Journal of Gerontological Nursing

Feature Article 

Insights to Identifying and Managing Pain in Persons With Dementia in Long-Term Care: A Mixed Methods Study Comparing the Abbey Pain Scale and Pain Assessment in Advanced Dementia Scale

Suzanne Parkman, PhD, RN, CNE; Beth Mastel-Smith, PhD, RN, FNGNA; Anthony McGuire, PhD, CCRN-K, ACNP-BC, FAHA; Gloria Duke, PhD, RN


Pain is usually identified by specific behaviors driven by the need for relief; however, persons with dementia present a unique challenge for nurses in assessing and managing pain. The aim of this mixed methods study was to explore the relationship between two observational pain scales, expressed need-driven behaviors, and likelihood of medication administration for persons with dementia. The qualitative strand examined nurses' perceptions regarding facilitators and barriers to pain scale use. Quantitative data analysis indicated the Abbey Pain Scale was significantly correlated with behaviors (r[26] = 0.41, p < 0.05) and approached significance with medication administration (r[26] = 0.35, p = 0.067). Qualitative analysis identified three core themes: (a) Challenges in Assessing Persons With Dementia for Pain; (b) Facilitators and Barriers to Pain Management; and (c) Difficulty Caring for Persons With Dementia. Clinical implications suggest the need for a systematic, consistent method of observing pain-related behaviors in persons with dementia. [Journal of Gerontological Nursing, 47(2), 21–30.]


Pain is usually identified by specific behaviors driven by the need for relief; however, persons with dementia present a unique challenge for nurses in assessing and managing pain. The aim of this mixed methods study was to explore the relationship between two observational pain scales, expressed need-driven behaviors, and likelihood of medication administration for persons with dementia. The qualitative strand examined nurses' perceptions regarding facilitators and barriers to pain scale use. Quantitative data analysis indicated the Abbey Pain Scale was significantly correlated with behaviors (r[26] = 0.41, p < 0.05) and approached significance with medication administration (r[26] = 0.35, p = 0.067). Qualitative analysis identified three core themes: (a) Challenges in Assessing Persons With Dementia for Pain; (b) Facilitators and Barriers to Pain Management; and (c) Difficulty Caring for Persons With Dementia. Clinical implications suggest the need for a systematic, consistent method of observing pain-related behaviors in persons with dementia. [Journal of Gerontological Nursing, 47(2), 21–30.]

Nurses caring for older persons with dementia face a tremendous challenge in meeting the needs of patients who are surrounded by the cocoon of dementia-related behaviors, which makes communication, assessment, and treatment decisions a delicate balance between the known and the unknown. Unable to rely on straight-forward communication to determine actions, nurses continuously seek better ways to assess basic needs, such as pain or hunger, in persons with dementia. Dementia is a syndrome in which there is deterioration in memory, thinking, behavior, and the ability to perform everyday activities (World Health Organization, 2019). The burden of dementia in the geriatric population is compounded by the presence of other painful conditions and comorbidities associated with aging that frequently necessitate care (Brecher & West, 2016). A need-driven behavior occurs in response to some real or perceived unmet need that spurs the person with dementia to do something to try to fulfill the need (Algase et al., 1996). Pain has been correlated to need-driven behaviors. In a randomized controlled trial (Husebo, Ballard, Fritze, et al., 2014), persons with dementia who expressed negative behaviors responded positively to pain treatment. Untreated pain can lead to need-driven behaviors, affects quality of life for persons with dementia, and serves as a potential source of stress for caregivers (Ahn et al., 2015; Ahn & Horgas, 2013; Kutschar et al., 2017).

One barrier to successful pain management is lack of standardized assessment tools (Cohen-Mansfield et al., 2015; Corbett et al., 2014; Lichtner et al., 2016). Other issues to pain management include the challenge of identifying pain in persons with dementia, lack of communication, workload pressures, time management, budgetary issues, and training and education of nurses (Burns & McIlfatrick, 2015). Much effort has been made over the past decade to improve pain management for this population, and these efforts have resulted in the development of more than 28 observational pain scales (Flo et al., 2014; Lichtner et al., 2016; Tsai et al., 2018). Many of these scales are psychometrically valid; however, scale application and score interpretation in daily clinical practice remain challenges (Husebo et al., 2016).

Although the American Society for Pain Management Nurses recommends a hierarchy of pain assessment techniques for persons with dementia, which includes incorporation of an observational tool (Herr et al., 2011), the literature provides no clear-cut guidelines for which pain assessment tool to use in this population. Pain assessment is further complicated by the limitations regarding the reliability and validity of instruments and clinical utility. A systematic review of systematic reviews (Lichtner et al., 2014) concluded that no recommendation for a specific observational pain scale can be made and concluded that the process of interpretation is only as good as the person using it. Synthesis of current research supported the conclusion that overall pain management for this population is inadequate (Tsai et al., 2018).

The incidence and prevalence of poor pain management in persons with dementia is well documented. However, many studies were conducted using secondary data or were retrospective in nature. Several international studies examined the effects of various observational pain scales and pain behaviors in cluster randomized controlled trials (Ahn & Horgas, 2013; Husebo, Ballard, Cohen-Mansfield, et al., 2014; Liu & Lai, 2017; Sandvik et al., 2014). Many pain protocols exist; however, the recommendations are based on expert opinion rather than empirical evaluation (American Geriatrics Society [AGS] Panel on Persistent Pain in Older Persons, 2002; AGS Panel on Pharmacological Management of Persistent Pain in Older Persons, 2009) and suggests the need for evidence to support pain assessment protocols for persons with dementia. Inability to accurately detect pain in this vulnerable population may result in either over- or under-medicating the person (Herr et al., 2011). It has been reported that nursing home residents with a dementia diagnosis were significantly less likely to have an order for an opioid, resulting in under-medicating for pain, which contributes to suffering and exacerbation of cognitive decline (Monroe et al., 2014).

For the past decade, nurses have been challenged to incorporate evidence-based strategies that uniquely reflect the needs of long-term care (LTC). According to Higuchi et al. (2015), there is an urgent need for evidence-based practice improvements in LTC; however, these improvements may be difficult to achieve due to the barriers listed above. Nurses play a central role in successful pain management for persons with dementia. However, nurses may lack the pain knowledge necessary to assess and manage pain in persons with dementia effectively (Sacoco & Ishikawa, 2014). Staff engagement has been acknowledged as a critical factor in sustaining change within health care organizations (McCormack et al., 2009). Accurate pain assessment in persons with dementia in the LTC setting has been associated with the ability of nurses to observe negative emotional expression (Lee et al., 2015). Thus, improving pain management is crucial to the comfort and well-being of persons with dementia living in nursing homes.

The aim of the current study was to explore the relationship between two observational pain scales, expressed need-driven behaviors, and pain medication administration for persons with dementia in LTC using the Need-Driven Behavior Model (NDBM). Nurses' perceptions regarding ease of and barriers to use of the scales were explored to help explain the likelihood of using pain scales as a key pain management tool with persons with dementia. The questions explored were: (1) Do the pain scales have a significant correlation with need-driven behaviors? (2) If behaviors were thought to be pain-related, did use of the scales prompt nurses to administer pain medication? (3) What were nurses' experiences using the pain scales? Broader findings and further details of this study regarding the aim, framework, and methodology are available in Scholar Works repository (Parkman, 2018).

Theoretical Framework

The NDBM (Algase et al., 1996) served as the foundation for the current study. The NDBM proposes that need-driven behaviors arise from the pursuit of a goal or expression of a need and are caused by proximal and background factors (Algase et al., 1996). Relatively stable individual characteristics (background factors) interact with current situational issues (proximal factors) to produce dementia-related behaviors. Background factors have established relationships with pain and problematic behaviors (Kutschar et al., 2017). Pain is a psychological and physiological need state (proximal factor) and, therefore, has a direct relationship with disruptive behaviors. The quantitative data in the current study were used to test the modified NDBM (Algase et al., 1996) that predicts assessment of pain in persons with dementia will be correlated with actual need-driven behaviors and increased pain medication administration.


One group of persons with dementia with severe cognitive impairment in LTC was recruited from a 50-bed memory care unit in northern New England. This LTC group served as its own control over an 8-week period. Using G*Power with a power of 0.80, alpha of 0.05, and an effect size of 0.5, calculated for difference between two dependent means, a total sample size of 27 residents was required (Faul et al., 2007). Following University Institutional Review Board approval, a sample of 31 residents met the criteria and were selected for the study; however, during the data collection period, three residents passed away, leaving the total sample size at 28 participants.

Proxy-consent was obtained for LTC participants. Eligibility criteria included: (a) age >65 years; (b) diagnosis of some type of dementia according to the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders; (c) scoring <48 on the Modified Mini-Mental State Examination (3MS), indicating severe cognitive impairment (Teng & Chui, 1987); and (d) at least one pain-related diagnosis. Exclusion criteria were: (a) medical condition for which individuals were frequently admitted (>2 times per month) to the hospital, such as heart failure; (b) comorbid psychiatric disorders, such as schizophrenia or bipolar disorder; (c) recent distressing social circumstances, such as death of a spouse; and (d) receiving palliative care. Data for participants who left the study were not included in analyses.

The first two research questions were explored using data from the LTC sample. The third question regarding nurses' experiences with pain scales was explored using qualitative data from nurses who used the scales. Nurses helped explain and provide some basis of explanation for the findings. A convenience sample of nurses was recruited by the first author (S.P.) for the qualitative strand. Inclusion criteria were: (a) full-time licensed nurse (RN or licensed practical nurse [LPN]) with >6-months' experience; and (b) cared for persons with dementia using the Abbey Pain Scale (APS) and Pain Assessment in Advanced Dementia Scale (PAINAD) during the 8-week phase. Exclusion criteria were float pool or registry nurses. A total of nine nurses participated in the study and met inclusion criteria. Of those nurses, six agreed to be interviewed, which provided data saturation.


The two pain scales chosen were the APS and PAINAD based on the literature (Lichtner et al., 2014; Moschinski et al, 2017) reporting rapid ease of administration, scale reliability, short time to complete, and applicability to the population. Both scales contain pain indicators for persons with dementia outlined by the AGS (2002, 2009).

The APS was used to measure discomfort in the sample of persons with dementia who had lost cognitive capacity, verbal communication abilities, and were dependent on caregivers (Abbey et al., 2004). Severity of six pain behaviors is rated on a 0 to 3 scale; total score ranges from 0 to 18. Severity of pain is interpreted as follows: 0 to 2 = absent, 3 to 7 = mild, 8 to 13 = moderate, and ≥14 = severe. The rater indicates type of pain: chronic, acute, or acute on chronic. The APS detected change in pain level before and after pain-relieving interventions and had a moderate level of correlation with nurses' proxy pain scores (Abbey et al., 2004). Scale reliability for the current study was α = 0.84.

The PAINAD was also administered to the sample. It is the current standard for this geographical region of the United States. The tool included five items: breathing, negative vocalization, facial expression, body language, and consolability. Each item is graded on a 3-point scale from 0 to 2 for intensity and summed for a total score of 0 to 10 (Warden et al., 2003). Scoring for pain severity is as follows: 0 = absent, 1 to 3 = mild, 4 to 6 = moderate, and 7 to 10 = severe. Scale reliability for the current study was α = 0.83.

The Cohen-Mansfield Agitation Inventory (CMAI) was used to measure need-driven behaviors to determine if there was a relationship to the pain scales. The CMAI is a caregiver rating questionnaire addressing frequency of 29 agitated behaviors (Cohen-Mansfield, 1996; Cohen-Mansfield et al., 1989). The total number of behaviors was recorded by nurses every 8 hours. Scale reliability for the current study was α = 0.87. Further, to understand the relationship between need-driven behaviors and pain medication administration, the Medication Quantification Scale version III (MQS III) was used; it quantifies pain medications according to dosage, pharmacological class, and detriment weight of a medication (Gallizzi et al., 2008). Scale reliability for the current study was α = 0.92.

Data Collection Procedures

Once the LTC participants were enrolled and before pain and behavioral data were collected, the researcher (S.P.) facilitated two 1-hour workshops for staff. The objective of the first workshop was for nurses and nurses' aides who were assigned to care for resident participants with dementia to understand and apply propositions of the NDBM to dementia care, specifically, behaviors associated with pain and appropriate responses. The second 1-hour workshop instructed nurses in appropriate use of the PAINAD, APS, and CMAI tools; behaviors assessed; and scoring practices using clinical video vignettes of persons with dementia in pain. The researcher gave a short didactic presentation about the PAINAD. Nurses were shown a video of a person with dementia in pain and completed the scales individually. They were then asked to complete the PAINAD in pairs and discuss the appropriate score. Afterwards, the group discussed correct responses. Another video was shown, and the same process was used with the APS. Nurses were given the video title or description, and videos were presented in random order.

LTC participants were randomly assigned to groups according to which scale would be used first. Pain assessment using the scales was done a minimum of every 8 hours and as needed. Nurses assessed one half of participants' pain using the APS (group 1) and used the PAINAD in the remainder of participants (group 2). Assessment tools were switched after 4 weeks: pain was assessed for group 1 using the PAINAD and the APS for group 2. This counterbalanced design reduced the chances that the order of treatment adversely influenced the results. Nurses documented need-driven behaviors using the 29 indicators of the CMAI every 8 hours and medications administered on the resident's medication administration record (MAR). All documents were located with patients' MAR and electronic medical records (EMRs) to avoid any confusion on which tool to use. Upon completion of the first 2-week data collection period, additional follow up with nurses was conducted to review the pain scales, behaviors assessed, and scoring practices to ensure treatment fidelity.

After quantitative data collection, qualitative data were collected to obtain nurses' views of their experience with the pain scales. Consenting nurses completed a demographic form followed by one-on-one interviews using a semi-structured interview guide. The guide was developed from a review of the literature and refined to suit the focus of the research questions: Do the pain scales have a significant correlation with need-driven behaviors? If behaviors were thought to be pain-related, did use of the scales prompt nurses to administer pain medication? What were nurses' experiences using the pain scales? The audio-recorded interviews lasted 30 to 45 minutes allowing sufficient time for participants to share essential information. The researcher kept field notes during or immediately after the interview to record thoughts, ideas, and reflections on the interview itself.


Quantitative Data Analysis

IBM SPSS Statistics 25 was used for LTC participants; and PAINAD, APS, MSQ III, and CMAI group data were pooled for statistical analysis. Participants were mostly female (67.9%), Caucasian (100%), retired blue collar workers (57.1%), and the majority had a diagnosis of vascular dementia (42.9%). Mean age was 81.9 (SD = 6.38 years, range = 67 to 91), number of painful comorbidities was 2.1 (SD = 1.11, range = 1 to 5), and total number of comorbidities was 13.4 (SD = 2.53, range = 6 to 18) according to chart audit (Table 1).

Description of Participants With Dementia (N = 28)Description of Participants With Dementia (N = 28)

Table 1:

Description of Participants With Dementia (N = 28)

Inferential Analysis

Table 2 presents a Pearson's r correlation examining the relationship between need-driven behaviors and APS/PAINAD sum total scores and MQS III mean scores and APS/PAINAD sum total scores. Question 1 explored the relationship between each pain scale and the occurrence of need-driven behaviors. Analysis showed a significant correlational relationship between the presence of need-driven behaviors relating to pain and the APS results (r[26] = 0.41, p < 0.05; d = 0.45). This finding indicates that the APS might be a satisfactory assessment tool for pain in persons with dementia. The PAINAD did not show a significant relationship with need-driven behavior scores. Neither pain scale showed a significant relationship to the pain medications administered to participants. However, more pain medications were administered during the APS condition with a level approaching significance (p = 0.07) based on nurses' perceived need for pain relief.

Pearson's r Correlations Between Pain Scales and Pain Indicator Scores

Table 2:

Pearson's r Correlations Between Pain Scales and Pain Indicator Scores

Qualitative Data Analysis

Qualitative data were reduced, managed, and analyzed using NVivo version 12. Giorgi's (2009) five-step process was used: (1) assume the phenomenological attitude, (2) read entire written account for a sense of the whole, (3) delineate meaning units, (4) transform the meaning units into sensitive statements of their lived meanings, and (5) synthesize a general psychological structure of the experience based on the constituents of the experience.

All interviews were done by the principal investigator (PI; S.P.) and transcribed verbatim by an experienced clerical assistant. To demonstrate validity, reliability, and rigor, all transcripts were checked for accuracy against the original digital recordings by two members (S.P., B.M.S.). Primary data analysis was performed by the PI, after the selection of transcripts were independently analyzed by two authors (B.M.S., G.D.). Following several re-readings of each transcript and discussion among authors, passages were assigned descriptive codes reflecting the concepts expressed by those data. An audit trail of analysis was kept, detailing steps in the development of the coding frame for each level of analysis. Finally, the core themes or concepts were shared with participants to ensure they reflected what was expected or felt.

Nurse Participants' Characteristics and Other Background Information

Nurse participants were mostly female (83.3%) and Caucasian (100%). Most participants were RNs (66.6%); 33.3% were LPNs. Mean age of nurses was 54 (SD = 9.4 years, range = 40 to 64), years of experience as a nurse was 24 (SD = 15.3 years, range = 4 to 40), and years of experience caring for persons with dementia was 14.8 (SD = 9.2 years, range = 2 to 25). Two thirds (66.7%) of nurses had additional certification in dementia care.

Themes Emerging From Qualitative Analysis

A total of six nurses participated in one-on-one interviews. Participants' experiences were characterized into three core themes: (a) Challenges in Assessing Persons With Dementia for Pain; (b) Facilitators and Barriers to Pain Management; and (c) Difficulty Caring for Persons With Dementia (Table 3). Participant comments are presented to support each theme.

Qualitative Themes in Assessment of Pain in Persons With Dementia

Table 3:

Qualitative Themes in Assessment of Pain in Persons With Dementia

Challenges in Assessing Persons With Dementia for Pain. Three challenges were lack of time, medication knowledge, and poor communication. This theme emerged as a result of nurses' comments regarding how they assess residents' pain, necessary knowledge of the resident, and the fact that assessment is an ongoing process but lamented that lack of time was a factor.

You have to know an individual pain and so their own individual strategies, it's a process that takes time and trust. Good pain management is to observe. Good observation before you just go ahead and medicate, but we don't really have the time.

Some nurses confessed they often under-medicate patients because they are unsure of pain or lack knowledge of the medication.

If I see someone acting out, I don't automatically think pain, with the opioid thing going on I don't want to give them too many meds.

Other comments reflected challenges with communication.

I get so frustrated with the other shifts, if they had just documented or at least told me in shift report, it would have saved a lot of time.

Difficulty assessing for pain was also a challenge. Nurses stated it was important to have good assessment techniques with this population, to know the person with dementia, recognize that every person is an individual, realize assessment is an ongoing process, provide consistent care with a good attitude, and the importance of staff education and knowledge. However, this is not always translated into practice.

I learned all this stuff in school, but I admit I have forgotten a lot and there is no training here especially about pain. I'm pretty sure there's no policy either.

Identifying the meaning behind behaviors proved challenging for nurses. Comments “Is it because you're hungry? Do you need to go to the bathroom? Are there too many people around?” indicated what they consider other causes of need-driven behaviors aside from pain and from experience using the APS and PAINAD tools with this population. The majority of nurses preferred the APS for use in persons with dementia.

I think that with dementia the Abbey works a little better because it gives you a little more play as far as where, what you're looking at to kind of assess—what's the norm and how is it different.

Facilitators and Barriers to Pain Management. The second theme encompassed comments regarding pain management for persons with dementia including factors that facilitated and served as barriers to pain management. Nurses addressed some positive ways in which they believed they were able to manage pain for residents. This management of pain included giving scheduled pain medications and having as-needed medications available for use and good documentation and communication from nurse to nurse, nurse to aide, and nurse to provider.

Having those specifically documented behaviors, if it's accurate, is a great way to be able to communicate and keep records for future reference.

Nurses also discussed the importance of enough staff, having a kind approach to care (i.e., “go in with a calm approach”), empowerment through education, ongoing training and practice development, knowing the residents, establishing consistent routines, and other strategies for pain management. Barriers to effective pain management were lack of time, poor or inaccurate documentation and communication, being short staffed, bad attitudes of nurses, and unfamiliar float nurses caring for residents. One nurse explained:

We don't always have adequate time—doing so many tasks that it's hard to stop; it becomes a challenge to have enough time, frankly, to adequately assess all the residents.

In addition, confusion over which behaviors were for what need, and unpredictable resident characteristics were expressed.

They can't totally tell you—this is what hurts and that's why I'm acting that way. Are they agitated because they're in pain or something else?

Difficulty Caring for Persons With Dementia. A third theme emerged from the data. Nurses expressed difficulties in caring for persons with dementia and the challenges of correctly interpreting residents' behaviors as persons with dementia cannot adequately express their needs. Statements included the need for patience, the emotionally and physically demanding nature of the work, and feeling invalidated by others.

It takes a lot of mental, emotional, and physical effort, which is tiring sometimes, and nobody seems to care.

However, some nurses believed that they made residents' lives better, providing a feeling of satisfaction.

I feel satisfied when I can sing and joke, and they are happy.


The current study corroborates and extends prior research showing the importance of regular pain assessments based on two tools designed for observation of persons with dementia. Although there is a plethora of studies on pain management and available observation pain scales in the literature, this mixed methods study adds another layer to the interpretation of behavior as a possible response to pain rather than another need. Three important findings were generated that could inform approaches to improving pain management for persons with dementia in LTC. First, answering the research question—Do the pain scales have a significant correlation with need-driven behaviors?—this study reports that the APS was correlated to need-driven behaviors at a significant level (p < 0.05); however, the PAINAD did not have a significant correlation. These findings suggest that nurses recognized the need to assess pain based on behaviors during the APS phase of the study.

Second, the study addressed whether using the observational pain scales and recognizing pain would prompt nurses to administer pain medication. The impact of pain medication administration was done by measuring the quantification of medication given by amount, pharmacological class, and the amount of detriment to the person with dementia using the MQS III scale. Neither the APS nor PAINAD showed significant correlation with medication administration. However, the APS was at a level approaching significance (p = 0.07). This finding indicates that residents received more pain medication during the APS condition of the study.

Third, all qualitative themes were consistent with current qualitative and mixed methods research studies, such as difficulty in assessing pain (Coker et al., 2010), relying on experience for assessments (Dowding et al., 2016), poor communication between nurses/providers (Burns & McIlfatrick, 2015), lack of pain recognition, workload pressures (Corbett et al., 2014), not knowing residents, poor staffing (Lichtner et al., 2016), and inadequate training/education (Brorson et al., 2014; Gilmore-Bykovskyi & Bowers, 2013). Nurses in the current study reported a preference for the APS because it provided a score for “deviation from the norm.” Nurses reported feeling emotionally, mentally, and physically drained; impatient; frustrated; and invalidated by others in the organization, as well as residents' families. Although lack of satisfaction and not meeting resident needs has been reported (Brorson et al., 2014), some nurses expressed feeling satisfied about their impact on patients' lives.

During qualitative interviews, nurses described lack of time, appropriate staff, education, and confidence in providing effective pain management. These described barriers were consistent with other similar qualitative studies (Burns & McIlfatrick, 2015; Dowding et al., 2016). They expressed confusion in understanding which behaviors belong to which need, responding to need-driven behaviors, and the burden placed on nurses (Brorson et al., 2014). The themes extracted validated important issues regarding facilitators, barriers, scale preference, and attitudes in providing care for persons with dementia. Other studies of pain perception instruments with a similar sample size and participant type as the current study have yielded varied results, such as the study by Natavio et al. (2020), which showed the Pain Assessment Checklist for Seniors with Limited Ability to Communicate (PACSLAC) scale to be more effective than the PAINAD. However, nurses showed a preference for the PAINAD possibly due to the length of the PACSLAC (Natavio et al., 2020).

Much evidence exists suggesting that pain management in persons with dementia remains a challenge for nurses (Gilmore-Bykovskyi & Bowers, 2013; Lichtner et al., 2015). There is also support for the idea that the various pain scales, although useful to assist in clinical judgment of pain, are still limited and only as good as the user (Tsai et al., 2018). There is much less research conducted directly with LTC participants. There is also no universally accepted tool for assessing pain in the dementia population, and the inconsistency and inadequacy of current tools does not fully address the challenges of pain assessment by nurses (Flo et al., 2014; Lichtner et al., 2016; Tsai et al., 2018). Therefore, it was important to investigate not only the use of observational pain scales for persons with dementia, but to also examine nurses' perceptions of the benefits and issues with these tools and their usefulness in pain management. The current study highlights two observational pain scales that have reliability and validity and reflect pain behaviors outlined in AGS (2002, 2009) recommendations, but more importantly they are simple and easy to use. Lack of consistent monitoring and accurate documentation were consistently described by nurses as barriers to effective pain management in this vulnerable population. In addition, nurses could reflect on their experiences, further strengthening the outcomes of pain observation and management. In the future, similar studies with larger, more diverse sample sizes can be implemented to collect stronger evidence and demonstrate the effectiveness of these particular observational pain scales in the improvement of pain assessment and management.


There are several limitations to the current study. Notably, there is potential that some data provided to researchers were not recorded in patients' medical records. The recording of behaviors using the CMAI was done at the end of an 8-hour shift based on nurses' recall. The sample size was fairly small, and attrition posed a threat. However, oversampling was done, and the final participant number (N = 28) was satisfactory for statistical conclusion validity. Another potential limitation was instrumentation. The use of the pain scales was performed by different nurse observers. To decrease this threat, independent pain observers (nursing staff) were trained by the researcher. Staff also practiced with the observation pain scales using clinical videos and an acceptable interrater reliability was established by comparing staff-rated practice pain with other staff and that of the PI. There were concerns of cross-contamination; therefore, the observational pain scales were imbedded in the EMR to prevent confusion. Nurses were also aware that they were part of a study, creating a possible Hawthorne effect. In addition, the results came from only one memory-care facility in a certain geographical region of the United States, which may not be generalizable to other facilities or geographical areas.

Clinical Implications

Clinical implications from the current study suggest the need for a systematic, consistent method of observing pain-related behaviors that is essential to decoding the meanings behind expressed behaviors. Incorporating an observational pain scale, such as the APS, into the EMR might prompt nurses to recognize behaviors, treat pain, and help overcome barriers, such as lack of time. At the conclusion of the study, the memory care unit incorporated the APS into their EMR system.

More critically, it is important to establish an institutional philosophy of dementia care. Using the NDBM was empowering for nurse participants as it helped validate and give structure to the care they intuitively provide. A theoretical structure might address other barriers, such as education and communication, and ultimately serve as a model that can be transferred to other facilities.

Due to the numerous barriers attributed to pain assessment, research to develop alternate assessment methods is needed. Additional research and education are warranted regarding how assessment outcomes are translated into clinical decisions to manage need-based behaviors. However, even research that purports to detect pain in persons with dementia should be approached cautiously as there are many variables, such as situational factors, that affect behaviors. An individualized approach to assessment may be recommended to determine if the person's behavior has deviated from baseline.

More research should be conducted in clinical practice to assess the feasibility and clinical utility of observational pain scales and their potential for use in everyday practice. In addition, research on pain protocols that incorporate the use of these scales is warranted. Further studies on types and degrees of dementia, as well as types of pain, should be explored. Qualitative findings indicated that LTC nurses experience many challenges in managing pain for people with advanced dementia. More in-depth research on these barriers are indicated as well as caregiver burden, compassion fatigue, and social support.


The current embedded mixed methods study examined the effects of implementing two observational pain scales, the APS and PAINAD, in the assessment and treatment of pain and correlation to need-driven behaviors. Findings revealed the APS had a significant correlation to need-driven behaviors and that nurses preferred the APS because it included behavior change from resident's baseline. Nurses must be sensitized to need-driven behaviors and recognize behaviors as symptoms of pain or other unmet needs. Raising awareness of pain in this population is a high priority. The goal of providing comfort and care for persons with dementia is pervasive and ongoing. Additional research in the area of pain management and associated need-driven behaviors is necessary for a more accurate differential assessment, and consequently, relief from pain for this vulnerable population.


  • Abbey, J., Piller, N., De Bellis, A., Esterman, A., Parker, D., Giles, L. & Lowcay, B. (2004). The Abbey pain scale: A 1-minute numerical indicator for people with end-stage dementia. International Journal of Palliative Nursing, 10(1), 6–13 doi:10.12968/ijpn.2004.10.1.12013 [CrossRef] PMID:14966439
  • Ahn, H., Garvan, C. & Lyon, D. (2015). Pain and aggression in nursing home residents with dementia: Minimum Data Set 3.0 analysis. Nursing Research, 64(4), 256–263 doi:10.1097/NNR.0000000000000099 [CrossRef] PMID:26126060
  • Ahn, H. & Horgas, A. (2013). The relationship between pain and disruptive behaviors in nursing home residents with dementia. Canadian Nursing Home, 13(4), 14 doi:10.1186/1471-2318-13-14 [CrossRef] PMID:23399452
  • Algase, D., Beck, C., Kolanowski, A., Whall, A., Berent, S., Richards, K. & Beattie, E. (1996). Need-driven dementia-compromised-behavior: An alternative view of disruptive behavior. American Journal of Alzheimer's Disease, 11(6), 10–18 doi:10.1177/153331759601100603 [CrossRef]
  • American Geriatrics Society Panel on Persistent Pain in Older Persons. (2002). The management of persistent pain in older persons. Journal of the American Geriatrics Society, 50(Suppl.), S205–S224 doi:10.1046/j.1532-5415.50.6s.1.x [CrossRef] PMID:12067390
  • American Geriatrics Society Panel on Pharmacological Management of Persistent Pain in Older Persons. (2009). Pharmacological management of persistent pain in older persons. Journal of the American Geriatrics Society, 57(8), 1331–1346 doi:10.1111/j.1532-5415.2009.02376.x [CrossRef] PMID:19573219
  • Brecher, D. B. & West, T. L. (2016). Under-recognition and undertreatment of pain and behavioral symptoms in end-stage dementia. The American Journal of Hospice & Palliative Care, 33(3), 276–280 doi:10.1177/1049909114559069 [CrossRef] PMID:25433066
  • Brorson, H., Plymoth, H., Örmon, K. & Bolmsjö, I. (2014). Pain relief at the end of life: Nurses' experiences regarding end-of-life pain relief in patients with dementia. Pain Management Nursing, 15(1), 315–323 doi:10.1016/j.pmn.2012.10.005 [CrossRef] PMID:23453467
  • Burns, M. & McIlfatrick, S. (2015). Nurses' knowledge and attitudes towards pain assessment for people with dementia in a nursing home setting. International Journal of Palliative Nursing, 21(10), 479–487 doi:10.12963/ijpn.2015.21.10.479 [CrossRef] PMID:26505082
  • Cohen-Mansfield, J. (1996). Conceptualization of agitation: Results based on the Cohen-Mansfield agitation inventory and the agitation behaviour mapping instrument. International Psychogeriatrics, 8(Suppl. 3), 309–315 doi:10.1017/S1041610297003530 [CrossRef] PMID:9154580
  • Cohen-Mansfield, J., Dakheel-Ali, M., Marx, M. S., Thein, K. & Regier, N. G. (2015). Which unmet needs contribute to behavior problems in persons with advanced dementia?Psychiatry Research, 228, 59–64 doi:10.1016/j.psychres.2015.03.043 [CrossRef] PMID:25933478
  • Cohen-Mansfield, J., Marx, M. S. & Rosenthal, A. S. (1989). A description of agitation in a nursing home. The Journals of Gerontology. Series A, Biological Sciences and Medical Sciences, 44(3), M77–M84 doi:10.1093/geronj/44.3.M77 [CrossRef] PMID:2715584
  • Coker, E., Papaioannou, A., Kaasalainen, S., Dolovich, L., Turpie, I. & Taniguchi, A. (2010). Nurses' perceived barriers to optimal pain management in older adults on acute medical units. Applied Nursing Research, 23, 139–146 doi:10.1016/j.apnr.2008.07.003 [CrossRef] PMID:20643323
  • Corbett, A., Achterberg, W., Husebo, B., Lobbezoo, F., de Vet, H., Kunz, M., Strand, L., Canstantinou, M., Tudose, C., Kappesser, J., de Waal, M. & Lautenbacher, S. (2014). An international road map to improve pain assessment in people with impaired cognition: The development of the Pain Assessment in Impaired Cognition (PAIC) meta-tool. BMC Neurology, 14, 229 doi:10.1186/s12883-014-0229-5 [CrossRef]
  • Dowding, D., Lichtner, V., Allcock, N., Briggs, M., James, K., Keady, J., Lasrado, R., Sampson, E. L., Swarbrick, C. & José Closs, S. (2016). Using sense-making theory to aid understanding of the recognition, assessment and management of pain in patients with dementia in acute hospital settings. International Journal of Nursing Studies, 53, 152–162 doi:10.1016/j.ijnurstu.2015.08.009 [CrossRef] PMID:26363705
  • Faul, F., Erdfelder, E., Lang, A.-G. & Buchner, A. (2007). G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39, 175–191 doi:10.3758/BF03193146 [CrossRef] PMID:17695343
  • Flo, E., Gulla, C. & Husebo, B. S. (2014). Effective pain management in patients with dementia: Benefits beyond pain?Drugs & Aging, 31, 863–871 doi:10.1007/s40266-014-0222-0 [CrossRef] PMID:25373921
  • Gallizzi, M., Gagnon, C., Harden, R. N., Stanos, S. & Khan, A. (2008). Medication Quantification Scale Version III: Internal validation of detriment weights using a chronic pain population. Pain Practice, 8(1), 1–4 doi:10.1111/j.1533-2500.2007.00163.x [CrossRef] PMID:18211588
  • Gilmore-Bykovskyi, A. L. & Bowers, B. J. (2013). Understanding nurses' decisions to treat pain in nursing home residents with dementia. Research in Gerontological Nursing, 6(2), 127–138 doi:10.3928/19404921-20130110-02 [CrossRef] PMID:23330944
  • Giorgi, A. (2009). The descriptive phenomenological method in psychology: A modified Husserlian approach. Duquesne University.
  • Herr, K., Coyne, P. J., McCaffery, M., Manworren, R. & Merkel, S. (2011). Pain assessment in the patient unable to self-report: Position statement with clinical practice recommendations. Pain Management Nursing, 12(4), 230–250 doi:10.1016/j.pmn.2011.10.002 [CrossRef] PMID:22117755
  • Higuchi, K. S., Edwards, N., Carr, T., Marck, P. & Abdullah, G. (2015). Development and evaluation of a workshop to support evidence-based practice change in long-term care. Journal for Nurses in Professional Development, 31(1), 28–34 doi:10.1097/NND.0000000000000120 [CrossRef] PMID:25608094
  • Husebo, B. S., Achterberg, W. & Flo, E. (2016). Identifying and managing pain in people with Alzheimer's disease and other types of dementia: A systematic review. CNS Drugs, 30, 481–497 doi:10.1007/s40263-016-0342-7 [CrossRef] PMID:27240869
  • Husebo, B. S., Ballard, C., Cohen-Mansfield, J., Seifert, R. & Aarsland, D. (2014). The response of agitated behavior to pain management in persons with dementia. The American Journal of Geriatric Psychiatry, 22, 708–717 doi:10.1016/j.jagp.2012.12.006 [CrossRef] PMID:23611363
  • Husebo, B. S., Ballard, C., Fritze, F., Sandvik, R. K. & Aarsland, D. (2014). Efficacy of pain treatment on mood syndrome in patients with dementia: A randomized clinical trial. International Journal of Geriatric Psychiatry, 29, 828–836 doi:10.1002/gps.4063 [CrossRef] PMID:24806873
  • Kutschar, P., Bauer, Z., Gnass, I. & Osterbrink, J. (2017). Does item overlap render measured relationships between pain and challenging behaviour trivial? Results from a multicentre cross-sectional study in 13 German nursing homes. Nursing Inquiry, 24, e12182 doi:10.1111/nin.12182 [CrossRef] PMID:28090716
  • Lee, K. H., McConnell, E. S., Knafl, G. J. & Algase, D. L. (2015). Pain and psychological well-being among people with dementia in long-term care. Pain Medicine, 16, 1083–1089 doi:10.1111/pme.12739 [CrossRef] PMID:25800276
  • Lichtner, V., Dowding, D., Allcock, N., Keady, J., Sampson, E. L., Briggs, M., Corbett, A., James, K., Lasrado, R., Swarbrick, C. & Closs, S. J. (2016). The assessment and management of pain in patients with dementia in hospital settings: A multi-case exploratory study from a decision making perspective. BMC Health Services Research, 16, 427–442 doi:10.1186/s12913-016-1690-1 [CrossRef] PMID:27553364
  • Lichtner, V., Dowding, D. & Closs, S. J. (2015). The relative meaning of absolute numbers: The case of pain intensity scores as decision support systems for pain management of patients with dementia. BMC Medical Informatics and Decision Making, 15, 111–112 doi:10.1186/s12911-015-0233-8 [CrossRef] PMID:26703244
  • Lichtner, V., Dowding, D., Esterhuizen, P., Closs, S. J., Long, A. F., Corbett, A. & Briggs, M. (2014). Pain assessment for people with dementia: A systematic review of systematic reviews of pain assessment tools. BMC Geriatrics, 14, 138 doi:10.1186/1471-2318-14-138 [CrossRef] PMID:25519741
  • Liu, J. & Lai, C. K. (2017). Implementation of observational pain management protocol for residents with dementia: A cluster-RCT. Journal of the American Geriatrics Society, 65(3), 56–63. doi:10.1111/jgs.14763 [CrossRef]
  • McCormack, B., Dewing, J., Breslin, L., Coyne-Nevin, A., Kennedy, K., Manning, M., Peelo-Kilroe, L. & Tobin, C. (2009). Practice development: Realising active learning for sustainable change. Contemporary Nurse, 32(1–2), 92–104 doi:10.5172/conu.32.1-2.92 [CrossRef] PMID:19697981
  • Monroe, T. B., Misra, S. K., Habermann, R. C., Dietrich, M. S., Cowan, R. L. & Simmons, S. F. (2014). Pain reports and pain medication treatment in nursing home residents with and without dementia. Geriatrics & Gerontology International, 14(3), 541–548 doi:10.1111/ggi.12130 [CrossRef] PMID:24020433
  • Moschinski, K., Kuske, S., Andrich, S., Stephan, A., Gnass, I., Sirsch, E. & Icks, A. (2017). Drug-based pain management for people with dementia after hip or pelvic fractures: A systematic review. BMC Geriatrics, 17, 54–69 doi:10.1186/s12877-017-0446-z [CrossRef] PMID:28196525
  • Natavio, T., McQuillen, E., Dietrich, M. S., Wells, N., Rhoten, B. A., Vallerand, A. H. & Monroe, T. B. (2020). A comparison of the Pain Assessment Checklist for Seniors with Limited Ability to Communicate (PACSLAC) and Pain Assessment in Advanced Dementia Scale (PAINAD). Pain Management Nursing. Advance online publication. doi:10.1016/j.pmn.2020.04.001 [CrossRef] PMID:32475696
  • Parkman, S. (2018).Pain, need-driven behaviors in dementia, and nurses' perceptions: An embedded mixed methods study [Doctoral dissertation]. University of Texas at Tyler, Tyler, TX.
  • Sacoco, C. & Ishikawa, S. (2014). Evidence-based practice for pain identification in cognitively impaired nursing home residents. Nursing Clinics of North America Journal, 49, 345–356 doi:10.1016/j.cnur.2014.05.008 [CrossRef] PMID:25155534
  • Sandvik, R. K., Selbaek, G., Seifert, R., Aarsland, D., Ballard, C., Corbett, A. & Husebo, B. S. (2014). Impact of a stepwise protocol for treating pain on pain intensity in nursing home patients with dementia: A cluster randomized trial. European Journal of Pain, 18(10), 1490–1500 doi:10.1002/ejp.523 [CrossRef]
  • Teng, E. L. & Chui, H. C. (1987). The Modified Mini-Mental State (3MS) examination. The Journal of Clinical Psychiatry, 48, 314–318 PMID:3611032
  • Tsai, I. P., Jeong, S. Y. & Hunter, S. (2018). Pain assessment and management for older patients with dementia in hospitals: An integrative literature review. Pain Management Nursing, 19(1), 54–71 doi:10.1016/j.pmn.2017.10.001 [CrossRef] PMID:29153920
  • Warden, V., Hurley, A. C. & Volicer, L. (2003). Development and psychometric evaluation of the Pain Assessment in Advanced Dementia (PAINAD) scale. Journal of the American Medical Directors Association, 4, 9–15 doi:10.1097/01.JAM.0000043422.31640.F7 [CrossRef] PMID:12807591
  • World Health Organization. (2019). Dementia fact sheet.

Description of Participants With Dementia (N = 28)

VariableMean (SD) (Range)
Age (years)81.89 (6.38) (67 to 91)
3MS16.8 (15.6) (0 to 46)
Painful comorbidities2.14 (1.11) (1 to 5)
Total comorbidities13.36 (2.53) (6 to 18)
n (%)
  Female19 (67.9)
  Male9 (32.1)
  White28 (100)
Marital status
  Married12 (42.9)
  Widowed12 (42.9)
  Divorced3 (10.7)
  Other1 (3.6)
  Catholic12 (42.9)
  Protestant7 (25)
  Born-again Christian4 (14.3)
  No religious affiliation5 (17.9)
Educational level
  Less than high school3 (10.7)
  High school diploma14 (50)
  Some college3 (10.7)
  College graduate8 (28.6)
Type of occupation
  Blue collar16 (57.1)
  White collar12 (42.9)
Type of dementia
  Vascular12 (42.9)
  Unspecified10 (35.7)
  Alzheimer's5 (17.9)
  Alcohol-related1 (3.6)

Pearson's r Correlations Between Pain Scales and Pain Indicator Scores

Pain Scale Sum TotalNeed-Based Behavior Scalep Value

Qualitative Themes in Assessment of Pain in Persons With Dementia

Challenges in Assessing Persons With Dementia for Pain
Difficulty with detection and assessment of pain due to cognitive decline.
Inadequate assessment techniques; lack of good observation skills.
Lack of knowledge about analgesics.
Insufficient documentation on effects of analgesics.
Lack of time spent with patients or not knowing patients.
Inadequate training of pharmacological and nonpharmacological pain relief.
Fear of over-medicating.
Lack of consistent practices about pain assessment and inconsistent policies.
Facilitators and Barriers to Pain Management
  Giving scheduled pain medications and having as-needed medications available for use.
  Good documentation and communication between nurse to nurse, nurse to aides, and nurse to provider.
  Adequate staffing.
  Calm approach.
  Consistent routines.
  Empowerment through education.
  Majority of nurses preferred the Abbey Pain Scale as an observational pain scale.
  Lack of time.
  Poor or inadequate documentation and communication.
  Bad attitudes of nurses.
  Float nurses or nurses unfamiliar with patients.
  Confusion over what behaviors are “pain” behaviors or another need.
  Unpredictable patient behavior.
Difficulty Caring for Persons With Dementia
  Challenges meeting the needs of patients due to inability to effectively communicate.
  Need for an abundance of patience.
  Emotionally and physically draining.
  Invalidation from peers, organization, and family members of patients.
  Some report a degree of satisfaction from making a patient happy.

Dr. Parkman is Assistant Professor of Nursing, and Dr. McGuire is Chair and Director of Nursing, Saint Joseph's College of Maine, Standish, Maine; and Dr. Mastel-Smith is Associate Professor, and Dr. Duke is Professor and Associate Dean, Office of Research, University of Texas at Tyler, Tyler, Texas.

The authors have disclosed no potential conflicts of interest, financial or otherwise.

The authors wish to express their deepest gratitude to Edward Latham and the nurses who participated in this study for their enthusiasm, inspiration, and insights.

Address correspondence to Suzanne Parkman, PhD, RN, CNE, Assistant Professor of Nursing, Saint Joseph's College of Maine, 278 Whites Bridge Road, Standish, ME 04084; email:

Received: January 27, 2020
Accepted: August 19, 2020


Sign up to receive

Journal E-contents