Chapter outline

Patients Who Are Critically Ill

Some critically ill patients are mistakenly believed to be unconscious or pain-free, but pain in critically ill patients has been well documented for some time. For example, patients with endotracheal tubes and those who have received neuromuscular blocking agents such as pancuronium (which does not alter sensitivity to pain) may be fairly alert, and some are fully capable of feeling pain and are able to self-report pain if given the appropriate opportunity. In a sample of 24 patients in an intensive care unit (ICU), interviews after transfer from the ICU revealed that all but 1 patient recalled their ICU stays, and 63% recalled moderate to severe pain (Puntillo, 1990). The patients described pain caused by surgical incisions, movement, coughing, endotracheal suctioning, and chest tube removal. Those who could not talk (80% had endotracheal tubes) described numerous behaviors they used in attempts to tell staff they were in pain, such as signaling with their eyes and moving their legs up and down. These patients are actually capable of giving self-reports of pain. Some can use writing materials and can be taught to use a pain rating scale. Some can use the call button, which should be placed within easy reach. Others can be asked questions and be taught to signal with their eyes that pain is present. For example, the clinician can establish with the patient that when asked about pain, squeezing the eyes once means yes and twice means no.

Clinicians are prone to overlook pain at rest as well as pain resulting from common procedures such as turning and suctioning. In a study of 30 critically ill, traumatically injured patients, pain was measured at rest in a supine position and after being turned onto their sides (Stanik-Hutt, Soeken, Belcher, 2001). On a visual analog scale (VAS) (0 to 100 mm), mean at-rest scores were 34.5, which indicates mild pain bordering on moderate pain. Immediately after the turns, the mean scores were 48.1 on the VAS, indicating moderate pain. Terms commonly used to describe pain at rest included throbbing, sharp, sore, hurting, and annoying. After turning, the pain was often described with the same words used for pain at rest (except for hurting) plus pressing, pulling, tender, tiring, and miserable. Of these patients, 13 were not included in the study because they refused to be turned, giving as the reason severe pain at rest or anticipated pain with turning. Their pain at rest was 47.1 on the VAS. Based on their review of the literature, the researchers concluded that pain intensity at rest in this study was in the same range as that in other studies of critically ill patients.

Endotracheal suctioning also causes considerable pain. In a study of 45 adults having cardiovascular (CV) surgery, patients rated their pain during endotracheal suctioning on a 0-to-10 scale (Puntillo, 1994). The mean pain intensity score was 4.5, and more than one third of the patients reported a pain intensity of 7 or greater, indicating severe pain. Patients described the pain as tiring-exhausting, stabbing, tender, sharp, and heavy.

The results of the Thunder Project II, a multisite study, provide extensive information about procedural pain in acute and critically ill patients (Puntillo, White, Morris, et al., 2001). In a total of 153 clinical sites, information was collected from 6201 patients, 5957 of whom were 18 years or older. Patients were alert and able to answer questions. Data were obtained about pain related to six procedures: turning, wound drain removal, tracheal suctioning, femoral catheter removal, placement of a central venous catheter, and nonburn-wound dressing change. Adults rated their procedural pain intensity as 2.65 to 4.93 on a 0-to-10 numerical rating scale (NRS). The most painful and distressing procedure for adults was turning; the mean pain intensity was 4.93 and the mean distress score was 3.47 on a scale of 0 to 10. After turning, the most painful procedures in adults, in descending order, with mean pain intensity in parentheses, were wound drain removal (4.67); wound care (4.42); tracheal suctioning (3.94); central line placement (2.72); and femoral sheath removal (2.65). Pain at rest was usually described as aching, and procedural pain was often described as sharp.

Thus, critically ill patients experience mild to moderate pain throughout the day. At rest, the pain is commonly mild, but turning, which usually occurs at least every 2 hours, is associated with moderate pain and, compared to other painful procedures, is the most distressing.

In a follow-up to the study just described, Puntillo, Wild, Morris, and others (2002) examined analgesic and local anesthetic administration in the 5957 adults they assessed prior to and during the six procedures. Undertreatment of procedural pain was apparent. More than 63% of the patients received no analgesics before or during the procedures, and less than 20% received opioids. The patients most likely to receive opioids were those undergoing femoral sheath removal (28.6%), and those least likely to were undergoing tracheal suctioning (3.7%). A total of 18.4% of patients received local anesthetics, and 89.5% of them were undergoing placement of a central venous catheter. Part of this undertreatment may be the result of failure to assess pain or anticipate pain. All of these patients were alert, and pain could have been assessed at least during the procedure or prior to the procedure for those who had already experienced the procedure, such as turning or tracheal suctioning.

Further analysis of data available from the previously mentioned study of 5957 critically ill adults revealed information about behavioral responses associated with the six procedures being studied (Puntillo, Morris, Thompson, et al., 2004). All the patients in this study were alert and able to self-report, but many critical care patients are unable to provide self-reports of pain because of intubation, motor impairments, cultural or language barriers, or altered levels of consciousness. Behavioral responses to pain may then become one of the most valuable contributions to pain assessment. A 30-item behavioral tool was used to observe patients’ behaviors before and during the procedures. Significantly more behaviors were exhibited by patients with procedural pain than by those without that pain. Behaviors specific to procedural pain were grimacing, rigidity, wincing, shutting of eyes, verbalizations, moaning, and clenching of fists.

The FLACC (Face, Legs, Activity, Cry, Consolability) Behavioral Scale has recently been studied in 29 critically ill adults and children who could not self-report pain (Voepel-Lewis, Zanotti, Dammeyer, 2010). This is the first study to provide support of the FLACC in this population, but further study is warranted.

Several behavioral scales for use with adult patients in critical care and unable to self-report are being developed, and the psychometric properties of six of these scales have been compared (Li, Puntillo, Miaskowski, 2008). They are listed here with references that include the actual tool:

Review of these assessment tools revealed that none had undergone vigorous validation or had been accepted as standardized measures, but two showed good evidence of validity and reliability—the BPS and the CPOT (Li, Puntillo, Miaskowski, 2008). Both are described here, and both are easy to use. However, we feel that the CPOT is a slightly better choice for clinical practice because it includes four behavioral domains instead of the three that are in the BPS, and because the CPOT divides the domain of patients on ventilators into two groups, patients who are intubated and patients who are not; whereas the BPS addresses only ventilated patients. Therefore, the actual tool for the CPOT is included (Form 4-1).

The CPOT evaluates four behavioral domains: facial expressions, movements, muscle tension, and ventilator compliance (Gelinas, Fillion, Puntillo, et al., 2006) . Each domain has three behaviors, which are scored from 0 to 2. The critical review of assessment tools used with seriously ill patients points out that a weakness of the CPOT is that generally low scores were reported in all patients during painful procedures (Li, Puntillo, Miaskowski, 2008). This should be kept in mind when using the tool, and clinicians should remember that pain behaviors are merely indicators of pain and that the number of behaviors do not equate with pain intensity (Pasero, McCaffery, 2005).

The sensitivity and specificity of the CPOT was studied in a sample of 105 adults in an ICU following cardiac surgery (Gelinas, Harel, Fillion, et al., 2009). Pain was evaluated during the painful procedure of turning. Sensitivity was high during turning, meaning that the tool could detect when pain was not present. The specificity of the CPOT was also high; that is, it correctly identified the patients who did have pain. This reduces the likelihood of administering an analgesic to a patient who does not have pain.

The BPS consists of three behavioral domains: facial expression, movements of upper limbs, and compliance with ventilation. Each domain contains four behaviors, each rated from 1 to 4 (Payen, Bru, Bosson, et al., 2001). In the critical review of assessment tools used with critically ill patients, the point is made that the observation of no body movement is equated with a pain-free state in the BPS (Li, Puntillo, Miaskowski, 2008). However, no body movement may simply reflect sedation, restraints, or other restrictions of movement. Another criticism concerns the numbering system, which starts with 1, meaning no pain behavior. It is suggested that starting with 0 more readily reflects no pain behaviors.

The American Society for Pain Management Nursing (ASPMN) published a position statement on pain assessment in the nonverbal patient (available at http://www.aspmn.org/Organization/documents/NonverbalJournalFINAL.pdf) that includes a section about intubated and unconscious patients (Herr, Coyne, Key, et al., 2006). Recommendations for assessment utilize the template for the hierarchy of pain measures, as presented previously (see Box 3-9, p. 123):

Physiologic indicators of pain such as changes in blood pressure are omitted from the hierarchy of pain assessment techniques. Their limited usefulness in assessing pain is discussed in Chapter 2, pp. 27-28. They are occasionally helpful but, especially in critically ill patients, these measures are often influenced by factors other than pain, such as medications, mechanical ventilation, change in level of consciousness, and patients’ illnesses or surgeries. In critically ill patients, behavioral indicators have been found to provide more valid information for pain assessment than physiologic indicators (Arbour, Gelinas, 2009; Gelinas, Arbour, 2009; Gelinas, Johnston, 2007).

For example, a study of 254 critically ill patients in intensive care units who were mechanically ventilated (144 conscious and 114 unconscious) examined potential behavioral and physiologic indicators of pain and their association with self-reports of pain (Gelinas, Arbour, 2009). These patients were observed at rest, during a nociceptive procedure (such as endotracheal suctioning), and 20 minutes postprocedure. Behaviors were measured with the CPOT, and physiologic indicators were obtained from the available monitoring. Patients able to self-report pain were asked if they had pain or not immediately after being scored with the CPOT. This prevented bias in the raters because the CPOT was used before requesting a self-report. Based on the patient’s self-report, only the CPOT score, not changes in vital signs, could predict the presence or absence of pain; consequently, only behavioral observation was recommended for pain assessment in unconscious patients. The researchers noted that changes in vital signs could be used as a cue to begin further assessment of pain.

Probably because the physiologic changes are so readily able to be observed in the ICU, critical care nurses are tempted to use them as indicators of pain, especially in patients who are unable to self-report. However, as mentioned above, many patients actually are able to self-report pain if the right approaches are used, such as eye blinking and the CPOT. Research strongly recommends use of behaviors to assess pain in unconscious patients and warns that vital signs should be used with caution (Gelinas, Arbour, 2009).

Patients Who Are Unconscious

Some critically ill patients appear to be unconscious due to injury such as brain trauma or disease, sedating medications, or neuromuscular blocking agents. But unconscious patients also exist outside the critical care setting. For example, patients in vegetative states may be found in long-term care facilities. Terminally ill patients may become unconscious as death approaches.

If unconsciousness is defined as no awareness of self or environment, how is it possible that an unconscious patient could feel pain and suffer from it? The problem is that unconsciousness is not easily determined, and there are varying levels of unconsciousness. A minimally conscious state is described as one in which there is some evidence of awareness of self and environment, and a persistent vegetative state is defined as wakefulness without awareness of self or environment (Boly, Faymonville, Schnakers, et al., 2008). Anecdotal evidence and some research strongly suggest that quite a few patients who appear to be unconscious and unresponsive to painful stimuli actually feel and recall pain. Thus, clinicians should assume that the unconscious patient may feel pain and should provide analgesia if anything known to be painful is present.

In a study of 100 critically ill patients whose records indicated that they were unconscious, interviews with the patients revealed that only about one quarter of them were actually unaware of themselves and their surroundings (Lawrence, 1995). The remaining three quarters had some awareness of themselves and their environments, and about 25% were able to hear and to feel pain.

During general anesthesia, some patients experience periods of awareness that they recall postoperatively. In a study of 26 patients who had experienced awareness during general anesthesia, hearing sounds and feeling paralyzed were reported by 89% and 85%, respectively (Moerman, Bonke, Oosting, 1993). Of those patients, 10% reported feeling pain. Aftereffects such as sleep disturbances, nightmares, and flashbacks occurred in 70%, and pain was recalled by one half of these patients. In another study of 45 patients who experienced awareness during general anesthesia, 8 patients reported feeling severe pain (Schwender, Kunze-Kronawitter, Dietrich, et al., 1998). A large study of 3921 patients undergoing general anesthesia identified an incidence of 1% (39 patients) who experienced awareness with recall (Errando, Sigl, Robles, et al., 2008). Pain was felt by 11 patients.

An anecdotal report illustrates the tragedy of assuming an unconscious patient does not feel pain. A terminally ill patient was receiving intravenous morphine for cancer-related pain; as her disease progressed, she gradually became unresponsive to voices and no longer grimaced during position changes, which previously had been painful for her. Because she appeared to be unconscious and unresponsive to painful stimuli, morphine was no longer administered. After 3 days the patient regained consciousness and immediately asked for morphine. The patient reported that during the apparent “coma,” she had heard voices and felt intense pain (Stephens, 1994).

Regarding pain in terminally ill patients who appear to be unconscious, a recommendation made almost 3 decades ago is worth repeating: “When patients are no longer able to verbally communicate whether they are in pain or not, the best approach is to assume that their cancer is still painful and to continue them on their regular medications…. a therapeutic narcotic level should be maintained…. continued narcotics simply ensure that the death will be as peaceful and painless as possible” (Levy, 1985, pp. 397-398).

Whether patients in persistent vegetative states feel pain remains controversial. This state is defined as complete unawareness of self and environment, including lack of the cortical capacity to feel pain. However, some have voiced reservations about assuming that patients in vegetative states do not feel pain and have suggested that, until it can be proved otherwise, clinicians should assume that such patients feel pain and should treat that pain (Bushnell, 1997; Klein, 1997).

This position is also advocated by researchers who studied brain activation during noxious stimulation consisting of electrical stimulation of the median nerve, comparing 5 patients in minimally conscious states, 15 normal controls, and 15 patients in persistent vegetative states (Boly, Faymonville, Schnakers, et al., 2008). No brain area was less activated in patients in minimally conscious states than in the controls. All areas of the cortical pain matrix showed greater activation in these patients than in patients in persistent vegetative states. The researchers concluded that this evidence confirmed the need for analgesic treatment for patients in minimally conscious states. They pointed out, however, that misdiagnosis of patients in persistent vegetative states and minimally conscious states is common and concluded that analgesic treatment is indicated in both groups. Analgesics for patients in persistent vegetative states are indicated also to prevent potentially damaging defensive hormonal reactions such as the production of adrenal stress hormones, despite the possible absence of feelings of pain.

For some unconscious patients the CPOT, discussed earlier (see Form 4-1, pp. 145-146), may be appropriate if they are capable of responding in the four behavioral domains: facial expressions, movements, muscle tension, and ventilator compliance (Gelinas, Fillion, Puntillo, et al., 2006). If these behaviors are absent, documenting procedures or pathologies that probably cause pain may be relied upon to assume pain is present (APP), as is suggested in Box 3-9 (p. 123). A behavioral tool is appropriate only with patients who are able to respond with the tool’s requisite behaviors.

Clinicians often wonder which dose of opioid to administer on an ongoing basis to a patient who cannot respond with any behaviors to indicate the effectiveness of a dose one way or another. In such cases, opioid therapy should be initiated and maintained at the recommended starting dose (e.g., 2.5 mg/hr morphine IV for adults in possibly severe pain). However, clinicians should not hesitate to increase doses if there are any behaviors whatsoever that might indicate pain (see Chapter 27 for more about analgesic administration in the ICU).

Patients Who Are Intellectually Disabled

The term developmental disabilities encompasses mental or physical disabilities that are evident in childhood and continue throughout life (Craig, 2006). Intellectual disability (ID), as defined for this chapter, is a type of developmental disability in which some form of intellectual disability or cognitive impairment is noted in childhood. The ID may or may not be accompanied by physical disability. Conversely, physical disabilities such as cerebral palsy are not always accompanied by mental disabilities. IDs are commonly defined by IQ scores. An IQ score of 50 to 70 indicates mild cognitive impairment, a score of 35 to 49 indicates moderate impairment, a score of 20 to 34 indicates severe impairment, and profound impairment is an IQ score below 20 (Bottos, Chambers, 2006; Grossman, 1983). Individuals with mild cognitive impairment make up 85% of those with IDs and usually acquire about a sixth grade level of academic skills (American Psychiatric Association [APA], 1994). Moderate cognitive impairment occurs in about 10% of cases of ID, and severe and profound impairment constitute 1% to 4% of cases of ID.

This chapter focuses on adults with IDs, that is, adults who have been cognitively impaired since birth or early childhood. This state is not to be confused with that of adults who have been cognitively intact for most of their lives but become cognitively impaired in their later years. Dementia is an example of a cognitive impairment that occurs late in life in individuals who were cognitively intact previously. The literature can be confusing because the publications that use the term cognitive impairment in their titles may be focusing on adults who have become impaired as they have grown older or on adults who have been cognitively impaired since childhood. Pain assessment strategies previously discussed in the section on cognitively impaired older patients probably are not useful in assessing pain in adults with mental retardation. Adults who have been cognitively impaired since childhood require assessment tools that have been specially designed for them because these individuals do not have the same understandings and experiences as adults who have been cognitively intact for many years and become cognitively impaired in later life.

Most of the research on pain and ID has been conducted in infants and children, but many of the issues raised in those studies, such as assessment, are regarded as being relevant to the care of adults with IDs (Symons, Shinde, Gilles, 2008). Consequently, some studies of children are reviewed here even though this book focuses on adults. Some pain assessment tools used with children with IDs have been adapted and studied in adults with IDs.

A long-standing misconception is that persons with IDs, or mental retardation, are insensitive or indifferent to pain (Symons, Shinde, Gilles, 2008). After reviewing several studies of children with IDs, Symons and others (2008) concluded that caregivers of children with IDs frequently underestimated pain intensity. Consequently, these individuals are undertreated. This conclusion appears to be based on observations of behavior, not measurements of sensitivity to pain. It is hypothesized that what may be confusing about behavioral observations of these patients is that behavioral indicators of pain may be delayed or unconventional, contributing to the belief that this population has a lowered sensitivity to pain or a greater tolerance. However, testing of patients with mild intellectual disabilities by measuring heat-pain thresholds showed that their pain threshold was significantly lower than that of normal controls (Defrin, Pick, Peretz, et al., 2004). Thus, these individuals not only are sensitive to pain but also may be more sensitive to some types of pain than persons without IDs.

Along the same lines, children with autism have also been thought to be insensitive to pain, but this is not supported by research. In a study comparing 21 children with autism with 22 without autism, all of them undergoing venipuncture, the children with autism manifested significant facial reactions quite similar to those in the unimpaired group but characterized by more facial activity (Nader, Oberlander, Chambers, et al., 2004). A perplexing finding was that the children who were the most reactive during the venipuncture had been identified by the parents as being less sensitive and less reactive to pain.

Understanding the misconceptions of caregivers of children with disabilities may be helpful in recognizing the possibility that the same misconceptions may occur in reference to adults with IDs. The degree of agreement between parent and child regarding pain intensity was examined in 68 children with spina bifida who were able to communicate (Clancy, McGrath, Oddson, 2005). Children rated their pain, and their parents completed a proxy report of the pain. Parents tended to underestimate the intensity of the pain their children felt. In another study, 65 caregivers of children with IDs completed questionnaires about their beliefs regarding pain in this population and revealed complex beliefs (Breau, MacLaren, McGrath, et al., 2003). Caregivers believed that children’s sensations of pain increased as the severity of the ID increased. They also believed that children with mild cognitive impairment might overreact to pain. Perhaps most alarming was the finding that the more caregivers had learned about ID, the more they believed that those with IDs experienced less pain than those without IDs. The authors suggested several reasons for this finding, including the possibility that the information that was being taught was inaccurate or misunderstood.

Educational materials for caregivers of individuals with IDs should be carefully examined. Caregivers may enter the learning situation already harboring misconceptions. If these are not addressed, they may persist. Underestimation of pain by caregivers of individuals with IDs may result in underreporting of pain to clinicians, leading to undertreatment.

In a literature review, Bottos and Chambers (2006) identified types of pain likely to occur in individuals with IDs. Pain in children with IDs, particularly cerebral palsy, includes gastrointestinal discomfort, such as gastroesophageal reflux and constipation, and musculoskeletal pain. Children with autism also tend to experience gastrointestinal disorders. Individuals with Down syndrome are prone to developing leukemia, hence experiencing all the usual pain associated with the disease, diagnostic tests, and treatment. They also tend to have hip abnormalities and oral health problems such as periodontal disease. In adults with severe to profound mental retardation, pain is more often chronic than acute. An extensive review of the literature about this population revealed that a significant percentage of these people experience significant intensities of pain on a daily basis (Bodfish, Harper, Deacon, et al., 2006).

A review of the literature found that in the adult population of individuals with cerebral palsy, as much as 84% reported one or more types of chronic pain; 56% experienced pain daily and 53% reported moderate to severe pain (Bottos, Chambers, 2006). In one study of 63 women with cerebral palsy, of the 53 who had pain, the most common sites were the head (28%), back (26%), and arms (23%) (Turk, Geremski, Rosenbaum, et al., 1997). Activities of daily living, such as assisted sitting, walking, stretching, toileting, and standing, are especially painful and common, being reported by 35% to 93% of individuals with cerebral palsy (Bottos, Chambers, 2006). Some of the musculoskeletal pain may worsen with aging. It is important to note that studies of the prevalence of pain in adults with cerebral palsy have usually been restricted to those with mild or no cognitive impairment, although those with greater cognitive impairment are not known to be less sensitive to pain or to have fewer types of pain. In fact, the prevalence of pain in children with IDs is higher than in healthy children, and children with greater cognitive impairment tend to experience the most intense and most frequent pain but are least able to communicate their pain (Bottos, Chambers, 2006). Thus, studies suggest that the prevalence of pain in adults with IDs is higher than in adults without IDs.

Assessment of individuals with IDs is extremely challenging and, although considerable research has gone into creating assessment tools for children with IDs, few studies have included adults with IDs. A large majority of adults with IDs achieve a sixth grade level of functioning, so many may be able to provide self-reports of pain, and others with greater cognitive impairment may also be able to self-report. The greatest challenge is the assessment of those without verbal ability. The following studies suggest what to look for as indicators of pain in adults with IDs who cannot self-report.

To identify differences in the pain indicators used by nurses to assess pain in children and adults with severe versus profound IDs, a questionnaire consisting of 158 possible indicators of pain was completed by 109 nurses (Zwakhalen, van Dongen, Hamers, et al., 2004). More than 50% of the nurses found the following seven indicators to be very important: moaning during manipulation, crying during manipulation, painful facial expression during manipulation, swelling, screaming during manipulation, not using the affected body part, and moving the body in “a specific way of behaving.” Pain appeared to be assessed differently in those with severe versus profound IDs. Physiologic indicators, such as gasping for breath, vomiting, and turning red in the face, were scored somewhat higher by the group of nurses specializing in individuals with profound IDs (N = 47). Nurses specializing in the care of persons with severe IDs (N = 42) gave the highest score to “unusual way of crying.” Other indicators used by nurses specializing in the care of those with severe IDs included seeking comfort and being grouchy. (The remaining 53 nurses were taking care of both groups of patients.)

Interviews with eight nurses caring for patients with learning disabilities, who might be in pain but who could not communicate their feelings verbally, identified numerous behaviors they used as indicators of pain (Donovan, 2002). The ages of the patients were not given but some were probably adults because mention was made of nurses’ developing close relationships with patients over a long period of time. Change in facial expression was the most frequently cited nonverbal behavior indicative of pain or distress. Crying was also mentioned frequently. Other behaviors the nurses thought were possible indicators of pain were pacing, out-stretched arms, self-harm such as picking at the skin, aggressive behaviors such as biting or kicking others, pallor, altered rates of breathing, wincing, clenching teeth, and changes in activities such as avoiding weight bearing. One nurse described a patient with limited verbal ability that used the word arm for pain in any part of the body. These nurses emphasized the importance of caring for patients over time so that changes in behavior could be noted. Because changes in usual daily activities are often mentioned as clues to the presence of pain, it is noteworthy that these nurses mentioned that the patients were often determined to carry out normal activities even when seriously injured.

In one study, 40 adults with IDs who were undergoing intramuscular injections were studied to identify the usefulness of facial expressions in detecting pain (LaChapelle, Hadjistavropoulos, Craig, 1999). Only 65% of the subjects were able to provide self-reports. Observing the intensity of facial expressions, specifically brow lowering and chin raising, proved to be useful for assessing pain in these adults. Further, the frequency and intensity of facial expressions did not vary with level of cognitive functioning. In other words, persons with levels of ID that prevented self-report of pain showed facial expressions similar to those shown by persons with less intellectual impairment, providing further evidence against pain insensitivity in persons with IDs.

Somewhat contrary to that study, which found that facial expression did not vary with level of cognitive functioning, another study of adults with IDs found that facial expressions differed in individuals with mild to moderate cognitive impairment compared to those with severe to profound ID (Defrin, Lotan, Pick, 2006). In a study of 159 adults with varying levels of mental retardation and a control group of 38 with normal cognition, behavior was observed before and during the painful stimulus of a vaccination. During the vaccination, facial expression increased in those with mild to moderate IDs and in those without impairment. However, in those with severe to profound impairment, 47% to 50% exhibited stillness, or a “freezing reaction” (face and body not moving for several seconds), suggesting that pain assessment tools based only on facial expressions might contribute to the misconception that these patients are insensitive to pain. An increase in cognitive impairment was associated with a significant increase in freezing. In a few patients, freezing was accompanied by a full-blown smile, and some laughed, reactions that could easily be misinterpreted as insensitivity to pain. Freezing was exhibited by only 8% to 13% of individuals with mild to moderate IDs.

One reason for the differing conclusions in the studies described may be related to the sizes of the samples. It is possible that the study of 40 adults (LaChapelle, Hadjistavropoulos, Craig, 1999) did not include a large enough sample to allow for the detection of differences in behavior among the levels of ID, whereas the study of 159 adults with IDs (Defrin, Lotan, Pick, 2006) did allow for such detection.

Two behavioral pain scales were used in the study of 159 adults with ID: the Non-Communicating Children’s Pain Checklist-Revised (NCCPC-R) and the Facial Action Coding System (FACS) (Defrin, Lotan, Pick, 2006). The FACS is impractical to use in daily clinical practice because it involves coding video tapes. The NCCPC-R involves observing and rating behaviors (over a 2-hour period) and was found to be more sensitive to changes occurring in individuals at all levels of cognitive functioning, whereas the FACS was more sensitive to those with mild to moderate IDs. The NCCPC-R consists of 30 behaviors divided into seven categories: vocal, eating/sleeping, social, facial, activity, body/limb, and physiologic signs (Breau, McGrath, Camfield, et al., 2002). An observer scores the frequency of occurrence of the behaviors on a scale of 0 to 3: not at all, just a little, fairly often, and very often. The NCCPC-R was found to be reliable for measuring acute pain in adults with IDs.

One problem with the NCCPC-R is that it was validated by using a 2-hour observation period, which is impractical in many clinical settings. For that reason Breau, McGrath, and Zabalia (2006) recommend using the Non-Communicating Children’s Pain Checklist-Postoperative Version (NCCPC-PV) over a 10-minute period when a 2-hour period is not feasible. Research using the NCCPC-PV in adults with IDs and acute pain is needed. A variety of versions of the NCCPC are currently being researched in various populations with IDs. So far, the NCCPC-R and the NCCPC-PV have the greatest support for use with children because of their psychometric properties.

Assessment of location of pain may be difficult, especially in those with Down syndrome. In a study involving cold stimuli applied to the wrists and the temples of 26 individuals with Down syndrome, all were able to express themselves verbally but had some difficulty locating the site of the cold stimuli (Hennequin, Morin, Feine, 2000). In a later study of parents’ perceptions of pain in their children with Down syndrome (ages 1 to 39 years), the parents reported having greater difficulty in identifying locations of pain in their children with Down syndrome than in their children without Down syndrome (Hennequin, Faulks, Allison, 2003). Their abilities improved as the children grew older.

Two of the pain assessment tools that have been researched in adults with severe to profound mental retardation are the Pain and Discomfort Scale (PADS) (developed from the NCCPC-R) and the Pain Examination Procedure (PEP), and it is recommended that they be used together (Bodfish, Harper, Deacon, et al., 2006). These tools have acceptable levels of reliability and validity. The research into the development of these tools is extensive and is summarized in the chapter by Bodfish and others (2006) cited earlier. Studies indicate that the PADS is sensitive to nonverbal signs of pain in adults with severe IDs and is sensitive to everyday pain, acute pain responses, chronic pain, and effects of treatments to relieve pain. The authors found that administering the PADS plus the PEP takes trained personnel approximately 10 minutes. However, training for administering the PADS and PEP is extensive. A CD-ROM that includes a manual and workbook for the PADS and a DVD training video are available by calling 919-966-4896.

Other researchers disagree with the soundness of using the PADS in adults with IDs (Burkitt, Breau, Salsman, et al., 2009). They state that research to date is insufficient to support the general use of the PADS with adults with IDs. Indeed, further research is needed to identify pain assessment tools for this population. Meanwhile, the PADS plus the PEP may be a good alternative to using no tool at all.

Another potential tool, the Chronic Pain Scale for Nonverbal Adults with Intellectual Disabilities (CPS-NAID) (Form 4-2), is being researched (Burkitt, Breau, Salsman, et al., in press; reported in Breau, Burkett, 2009). This scale was arrived at by having two observers use the NCCPC-R to observe 16 nonverbal adults with IDs and chronic or recurrent pain conditions for a period of 5 minutes under conditions of pain and no pain. The observers also rated the pain using a VAS. Analyses indicated that six items should be removed from the NCCPC-R, resulting in a 24-item scale. This yielded 94% sensitivity to pain and 87% specificity for the presence of pain. These researchers concluded that the CPS-NAID has sound psychometric properties and can be used to assess chronic pain in adults with IDs. They think that this tool is superior to the original NCCPC-R, but further studies are needed to compare the two.

An attempt was made to establish cutoff points for pain on the basis of the total score of the CPS-NAID. However, it was based on proxy pain reports. As mentioned previously, in the section on behavioral scales for cognitively impaired adults, given current knowledge, scores on behavioral pain scales do not equate with pain-intensity scores (Pasero, McCaffery, 2005). In most studies of pain assessment tools, there are low correlations between proxy judgments of pain severity and patients reports of severity. This raises concern about comparing pain-behavior scores with pain-intensity scores. If clinicians could reliably determine the intensity of pain in patients who cannot self-report, there would be no need for a behavioral assessment tool (Zwakhalen, Hamers, Abu-Saad, et al., 2006).

The hierarchy of pain assessment techniques (see Box 3-9, p. 123) is helpful in the assessment of patients with IDs. Following are the steps in that hierarchy, with comments related to assessing adults with IDs, starting with attempts to obtain self-reports.

Patients Who Are Mentally Ill

Most of this part of the chapter focuses on assessment of pain in patients who have one of two mental disorders: (1) schizophrenia, in which decreased sensitivity to pain (hyppoalgesia) or impaired pain responsiveness has been documented in some patients (Singh, Giles, Nasrallah, 2006); and (2) posttraumatic stress disorder (PTSD), in which both decreased and increased sensitivity to pain have been reported(Geuze, Westenberg, Jochims, et al., 2007). The occurrence and the perception of pain in persons who have these two mental disorders are examined. Depression and anxiety in relation to pain are discussed briefly in Chapter 2 (pp. 31-32).

Stay updated, free articles. Join our Telegram channel

Join