In older patients with dementia, behavioral changes don’t always signal a decline in health. Sometimes, a treatable pain condition is to blame.

Mr. M, an 87-year-old patient with a history of severe dementia (Alzheimer type), hypertension, diabetes, and persistent low back pain, was brought to the clinic by his daughter because of his increasingly frequent vocalizations.
 
At his last clinic visit, Mr. M could transfer from a sitting to a standing position independently. He was dependent in all other activities of daily living except self-feeding. While he could respond to verbal commands, he couldn’t speak well enough to make his needs known. 
 
However, 2 weeks after that visit, Mr. M became very weak and required assistance with transfers. Shortly thereafter, he became bedbound. A urinary tract infection was diagnosed and antibiotics were prescribed. Mr. M became resistant to care and developed repetitive unintelligible vocalizations. His physician prescribed quetiapine 25 mg twice daily, but Mr. M’s behavior did not improve.
 
So Mr. M’s daughter brought him in to the clinic for further evaluation. During the visit, Mr. M appeared agitated and more confused than on previous visits; physical examination was unremarkable. In addition to the quetiapine that Mr. M had recently been prescribed, he was also taking lisinopril, metformin, donepezil, memantine, sertraline, and acetaminophen.

A full work-up was performed to determine the cause of his presumed delirium. However, the results of a urinalysis, complete blood count, metabolic profile, electrocardiogram, and chest x-ray were normal. 
 
How would you proceed with this patient’s care?
Evaluating patients like Mr. M can be challenging—not only because they are unable to provide a reliable history, but also because an initial work-up may not help identify the underlying clinical issue.
 
In cases like this, it’s advisable to include a painful condition in the differential diagnosis and consider empirical pain therapy. Our experience in treating geriatric patients suggests that there is a high probability that Mr. M has an undetected painful condition, and that treating the presumed pain may result in reduced agitation.
 
Consider, for instance, one recent study that demonstrated that 1 in 5 nursing home patients with severe dementia became significantly less agitated and aggressive after treatment with pain medications.¹ Although it is possible that the neuropsychiatric symptoms improved due to sedation from opioid analgesics, neither activities of daily living nor cognition worsened in the treatment group.¹

Unique issues arise when treating pain in the elderly
About 25% to 50% of community-dwelling older adults experience persistent pain.² As there is no objective measurement for pain, clinicians typically rely on a patient’s self-report. However, elderly patients often underreport their pain because of various misconceptions and/or fears. For example, older patients may view pain as a natural part of aging, may fear being judged negatively, or may expect the clinician to give a low priority to pain compared with their other medical problems.³
 
The presence of cognitive impairment poses additional challenges. While patients with mild to moderate cognitive impairment may be able to communicate their pain symptoms, those with more severe cognitive deficits often cannot. Pain assessment scales are useful tools used to help quantitatively assess pain. Commonly used scales include the Wong-Baker Faces Pain Rating Scale and the numeric pain intensity rating scale. (Some sample pain scales are available at: http://painconsortium.nih.gov/pain_scales/).  
                 
When self-report is not an option
When a patient cannot provide a reliable self-report, changes in the following items may indicate the presence of pain: facial expressions; verbalizations/vocalizations; body movements, such as rubbing a body part, fidgeting, pacing, or rocking; changes in interpersonal interaction; changes in activity patterns, such as not participating in routine activities or refusing to eat; and mental status changes.⁴ Behavioral observation-based assessments are critical for detecting underlying clinical issues in patients with dementia who are unable to communicate symptoms.
 
A validated tool. If you are caring for an older adult with advanced dementia, you may want to consider using the Pain Assessment in Advanced Dementia (PAINAD) scale (http://web.missouri.edu/~proste/tool/cog/painad.pdf).⁵
 
The PAINAD scale is a validated observational tool that screens for 5 pain-related behaviors:

• breathing independent of vocalization (normal, occasional labored, long period of hyperventilation)
• negative vocalization (none, occasional moan, crying)
• facial expression (smiling, frowning, grimacing)
• body language (relaxed, tense, rigid)
• consolability (no need to console, reassured by voice, unable to console).

Each behavior is scored on a scale from 0 to 2. The total score can range from 0 to 10, with a higher number indicating more severe pain. Interrater reliability of the PAINAD scale is high (Pearson r = 0.82-0.97).⁶
 
Clinical observation must be supplemented with the caregiver’s observations, when available. Caregivers are familiar with the patient’s typical behavior and may more easily identify the subtle and not-so-subtle changes that indicate a painful condition. 

When pain is to blame, start low, go slow with meds
Once the presence of pain is identified or considered a cause of behavioral symptoms, appropriate pain management is indicated. (FIGURE)

FIGURE: Algorithm for pain assessment and management

Data source: AGS Panel on Persistent Pain in Older Persons. The management of persistent pain in older persons.
J Am Geriatr Soc. 2002;50(6 suppl): S205-S224.⁴
 
The choice of analgesic treatment depends on the cause and intensity of the pain, as well as such factors as the presence of comorbidities, drug-drug interactions, and cost.
 
The altered physiology of older patients increases the risk of drug toxicity and affects the efficacy of pain medications. Moreover, the presence of reduced cognitive reserve or dementia increases the susceptibility to potential adverse central nervous system or cognitive effects. Safety concerns dictate that initial drug doses should be lower and titration should be slower for all older patients, and this approach is even more important for those with cognitive impairment.
 
The American Geriatrics Society (AGS) guideline for management of persistent pain in older adults may be used to guide pain control.⁷ Acetaminophen may be prescribed as initial pharmacotherapy for the treatment of mild musculoskeletal pain. Nonsteroidal anti-inflammatory drugs (NSAIDs) may be considered, but patients treated with these agents should be monitored for gastrointestinal and renal toxicity.
 
The AGS guideline also encourages nonpharmacologic interventions for pain management, including physical exercise, cognitive-behavioral therapy, patient pain education, acupuncture, transcutaneous nerve stimulation, chiropractic care, heat, cold, massage, relaxation, and distraction techniques.⁴ For patients with severe dementia, sensory-calming and sensory-stimulating activities may also be effective. Reading to the patient or playing soothing music are examples of appropriate sensory-stimulating activities.

Consider opioids for severe pain 
Opioid analgesics are widely accepted for first-line treatment of severe acute pain and chronic pain related to cancer or at the end of life.⁸ Several studies have also shown the efficacy of various opioids in the treatment of persistent pain associated with musculoskeletal conditions, including low back pain and osteoarthritis.⁷
 
Review and revise the regimen. The incidence of addictive behavior and misuse and abuse of opioids are significantly lower in the older-age population.⁷ Nonetheless, it’s essential that you periodically reassess a patient’s response to opioid therapy and revise the pain regimen as indicated, including tapering or substituting medications.
 
Was pain to blame for Mr. M’s increased agitation?
Based on the PAINAD assessment tool, Mr. M’s score during his clinic visit was 8. For breathing independent of vocalization, his score was 1 (occasional labored breathing); for negative vocalization, 2 (loud moaning/groaning); for facial expression, 2 (grimacing); for body language, 1 (fidgeting); and for consolability, 2 (unable to console/distract).
 
A thorough physical examination revealed no physical injury and no cardiothoracic or intra-abdominal concerns. However, given that Mr. M had recently become bedbound with a urinary tract infection, it seemed likely that the prolonged bed rest had exacerbated his chronic low back pain.
 
He was given a trial dose of morphine 5 mg by mouth and became calm but remained alert. His persistent vocalizations subsided.
 
Mr. M was sent home with a prescription for oxycodone 5 mg 3 times daily, along with docusate sodium/sennosides once daily, as needed for opioid-induced constipation. His daughter was instructed to assist in changing Mr. M’s position every 2 hours while he was awake, and physical therapy was prescribed. In addition, Mr. M’s daughter was advised about relaxation and distraction techniques, such as giving him a back rub or playing his favorite music.
 
After being on his oxycodone regimen for 2 weeks, Mr. M revisited the clinic without vocalizations. He responded to verbal commands, and was able to transfer with only minimal assistance. His daughter reported that she sometimes found him asleep in his chair during daylight hours, which was atypical for Mr. M. The oxycodone dose was therefore decreased to 2.5 mg 3 times daily.
 
Take-away points
Mr. M’s case illustrates the need to assess and manage pain effectively as part of the overall treatment of neuropsychiatric symptoms in patients with dementia. Treating pain may reduce the use of antipsychotics and other psychotropic medications in this population.
 
Pain management should be tailored to the type and intensity of the pain as well as the patient’s physiologic status and psychosocial characteristics.


Disclosures
The authors reported no potential conflict of interest relevant to this article.
 

References
1. Husebo BS, Ballard C, Sandvik R, et al. Efficacy of treating pain to reduce behavioural disturbances in residents of nursing homes with dementia: cluster randomised clinical trial. BMJ. 2011; 343:d4065.