Current Diagnosis & Treatment of Pain, 1st Edition

21. Pain in the Elderly

Joshua M. Hauser MD

Essentials of Diagnosis

  • Elderly persons have a less predictable pain reaction to a specific disease or injury than do younger persons.
  • Obtain an accurate medication history by asking patients to bring their medication to the appointment.
  • Unidimensional pain assessment scales (eg, Visual Analogue Scale or Numeric Rating Scales) are particularly helpful in elderly patients, especially those who have cognitive impairment.

General Considerations

Pain in the elderly can be a challenging diagnostic and therapeutic problem for clinicians of many disciplines. The majority of elderly patients in pain will not be seen by geriatricians but rather by internists, family practitioners, oncologists, surgeons, and palliative care physicians. It is critical, therefore, that physicians within these specialties be able to recognize, treat, and consider the special circumstances of the elderly with pain.

  1. Epidemiology

In the Assessing Care of Vulnerable Elders (ACOVE) project, which began in 2000, an expert panel of geriatricians, epidemiologists and health services researchers convened to identify geriatric conditions as optimal targets for quality improvement. After the panel members initially identified 78 common conditions among the elderly, they reduced the list to 35 on the basis of prevalence, impact on health, effectiveness of interventions, and disparities in the quality of care for these conditions among providers. Of these 35 conditions affecting the elderly, pain management ranked fifteenth in importance, ahead of pneumonia, influenza, malnutrition, and osteoarthritis.

The prevalence of pain in the elderly varies according to the setting. Consider whether the elderly person lives in the community or in a long-term care facility, or whether he or she is hospitalized with an acute illness. The studies reviewed below generally include any type of pain (eg, neuropathic or nociceptive) occurring at any anatomic site. The general goal is to give a sense of the overall burden of pain in the elderly, not to characterize specific diagnoses or management strategies.

  1. In the Community—

The prevalence of pain in the elderly who live in the community has been shown to range from 25 to 56%. The sources of pain include back pain (21 to 49.5 %), joint pain, and headaches. A study of community-dwelling elderly using the Minimum Data Set (MDS) for Home Care (3046 patients) found that between 39 of age and 41% of patients older than 65 years reported daily pain. Of those who reported daily pain, 25% received nonopioid therapy for mild pain, 6% received opioids plus nonopioids for moderate pain, and 3% received opioids plus nonopioids for severe pain (see Figure 3-1). Two specific groups within the elderly appeared to confer added vulnerability to pain: Both the oldest old and those with diminished cognitive ability were at increased risk for not receiving analgesia. In a separate study of over 300 elderly patients in California, undertaken as part of the ACOVE project, the prevalence of pain was documented to be 33%. Even more alarming, 40% of these patients reported being screened for pain.

  1. In the Nursing Home Setting—

Since significant numbers of elderly persons live in nursing homes, the problem of pain prevalence and control in this setting is crucial to consider. In an examination of data from the MDS, Teno and colleagues found that almost 15% of nursing home residents had persistent pain on two separate assessments and 41.2% of residents in pain at first assessment were in severe pain 60 to 180 days later. This rate was fairly uniform across the country, varying from 37.7 to 49.5% in different states and suggests that even when pain is recognized, it is frequently not addressed. A more recent study, which also used the MDS to analyze 21,380 nursing home residents, found that 49% of residents had persistent pain. The differences may be due to differing standards for “persistent pain.”


A more recent study of a national sample of nursing home patients showed a relatively small subset of patients (4%) reported “daily pain that was excruciating at some time in the previous week”; nearly half of these patients had a similar report in a follow-up assessment a week later. All of these studies suggest significant problems not only in the recognition of pain but also in its treatment once recognized. Clearly, pain in nursing homes is frequently underrecognized, undertreated, and the source of significant morbidity.

There are multiple quality improvement projects that have attempted to address this issue through both educational and systems interventions. For example, Miller and colleagues compared pain management of nursing home patients receiving hospice with pain management of those not receiving hospice in over 800 nursing homes. They found that hospice patients had more favorable pain treatment by a number of measures:

  1. 15% of hospice residents and 23% of nonhospice residents in daily pain received no analgesics.
  2. A lower proportion of hospice residents (21%) than of nonhospice residents (29%) received analgesics not recommended by the American Medical Directors Association.
  3. Controlling for clinical confounders, hospice residents were twice as likely as nonhospice residents to receive regular treatment for daily pain.

Studies such as these suggest that a concerted clinical or educational effort can impact pain assessment and treatment in the nursing home setting.

However, there is one important caveat in these studies: the source of the data (the MDS) has been shown to significantly underestimate pain in the elderly. In a recent study, Cadogan and colleagues interviewed patients concerning pain and compared their findings with the patients' scores on the MDS, which is filled in by health care providers. They found that the MDS reported prevalence was between 15 and 30% in the nursing homes sampled, but 27 to 47% of residents reported pain when interviewed. Another study compared MDS scores to reports of clinical nursing assistants caring for residents. Similar problems with the MDS were found: Clinical nursing assistants who used a standard measurement of pain for patients with dementia found a prevalence of 48%, compared with a prevalence of 20% found using the MDS.

The implications of this measurement mismatch is that even the relatively high levels of pain among the elderly reported in the MDS may be underestimating the burden of pain in elderly nursing home residents. Therefore, the information obtained from MDS data must be treated as a fairly crude estimate.

  1. In the Hospital Setting—

In a comprehensive study of pain in hospitalized patients in a tertiary care hospital, Whelan and colleagues found an overall prevalence of 49.3% in elderly patients reporting moderate or severe pain. A larger study of elderly hospitalized patients over 80 years of age, a part of the SUPPORT (Study to Understand Patient Preferences for Outcomes and Treatment) study, showed a prevalence of pain ranging from 43% in patients with congestive heart failure to 60% in patients with colon cancer.

These aforementioned studies focused on elderly patients hospitalized on medical services. A parallel group of studies has looked at patients on surgical services and found that postoperative pain has also been shown to be inadequately managed in the elderly. Sauaia and colleagues evaluated patient satisfaction with postoperative pain management in elderly persons who underwent a variety of procedures. They found that 62% reported severe pain and documented inconsistent strategies to address it in eight hospitals. Interestingly, 87% of patients reported being “satisfied” with treatment, suggesting that patients may have low expectations of pain treatment. A recent review of pain management for the elderly in the orthopedic setting suggests that the weight of multiple studies in this area show between 50 and 75% of older adults do not have pain adequately managed. A failure to assess pain, low knowledge about assessment and management, a view that pain is a natural consequence of aging, and concerns about the use of analgesics in patients with cognitive dysfunction or other comorbid illnesses all play a role in the undertreatment of pain in the elderly.

  1. Etiology and Classification

The elderly are vulnerable to both acute and chronic pain. Acute pain is defined as distinct in onset, with a clear cause, and often a limited duration. This type of pain is associated with an injury, an acute illness, or an acute exacerbation of a chronic illness. Very frequently, it has a much clearer underlying cause than chronic pain.

In contrast to acute pain, chronic pain (also called persistent pain) has a duration of at least 3 months, a less predictable pattern of beginning and ending, and a less clear etiology than acute pain.

The etiology of pain in the elderly may be classified by pathophysiology (eg, nociceptive pain caused by inflammation, trauma, or tumor; or neuropathic, caused by diabetic neuropathy, postherpetic neuropathy, or medication-induced neuropathy). It may also be classified by the condition that is causing the pain (eg, cancer, diabetes mellitus, musculoskeletal disorders). In general, the prevalence of chronic diseases is higher in the elderly and the prevalence of musculoskeletal disorders, such as osteoarthritis and chronic low back pain, rises significantly in the elderly (Table 21-1).

Table 21-1. Pain Syndromes in the Elderly.




Polymyalgia rheumatica
Rheumatoid arthritis
Temporal arteritis


Diabetic neuralgia and neuropathy
Trigeminal neuralgia
Postherpetic neuralgia


Pain secondary to cancerous involvement of bowel, liver, bones


Peripheral arterial disease


Pain secondary to fracture or fall

Reproduced, with permission, from Ferrell BA. Acute and Chronic Pain. In: Cassel C K et al (editors). Geriatric Medicine. 4th ed. New York: Springer; 2003.



Cadogan MP et al. A minimum data set prevalence of pain quality indicator: is it accurate and does it reflect differences in care processes? J Gerontol A Biol Sci Med Sci. 2004;59:281. [PMID: 15031314]

Chodosh J et al. The quality of medical care provided to vulnerable older patients with chronic pain. J Am Geriatr Soc. 2004;52:756. [PMID: 15086657]

Desbiens NA et al. Pain and suffering in seriously ill hospitalized patients. J Am Geriatr Soc. 2000;48:S183. [PMID: 10809473]

Ferrell BA. The management of pain in long-term care. Clin J Pain. 2004;20:240. [PMID: 15218408]

Fisher SE et al. Pain assessment and management in cognitively impaired nursing home residents: association of certified nursing assistant pain report, Minimum Data Set pain report, and analgesic medication use. J Am Geriatr Soc. 2002;50:152. [PMID: 12028260]

Karani R et al. Systemic pharmacologic postoperative pain management in the geriatric orthopaedic patient. Clin Orthop Rel Res. 2004;(425):26. [PMID: 15292784]

Landi F et al. Pain management in frail, community-living elderly patients. Arch Intern Med. 2001;161:2721. [PMID: 11732938]

Miller SC et al. Does receipt of hospice care in nursing homes improve the management of pain at the end of life? J Am Geriatr Soc. 2002;50:507. [PMID: 11943048]

Sauaia A et al. Postoperative pain management in elderly patients: correlation between adherence to treatment guidelines and patient satisfaction. J Am Geriatr Soc. 2005;53:274. [PMID: 15673352]

Sloss EM et al. Selecting target conditions for quality of care improvements in vulnerable older adults. J Am Geriatr Soc. 2000;48:363. [PMID: 10798460]

Teno JM et al. Daily pain that was excruciating at some time in the previous week: prevalence, characteristics, and outcomes in nursing home residents. J Am Geriatr Soc. 2004;52:762. [PMID: 15086658]

Teno JM et al. Persistent pain in nursing home residents. JAMA. 2001;285:2081. [PMID: 11311096]

Whelan CT et al. Pain and satisfaction with pain control in hospitalized medical patients: no such thing as low risk. Arch Intern Med. 2004; 164:175. [PMID: 14744841]

Won AB et al. Persistent nonmalignant pain and analgesic prescribing patterns in elderly nursing home residents. J Am Geriatr Soc. 2004;52:867. [PMID: 15161448]

Assessment & Diagnosis

The assessment and diagnosis of pain in the elderly should begin with a thorough history and physical examination. Information from family caregivers may be particularly helpful, especially when the patient has coexisting cognitive deficits. Because of the complexities of patient and family reporting of pain, clinicians should assess patients both in the presence of family members and, if possible, alone. Family members often provide key information about a patient's condition, but the phenomenon of clinicians deferring to family members and ignoring the patient is well documented.

During the assessment of elderly patients with pain, five key issues need to be addressed: changes in pain perception, polypharmacy, functional status, pain measurement scales, and assessment of pain in persons who are cognitively impaired.

  1. Changes in Pain Perception

In general, age-related changes in the elderly alter their perception of pain. Although many studies of neurologic decline have not been specific to pain perception, the neurologic changes that have been documented include decreases in pain receptors in the skin, decreased density and conduction of both myelinated and nonmyelinated neurons, and loss of neurons in the dorsal horn of the spinal cord. Clinically, investigators have observed that myocardial infarction or abdominal illness presents with less pain in the elderly than in younger patients. Furthermore, a study that directly measured pain perception by inserting an intravenous line in elderly and younger persons showed that elderly persons reported significantly less pain in response to this procedure than younger patients. Therefore, evidence suggests that a patient who is elderly has a less predictable pain reaction to a specific disease or injury than a younger patient.

  1. Polypharmacy

Many elderly persons take multiple medications that put them at risk for adverse drug events as well as missing medications and doses. It is not clear that polypharmacy in itself is a cause of pain, but it is clear that the more


medications a patient takes, the higher his or her likelihood is of missing doses. Therefore, a “treatment failure” may be that the person is not taking the medication that the physician thinks he or she is taking. There may have been no change in a patient's underlying pathophysiology to explain the pain; the only change may be in the missing of medication. A thorough medication history is vital in all patients and even more so in this population. One technique that is encouraged is the so-called “brown bag” test in which a patient is asked to bring all of his or her medication to an appointment in a brown bag; examining the contents of the medicine bottles is often more accurate than the list of medications in a patient's medical record that he or she is “supposed to be taking.”

  1. Functional Status

Comprehensive assessment of the elderly is the cornerstone of geriatrics and stems from the realization of the close correlation between functional status and quality of life. The elderly are at higher risk than younger patients for having both diminished and diminishing functional status, and pain has been shown to be correlated with worsened functional status. Its relief, therefore, has the potential to result in improved functional status.

  1. Measurement Scales

In addition to the usual tools of history taking, physical examination, laboratory tests, and radiologic studies, a variety of pain scales are available to assess pain in the elderly. Since pain is, by definition, a subjective symptom, efforts to measure it have been widespread. There are multiple pain assessment scales used in clinical care and research. The sheer number of these is less important than how they have been applied clinically and which are specific to the elderly. Some of these have been validated in multiple populations with multiple types of pain (eg, the McGill Pain Questionnaire) while others are more specific to cancer pain (eg, Wisconsin Brief Pain Inventory, the Memorial Sloan-Kettering Cancer Center Pain Scale). Ferrell and colleagues have developed a validated scale for use in the elderly that consists of 22 yes or no items and two 0 to 10 scales.

In the elderly and especially in the cognitively impaired, it may be more feasible to use a unidimensional scale. These include a Visual Analogue Scale, which is a 10-cm line that is either horizontal or vertical and has clear end points where a patient can indicate the level of pain; scales with pictures, such as the Memorial Sloan-Kettering Pain Assessment Card; or a verbal 0 to 10 scale. For patients with visual difficulties, a spoken 0 to 10 scale may be most effective at assessing their pain. For patients with hearing difficulties, a visual scale is most appropriate.

  1. Pain in the Setting of Cognitive Impairment

Although standardized pain scales are useful in multiple populations of patients in pain, it is in patients with cognitive impairment that they are probably most valuable. In patients with cognitive impairment, assessment is a particular challenge. The challenge is one of accurate measurement and assessment in patients whose ability to interact has been compromised. A recent review of available assessment tools identified 39 instruments used to assess pain in the setting of cognitive impairment. Of 30 that met minimum standards, 18 were self-reported and 12 were staff reported. Of particular note, no instruments met all major tests of validity and reliability. In general, the simpler instruments had more extensive testing.

In one study that compared multiple assessment scales among cognitively impaired patients, Krulewitch and colleagues assessed cognitively impaired patients using a Visual Analogue Scale; a FACES Pain Scale; and the Philadelphia Pain Intensity Scale, a six-item self-report scale. They found that the Philadelphia Pain Intensity Scale was most likely to be filled out by patients and their caregivers. There was, importantly, relatively high correlation between the FACES and Visual Analogue Scale in their study.

The need for scales is compounded by recent findings that there is some evidence that the accuracy of physician assessment decreases as cognitive status worsens. A study that compared the pain assessments of patients' personal geriatricians with those of experts found that although assessments for patients with moderate impairment were accurate, the assessment in the most cognitively impaired group (average Mini-Mental Status Examination score of 1.91 out of a possible 30) was poor.

In severely cognitively impaired patients, scales of pain assessment consist of observed behaviors of the patient. One example that is fairly widely used is the Hurley Discomfort Scale. This scale consists of a trained examiner observing the breathing, vocalizations, facial expressions, and body movements in patients with dementia (Table 21-2). It has been reported to have moderate reliability. Even in cases where the full scale cannot be used, nonverbal cues to pain (eg, furrowing of the brow, moaning, calling out, restlessness, increased agitation) are critical to a thorough assessment. Physicians, nurses, and other health care professionals should routinely note such observations in the assessments. The presence of any of these nonverbal cues should prompt a therapeutic trial of analgesic medications. Family caregivers can also be asked about these signs of pain and also about positions, manipulations, or other actions that they have seen relieve or exacerbate pain.

Another example of a tool used to assess patients with cognitive impairment uses a mnemonic called BODIES. This tool was recently developed by Snow and colleagues


for nursing assistants to help in the assessment of pain in nursing home residents with dementia. This mnemonic consists of the following fields:

  • B-What behaviorsdid you see?
  • O-How oftendid the behaviors occur?
  • D-What was the durationof the behaviors?
  • I-How intensewere the behaviors?
  • E-How effectivewas treatment, if given?
  • S-What made the behaviors startand stop?

Table 21-2. Pain Assessment in Advanced Dementia-PAINAD.




                2                                                               Scorea

Breathing independent of vocalization


Occasional labored breathing
Short period of hyperventilation

Noisy labored breathing
Long period of hyperventilation
Cheyne-Stokes respirations

Negative vocalization


Occasional moan or groan
Low-level speech with a negative or disapproving quality

Repeated troubled calling out
Loud moaning or groaning

Facial expression

Smiling or inexpressive



Body language


Distressed pacing

Fists clenched
Knees pulled up
Pulling or pushing away
Striking out


No need to console

Distracted or reassured by voice or touch

Unable to console, distract, or reassure





aPain score of 4 or greater requires intervention. Pain score of 3 or less requires careful monitoring (screening with vital signs).

Reproduced, with permission, from Warden V et al. Development and psychometric evaluation of the Pain Assessment in Advanced Dementia (PAINAD) scale. J Am Med Dir Assoc. 2003;4:9-15.

Although it uses relatively nonspecific items for care-givers to assess, this highlights important areas for further development of valid scales.

Cohen-Mansfield J et al. Pain in cognitively impaired nursing home residents: how well are physicians diagnosing it? J Am Geriatr Soc. 2000;48:1607. [PMID: 12110063]

Ferrell BA et al. The Geriatric Pain Measure: validity, reliability and factor analysis. J Am Geriatr Soc. 2000;48:1669. [PMID: 11129760]

Gibson SJ et al. Age-related differences in pain perception and report. Clin Geriatr Med. 2001; 17:433. [PMID: 11459714]

Goulding MR. Inappropriate medication prescribing for elderly ambulatory care patients. Arch Intern Med. 2004; 164:305. [PMID: 14769626]

Krulewitch H et al. Assessment of pain in cognitively impaired older adults: a comparison of pain assessment tools and their use by non-professional caregivers. J Am Geriatr Soc. 2000;48:1607. [PMID: 11129750]

Li SF et al. Effect of age on acute pain perception of a standardized stimulus in the emergency department. Ann Emerg Med. 2001;38:644. [PMID: 11719743]

Snow L et al. Pain management in persons with dementia. BODIES mnemonic helps caregivers relay pain-related signs, symptoms to physicians and nursing staff. Geriatrics. 2005;60:22. [PMID: 15877481]

Stolee P et al. Instruments for the assessment of pain in older persons with cognitive impairment. J Am Geriatr Soc. 2005;53:319. [PMID: 15673359]


  1. Analgesic Therapy

The principles of analgesic treatment in the elderly parallel the principles for all patients. A thorough history


and physical examination as well as the appropriate use of laboratory and radiologic testing are critical to identify the cause of pain. In some cases, such as lower back pain, a specific cause will be elusive. In other cases, a specific anatomic cause or injury can be identified. In all cases, it is crucial to define the extent of diagnostic workup by the goals of the patient and his or her family. In the elderly, especially those who may be close to the end of life, issues of goals of care can help determine the level of diagnostic testing that is desired, which means that clinicians may not always be able to discover a clear cause to help guide treatment. For example, for a patient who has dementia and is severely debilitated, contracted, and bedbound, a simple x-ray film of the lumbar spine may be a significant burden-far greater than it might be for a 60-year old. This does not mean that clinicians should hesitate to treat pain until an etiology can be found; quite the contrary, clinicians should feel comfortable treating pain even if a clear etiology is not known.

Both the American Geriatrics Society (AGS), the main professional society for geriatricians, and the American Medical Directors Association, the main professional society for physicians in long-term care, have position statements and clinical guidelines concerning geriatric pain. These guidelines stress a step-wise approach to pain treatment that is consistent with the original pain ladder established by the World Health Organization (see Figure 3-1). This approach is applicable to the setting of acute and chronic pain. It consists of beginning with a nonsteroidal anti-inflammatory drug (NSAID) or acetaminophen and then moving to different strength opioids if the pain is not relieved. Adjuvant medications, especially for neuropathic pain, are efficacious in the elderly. Their side effects should be carefully monitored: tricyclic antidepressants have more significant anticholinergic effects in the elderly than in younger patients and the elderly are at increased risk for glycemic changes and mood changes from corticosteroids. While this does not mean that these medications should be withheld, it does mean that their risk-benefit profile needs careful consideration.

  1. Nonsteroidal anti-inflammatory drugs—

These have been shown to be effective in a variety of conditions, particularly osteoarthritis. NSAIDs may be combined with opioids or used as single agents. Acetaminophen can be given in doses up to 4000 mg/d but should be reduced in patients with concurrent alcohol use or liver dysfunction. NSAIDs, such as ibuprofen, have a higher incidence of bleeding complications in the elderly compared with the nonelderly. Although gastroprotective drugs offer some protection, this is not complete. They also do not change the known renal toxicity of long-term NSAID use in the elderly.

  1. Opioids—

Opioids have been found in repeated trials to be safe and effective for the elderly in both chronic and acute pain. The issue of opioid use is one of slow and careful titration of dosage rather than avoidance. For example, Roth and colleagues found around-the-clock, controlled-release oxycodone therapy to be safe and effective treatment for patients with moderate to severe pain due to osteoarthritis. The most common opioids for use in the elderly include morphine, oxycodone, and hydrocodone. In patients with renal dysfunction, hydromorphone is a preferred opioid. Starting doses are shown in Table 21-3.

  1. Opioids to be avoided—

The AGS makes a specific recommendation against the use of several opioid medications, including methadone (whose variable half-life is especially problematic in the elderly) and tramadol (which can lower seizure threshold). Although not specifically mentioned in the AGS guidelines, the use of meperidine is not recommended in the elderly because of the neuroexcitatory effects of its metabolite normeperidine. In response to adverse events concerning meperidine, some hospitals have taken it off their formulary. In addition, propoxyphene is not recommended in any person because of its poor analgesia, high acetaminophen content and, of particular concern in the elderly, the neurotoxic effects of the metabolite norpropoxyphene.

  1. Nonpharmacologic Approaches

Given the presence of increased vulnerability to side effects, many clinicians in the geriatrics community have called for the integrating of nonpharmacologic approaches to pain. These are most commonly used in conjunction with pharmacologic approaches and can be appropriate both in the setting of acute and chronic pain.

One approach has been to use exercise as an adjunctive pain relief technique. Ettinger and colleagues conducted a randomized trial comparing aerobic exercise and resistance exercise in the treatment of pain in older adults with osteoarthritis. They found that both types of exercises improved pain and disability scores, with pain scores showing more improvement in the resistance exercise model. Iversen and colleagues found improvements in pain and quality of life in a small sample of elderly patients with chronic low back pain who underwent a bicycle exercise program three times a week for 12 weeks. In their clinical practice guidelines, the American Geriatrics Society recommends an exercise prescription for adults with osteoarthritis. Although it is tempting to extrapolate that all elderly persons with pain will benefit from exercise, some caution is warranted. For example, it is likely only those with a functional status to allow appropriate participation will benefit from exercise. There is some suggestion that these benefits from exercise may be isolated to elderly with relatively preserved functional




status. In a small trial among elderly persons with dementia and incontinence, a controlled exercise program failed to show improvement in pain scores.

Table 21-3. Selected Opioid Analgesic Medications for Pain.


Starting dose (oral)




15 mg q4h

Short-intermediate half-life
Older people are more sensitive than younger people to side effects

Titrate to comfort; continuous use for continuous pain; intermittent use for episodic pain; anticipate and prevent side effects


30-60 mg q4-6h

Acetaminophen or NSAIDs limit dose
Constipation is a major issue

Begin bowel program early; do not exceed maximum dose for acetaminophen or NSAIDs.


5-10 mg q3-4h

Toxicity similar to morphine, acetaminophen, or NSAID combinations
Limit maximum dose

Same as above


20-30 mg q3-4h

Toxicity similar to morphine, acetaminophen, or NSAID combinations
Limit maximum dose
Oxycodone is available generically as a single agent

Same as above


2 mg q3-4h

Half-life may be shorter than morphine
Toxicity similar to morphine

Similar to morphine

Sustained-release morphine (MS Contin, Oramorph, Kadian)

MS Contin 15-30 mg q12h
Oramorph 15-30 mg q12h
Kadian 15-30 mg q24h

Morphine sulfate in a wax matrix tablet or sprinkles
MS Contin and Oramorph should not be broken or crushed
Kadian capsules can be opened and sprinkled on food but should not be crushed

Titrate dose slowly because of drug accumulation
Immediate-release opioid analgesic often necessary for breakthrough pain

Sustained-release oxycodone

15-30 mg q12h

Similar to sustained-release morphine

Similar to sustained-release morphine

Transderm fentanyl

25-mcg patch q72h

Reservoir for drug is in the skin, not in the patch
Equivalent dose compared with other opioids is not very predictable (see package insert)
Effective activity may exceed 72 h in older patients

Titrate slowly using immediate-release analgesics for breakthrough pain
Peak effect of first dose may take 18-24 h
Not recommended for opioid-naïve patients

Fentanyl lozenge on an applicator stick

Rub on buccal mucosa until analgesia occurs, then discard

Short half-life
Useful for acute and breakthrough pain when oral route is not possible

Absorbed via buccal mucosa
Not effective orally

Ferrell BA. Acute and Chronic Pain. In:Cassel CK et al (editors). Geriatric Medicine. 4th ed. New York: Springer; 2003.

Another nonpharmacologic approach that has shown some promise is acupuncture. In a recent trial of acupuncture for patients with chronic low back pain, Meng and colleagues demonstrated significant improvements in disability measured by the Roland Disability Questionnaire, which includes pain. A larger study by Berman and colleagues compared true acupuncture with sham acupuncture in 570 patients. Although they had some limitations in follow-up in both groups, they found significant improvements in pain scores among the patients who received true acupuncture compared with the sham group.

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. [PMID: 12067390]

American Geriatrics Society Panel on Exercise and Osteoarthritis. Exercise prescription for older adults with osteoarthritis pain: consensus practice recommendations. A supplement to the AGS Clinical Practice Guidelines on the management of chronic pain in older adults. J Am Geriatr Soc. 2001;49:808. [PMID: 11480416]

Berman BM et al. Effectiveness of acupuncture as adjunctive therapy in osteoarthritis of the knee: a randomized controlled trial. Ann Intern Med.2004;141:901. [PMID: 15611487]

Ettinger WH Jr et al. A randomized trial comparing aerobic exercise and resistance exercise with a health education program in older adults with knee osteoarthritis. The Fitness Arthritis and Seniors Trial (FAST). JAMA. 1997;277:25. [PMID: 8980206]

Iversen MD et al. Enhancing function in older adults with chronic low back pain: a pilot study of endurance training. Arch Phys Med Rehabil. 2003;84:1324. [PMID: 13680569]

Meng CF et al. Acupuncture for chronic low back pain in older patients: a randomized, controlled trial. Rheumatology (Oxford). 2003;42:1508. [PMID: 12890859]

Roth SH et al. Around-the-clock, controlled-release oxycodone therapy for osteoarthritis-related pain: placebo-controlled trial and long-term evaluation. Arch Intern Med. 2000; 160:853. [PMID: 10737286]

Simmons SF et al. Effects of a controlled exercise trial on pain in nursing home residents. Clin J Pain. 2002; 18:380. [PMID: 12441832]

Special Issues

  1. Depression and Pain

Depression has been shown to influence pain in both elderly and nonelderly populations. In a prospective random sample of more than 18,000 persons, which included elderly and nonelderly, chronic pain was shown to be strongly associated with major depression. The message is clear that patients with pain should be evaluated for depression and vice versa. The coexistence of these two diagnoses can make both assessment and treatment of pain more complex for the elderly.

A recent study reviewed the effect of two different depression care interventions on pain reports in a large sample of elderly with arthritis. The findings showed that an intervention consisting of antidepressants and problem-solving psychotherapy not only improved patients' depression but also improved the patients' pain reporting. Unutzer and colleagues examined the prevalence of pain and functional decline in a sample 1801 patients with depression. They found that 79% of patients reported functional impairment from pain in the previous month, and 57% reported a diagnosis of treatment for chronic pain in the previous 3 years. Yet, only 51% reported any analgesic use.

What might be the mechanism by which depression and chronic pain influence each other? In a prospective cohort study, investigators monitored 226 patients with disabling musculoskeletal pain and examined levels of depressive symptoms and self-efficacy. Both depression and a measure of low self-efficacy were correlated with higher pain scores.

  1. Pain and Quality of Life

Depression is a relatively specific clinical diagnosis. Its correlation with pain has naturally led investigators and clinicians to wonder how pain more generally influences quality of life and self-reported health. The intuitive concern that pain always decreases quality of life is often but not always borne out in the literature. In a small study at one center, Kong and colleagues recently studied the impact of pain on health-related quality of life for stroke survivors. Using one of the most widely accepted quality of life measures, the SF-36 (Short Form-36), they found that although pain was common in poststroke patients (42% prevalence), there was no difference in scores (with the exception of the subscale of the SF-36 that concerns pain) between patients with persistent poststroke pain and those without persistent pain.

Mantyselka and colleagues examined the relationship between chronic pain and self-rated health in more than 6500 patients in Finland. Their study included adults 15 to 74 years of age, and found the prevalence of chronic pain increased consistently with age, with 30% of those in the oldest group (70 to 74 years) reporting daily pain and 15% reporting pain several times a week. There was a similar consistent increase in the numbers of patients who reported poor perceived health.

The correlation is not just between self-reported health and pain but to a variety of objective measures of poor health. Leveille and colleagues, for example, have shown that pain is associated with increased falls in the elderly. In a prospective cohort study of over 1000 patients in Baltimore, Maryland, they found that women with pain were 1.66 times more likely to have reported falls. However, they found that those treated for pain were


less likely to have experienced falls. Since falls themselves are associated with significant, morbidity and mortality in the elderly, the role of pain in contributing to their increased incidence is crucial to understand.

Finally, Won and colleagues assessed more than 49,000 nursing home residents and examined the associations between the presence of pain and activities of daily living, mood, and activity involvement. In addition to a high prevalence of daily pain (26%), they found that pain was correlated with decreased ability to perform activities of daily living, low mood, and less involvement in activity.

None of these studies are able to elucidate a mechanism by which pain is related to quality of life, activities of daily living, and depression. However, they do illustrate a clear correlation. The lesson for clinicians is that pain does not occur as an isolated phenomenon but has myriad connections with other parts of patients and families lives.

  1. Family Caregivers

The elderly are more likely to have family members involved in assessing and treating their pain than are younger patients. Therefore, it is vital to understand how family caregivers assess pain among patients. Studies in this area have shown that family caregivers have extensive involvement and confidence in their ability to judge patients' pain yet also find this role distressing.

Focus groups of caregivers and patients suggest that improvements in communication between clinicians and family members, increasing participation of family members by clinicians in care, and addressing fears of using pain medications are key interventions that health systems need to consider. More recently, a group has developed instruments such as the Caregiver Pain Medicine Questionnaire to measure concerns about reporting pain and administering medications. In an initial study of this 22-item instrument, investigators found that although very few caregivers had concern about overall communication, more than 25% had concerns about addiction to pain medications and about their ability as a caregiver to decide how much medication to administer.

When caregiver and patient reports are compared, caregivers have been shown to overestimate pain prevalence in the elderly. Redinbaugh and colleagues explored this phenomenon of noncongruence between patients and their family caregivers. They examined 31 patient-caregiver pairs and found that caregiver knowledge of cancer pain management was not significantly associated with accuracy of ratings. However, the caregivers' experience of pain was significantly associated with the accuracy of their ratings. For example, caregivers who reported their loved one to be in distress secondary to his or her pain more accurately reported the pain, and caregivers who themselves were more distressed more accurately reported the pain.

Kimberlin C et al. Cancer patient and caregiver experiences: communication and pain management issues. J Pain Symptom Manage. 2004;28:566. [PMID: 15589081]

Kong KH et al. Prevalence of chronic pain and its impact on health-related quality of life in stroke survivors. Arch Phys Med Rehabil. 2004;85:35. [PMID: 14970965]

Letizia M et al. Barriers to caregiver administration of pain medication in hospice care. J Pain Symptom Manage. 2004;27:114. [PMID: 15157035]

Leveille SG et al. Musculoskeletal pain and risk for falls in older disabled women living in the community J Am Geriatr Soc. 2002;50:671. [PMID: 11982667]

Lin EH et al; IMPACT investigators. Effect of improving depression care on pain and functional outcomes among older adults with arthritis: a randomized controlled trial. JAMA. 2003;290:2428. [PMID: 14612479]

Mantyselka PT et al. Chronic pain and poor self-rated health. JAMA. 2003;290:2435. [PMID: 14612480]

Ohayson MM et al. Using chronic pain to predict depressive morbidity in the general population. Arch Gen Psychiatry. 2003;60:39. [PMID: 12511171]

Redinbaugh EM et al. Factors associated with the accuracy of family caregiver estimates of pain. J Pain Symptom Manage. 2002;23:31. [PMID: 11779666]

Reid MC et al. The relationship between psychological factors and disabling musculoskeletal pain in community-dwelling older persons. J Am Geriatr Soc.2003;51:1092. [PMID: 12890071]

Shega JW et al. Pain in community-dwelling persons with dementia: frequency, intensity and congruence between patient and caregiver report. J Pain Symptom Manage. 2004;28:585. [PMID: 1558908HD]

Unutzer J et Al. Pharmacotherapy of pain in depressed older adults. J Am Geriatr Soc. 2004;52:1916. [PMID: 15507072]

Won A et al. Correlates and management of nonmalignant pain in the nursing home. SAGE Study Group. Systematic Assessment of Geriatric drug use via Epidemiology. J Am Geriatr Soc. 1999;47:936. [PMID: 10443853]

Yates P et al. Family caregivers' experiences and involvement with cancer pain management. J Palliat Care. 2004;20:287. [PMID: 15690831]

Barriers to Pain Control

Barriers that May explain The overall low level of recognition and treatment of pain in the elderly include deficiency of knowledge, inadequate systems to assess and monitor pain relief, and attitudinal barriers. A study of six community-based and one Veterans Affairs long-term care facilities, explored the barriers among both residents and staff to pain treatment. The study compared residents, nurses, and clinical nursing assistants and found that among residents, attitudes that interfered with adequate pain control included the belief that chronic pain does not change, a fear of addiction, and fear of


dependence. The major attitude that nurses identified was that complaints from residents to staff frequently went unheard. Clinical nursing assistants believed lack of time and complaints being unheard were the major barriers

Other investigators have identified barriers that are clinician based: a failure to assess for pain, inadequate knowledge about management principles, a perception that pain is an inevitable part of growing old, and concerns about the use of pain medications in patients with cognitive difficulties or multiple comorbidities. Many of these barriers are amenable to education. Other barriers, such as the lack of availability of opioids in pharmacies that serve lower income patients, have economic and cultural causes and will require policy, legislative, and economic solutions.

Karani R et al. Systemic pharmacologic postoperative pain management in the geriatric orthopaedic patient. Clin Orthop Rel Res. 2004;(425):26. [PMID: 15292784]

Morrison RS et al. “We Don't Carry That”-failure of pharmacies in predominantly nonwhite neighborhoods to stock opioid analgesics. N Engl J Med. 2000;342: 1023. [PMID: 10749965]

Weiner DK et al. Attitudinal barriers to effective treatment of persistent pain in nursing home residents. J Am Geriatr Soc. 2002;50:2035. [PMID: 12473018]