- Top of page
- PATIENTS AND METHODS
It is a common clinical observation that chronic pain patients often report cognitive deficits including concentration and memory problems. Few studies have attempted to quantify the extent of cognitive deficits in chronic pain patients. Kewman et al (1) used a brief standardized computer screening test and found that approximately one-third of their pain patients had at least one index score in the impaired range. Grigsby et al (2) compared the performance of chronic pain patients with pain-free head injured patients on a variety of computerized tasks designed to assess speed of information processing and short-term memory. They found that the chronic pain patients performed poorly compared with the head injured patients on 33% of the speed of information processing tasks, and equally or worse than head injured patients on all other measures. Eccleston and Crombez (3) documented that patients with chronic pain who reported higher levels of both pain and somatic awareness showed significant performance decrements on an attentionally demanding switching task compared with other chronic pain patients.
Patients with fibromyalgia (FM) typically experience widespread musculoskeletal (MSK) pain, stiffness, and fatigue (4). These patients generally also report high levels of somatic awareness (5). Although data are scarce regarding cognitive deficits in FM patients, many of these patients report attention and memory deficits (6). Sletvold et al (7) investigated attention and memory deficits in FM patients and found only attentional deficits in FM patients compared with healthy controls on 2 standardized attention tests. In a well-designed study, Grace et al (8) reported that compared with matched healthy controls, FM patients showed significant attentional and memory deficits on a neuropsychological test battery. The attentional deficits were found between groups on the Paced Auditory Serial Additions Test (PASAT) but not on the attention/concentration index of the revised Wechsler Memory Scale. It is notable that these reported attentional deficits were observed only when task performance had a high attentional demand. Unfortunately, because both studies (7, 8) tested only FM patients and pain-free controls, it is not possible to conclude whether the reported deficits were primarily related to FM, or whether patients with FM experience greater attentional disruption in comparison with other chronic pain patients. There is some evidence to suggest that patients with a diagnosis of FM will demonstrate greater deficits due to a decreased tolerance to pain (and possibly to other stressors), a heightened attention to pain, and an increased interruptibility (6, 9, 10). However, no single study has compared these deficits in patients with different rheumatologic diagnoses.
This study examines attentional functioning between groups of chronic pain patients, using a standardized neuropsychological test with a high level of ecological validity. The study was designed to test 2 related hypotheses: 1) All chronic pain patients, regardless of diagnosis, will exhibit performance decrements on all measures of attention compared with pain-free controls, and 2) FM patients will show significantly greater impairment on attentional tasks compared with rheumatoid arthritis (RA) and musculoskeletal pain patients.
- Top of page
- PATIENTS AND METHODS
Descriptive statistics of the self-report measures are reported in Table 1. Group differences on all descriptive measures were tested using pairwise comparisons with Bonferroni corrections to control for multiple testing. Compared with the pain-free control group, the 3 patient groups had significantly more pain, depression, pain-related disability, higher levels of somatic awareness, and catastrophic thinking about pain. FM patients were significantly younger and reported more anxiety than all other groups. They also reported a significantly higher level of somatic awareness than the RA patients. There were no differences between inpatients and outpatients on any of the measures.
Correlational analyses were performed with demographic measures, self-report measures, and TEA scores. Patients' self-report measures of pain at the time of neuropsychological testing, depression, anxiety, pain disability, somatic awareness, pain catastrophizing, and sleep (number of hours of sleep and number of wakings per night) tended to be significantly correlated (see Table 2). However, none of these variables were significantly correlated with any of the TEA index results.
Table 2. Two-tailed Pearson's correlation between self-report measures for all patients*
|Variable||Pain VAS||Hrs sleep||Sleep disrupt||HAD-Anx||HAD-Dep||PDI||MSPQ||PCS|
|Hrs sleep|| ||—||NS||NS||NS||NS||−0.27†||−0.32†|
|Sleep disrupt|| || ||—||NS||0.30†||0.36‡||0.39‡||0.46§|
|HAD-Anx|| || || ||—||0.63§||0.33†||0.40‡||0.70§|
|HAD-Dep|| || || || ||—||0.52§||0.32†||0.62§|
|PDI|| || || || || ||—||0.48§||0.56§|
|MSPQ|| || || || || || ||—||0.48§|
A series of analyses of variance were performed to examine group differences in performance on the overall TEA, and on the summary scores for the subscales of selective attention, sustained attention, auditory-verbal working memory, and attentional switching. Specific group differences were tested using pairwise comparisons with Bonferroni corrections to control for multiple testing. The means and standard deviations of the overall TEA and the domains of cognitive functioning are included in Table 3. All TEA score results are derived using the TEA's individual age-normed scaled scores. All significant between-group differences in attentional functioning remained significant when age, depression, anxiety, somatic awareness, and catastrophizing were included as covariates. Scores from patients who received opioid medication did not differ significantly from the scores of patients who did not receive opioids on any of the TEA measures.
Table 3. Results of overall performance and subtest scaled scores on the Test of Everyday Attention (TEA) for patients with chronic pain associated with rheumatoid arthritis, fibromyalgia, musculoskeletal disorders, and pain-free control participants*
| ||Rheumatoid arthritis group||Fibromyalgia group||Musculoskeletal disorders group||Pain-free control group|
|Overall TEA score (mean of all standard scores)||8.0 ± 1.1†||8.7 ± 1.6†||8.6 ± 2.0†||10.5 ± 1.4|
|Selective attention‡||24.8 ± 6.1†||25.2 ± 6.6†||25.1 ± 10.1†||33.4 ± 8.0|
|Sustained attention‡||23.8 ± 4.6†||27.4 ± 3.5||25.6 ± 5.9||29.3 ± 3.7|
|Attention switching||7.4 ± 3.5||9.2 ± 3.6||7.6 ± 2.9||9.5 ± 3.1|
|Working memory‡||15.0 ± 4.4†||17.9 ± 5.5†||18.5 ± 4.0||22.2 ± 4.4|
Overall TEA score.
There was a significant effect for group on mean TEA standard scores (F[3,76] = 9.87, P < 0.0001). Pairwise comparisons revealed that pain-free controls performed better than each of the 3 patient groups. The FM group had the same performance as the other patient groups. No other differences were significant.
There was a significant effect for group on the composite score (sum of scaled scores of subtests for this index) of selective attention (F[3,74] = 5.42, P < 0.003). Pairwise comparisons revealed that pain-free controls had a significantly higher level of performance than each of the patient groups. The FM group did not perform worse than the other patient groups. No other between-group differences were observed.
There was a significant effect for group on the composite score (sum of scaled scores of subtests for this index) of sustained attention (F[3,73] = 5.43, P < 0.003). Pain-free controls were more able to sustain attention than patients in the RA group. There was no significant difference between the pain-free controls and the other two patient groups. FM patients did not perform worse on sustained attention than the 2 other patient groups. No other between-group differences were observed.
No between-group differences were observed on the summary score related to attention switching (F[3,73] = 2.01, P not significant).
There was a significant main effect for group on a composite score (sum of scaled scores of subtests for this index) of auditory-verbal working memory (F[3,72] = 8.13, P < 0.0001). Pain-free controls had higher scores than the FM and RA patients, but they were not significantly different from the MSK patients. No other between-group differences were observed.
A further analysis showed that 60% of patients had at least one TEA subtest score in the clinically impaired range compared with 20% of healthy controls. Further, 38% of patients and only one (5%) of the healthy control participants had more than one subtest score in the impaired range.
- Top of page
- PATIENTS AND METHODS
In this study a majority of patients with chronic pain exhibited cognitive problems. Sixty percent of the patients had at least one score in the “clinically impaired” range (scaled score ≤ 5) compared with 20% of the pain-free controls. This suggests that although at the group level mean TEA score differences would place chronic pain patients in a mildly impaired range in terms of attentional functioning (see Table 3 for group means), at the individual level, the majority of chronic pain patients showed severe impairment on at least one attentional index. Attentional deficits in the patient groups were significant on tasks of selective and sustained attention, and auditory-verbal working memory. These results further document the extent of concentration and memory problems of chronic pain patients. Because these deficits are found with a test battery specifically designed to be ecologically valid, it is reasonable to expect that the observed deficits are clinically relevant for everyday cognitive functioning.
It may be interesting to note that although expected correlations existed between a number of the mood and pain-related self-report measures, there was no correlation between attentional functioning and any of these factors. This is in partial contrast to the findings of Grace et al (8), in which their patients diagnosed with FM showed modest correlations between trait anxiety and performance, and pain severity and performance on the PASAT.
The finding that TEA results were not correlated to the pain VAS measures suggests that attentional disruption may not be related to the level of pain experienced at a given time. It may be that the observed attentional disruption in chronic pain patients was less a result of the pain they were experiencing during testing and more a result of having a history of chronic attentional disruption by pain, as discussed by Eccleston and Crombez (3). Another objective of this study was to investigate whether FM is characterized by more severe cognitive problems, or by a specific attentional problem. Although FM patients showed a significantly higher level of anxiety than the other 3 groups, this study did not reveal that FM patients had more severe attentional problems than other chronic pain patients. Furthermore, FM patients did not show poorer performance than patients with RA or MSK in any of the investigated domains of attentional and cognitive functioning. These results are the first to show that patterns of attentional performance are comparable among chronic pain patients with different rheumatologic disorders. They suggest that the observed attentional deficits may be related more to the presence of chronic pain than to any specific disease-related factor. There are a number of issues that could be improved upon in future studies. First, further studies could explore the possible mediating role of psychological and medical variables in producing attentional deficits of this order. A larger patient sample will allow regression modeling and an exploration of the potential mediating role of specific psychological variables (3).
Second, due to the considerable variability in types and doses of medications taken, it was not possible to fully control for the putative effects of medication. The RA and FM groups reported greater use of opioid medications compared with the MSK group. However, although a variety of medications, including opioids, are known to affect cognitive function, recent data have shown that attentional functioning is either not disrupted or improves in individuals, including chronic pain patients, following opioid consumption using both performance and electrophysiological measures (17–22). In our study, patients who received opioid medications did not perform significantly differently from patients who did not receive opioids.
Third, in this study the pain-free control group contains a majority of male participants, although the patient groups contain a majority of females. This difference arose due to our use of a pragmatic consecutive sample design for both the patient and pain-free groups. Although the TEA is gender neutral and not normed on the basis of sex, a future study could ideally use a sampling technique that matches both age and sex across samples.
Finally, these findings could be strengthened in future work with the addition of self-report measures of perceived performance, cognitive effort, and test anxiety. More specific measures of the attentional presentation of patients may also help to clarify the source of the pain-related decrement in performance (23). The addition of a computerized version of this test battery will allow for the micro-analysis of specific attention and memory functions.
This study supports previous findings by reporting that many chronic pain patients have significant attentional dysfunction. An important, novel finding is that regardless of disease status, chronic pain patients demonstrate cognitive impairment when performing everyday attentional tasks when compared with matched pain-free controls. Further investigation and clarification are warranted to more fully understand the nature of these attentional deficits, their underlying mechanisms, and their clinical implications.