Journal of Gerontological Nursing

Clinical Outlook 

Pain Assessment in Older Adults Who Are Cognitively Impaired

Sharon Galloway, RN, MSN; Linda Turner, RN, MN


Dementia may cause a barrier to communication; therefore, behavioral indicators and familiarity with patients are valuable in assessing and treating pain in this population.


Dementia may cause a barrier to communication; therefore, behavioral indicators and familiarity with patients are valuable in assessing and treating pain in this population.

Pain is a highly subjective experience affected by numerous factors, including personality, cognition, cultural background, and past experiences with pain. There are no biological markers for pain, and as a result, assessment is based solely on patients' perceptions. and selfreports (Ferrell, Ferrell, & Osterweil, 1990; Parke, 1992). Pain is defined as an

unpleasant sensory or emotional experience associated with actual or potential tissue damage or described in terms of such damage (International Association for the Study of Pain, 1979, p. 250).

McCaffery's (1972) definition of pain is

whatever the experiencing person says it is and exists whenever he says it does (p. 8).

Both of these definitions rely heavily on the subjective component of sufferers, and if patients are unable to report their pain, it ceases to exist.

Principles of pain management are the same for elderly patients as for any patient in pain. However, there are several specific areas of concern regarding the management of pain in the elderly population (Hofland, 1992). Elderly patients may perceive and report pain differently from younger patients because of physical and psychological changes associated with aging. Many patients are at serious risk for undertreatment or overtreatment with analgesics because of unpredictable age-related changes in response to these drugs (Gloss, 1994). Health care workers in longterm care facilities exaggerate concerns for respiratory depression and fear of addiction and, thus, may not use narcotic analgesics (Ferrell, McCaffery, & Grant, 1991).

Dementia and other types of cognitive impairments present serious barriers to pain assessment. Dementia is an impairment in brain functioning. It is characterized by memory loss, personality changes, and loss of other cognitive functions such as judgment, problem solving, abstract thinking, and language skills (Canadian Study of Health and Aging Working Group, 1994). In Canada, it is estimated that 8% of Canadians age 65 or older have dementia. This represents approximately 252,600 individuals (Canadian Study of Health and Aging Working Group, 1994). If prevalence estimates remain constant, the number of Canadians with dementia will rise to 592,000 by 2021. In most situations, the presence of a cognitive impairment will interfere with the caregivers* administration of analgesics.

In a study of 60 patients with Alzheimer's disease, it was shown that although 26 of these patients had diagnoses commonly associated with painful conditions, only 3 patients were given analgesics (Marzinski, 1991). Parmalee, Smith and Katz (1993) reviewed 758 patients in nursing homes and studied the hypothesis that perhaps cognitively impaired patients used pain as a socially acceptable manner in which to mask early cognitive decline. The authors concluded that although patients who are cognitiveIy impaired slightly underreported pain, their reports of pain were as valid as those who were cognitively intact. They also found that as the cognitive impairment increased in severity, there was a corresponding decrease in reporting of pain.

Statistics indicate pain is two times more likely to occur in individuals age 60 and older (Ferrell, 1991; Ferrell & Ferrell, 1990). Forrest (1995) reports that 70% to 83% of nursing home residents describe having significant pain and that approximately one third of these residents may be in constant pain. Causes are varied, but it appears that arthritis, other musculoskeletal conditions such as old fractures, and neuropathies are the major causes of such pain (Ferrell, 1995; Ferrell, Ferrell, & Rivera, 1995).

Limited research has focussed on the ability of elderly individuals to use various pain assessment scales. It has been suggested that elderly individuals may have deficits in abstract ability and visual disturbances and so may have difficulty using the Visual Analogue Scale (VAS) (Herr & Mobily, 1993). Unfortunately, this is the tool most often used in hospitals and long-term care facilities. Recently, Ferrell et al. (1995) assessed pain in residents with cognitive impairment by using four unidimensional pain scales. They concluded that 83% of the 217 residents could complete at least one of the pain assessment scales. They noted that the conditions under which the tool was administered were important (e.g., adequate lighting, use of visual cues, ensuring that residents had time to respond).

One objective scale was developed by Hurley, Volicer, Hanrahan, Houde, and Volicer (1992) to measure discomfort in patients with advanced dementia of the Alzheimer's type. They generated a list of pain indicators from RNs and care aides. This list was refined further through interrater reliability to include nine behavioral indicators for discomfort. This work suggests that care workers who have experience and are familiar with their patients are able to use behavioral observations to identify pain.

Marzinski (1991) also has reported that health care workers with experience and familiarity with their patients are able to detect differences in residents when they are in pain. What appears to be intuition is actually a complex recognition of verbal and nonverbal cues and the arrangement of these cues into a meaningful pattern.

The purpose of this research was twofold: first, to learn how skilled health care workers assess pain in residents with whom they are familiar and, second, to understand the treatment given to those residents.


Separate focus groups were conducted with RNs and patient care aides (PCAs) to elicit information to answer the research questions. Particular emphasis was given to recruiting individuals who were representative of the population of nurses at the authors* hospital caring for geriatric patients.

Inclusion criteria included the following:

* 1 year or more employment in the hospital.

* Permanent member of the staff on the particular patient unit.

* Identification by the clinical nurse specialist (CNS) and a nurse manager as being representative of the staff nurses on that particular unit with respect to their knowledge of pain.

The CNS contacted the staff member, determined whether the individual was willing to participate, and then informed the individual of the time and date of the focus group. The interviewer, an independent practitioner skilled in the interview process, obtained written informed consent from the participants prior to the beginning of each interview. Five separate focus groups of RNs only and PCAs only were conducted. The interviews were conducted between the dates of November 6, 1995 and January 23, 1996. Interviews were typed verbatim by a skilled transcriber. Each interview was reviewed by each of the investigators, and a thematic content analysis then was conducted by the investigators (Giorgi, 1985; Ornery, 1983).

After the content was analyzed and collated, the investigators scheduled meetings with the original focus group participants to confirm that their analysis made sense to the participants. Three meetings were held with the original five focus groups and two of the investigators.


Approval was obtained from the Hospital Research Committee as well as the University of British Columbia's Behavioral Research Ethics Committee. Informed consent was obtained from all participants in this study. All study results were kept confidential and anonymous. Results will be destroyed after completion of the study.


Determine What Patient Behaviors Trigger Treatment for Pain

The RNs and PCAs described numerous indicators of pain in patients who are cognitively impaired. The data were rich in detailed explicit indicators of expressions of pain in the elderly patients with cognitive impairment. These pain cues were categorized into verbal, facial, and behavioral expressions and physiological parameters. There was little question that staff were the experts in interpreting and identifying pain in this population of patients.

A summary of comments follows: "You can have different types of pain." You can have "physiological pain or emotional pain" but it usually "prevents patients from performing at their optimal level." "Pain is a discomfort and breaks one's body harmony."

"Pain in all people occurs in degree and how people deal with pain enters into their expression of pain." "Anything that varies from the normal state, be it physical or mental that would cause a change in behavior." Sometimes people "won't eat or you know by their facial expressions and forms of depression." When you touch them "they will cry" or they will "wince and move away." Some patients have a very "distressed face and whimper and cry." Sometimes their behavior becomes "childlike." On occasion, patients will "suddenly withdraw from the group." It is important to "monitor their activities and their reactions." You need to ask, "Has the patient changed from the last time you saw them, and is there something different?". The thematic content analysis was compiled in a table of indicators contained in the Figure.

Determine What Processes or Situations Promote Assessment and Treatment of Pain

The RNs and PCAs had similar ideas regarding how to assess the pain. They both believed you "had to get to know the person." It is helpful first to "talk to a family member." "If you know that the person's diagnosis is painful, then you know to look for specific symptoms" (e.g., "does the person have arthritis?"). You have to ask, "Will this kind of situation normally cause a lot of pain in an older person? That's kind of an indicator as to whether a person who is cognitively impaired will have pain."

The RNs and PCAs develop a care plan to handle the pain. "Whatever is causing this person to suffer, there might be some way to relieve it as well." It takes a "couple of people getting together and.. .seeing what we [sic] can do about it." The other thing they identified is that "geriatric patients have multisystem problems. [Acetaminophen] will work for something but not touch another pain elsewhere."

Determine How This Information Is Being Communicated Among the Health Care Professionals Caring for Older Adults Who Are Cognitively Impaired

Communication regarding pain took several different forms. In total, 1 1 different places were identified where information could be found on the pain status of the patients. Several places in the charts, the nursing care plans, flow sheets, as well as audiotaped, verbal, and written reports were used. "We even put a sticker on our cart - on top of the cart - that you have to read before you pass your medications, and then the one who puts that on must say, 'assess in how many [sic] days/ and puts it on the calendar." Notes attached to medication records and the unit communication books were other places where information was exchanged. The 1 1 different places may reflect the different types of documentation used in the different units of the hospital. As well, the 1 1 different places may reflect some frustration with and reluctance to use the current pain flow sheets. One respondent stated:

The pain flow sheets don't work with these cognitively impaired people so that means spelling it out on the care plan and then also documenting what we see, what we think is wrong, and then what the results are.

Several different individuals were identified as being part of the communication process. In extended care, the PCAs for the patients in pain would communicate to the RNs, who could then inform the clinicians. The clinicians would inform other team members about the pain and either the clinicians or primary nurses would contact the physicians. The PCAs expressed a great deal of frustration regarding the lack of feedback and direct contact they had with the decisionmakers in this process.

Explore the Process RNs and PCAs Currently Are Using to Assess Pain in Older Adults Who Are Cognitively Impaired

The RNs identified a two-step problem solving process to establish the pain status of older adults who are cognitively impaired. They described in detail how they first gathered information about older patients who were cognitively impaired and then interpreted that information to determine if these patients were experiencing pain.

Information was gathered in the following three ways:

* By direct observation and assessment of the patients.

* Through the written records.

* By conferring with others.

In using their observational and assessment skills, nursing staff would try to ask the patients, using direct questioning, to determine if the patients were in pain. One respondent stated:

Sometimes, you can talk with them and say, 'where does it hurt, are you hurting?', not that they can tell you but you never know what kind of expression you might get from them.

The nursing staff also looked for the presence or absence of specific symptoms associated with pain. They looked for any change in behavior from baseline as a possible indicator that the individuals were in pain. For many patients, a change in baseline behavior was detected in response to nursing care or in the patients' ability to participate in therapeutic activities. One respondent stated:

Sometimes they don't look like themselves, like one day they are so active and the next day, they are kind of limp and their face is withdrawn. Their color changes, they hang their head.

Figure. Amy's guide. This guide is for assessment of pain in cognitively impaired older adults or in those individuals who temporarily have altered mental status or who do not communicate clearly. It is dedicated to Amy McAuley, Clinical Nurse Specialist, Gerontology, Vancouver Hospital and Health Sciences Centre, Vancouver, British Columbia, Canada, who was one of the original researchers and who died on October 11, 1996.

Figure. Amy's guide. This guide is for assessment of pain in cognitively impaired older adults or in those individuals who temporarily have altered mental status or who do not communicate clearly. It is dedicated to Amy McAuley, Clinical Nurse Specialist, Gerontology, Vancouver Hospital and Health Sciences Centre, Vancouver, British Columbia, Canada, who was one of the original researchers and who died on October 11, 1996.

The acute care nurses noted how disadvantaged they were because they frequently did not know the baseline behavior of their patients. Their assessment of pain was based on the respondents* understanding of the patients' medical diagnoses and the nursing staff's understanding of pathophysiology. One respondent stated:

A male patient had osteomyelitis in one of his toes. He was also very cognitively impaired and had restless, agitated behavior. Finally, one of the nurses said he must be having a lot of pain. Then they started looking at the pain that he was having from the bone problem.

Information also was gathered from written records. Besides identifying specific diagnoses or procedures that may be associated with pain, these records were used to identify what, if any, treatment had been used to treat pain in the patients, and to evaluate the outcome.

Formal and informal discussions with other individuals were identified as the third source of information. Family members of the individuals who are cognitively impaired as well as other members of the interdisciplinary team provided input to the RNs and PCAs in the assessment of pain in patients who are cognitively impaired. While some of this information was discussed in the formal care conferences, other discussions occurred on a less formal basis. One RN said:

It usually starts out with an assessment and a couple of people getting together and saying this person is really restless.-.let's see what we can do about it.

Three factors were identified that influenced how the study participants interpreted the information they gathered. Personal experience with pain, cultural influences, and previous professional experiences in handling pain influenced how RNs reached a decision on the pain status of patients who are cognitively impaired. One RN stated how her husband's recent experience with pain after knee surgery made her more aware of how pain can present in different ways. Cultural norms of the patients were discussed as influencing how the patients may express pain as well as affecting how pain behaviors were interpreted. Acute and extended care workers used their experience with other patients with similar diseases or surgeries to determine if individuals who were cognitively impaired were in pain.


These findings further support that caregivers experienced in working with elderly individuals who are cognitively impaired are skilled in identifying pain in this population. This finding also was supported by the depth and detail of descriptive information gathered from personal care attendants familiar with individual residents. This finding is consistent with those of Marzinski (1991) and Hurley et al. (1992) who identified behaviors as possible indicators of pain.

The information obtained in this study indicates that a thorough baseline knowledge of patients who are cognitively impaired is imperative when assessing pain. This knowledge may be obtained from family and friends who know the patients well and also from caregivers who have worked with the patients for a length of time.

As concluded by other researchers (Ferrell et al., 1995; Parmalee et al., 1993), patients who are cognitively impaired, depending on the severity of the cognitive impairment, may be able to respond to questions about pain. Although direct questioning may not always be appropriate, this method was not automatically excluded by the nurses.

Ferrell et al. (1995) discussed the type of measurement tool that may be used for assessing pain in patients who are cognitively impaired. They reported that by using a variety of measurement tools, 83% of patients in their sample were able to communicate levels of pain. Conditions when administering the tools are important (e.g., lighting, print size). It seems that any elucidation regarding reporting pain in this population is valuable. The authors believe observations of behaviors may contribute to pain assessment and that future research is needed in this area.

Although the Agency for Health Care Policy and Research (1992) guidelines recommend use of an objective measure for assessing pain, this may not be possible with individuals who are cognitively impaired. Nurses must be aware that use of behavioral indicators may be valuable in the assessment and treatment of pain in this population. Experienced geriatric practitioners were able to clearly identify behaviors which indicated pain but did not believe their assessments were valid. Consequently, pain was not assessed at all. The detailed description of behaviors was particularly useful for acute care nurses. In acute care, nurses were familiar only with using a VAS, so when patients could not respond, pain was not assessed. The introduction of "Amy's Guide" (Figure) provided acute care nurses with assessment parameters.

The second problem was encountered in communicating the pain assessment and developing comprehensive plans of care. In cases in which patients cannot respond using an objective measurement, it is preferable to observe behaviors rather than not assess at all. It is important to note that behavioral assessments can be used to evaluate outcomes of interventions to control pain.

Because of numerous factors, assessment of pain in the cognitively impaired population is not conducted well. The identified behaviors raise awareness regarding how pain may present in individuals who are cognitively impaired. The authors hope the use of this guide will cause nurses to be more aware of the behavioral signs of pain and improve management of pain in this population. Currently, this guide and a flowsheet are involved in trials in three hospital units, including extended and acute care areas. The authors believe this trial will provide insight into its utility.

A copy of the flowsheet is available by contacting Vancouver Hospital and Health Sciences Centre at (604) 875-5666 extension 62400.


  • Agency for Health Care Policy and Research. (1992). Acute pain management operative and medical procedures and trauma (AHCPR Publication No. 92-0032). Rockville, MD: United States Department of Health and Human Services.
  • Canadian Study of Health and Aging Working Group. (1994). Canadian study of health and aging: Study methods and prevalence of dementia. Canadian Medical Association Journal, 150(6), 899-913.
  • Gloss, SJ. (1994). Pain in elderly patients: A neglected phenomenon? Journal of Advanced Nursing, 19(6), 1072-1081.
  • Ferrell, B. (1991). Pain management in elderly people. Journal of the American Geriatrics Society, 39, 164-173.
  • Ferrell, B.A. (1995). Pain evaluation and management in the nursing home. Annals of Internal Medicine, 123(9), 681-687.
  • Ferrell, B., Ferrell, B., & Osterweil, D. (1990). Pain in the nursing home. Journal of the American Geriatrics Society, 38, 409-414.
  • Ferrell, B.A., Ferrell, B.R., & Rivera, L. (1995). Pain in cognitìvely impaired nursing home patients. Journal of Pain and Symptom Management, 10, 591-598.
  • Ferrell, B.R., & Ferrell, B.A. (1990). Easing the pain. Geriatric Nursing, i/(4), 175178.
  • Ferrell, B.R., McCaffery, M., & Grant, M. (1991). Clinical decision making and pain. Cancer Nursing, 14, 289-297.
  • Forrest, J. (1995). Assessment of acute and chronic pain in older adults. Journal of Gerontological Nursing, 2i(10), 15-20.
  • Giorgi, A. (1985). Phenomenology and psychological research. Pittsburgh, PA: Duquesne University Press.
  • Herr, K., Oc MobÜy, P. (1993). Comparison of selected pain assessment tools for use with the elderly. Applied Nursing Research, 6(1), 39-47.
  • Hofland, S. (1992). Elder beliefs: Blocks to pain management. Journal of Gerontological Nursing, 18(6), 19-24.
  • Hurley, A.C., Volicer, B.J., Hanrahan, P.A., Houde, S., & Volicer, L. (1992). Assessment of discomfort in advanced Alzheimer's patients. Research in Nursing and Health, IS, 369-377.
  • International Association for the Study of Pain. (1979). Pain terms: A list with definitions and notes on usage. Pain, 6, 249-252.
  • Marzinski, L. (1991). The tragedy of dementia: Clinically assessing pain in the confused, nonverbal elderly. Journal of Gerontological Nursing, 17(6), 25-28.
  • McCaffery, M. (1972). Nursing management of the patient with pain. Toronto: Lippincott.
  • Ornery, A. (1983). Phenomenology: A method for nursing research. Advances in Nursing Science, 5(2), 49-63.
  • Parke, B. (1992). Pain in the cognittvely impaired elderly. The Canadian Nurse, 8, 17-20.
  • Parmalee, P., Smith, B., & Kate, I. (1993) Pain complaints and cognitive status among elderly institution residents. Journal of the American Geriatrics Society, 41, 517-522.


Sign up to receive

Journal E-contents