Many nursing home residents endure prolonged efforts at disease modification, but supportive, symptomatic care is not approached aggressively enough. Residents of nursing homes often believe that pain is to be expected with aging and that complaining may negatively affect their care [3]. Even when pain is identified, it is often not optimally managed [3, 4].

Effective pain management has important implications for improving functional status, quality of life, and quality of care in nursing homes. Comfort and maximum independence are the most important goals for most nursing home residents [5]. Pain management has become more successful in younger patients [6] and in patients with cancer [7], and it can also be improved in nursing home residents [8].

It has been suggested that age-related changes in pain sensation occur within the complex processing of nociperception by the central nervous system and in psychological responses to painful stimuli [9-12]. Elderly persons are known to present with painless myocardial infarctions [13, 14] and intra-abdominal catastrophes [15, 16]. Whether these clinical events result from altered pain reporting or from age-related changes in pain receptors, nerve transmission, or central nervous system processing remains to be seen. Studies using various methods to experimentally induce pain in volunteers have shown mixed results [8, 11, 17]. Moreover, the clinical relevance of these studies is questionable because experimentally induced pain may not be analogous to pain associated with disease. In the final analysis, the widespread belief of many clinicians that aging itself decreases pain sensitivity or increases pain tolerance lacks scientific support [6].

Finally, the Agency for Health Care Policy and Research [6, 18] has stated that frail elderly persons, especially those in nursing homes, have special needs for acute and postoperative pain management as well as for pain management in chronic cancer. The Agency's guidelines have highlighted the fact that most pain research has systematically excluded elderly persons and that relatively little attention has been paid to pain management in geriatric medical and nursing textbooks and medical and nursing school curricula. Given the growing number of elderly persons requiring nursing home care, more research is clearly needed [6-18].

Epidemiology

In several selected nursing home populations, the overall prevalence of pain has been reported to be as high as 45% to 80% [2-419, 20]. Most pain in the nursing home is related to arthritis and musculoskeletal problems, including degenerative arthritis, lower-back disorders, and crystal-induced arthropathies [3, 19-21]. The neuropathic pain syndromes, including diabetic neuropathy and herpes zoster, are also common [3]. Cancer may be less common, but it is a source of severe pain in this setting [3, 19]. Other common pain problems include leg cramps, headaches, and claudication [3].

Complications of unrelieved pain are also widespread in the nursing home. Depression, decreased socialization, sleep disturbance, impaired ambulation, and increased health care use and costs have all been associated with the presence of pain in elderly patients and patients in nursing homes [3, 22, 23]. Deconditioning, gait disturbances, falls, slow rehabilitation, polypharmacy, cognitive dysfunction, and malnutrition are among the many geriatric conditions potentially worsened by the presence of pain [8]. These facts highlight the need to recognize pain as a complication with substantial potential to disrupt treatment goals and overall quality of life for nursing home residents.

Assessment of Pain in the Nursing Home

Typical nursing home residents present many challenges to an adequate assessment of pain. Multiple concurrent illnesses, under-reporting of symptoms, and a high prevalence of cognitive impairment make pain evaluation much more difficult in this compared with other adult populations [24]. With no objective biological markers for pain, physicians must rely on patients' self-reports. Accurate pain assessment begins when the physician believes patients and takes their complaints of pain seriously.

Some elderly patients do not complain, despite severe pain that affects their mood and functional status. They see others around them that are worse off than themselves, and they expect pain to be associated with aging [3]. They may fear the meaning of pain. For example, among patients with cancer, pain is a metaphor for advancing disease and approaching death [25]. Elderly patients may also think that pain cannot be relieved. Nursing home residents often try to avoid being labeled a "complainer" because of the negative effects that this might have on their overall care [3]. In this setting, functional decline, mood disturbances, and changes in behavior may be important in the evaluation of pain. The importance of nursing staff and family caregivers as sources of information about elderly patients cannot be overemphasized.

It has been estimated that more than 50% of nursing home residents have substantial cognitive impairment or dementia [19, 26]. Pain-assessment instruments such as visual-analog, word-descriptor, and numerical scales have been only partially validated in elderly populations [19, 27]. A high prevalence of visual, hearing, motor, and cognitive impairments may impede the direct adaptation of many of these instruments in nursing home populations. Behavioral scales based on facial grimace and posturing have been investigated in infants in postoperative recovery rooms [28], but they have not been well established for clinical use in patients with Alzheimer disease or other dementias common in the nursing home. Recently, Hurley and colleagues [29] presented a new multidimensional discomfort scale for use in noncommunicative patients with Alzheimer disease. Although this scale may eventually prove useful, it requires substantial training to administer.

Despite the difficulties involved in quantifying pain in patients with delirium or dementia, Parmelee and coworkers [30], in a study of more than 750 nursing home residents, found no evidence that cognitive impairment "masked" pain complaints. Data indicated that although cognitively impaired patients tended to under-report pain, the self-reports of such patients were no less valid than those of cognitively intact patients [30]. We recently reported our experience with five unidimensional pain-intensity scales that had previously been established for younger patients [19]. These included a visual-analog scale, three word-descriptor scales, and a graphic pictorial scale. Our results suggested that only one third of the patients who had pain could complete all of the scales but that 83% could complete at least one scale. These observations indicate that pain reports, even those of persons with moderate to severe cognitive impairment, are usually valid and reliable. Patients can report pain using existing scales that are tailored to individual patients' disabilities and preferences when questions are framed at the moment (for example, How much pain are you having right now?). On the other hand, whether these patients can accurately recall, integrate, and report pain over time (for example, in answers to questions such as, How much pain have you had over the last week?) has not been fully evaluated and needs further study.

Many nursing home residents with chronic pain will have substantial anxiety or depressive symptoms at some time and may benefit from psychological or psychiatric intervention [22]. However, care must be taken to avoid attributing pain entirely to depression or psychogenic causes. Psychogenic pain (pain for which no cause other than a psychological origin can be identified) is unusual in this setting [3, 19, 22].

Nursing home residents often have multiple potential sources of pain [3, 19, 22], and care must be taken to avoid attributing acute pain to preexisting illness. Exacerbating this problem is the fact that chronic pain is usually not constant; both the character and the intensity of chronic pain fluctuate with time. Frail nursing home patients are particularly prone to falls and occult traumas that can be easily overlooked. Acute gout and calcium pyrophosphate can be mistaken for osteoarthritis [21]. Therefore, new pain complaints or changes in the character of old pain complaints require careful evaluation in this population.

Physical examination in the evaluation of nursing home residents with pain should concentrate on the musculoskeletal and nervous systems because diagnoses in these systems are so common. It is important to palpate for trigger points and inflammation. Trigger points resulting from tendonitis, muscle strain, or nerve irritation may benefit from local injections or specific physical therapy. Maneuvers that reproduce the pain, such as straight-leg raising and joint motion, are often useful in both diagnosis and functional assessment. Neurologic examination should include attention to signs of autonomic, sensory, and motor deficits suggestive of neuropathic conditions and nerve injuries that may require specific treatments [24].

The evaluation of functional status is important as an outcome measure for pain management so that mobility and independence are maximized. Functional assessment may include information from the history and physical examination as well as the use of several available functional-assessment scales. Assessment of activities of daily living used in routine geriatric evaluation may be useful [31]. For ambulatory patients, advanced activities of daily living and "elective" activities, such as ambulation, psychosocial function, and quality of life, may correlate clinically with the presence and severity of pain [8].

When risky or highly technical procedures are required to evaluate and treat severe pain problems, brief hospitalizations may be appropriate, because some nursing homes are not equipped to manage technical procedures or their acute complications [32]. Most nursing homes do not have on-site laboratories, diagnostic facilities, or pharmacies; this frequently tempts physicians to send patients with acute distress to emergency rooms or distant facilities. It should be remembered that transportation to other facilities often results in missed meals and medications, making the trips physically exhausting, emotionally disruptive, and fraught with the potential for iatrogenic illness for most nursing home residents.

Pain Management

Physicians have paid little attention to pain management in the nursing home; this is exemplified by the dearth of data-based information on this topic in medical textbooks [8]. Fewer than 1% of the more than 4000 papers published on pain each year focus on pain in elderly persons [12], and almost no English-language studies describe the effectiveness of available pain-management strategies in the nursing home population. Pharmacologic pain research has largely been limited to single-dose studies or trials in young or middle-aged adults. Most reports from rehabilitation programs or specialized pain centers are flawed by biased referral sources for elderly patients that include only patients with easy access to transportation and other support [6, 33, 34].

Most pain problems encountered in the nursing home can be managed with the careful use of medications and effective nonpharmacologic pain-management strategies. A combination of pharmacologic and nonpharmacologic techniques results in more effective pain control and less reliance on medications that have major side effects in elderly persons [35]. Nursing home residents may benefit most from physicians, nurses, and restorative care personnel who use an interdisciplinary approach to these complex problems.

Prescribing Analgesic Drugs

The analgesic drugs of choice for nursing home residents are those with the lowest side-effect profiles. Adverse effects of drugs are more common in elderly persons and nursing home residents, probably because these patients often have multiple medical problems and require multiple medications [36]. Therefore, several things should be considered when prescribing analgesic drugs for nursing home residents.

Acetaminophen is the analgesic most often prescribed for elderly nursing home patients [3]. Although it is safe in most cases, a recent case–control report suggested that this drug causes a cumulative dose-dependent increase in the risk for end-stage renal disease [37]. The investigators reported that, when persons who received less than 1000 acetaminophen pills in a lifetime were used as a reference, the odds ratio for developing end-stage renal disease was 2.0 for those who received 1000 to 4999 pills in a lifetime and 2.4 for those who received 5000 or more pills in a lifetime. Although this study can be criticized for its case–control design and for the investigators' reliance on self-reported past medication use, the risk for end-stage renal disease among long-term users of acetaminophen deserves consideration. Alternatives to acetaminophen (such as salicylates and pyrazolones) may induce less end-stage renal disease, but their own side-effect profiles for gastrointestinal bleeding, platelet dysfunction, and other abnormalities must be considered.

Nonsteroidal anti-inflammatory drugs often work well for nursing home residents, whether given alone or in combination with opioid analgesics for metastatic bone pain and inflammatory conditions. However, these drugs have been associated with various adverse effects, including peptic ulcer disease, renal insufficiency, and bleeding diathesis [37-39], that may be more common in typical nursing home patients. Among frail elderly persons, these drugs have occasionally been reported to cause constipation, cognitive impairment, and headaches [38].

A recent review [40] has pointed out that older persons have generally been omitted from clinical trials of nonsteroidal anti-inflammatory drugs. Between 1987 and 1990, 83 randomized controlled trials including more than 9600 participants were found to have no patients more than 85 years of age. Only 2.3% of participants were more than 65 years of age [39]. This is disturbing in light of the particularly high incidence of peptic ulcer disease and upper gastrointestinal bleeding in older persons. Griffin and colleagues [41] estimated that the relative risk for peptic ulcer disease among elderly persons who used nonsteroidal anti-inflammatory drugs other than aspirin was 4.1 (95% CI, 3.5 to 4.7) compared with persons who did not use them. These investigators showed that the relative risk increased with dose from 2.8 for the lowest dose to 8.0 for the highest dose.

Finally, the analgesic activity of nonsteroidal anti-inflammatory drugs is limited by a low ceiling effect. An agent that has a ceiling effect has a level beyond which increasing the dose of the agent does not further increase analgesia. Bradley and colleagues [42] showed that acetaminophen (4000 mg/d) resulted in analgesia similar to that of ibuprofen, whether administered as an analgesic dose (1200 mg/d) or as an anti-inflammatory dose (2400 mg/d) to patients with chronic osteoarthritis of the knee. Thus, acetaminophen may be the preferred choice for patients without substantial inflammation because of its lower side-effect profile.

Opiate drugs, such as morphine, have no ceiling and have been shown to relieve all types of pain [43]. Short-term studies [44, 45] have shown that elderly patients are more sensitive than younger patients to the pain-relieving properties of these drugs. Advanced age is associated with a prolonged serum half-life for most opiate drugs [46]. Thus, elderly patients may achieve pain relief from doses of opiate drugs that are smaller than those required by younger patients.

The potential of opiate drugs to cause cognitive disturbances, respiratory depression, and constipation is increased in typical nursing home residents. These drugs may also produce paradoxical excitement and agitation. Morphine remains the standard with which other opiate drugs should be compared in elderly persons because its effects are the best understood and the most predictable. Thus, morphine is the opiate of choice for severe pain in most nursing home residents [6, 18]. In the nursing home setting, issues of drug dependency and drug tolerance are usually irrelevant [47]. This is not to suggest that morphine and other opiates can be used indiscriminately; it means only that dependency and other side effects do not justify withholding effective pain relief.

Tolerance to some side effects of opiates has the beneficial effect of reducing the risk for respiratory depression and drowsiness. For this reason, these drugs should be administered on a continuous basis (as opposed to prore nata or "as needed") whenever possible. Regular dosing results in reduced overall drug consumption, continuous analgesia, and tolerance to drowsiness and respiratory depression [35]. On the other hand, some side effects of opiates, such as constipation and possibly nausea, do not diminish with time and make overall pain management more difficult. In nursing home patients, it is important to begin bowel regimens early, when opiate therapy is first started. Increased fluids, bulk agents, lubricating agents, and bowel stimulants may be required [8]. Although antiemetic drugs such as antihistamines and phenothiazines have been mainstays in the prevention of opiate-induced nausea, no clinical trials on this point have been done in elderly or nursing home populations. It is important to remember that patients in these populations are especially sensitive to the anticholinergic side effects of many antiemetic drugs, including bowel or bladder dysfunction, delirium, and movement disorders. Thus, antiemetic drugs should be chosen with an eye to which ones have the lowest side-effect profiles.

Some opiates require special attention when used in the nursing home setting. Propoxyphene is a controversial drug that is probably overprescribed in elderly persons. Reports suggest that its efficacy is no better than that of aspirin or acetaminophen, and it has substantial potential for dependency and renal injury [48]. Pentacozine is an opiate that should be avoided because it frequently causes delirium and agitation in elderly persons. This effect seems to be related to the drug's mixture of agonist and antagonist opiate-receptor activity [49]. Meperidine is also particularly hazardous in the elderly; its unique toxicity has been seen in patients with renal impairment and in those receiving antidepressants of the monoamine oxidase-inhibitor class. The active metabolite normeperidine is particularly prone to accumulation and is often associated with delirium and seizure activity [50]. Methadone should be used with caution because of its propensity to accumulate. More importantly, it may be a poor choice because its analgesic effect may be short in comparison with its serum half-life [51]; this increases the potential for accumulation or overdosage in elderly persons. Finally, transdermal fentanyl citrate is an extremely potent drug (perhaps 50 times as potent as morphine) with a potential for complications [52]. The transdermal delivery system for this drug forms a tissue reservoir that results in a serum half-life of 36 hours [53]. Because of the drug's extreme potency and the potential for overdosage, transdermal fentanyl citrate should not be used in "opiate-naive" elderly patients or in those unaccustomed to the respiratory depression caused by opiates.

Some adjuvant drugs may be helpful in recalcitrant pain problems. It is important to remember that these drugs are usually only partially successful and that they work best when used with other analgesic drug and nondrug pain strategies [54]. Various antidepressant drugs have been shown to alleviate the pain of diabetic and other neuropathies; it has been suggested that this occurs because pain relief is mediated by the prolonged synaptic activity of norepinephrine and serotonin, which inhibits neurons in pain transmission. Recent reports [55, 56] have suggested that drugs with more selective norepinephrine inhibition (such as amitriptyline and imipramine) may be more effective than drugs that have mixed serotonergic inhibition (such as desipramine) or those that have more selective serotonergic inhibition (such as paroxetine and fluoxetine) [55, 56]. One recent study [56] found desipramine to be as effective as amitriptyline in the management of diabetic neuralgia; thus, desipramine is an alternative for patients unable to tolerate the high side-effect profile of amitriptyline. The same study found fluoxetine to be no better than placebo.

Some neuropathic pain syndromes, such as trigeminal neuralgia, have been found to respond to antiepileptic drugs, including carbamazepine, valproate, and clonazepam [54, 57]. Intravenously administered local anesthetics, such as lidocaine and procaine, may also ameliorate the neuropathic pain syndromes independently of conduction blockade [57-59]. Randomized trials have found oral tocainide to be effective in trigeminal neuralgia [58] and mexiletine to be effective in diabetic neuralgia [59]. Each of these drugs has a substantial risk for toxicity in nursing home residents and should be reserved for patients with severe pain in whom other treatments have failed.

Capsaicin is applied topically and has been shown to deplete free nerve endings of substance P by blocking its re-uptake. It may be useful as an anesthetic for herpes zoster [60], diabetic neuropathy [61], and postoperative neuropathies [62]. Although now available without a prescription, its overall efficacy for arthritis and other painful syndromes remains controversial. It normally causes a burning sensation that may be intolerable to some patients and that has led to speculation that a gate-control mechanism may contribute to its action.

Finally, tramadol, a new analgesic recently released in the United States (Ultram, McNeil Consumer Products Company, Fort Washington, Pennsylvania) may be helpful to some nursing home residents. This non-narcotic analgesic may have a potency similar to that of codeine or oxycodone. It has some opiate-receptor activity and has been reported to cause drowsiness, nausea, and—rarely—respiratory depression. Tramadol has been available for some time in Europe, but long-term studies and experience in older populations have been only partially explored [63, 64].

Physicians who care for nursing home residents must help to establish a plan of care that is reasonable given the resources and skills available in the nursing home setting. Medication regimens can often be simplified. Long-acting analgesics should be used to provide a longer duration of comfort for patients and fewer doses for nurses to administer. Pain should be prevented by routine analgesia, and pro re nata medications should be avoided if possible. Short-acting analgesics should be prescribed for breakthrough pain or for pain associated with physical therapy, bathing, or other potentially painful activities. Treatments should be simplified so that nighttime monitoring requirements are minimized. It is important to remember that nursing homes usually have limited pharmacy resources that are not available on a 24-hour basis. Contingency plans for pain management must be anticipated so that delays do not occur during medication changes or dosage adjustments. Finally, state regulatory requirements for multiple-copy prescriptions may be a substantial barrier to effective pain management in this setting; careful planning is the only solution [65].

Nonpharmacologic Pain-Management Strategies

Many nonpharmacologic pain-management strategies are effective, especially when used in combination with drug strategies. Education programs for patients have been shown to significantly improve pain management and quality of life [66, 67]. The benefits of physical methods, including heat, cold, and massage, should not be underestimated. These methods are effective for many patients [67]; they relax tense muscles and soothe many problems. Some of these techniques can be applied by the patient, providing a sense of control over symptoms and treatment, although precautions should be taken to avoid injury from the use of heat or ice by patients with cognitive impairment. Physical therapy directed at stretching and strengthening specific muscles and joints and maintenance exercise programs are available in most nursing homes; these are useful in improving muscle strength, reducing muscle spasm, and enhancing functional activity. Consultation and treatment by skilled therapists (available in many nursing homes) are appropriate for safe and effective rehabilitation from many painful conditions.

Transcutaneous nerve stimulation has been used successfully in various chronic pain conditions in older patients [68]. Painful diabetic neuropathies, shoulder pain or bursitis, and fractured ribs have been shown to respond to transcutaneous nerve stimulation therapy. Although this therapy has relieved some patients for years, its effectiveness usually diminishes with time, and strong placebo effects have been associated with its use [69]. The appropriate placement of electrodes and current adjustment are important to the success of this therapy. This involves meticulous searching, with the help of a trained physical therapist, for the best settings for an individual patient's optimum comfort. Care must also be taken to avoid skin irritation from the electrodes.

Some psychological maneuvers may be effective in controlling pain. Biofeedback, relaxation, and hypnosis may help some patients. These methods usually require high levels of cognitive function and may not be suitable for patients with substantial cognitive impairment. A trained psychologist or therapist should be consulted about these techniques.

Finally, various activities provided in the nursing home may be effective in decreasing the perception of pain. Many patients find comfort in prayer, meditation, or music. Activities, exercise, and recreation should be encouraged insofar as they can be tolerated. Inactivity and immobility may contribute extensively to depression and worsening of pain.

"High-Tech" Pain-Management Strategies

Recent developments in pain management have focused on various "high-tech" drug-delivery systems for the management of pain; these systems are being used in some nursing homes. With appropriate supervision and patient education, many techniques have become feasible for selected nursing home patients [70, 71]. A randomized trial [72] found that patient-controlled analgesia using morphine infusions is safe and effective for postoperative pain management among nondemented, frail elderly men. However, parenteral morphine infusions for chronic cancer pain are expensive and may cost several thousands of dollars a month to maintain [73].

These procedures have been effective in selected cases, but more work needs to be done to define expanded roles for these technologies in nursing home residents with pain. Because of cost and potential side effects, it is usual to consider "high-tech" strategies only after oral medications have failed or after all other treatments have been tried. Further study is needed to determine whether these technologies have risk-to-benefit ratios sufficient to justify their routine use for nonmalignant pain or the less intense pain syndromes. Although most of these techniques are expensive, they are often partially reimbursable by Medicare and other insurers. These issues have raised ethical questions about the application of "high-tech" pain-management technologies in patients who might be equally well managed with oral medications that, unfortunately, are not reimbursable [74, 75].

Summary and Future Directions

Pain is a common problem that has tremendous potential to influence the physical function and quality of life of nursing home residents. It is unfortunate that so little geriatric research and education has focused on this important topic. Pain management in nursing homes can be improved, particularly through the careful use of appropriate pharmacologic and nonpharmacologic pain-management strategies. The incorporation of pain management into geriatric nursing and medical education at all levels will produce long-term-care professionals who are skilled in pain management and in comforting persons with chronic pain.

Substantial research is still needed to further our understanding of pain and its management among elderly persons. Valid and reliable pain measures, such as pain-intensity scales, functional scales, and behavioral observations need to be established for cognitively impaired persons. New drugs with milder side-effect profiles are urgently needed. Nondrug pain-management strategies, such as exercise and other physical methods, should be investigated. Indications for "high-tech" pain-management strategies, such as morphine pumps and chronic spinal infusions, need to be clarified in this population. And, finally, comparative studies of the long-term outcomes of pain-management strategies need to be done.

As the need for nursing home care continues, many residents will require effective pain management to maintain their dignity and quality of life. It is our obligation to do everything possible to provide comfort and effective pain management for these persons during their remaining years.