Dick, Bruce D. PhD*†‡; Rashiq, Saifudin MB, MSc, FRCPC*‡
Chronic pain needs to be recognized as a public health priority. The cost of chronic pain to society is great (1). Prevalence studies indicate that as much as 44% of the population experience pain on a regular basis, and that in one-quarter of this group the pain is severe (2,3). Chronic pain affects the ability to work, sleep, and perform other activities essential to leading a full life. These changes generally worsen over time (4).
Chronic pain disrupts attentional functioning (5–12), which in chronic pain patients has been found to be improved by treatments resulting in analgesia using electrophysiological measures (11,13). Patients with chronic pain report widespread disruption of attention and memory (14). An outstanding review by Hart et al. (15) describes the complex nature of attention and memory deficits in individuals with chronic pain. That review discusses the importance of many factors that contribute to the disruption of cognitive processes. Individuals who experience cognitive deficits related to pain are at increased risk for poorer mental health including mood (16). In summary, chronic pain has a significant negative impact on a variety of factors related to an individual’s quality of life (16).
Although it is clear that many individuals who suffer from chronic pain experience attentional and other deficits, the cognitive mechanisms that are affected by chronic pain are not clear. In this study, we sought to discover performance factors that would differentiate individuals with chronic pain whose attentional function is disrupted by their pain from those whose attentional function is not affected by pain. We also examined whether short-term analgesia would improve attentional functioning.
To our knowledge, the impact of attentional resource availability, as measured by working memory capacity, on the degree of disruption by pain has not previously been investigated. Understanding this could help predict which individuals are at risk for cognitive disruption by chronic pain, as well as the tasks that would be most difficult for them to perform. This knowledge could be of great help in treatment planning.
This study was approved by the Health Research Ethics Board at the University of Alberta.
Written informed consent was obtained from all participants in accordance with the University of Alberta’s Health Research Ethics Board requirements before study inclusion. Following power calculations based on previous findings (10), we recruited a consecutive sample of 24 participants (18 females) from the Multidisciplinary Pain Centre at the University of Alberta Hospital. Participants were recruited from those patients who were ordinarily scheduled to receive invasive analgesic procedures as part of their ongoing treatment. Participants had a baseline pain intensity of at least 4/10 on a numerical pain rating scale (NRS) anchored at 0 and 10 (described below) and had had pain for 6 months or longer. Individuals with a medical history that included significant head injury, neurological disorder, or disease known to impair attentional functioning (e.g., epilepsy, multiple sclerosis, Parkinson’s disease, chronic migraine/chronic daily headache) were excluded. Demographic, sleep, and medical information for participants can be found in Table 1. Participants’ questionnaire data are found in Table 2.
The following measures were used to assess variables of interest based upon previous research and have been standardized and validated to examine the variables being studied.
Age, sex, and number of years of education.
Medical and Pain History
Diagnosis, etiology of pain, location of pain, and a detailed list of medications being taken at each time of testing.
We measured pain using a 11-cm NRS for current pain intensity and the McGill pain questionnaire [MPQ; (17)]. The NRS is completed by drawing a vertical line along a 11-cm horizontal line to represent current pain level. The horizontal line is labeled at each end with 0 (no pain) and 10 (worst pain imaginable). The MPQ provides measures of the sensory, affective, and evaluative aspects of pain. This questionnaire is widely used because of its ability to measure these different aspects of pain, its sensitivity to differences in pain levels, and its sensitivity to differences in different qualities of an individual’s pain.
The pain catastrophizing scale (18) was administered to collect information related to participants’ levels of pain catastrophizing. The frequencies of occurrence of 13 pain-related thoughts during pain are evaluated by this measure.
The test of everyday attention [TEA; (19)] is a standardized neuropsychological test that provides a profile of attentional functioning in the domains of sustained and selective attention. It also provides a measure of attentional switching and an index of auditory-verbal working memory. The TEA was chosen because of the variety of attentional indices evaluated and its strong ecological (relevance to everyday tasks) and face validity. Another major benefit associated with using this test is that there are multiple forms of each subtest (i.e., versions A and B), allowing for retesting of skills without the potential confound of test–retest effects. Subtests from this test include tasks such as selective attention tasks, divided attention tasks, memory tasks that tax sustained attention, and both auditory and visual attention switching tasks.
Attentional/Working Memory Capacity
The reading span test [RST; (20)] assesses abilities related to linguistic and nonlinguistic cognitive abilities. More specifically, it has been proposed to reflect the efficiency of basic cognitive skills and the availability of cognitive resources (21). This measure was chosen because it has been widely used and validated as a measure of verbal working memory. On each trial of this test, participants were asked to read a set of sentences while maintaining target words in memory, one target word for each sentence. Three trials are administered at each set size where set sizes range from two to six sentences.
In addition to the RST, the spatial span test [SST; (22)] was also used. The SST has been found to tax processing and storage components of spatial working memory. The use of the SST was also important in this study because a number of the subtests on the TEA include tasks that test nonverbal and spatial cognitive skills.
Both the RST and SST were presented in a standardized format using a Toshiba Satellite™ (New York, NY) laptop. The RST was completed using the standardized instructions found in Daneman and Carpenter’s (20). Participants were instructed to read each sentence aloud and report whether or not each sentence was grammatically correct while remembering the terminal word of each sentence (Fig. 1). At the end of each set, participants were prompted to report the terminal words of all sentences in the set. The set sizes of stimulus trials ranged from two to six sentences and set size increased as the paradigm progressed. As the number of sentences in each set becomes larger, the cognitive load of each set becomes greater, increasingly taxing working memory.
The SST was completed using a paradigm previously described in a number of reports including Shah and Miyake’s (22). This paradigm was generated using E-Prime™ (Pittsburgh, PA) software. On this task, participants are required to report whether letter stimuli are presented in normal or mirror-image orientations while remembering the orientation of letters presented in the set (Fig. 2). Participants were required to report the orientations of all letter stimuli in each set at the conclusion of each set using a spatial grid presented on the computer screen. The set size of this task also increased as the paradigm progressed, increasingly taxing participants’ working memory. The SST orientation score provides a measure of spatial working memory. An additional score is also generated for participants’ ability to perform the mirror task, a reflection of their ability to maintain and mentally manipulate a memory trace while performing the spatial working memory task.
Mental Health (Anxiety and Depression)
The hospital anxiety and depression (HAD) scale (23) is a brief (14-item) measure that was used to assess participants’ current levels of depression and anxiety.
Average number of hours of sleep per night and average number of wakings per night were recorded using participants’ self-reports.
Invasive analgesic procedures are often performed in chronic pain patients attending pain clinics. These included epidural injection, sympathetic blockade, somatic nerve blockade, pulsed radiofrequency rhizotomy of the medial branch nerves, and trigger point therapy. These procedures are performed as diagnostic or therapeutic measures, or to allow the patient to undertake physical therapy and rehabilitation. The effect of these interventions is usually at its maximum level within the first few days after administration. Participants were recruited if their interventional procedures reduced their pain by at least four points on the NRS.
Participants attended two testing sessions on separate “pain” and “less pain/control” days. On the less pain/control day, participants were tested after receiving a pain-reducing procedure (as noted above) as part of their ongoing treatment at the Multidisciplinary Pain Centre. The timing of this visit was made such that participants were experiencing a high level of pain relief from the procedure but after a diminution of any associated negative side effects secondary to the procedure (e.g., nausea, drowsiness). These sessions were scheduled as close together as possible within these constraints. Oral medication use was the same on both days of testing, and testing was scheduled at the same time of day for each individual’s two tests to control for medication effects. On the pain day, participants were tested without having received a recent blockade procedure when their pain was reported to be at a high level.
Testing was administered by two research assistants trained in the use of the standardized testing tasks. One research assistant recruited participants and administered study questionnaires and the computer-generated RST and SST tasks on the pain testing day. There are no alternate forms of the RST and SST that have been standardized and normed to account for test–retest effects, so these tests were only administered once. The questionnaires were only provided on the pain day as we were primarily interested in the results of these forms when participants were experiencing significant pain and to reduce participant burden. The pain testing day was chosen for the administration of the RST and SST, as our primary hypothesis was that working memory performance would predict attentional performance and the level of attentional disruption by pain. The research assistant who administered the TEA during each testing session was blinded to the pain status of the participants. Testing session order was randomized and balanced across participants.
Participants’ TEA scores on subtests were added to create an overall performance score using the sum of all subtest scores as previously reported in Dick et al. (10) Index scores of selective attention and sustained attention were calculated using the sum of subtest scores found to be predictive of each domain of attention as outlined by Robertson et al. (19) Using SPSS (Chicago, IL) version 11.5, paired t-tests and one-way analyses of variance (ANOVA) were used to compare results between conditions.
After these analyses, further planned analyses were performed to examine how working memory performance would predict attentional function. These analyses were performed using a median split method as reported by Eccleston (5,6). In the current study, participants were categorized according to their level of attentional disruption. Eight participants showed no clinical impairment because of pain on their pain testing day. The remaining 16 participants had at least one subtest score in the clinically impaired range on the TEA. Those below the median response (i.e., one score in the clinically impaired range) were categorized as the less disrupted group, whereas those above the median response (i.e., two or more subtest scores in the clinically impaired range) were classified as the highly impaired group. Planned comparisons were performed using Bonferroni corrections to control for the inflation of α.
No test–retest effects were found for participants across testing sessions. As well, no differences were found in participants’ performances on versions A or B of the TEA for both pain and less pain/control conditions. Thus, participants’ performances did not improve because of taking TEA for the second time, and participants’ performances were not significantly better on either version of the test.
All participants reported a significant reduction in their pain after their analgesic procedures. None of the participants randomized to the condition of being tested in pain for their second testing session were excluded because of excessive long-term pain relief. A paired t-test found that participants’ reported pain levels were significantly reduced by their pain treatments [t(23) = 5.90, P < 0.0001]. However, despite this reduction in pain, participants’ performances were not improved on measures of overall TEA performance [t(23) = −0.05, not significant], selective attention [t(23) = 0.03, not significant] or sustained attention [t(23) = 0.74, not significant]. These findings suggest that the short-term pain-reducing interventions described above did not improve attentional performance.
These three groups did not differ on measures of age [F(2,23) = 0.23, not significant], education [F(2,23) = 2.05, not significant], sleep [hours of sleep per night: F(2,23) = 0.54, not significant; number of wakings F(2,23) = 0.62, not significant] overall pain scores [NRS: F(2,23) = 3.29, not significant; MPQ S: F(2,23) = 0.10, not significant; MPQ A: F(2,23) = 0.01, not significant; MPQ PRI: F(2,23) = 0.04, not significant], anxiety [HAD A: F(2,23) = 0.01, not significant], depression [HAD D: F(2,23) = 0.11, not significant], or pain catastrophizing [pain catastrophizing scale: F(2,23) = 1.47, not significant]. Group data on medications taken on the day of testing and pain location are also available in Table 1. These groups also did not differ in their verbal working memory [RST: F(2,23) = 2.45, not significant] performance or on the spatial working memory orientation task [SST orientation: F(2,23) = 0.02, not significant] scores.
However, some important group differences were discovered in participants’ performance on the SST mirror task. As previously noted, this task provides a reflection of one’s ability to maintain and mentally manipulate a memory trace while performing the spatial working memory task. The performance of the nonimpaired and slightly impaired groups was not significantly different on the SST mirror task [F(1,15) = 0.38, not significant]. However, the highly impaired group’s performance on the everyday attentional tasks measured by the TEA was significantly worse than the less impaired groups [F(1,23) = 4.36, P < 0.05] on the SST mirror task. This finding suggests that those participants who experienced the most difficulty with the added demands of the SST mirror dual task during maintenance of information in the spatial working memory store showed the most impairment in the everyday attentional tasks.
This study was carefully designed to consider as many factors as possible that could play a role in the disruption of attention and memory in individuals with chronic pain. Given the potent analgesic properties of the medications that many patients with chronic pain are prescribed (e.g., opioids), we took special care to record and control for any possible effects of medication on cognitive performance in our participants. Fortunately, there is strong evidence that performance on neuropsychological tests is either not affected nor improved in individuals who are prescribed opioids for their chronic pain, even after 1 year of the use of these powerful analgesics (13,24). Despite this evidence, it is important to be aware that the potential effects of opioid medications on cognitive function in individuals with chronic pain are complex. Further research is warranted in this area.
Our findings provide additional insight into the specific cognitive mechanisms disrupted in individuals with chronic pain. Two-thirds of the participants in this study showed clinically significant disruption of attention associated with impaired working memory processes. This disruption was not found to be associated with sleep problems, psychological distress, or age. At the outset of this study, we expected that measures of attentional resource capacity would predict the magnitude of attentional disruption on standardized everyday tasks. We also predicted that short-term analgesia could result in improved attentional function. Although the latter prediction was not borne out in our data, our primary prediction was confirmed. However, our findings point to disruption by pain in a specific process that we did not anticipate.
This study extends a line of research founded on Eccleston’s (5,6) seminal studies. That work found that a high level of pain interferes with attentional processes during the completion of demanding tasks. Dick et al. (10) extended that work in studies that showed attentional disruption across a variety of chronic pain groups during the completion of everyday tasks assessed by the TEA. That work has led to further studies that examined and elucidated more basic mechanisms of cognitive disruption (11,25). Taken together, this line of research has complemented a variety of other neurocognitive studies of cognitive disruption by chronic pain. Specifically, the present study points to a specific cognitive process that is disrupted by chronic pain, the maintenance of a memory trace during the completion of a complex task. Memory traces are important in the unconscious completion of many everyday tasks and are usually taken for granted. It is possible that the interference of memory traces by pain could play an important role in the general disruption of attention and memory that is reported by many individuals with chronic pain.
For example, Grisart et al. (8,9) demonstrated that chronic pain from fibromyalgia adversely affects memory functioning. Their work suggests that cognitive disruption by pain is a reflection of a disruption of normal attentional resource allocation. They investigated the effects of chronic pain on tasks requiring controlled intentional processing and compared those findings with pain’s effects on tasks that only require automatic processing. Although their primary interest was chronic pain’s effects on memory, they proposed that the underlying cognitive mechanism impaired by pain was the availability of attentional resources required during controlled processing. They also predicted that this disruption of attention resource allocation would affect parallel processing, exactly as demonstrated by our subjects in the dual task requirements of the SST.
In a recent study on patients with chronic pain from fibromyalgia, Leavitt and Katz (12) proposed that when cognitive tasks are performed in conditions where there is stimulus competition, chronic pain interferes with information processing and retention of information. Using standardized neuropsychological tests, they found short-term memory disruption that appears to have resulted from a source of distraction during the completion of a dual task paradigm. They used the auditory consonant trigram (ACT), a test that detects the loss of very recently acquired information (a memory trace) because of momentary distraction (26). It has been suggested that disrupted performance on the ACT occurs because this task requires retention of information while a second task is being performed, making it a very attentionally demanding task (27). Leavitt and Katz (12) found that the distracter task in the ACT resulted in impaired retention of new information in their participants. Their findings provide evidence of attentional disruption by challenging tasks that divide attention, leading to functional impairment. Our current findings with the SST provide complementary support to Leavitt and Katz’s results in another patient population with chronic pain. Our findings suggest that pain may disrupt the maintenance of the memory trace that is required to hold information for processing and to later retain it for storage in longer-term memory stores. This finding provides an additional possible explanation for previous findings related to the disruptive effects of chronic pain on attention.
We recognize that the present findings are based on a small sample of individuals with chronic pain. However, during the course of our analysis, it was clear that the statistically significant findings cited here were robust, despite this sample size. Those findings that were not statistically significant had such small effect sizes that an enormous sample size would be required to ascertain other secondary effects. In a study such as this, such small effects, even in a large group, would be relatively meaningless and have minimal relevance from a clinical standpoint.
In addition, a further limitation to this study was the fact that oral medication intake was not reduced in participants between the pain and less pain/control sessions, although their NRS scores were significantly reduced. This could call into question the analgesic efficacy of the procedures received by participants. However, this is a difficult issue to clarify, as patients’ verbal reports of NRS scores suggested clear reductions in their pain.
Further, some might argue that the effects of medications being taken by participants could have obscured the beneficial effects of these procedures on cognitive function. It is more likely that the complex and severe nature of the chronic pain syndromes experienced by these participants made it difficult to detect any subtle differences that may have occurred. Eccleston and Crombez (28) have described chronic pain as a chronic disruption of attention. Therefore, when a complex chronic pain syndrome has been present for months or years, the likelihood of seeing an immediate improvement in memory and attention function soon after a short-term pain reduction procedure may be very low. It may be unrealistic to expect that such a procedure would produce a marked improvement in cognitive function. Previous studies have also suggested that central changes in neural mechanisms related to attentional processing, as well as pain processing, could account for the attentional disruption that occurs in individuals with chronic pain (25,29).
The clinic population studied was quite heterogeneous with regard to pain etiology and the short-term interventions used to treat participants’ pain. These factors were not controlled more rigidly in this study to allow study results to be more reflective of a general clinic population. Much remains to be studied with regard to these issues.
Chronic pain disrupts attention, and earlier studies provided evidence that it negatively affects mental health and the ability to perform everyday tasks. However, the relationship between the level of attentional disruption and quality of life has not been carefully studied. We are presently studying these relationships.
Psychological treatment using patient education and cognitive-behavioral therapy is effective for patients with chronic pain (30). The extent to which cognitive deficits experienced by individuals with chronic pain affect their ability to benefit from this therapy has not been studied. It remains to be seen whether attentional training, such as the work by Sohlberg and Mateer (31), could improve cognitive function in patients with chronic pain, and thereby decrease functional disability and increase the capacity to benefit from other therapies.
Finally, recent advances in functional neuroimaging may permit identification of the anatomical locations associated with pain processing and the deficits in attention and memory processing found in this study.
In summary, although there is good evidence that chronic pain disrupts attention, until recently there has been a lack of data to clarify which cognitive mechanisms are interrupted by chronic pain. This study has identified the working memory trace as a specific cognitive process that appears to be disrupted by chronic pain. If we can better understand which cognitive mechanisms are disrupted by chronic pain and how they are disrupted, we can continue to move toward improving our ability to predict which individuals are most at risk for cognitive disruption by chronic pain and tailor their treatment accordingly.
We thank Dr. Adrienne Witol, neuropsychologist, for her consultation on some of the neuropsychological aspects and findings of this study. We also thank Michelle Verrier for her assistance with data collection and in the preparation of this manuscript.
1. Turk DC. Clinical effectiveness and cost-effectiveness of treatments for patients with chronic pain. Clin J Pain 2002;18:355–65.
2. Birse TM, Lander J. Prevalence of chronic pain. Can J Public Health 1998;89:129–31.
3. Smith BH, Elliott AM, Chambers WA, et al. The impact of chronic pain in the community. Fam Pract 2001;18:292–9.
4. Wincent A, Liden Y, Arner S. Pain questionnaires in the analysis of long lasting (chronic) pain conditions. Eur J Pain 2003;7:311–21.
5. Eccleston C. Chronic pain and attention: a cognitive approach. Br J Clin Psychol 1994;33:535–47.
6. Eccleston C. Chronic pain and distraction: an experimental investigation into the role of sustained and shifting attention in the processing of chronic persistent pain. Behav Res Ther 1995; 33:391–405.
7. Grace GM, Nielson WR, Hopkins M, Berg MA. Concentration and memory deficits in patients with fibromyalgia syndrome. J Clin Exp Neuropsychol 1999;21:477–87.
8. Grisart JM, Van der Linden M. Conscious and automatic uses of memory in chronic pain patients. Pain 2001;94:305–13.
9. Grisart JM, Van der Linden M, Masquelier E. Controlled processes and automaticity in memory functioning in fibromyalgia patients: relation with emotional distress and hypervigilance. J Clin Exp Neuropsychol 2002;24:994–1009.
10. Dick BD, Eccleston C, Crombez G. Attentional functioning in fibromyalgia, rheumatoid arthritis and musculoskeletal pain patients. Arthritis Care Res 2002;47:639–45.
11. Dick BD, Connolly JF, McGrath PJ, et al. The disruptive effect of chronic pain on mismatch negativity. Clin Neurophysiol 2003;114:1497–1506.
12. Leavitt F, Katz RS. Distraction as a key determinant of impaired memory in patients with fibromyalgia. J Rheumatol 2006;33: 127–32.
13. Lorenz J, Beck H, Bromm B. Cognitive performance, mood and experimental pain before and during morphine-induced analgesia in patients with chronic non-malignant pain. Pain 1997;73:369–75.
14. Jamison RN, Sbrocco T, Parris WCV. The influence of problems with concentration and memory on emotional distress and daily activities in chronic pain patients. Int J Psychiatry Med 1988;18:183–91.
15. Hart RP, Martelli MF, Zasler ND. Chronic pain and neuropsychological functioning. Neuropsychol Rev 2000;10:131–49.
16. Shifren K, Park DC, Bennett JM, Morrell RW. Do cognitive processes predict mental health in individuals with rheumatoid arthritis? J Behav Med 1999;22:529–47.
17. Melzack R. The McGill pain questionnaire: major properties and scoring methods. Pain 1975;1:277–99.
18. Sullivan MJL, Bishop SR, Pivik J. The pain catastrophizing scale: development and validation. Psychol Assess 1995;7:524–32.
19. Robertson IH, Ward T, Ridgeway V, Nimmo-Smith I. The test of everyday attention. Suffolk: Thames Valley Test Company, 1994.
20. Daneman M, Carpenter PA. Individual differences in working memory and reading. J Verb Learn Verb Behav 1980;19:450–66.
21. Rosen W, Engle RW. The role of working memory capacity in retrieval. J Exp Psychol Gen 1997;126:211–27.
22. Shah P, Miyake A. The separability of working memory resources for spatial thinking and language processing: an individual differences approach. J Exp Psychol Gen 1996;125:4–27.
23. Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand 1983;67:361–70.
24. Tassain V, Attal N, Fletcher D, Brasseur L, Degieux P, Chauvin M, Bouhassira D. Long term effects of oral sustained release morphine on neuropsychological performance in patients with chronic non-cancer pain. Pain 2003;104:389–400.
25. Dick BD, Connolly JF, Houlihan ME, et al. Effects of experimentally induced pain on MMN. J Psychophysiol 2006;20:21–31.
26. Crowder RG. Some relationships between global amnesias and the memory impairments in Alzheimer’s disease. In: Corkin S, Davis KL, Groudin JH, et al., eds. Alzheimer’s disease: a report of progress in research. Hillsdale, NJ: Lawrence Erlbaum Associates, Inc., 1982.
27. Vallar G, Baddeley AD. Fractionation of working memory: neuropsychological evidence for a phonological short-term store. J Verb Learn Verb Behav 1984;23:151–61.
28. Eccleston C, Crombez G. Pain demands attention: a cognitive affective model of the interruptive function of pain. Psychol Bull 1999;125:356–66.
29. Woolf CJ, Salter MW. Neuronal plasticity: increasing the gain in pain. Science 2000;288:1765–9.
30. Morley S, Eccleston C, Williams A. Systematic review and meta-analysis of randomized controlled trials of cognitive behaviour therapy and behaviour therapy for chronic pain in adults, excluding headache. Pain 1999;80:1–13.
31. Sohlberg MM, Mateer CA. Effectiveness of an attention-training program. J Clin Exp Neuropsychol 1987;9:117–30.