 31(4), 186–193. [DOI] [PubMed] [Google Scholar]
  2. Alexopoulos G, Abrams R, Young R, & Shamoian C (1988). Cornell scale for depression in dementia. Biology and Psychiatry, 23, 271–284. [DOI] [PubMed] [Google Scholar]
  3. Arbuckle J (1997). Amos users' guide version 3.6. Chicago: Small Waters Corporation. [Google Scholar]
  4. Auvray M, Myin E, Spence C (2010). The sensory-discriminative and affective-motivational aspects of pain. Neuroscience and Biobehavioral Review, 34(2), 214–23. [DOI] [PubMed] [Google Scholar]
  5. Baeta-Corral R, Defrin R, Chagi G, & Gimenez-Liort L (2015). Tail-flick test response in 3 x Tg-AD mice at early and advanced stages of disease. Neuroscience Letters, 600, 158–163. [DOI] [PubMed] [Google Scholar]
  6. Barca M, Engedal K, Selbaek G (2010). A reliability and validity study of the Cornell Scale among elderly inpatients, using various clinical criteria. Dementia Geriatrics and Cognitive Disorders, 29(5), 438–447. [DOI] [PubMed] [Google Scholar]
  7. Bartley E, & Fillingim R (2013). Sex differences in pain: A brief review of clinical and experimental findings. British Journal of Anaesthesia, 111, 52–58. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Beach P, Huck J, Miranda M, & Bozoki A (2015). Autonomic, behavioral, and subjective pain responses in Alzheimer’s Disease. Pain Medicine (United States), 16(10), 1930–1942. [DOI] [PubMed] [Google Scholar]
  9. Beerens H, Zwakhalen S, Verbeek H, & et al. (2013). Factors associated with quality of life of people with dementia in long term care facilities: A systematic review. International Journal of Nursing Studies, 50, 1259–1270. [DOI] [PubMed] [Google Scholar]
  10. Berterame S, Erthal J, Thomas J, et al. (2016). Use of and barriers to access to opioid analgesics: A worldwide, regional,and national study. Lancet, 387, 164–168. [DOI] [PubMed] [Google Scholar]
  11. Bollen KA (1989). Structural equations with latent variables: Wiley-Interscience. [Google Scholar]
  12. Brennan P, Greenbaum M, Lemke S, & Schutte K (2019). Mental health disorder, pain and pain treatment among long-term care residents: Evidence from the Minimum Data Set 3.0. Aging and Mental Health, 23(9), 1146–1155. [DOI] [PubMed] [Google Scholar]
  13. Brennan PL (2020). Life stressors: Elevations and disparities among older adults with pain. Pain Medicine, 21(10), 2123–2136. . [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Cohen-Mansfield J (1986). Agitated behaviors in the elderly. II. Preliminary results in the cognitively deteriorated. Journal of the American Geriatrics Society, 34(10), 722–727. [DOI] [PubMed] [Google Scholar]
  15. Cole L, Gavrilescu M, Johnston L, Gibson S, Farrell M, & Egam GF (2011). The impact of Alzheimer's disease on the functional connectivity between brain regions underlying pain perception. European Journal of Pain, 568, e1–e6. [DOI] [PubMed] [Google Scholar]
  16. Corbett A, Husebo B, Malcangio M, & et al. (2012). Assessment and treatment of pain in people with dementia. National Review of Neurology, 8, 264–274. [DOI] [PubMed] [Google Scholar]
  17. Corbett A, Nunez K, Smeaton E, Testad I, Thomas A, Closs S, et al. (2016). The landscape of pain management in people with dementia living in care homes: a mixed methods study. International Journal of Geriatric Psychiatry, 31(12), 1354–1370. [DOI] [PubMed] [Google Scholar]
  18. Coutinho C, Santos S, Queiros A, & Silva A (2019). Capacity measures explain only a small part of self-reported disability in institutionalized ambulatory older adults: An observational study. Physical & Occupational Therapy in Geriatrics, 37(2), 108–122. [Google Scholar]
  19. Cowana RL, Beach PA, Atallac SW, et al. (2017). Sex differences in the psychophysical response to contact heat in moderate cognitive impairment Alzheimer’s disease: A cross-sectional brief report. Journal of Alzheimer’s Disease, 60, 1633–1640. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Dahlhamer J, Lucas J, Zelaya C, Nahin R, Mackey S, DeBar L, et al. (2018). Prevalence of chronic pain and high-impact chronic pain among adults — United States, 2016. Morbidity and Mortality Weekly Report, 67(36), 1001–1006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. DeWaters T, Faut-Callahan M, McCann J, Paice J, Fogg L, Hollinger-Smith L, Stanatis H (2008). Comparison of self-reported pain and the PAINAD scale in hospitalized cpognitively impaired and intact older adults after hip fracture surgery. Orthopedic Nursing 27(1), 21–28. [DOI] [PubMed] [Google Scholar]
  22. Galik E, Resnick B, Hammersla M, & Brightwater J (2014). Optimizing function and physical activity among nursing home residents with dementia: Testing the impact of function focused care. The Gerontologist, 54(6), 930–943. [DOI] [PubMed] [Google Scholar]
  23. Galik E, Resnick B, & Pretzer-Aboff I (2009). Knowing what makes them tick: Motivating cognitively impaired older adults to participate in restorative care. International Journal of Nursing Practice,15(1), 48–55. [DOI] [PubMed] [Google Scholar]
  24. Galik E, Resnick B, Vigne E, Holmes S, & Nalls V (2017). Reliability and validity of the Resistiveness to Care Scale among cognitively impaired older adults. Journal of the American Medical Directors Association,18(1), 59–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Green CR, Hart-Johnson T, Green CR, Hart-Johnson T (2010). The impact of chronic pain on the health of black and white men. Journal of the National Medical Association, 102(4), 321–331. [DOI] [PubMed] [Google Scholar]
  26. Grewal R, Cote J, & Baumgartner H (2004). Multicollinearity and measurement error in structural equation models: Implications for theory testing. Marketing Science, 23(4), 519–529. [Google Scholar]
  27. Hu LT, & Bentler PM (1995). Evaluating model fit. In Hoyle RH (Ed.), Structural equation modeling: Concepts, issues, and applications (pp. 76–99). Thousand Oaks, CA: Sage. [Google Scholar]
  28. Jeon Y, Tudball J, & Nelson K (2019). How do residents of aged care homes perceive physical activity and fucntional independence? A qualitative study. Health & Social Care in the Community, 27(5), 1321–1332. [DOI] [PubMed] [Google Scholar]
  29. Kaiho Y, Sugawara Y, Sugiyama K, & et al. (2017). Impact of pain on incident risk of disability in elderly Japanese: Cause-specific analysis. Anesthesiology, 126(4), 688–696. [DOI] [PubMed] [Google Scholar]
  30. Kales H, Gitlin L, Lyketsos C, & Detroit Expert Panel on Assessment and Management of Neuropsychiatric Symptoms of Dementia. (2014). Management of neuropsychiatric symptoms of dementia in clinical settings: Recommendations from a multidisciplinary expert panel. Journal of the American Geriatrics Society, 62(4):762–769. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Kalinowski S, Drager D, Kuhnert R, Kreutz R, & Budnick A (2019). Pain, fear of falling and functional performance among nursing home residents: A longitudinal study. Western Journal of Nursing Research, 41(2), 191–216. [DOI] [PubMed] [Google Scholar]
  32. Kolanowski A, Mogle J, Fick D, Hill N, Mulhall P, Nadler J, et al. (2015). Pain, delirium, and physical function in skilled nursing home patients with dementia. Journal of the American Medical Directors Association 16(1), 37–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Kooten J, Van der Wouden J, Sikkes S, Smalbrugge M, Hertogh C, & Stek M (2017). Pain, neuropsychiatric symptoms and quality of life of nursing home residents with advanced demtia in the Netherlands. Alzehiemr’s Disease and Associated Diisorders, 31, 315–321. [DOI] [PubMed] [Google Scholar]
  34. Linn B, Linn M, & Gurel L (1968). Cumulative illness rating scale. Journal of the American Geriatrics Society, 16(5), 622–626. [DOI] [PubMed] [Google Scholar]
  35. Mahoney E, Hurley A, Volicer L, Bell M, Gianotis P, Hartshorn M, et al. (1999). Development and testing of the Resistiveness to Care Scale. Research in Nursing and Health, 22(1), 27–38. [DOI] [PubMed] [Google Scholar]
  36. Mahoney F, & Barthel D (1965). Functional evaluation: The Barthel Index. Maryland State Medical Journal, 14(2), 61–66. [PubMed] [Google Scholar]
  37. Mansbach W, Macea R, & Clarka M (2014). Differentiating levels of cognitive functioning: A comparison of the Brief Interview for Mental Status (BIMS) and the Brief Cognitive Assessment Tool (BCAT) in a nursing home sample. Aging & Mental Health, 18(7), 921–928. [DOI] [PubMed] [Google Scholar]
  38. Monroe T, Carter MA (2010). Retrospective pilot study of African-American and Caucasian nursing home residents with dementia who died from cancer. Journal of Pain Symptom Management, 40(4), e1–e3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Nevitt J, Hancock G, (2000). Improving the root mean square error of approximation for nonnormal conditions in structural equation modeling. The Journal of Experimental Education, 68(3), 251–268. [Google Scholar]
  40. Nortonab MJ, Allenab RS, Snowabc AL, Hardinbd JM, Burgio LD (2010). Predictors of need-driven behaviors in nursing home residents with dementia and associated certified nursing assistant burden. Aging & Mental Health, 14(3), 303–309. [DOI] [PubMed] [Google Scholar]
  41. Resnick B, Gruber-Baldini A, Aboff-Petzer I, Galik E, Russ K, & Zimmerman S (2007). Reliability and validity of the evaluation to sign consent meausre. The Gerontologist, 47(1), 69–77. [DOI] [PubMed] [Google Scholar]
  42. Resnick B, Kolanowski A, Van Haitsma K, Galik E, Boltz M, Ellis J, et al. (2018). Testing the evidence integration triangle for implementation of interventions to manage behavioral and psychological symptoms associated with dementia: Protocol for a pragmatic trial. Research in Nursing and Health, 41(3), 228–242. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Resnick B, Kolanowski A, Van Haitsma K, Galik E, Boltz M, Ellis J, et al. (2018). Reliability and validity testing of the Quality of Life in Late-stage Dementia Scale. American Journal of Alzheimer's Disease and Other Dementias, 33(5), 277–283. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Romanoa RR, Andersona AR, Faillaf MD, et al. (2019). Sex differences in associations of cognitive function with perceptions of pain in older adults. Journal of Alzheimer’s Disease 70, 715–722. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Seidel S, Aigner M, Ossege M, Pernicka E, Wilden B, & Sycha T (2013). Antipsychotics for acute and chronic pain in adults. Cochrane Databases Systematic Review, 29(CD004844). [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Torvik K, Kaasa S, Kirkevold O, & et al. (2010). Pain and quality of life among residents of Norwegian nursing homes. Pain Management Nursing, 11, 35–44. [DOI] [PubMed] [Google Scholar]
  47. Ulbricht C, Hunnicutt J, Gambassi G, Humer A, & Lapane K (2019). Nonmalignant pain symptom subgroups in nursing home residents. Journal of Pain and Symptom Management, 57(3), 535–544. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. van Kooten J, Van der Wouden J, Sikkes S, Smallbrugge M, Hertogh C, & Stek M (2017). Pain, neuropsychiatric symptoms and quality of life of nursing home residents with advanced dementia in the Netherlands. Alzheimer's Disease and Associated Disorders, 31, 315–321. [DOI] [PubMed] [Google Scholar]
  49. Wandner LD, Hirsh AT, Torres CA, et al. (2013). Using virtual human technology to capture dentists' decision policies about pain. Journal of Dental Research, 92(4), 301–305. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Wandner LD, Scipio CD, Hirsh AT, Torres CA, Robinson ME (2012). The perception of pain in others: how gender, race,and age influence pain expectations. The Journal of Pain, 13(3), 220–227. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Warden V, Hurley A, & Volicer L (2003). Development and psychometric evaluation of the pain assessment in advanced dementia (PAINAD) Scale. Journal of the American Medical Directors Association, 4(1), 9–15. [DOI] [PubMed] [Google Scholar]
  52. Weiner M, Martin-Cook K, Svetlik D, Saine K, Foster B, & Fontaine C (2000). The quality of life in late- stage dementia (QUALID) scale. Journal of the American Medical Directors Association, 1, 114–116. [PubMed] [Google Scholar]
  53. Wróblewska I, Talarska D, Wróblewska Z, Susło R, & Drobnik J (2019). Pain and symptoms of depression: International comparative study on selected factors affecting the quality of life of elderly people residing in institutions in Europe. BMC Geriatrics, 19(1), 1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES