Predicting Complaints of Impaired Cognitive Functioning in Patients with Chronic Pain

Transcription

1 392 Journal of Pain and Symptom Management Vol. 21 No. 5 May 2001 Original Article Predicting Complaints of Impaired Cognitive Functioning in Patients with Chronic Pain Lance M. McCracken, PhD and Grant L. Iverson, PhD Departments of Psychiatry and Anesthesia & Critical Care (L.M.M.), The University of Chicago, Chicago, Illinois, USA; and Department of Psychiatry (G.L.I.), University of British Columbia, Vancouver, British Columbia, and Neuropsychiatry Unit (G.L.I.), Riverview Hospital, Port Coquitlam, British Columbia, Canada Abstract Patients with chronic pain often complain of difficulties with cognitive functioning. Previous studies suggest that these occur with no history of head trauma or neurological disease. This study examined potential predictors of cognitive complaints in 275 consecutive patients referred to a university pain management center. Patients completed a brief set of self-report measures of problems with cognitive functioning, biographical information, pain severity, pain location, depression, anxiety, sleep quality, medication use, and litigation status during their first visit to the clinic. The most frequently reported cognitive complaints included forgetfulness (23.4%), minor accidents (23.1%), difficulty finishing tasks (20.5%), and difficulty with attention (18.7%). Fifty-four percent of patients reported at least one problem with cognitive functioning. Correlation analyses showed that using antidepressants, pain severity, pain-related anxiety, and depression were moderately associated with total cognitive complaints. Regression analyses showed that depression accounted for the largest unique proportion of variance in cognitive complaints ( R 2 29%). Given the high frequency of complaints of impaired cognitive functioning, this realm of functioning deserves routine assessment. When these complaints are encountered, a careful evaluation considering a range of neurological, social, and emotional influences is in order. J Pain Symptom Manage 2001; 21: U.S. Cancer Pain Relief Committee, Key Words Chronic pain, complaints, cognitive functioning, depression, emotional distress Address reprint requests to: Lance M. McCracken, PhD, The Pain Management Unit, The Royal National Hospital for Rheumatic Diseases, Upper Borough Walls, Bath BA1 1RL, United Kingdom. Accepted for publication: June 30, Introduction Patients seeking treatment for chronic pain often report difficulties with memory, concentration, and other cognitive functions. Schnurr and MacDonald 1 showed that patients seeking treatment for chronic pain report more memory problems than control subjects seeking general medical care, dental care, or psychotherapy. They also found that patients with pain from motor vehicle accidents who may have suffered mild head injuries did not report more memory problems than patients with pain from accidents unlikely to include head injury. Iverson and McCracken 2 found that 42.0% of patients seeking treatment for chronic pain who had no history of head injury reported at least one cognitive complaint (forgetfulness, difficulty maintaining attention, or difficulty with attention and thinking). A large U.S. Cancer Pain Relief Committee, /01/$ see front matter Published by Elsevier, New York, New York PII S (01)

2 Vol. 21 No. 5 May 2001 Impaired Cognitive Functioning in Chronic Pain Patients 393 majority (80.6%) of the total sample reported three or more relevant non-cognitive symptoms (i.e., fatigue, sleep problems, dizziness, irritability, etc.). Thirty-nine percent of the sample showed the symptom features of DSM-IV Postconcussional Disorder. 3 Cognitive complaints may mark the presence of truly impaired cognitive functioning or they may represent the patient s perception of impairment where none exists. Similarly, they may or may not result from some neuropathology. A report by Anderson and colleagues 4 showed that 7 of 67 consecutive patients referred to a pain rehabilitation program had undiagnosed brain injury. Patients with brain injury and chronic pain may require longer treatment 5 or may be more susceptible to treatment failure unless treatment addresses both the consequences of the brain injury and the chronic pain. 4 A range of causes may lead to reports of cognitive deficits. They may be a product of a primary neurological disorder, medications, the pain itself, preoccupation with injuries or ill health, sleep deprivation, stressful circumstances, some other emotional state, or a combination of these factors. Previous results suggest that depression may account for some but not all of these complaints. 1 Results from a study of general, physical, symptom complaints in persons with chronic pain showed that they are related to pain, anxiety, and depression. 6 Other results also suggest that pain itself may lead to cognitive complaints. These results come from a study in which pain relief from opioid therapy was associated with improved cognitive performance. 7 It is clear that patients seeking treatment for chronic pain are subjected to numerous medical, social, and psychological influences that may lead them to complain of difficulties with cognitive functioning. The purpose of this study is to examine a selected set of these factors to determine which best predict cognitive complaints. We explicitly examined a group of patients seeking treatment for chronic pain with no history of head injury or dementing illness to control those potential causes. We sampled a broader array of potential predictors than previous studies. These include demographic variables, pain severity, pain location, depression, anxiety, sleep quality, medication use, and litigation status. Based on previous studies we predicted that cognitive complaints would be uniquely associated with variables assessing emotional distress. Methods Subjects Subjects for this study were 275 consecutive patients referred to a university pain management center. Data were gathered during standard clinical assessment procedures. Patients who were at least 18 years old, could read and write English, and completed all measures for this study were selected. Twelve patients who reported a history of head trauma (n 9) or stroke (n 3) were excluded. Subject characteristics are provided in Table 1. Measures All subjects completed a number of selfreport measures as part of their initial evaluation in the clinic. They provided information about demographic and pain-related variables including chronicity, severity, location, medication use, and pain treatment history. Pain severity was assessed with a 100-mm visual analog measure with ends labelled no pain and worst pain possible. All subjects also completed the standard questionnaires listed below. Table 1 Sample Characteristics N 275 Gender 62.0% women Age M 46.6 years (SD 13.8) Marital status 53.8% married 22.9% never married Race/Ethnicity 75.3% White 20.7% Black 1.8% Asian 1.8% Hispanic 0.4% other Education M 13.9 years (SD 2.7) 59.3% 12 years Employment status 41.5% not working due to pain 22.2% working full time 10.2% retired 8.0% working part time due to pain 5.5% homemaker 12.6% other Pain location 54.9% low back 13.6% lower limbs 8.8% shoulder/upper limbs 7.3% cervical region 15.4% other Pain chronicity M 63.6 months (SD 80.3) Median 32.5 (range 3 372)

3 394 McCracken and Iverson Vol. 21 No. 5 May 2001 Table 2 Frequencies of Cognitive Complaints Forgetting a lot, recent things, appointments 23.4 Minor accidents 23.1 Not finishing things started 20.5 Not keeping attention on activity 18.7 Difficulty with concentration and thinking 17.2 Making mistakes 15.4 Difficulty reasoning and problem-solving 12.8 Confusion 9.9 Reacting slowly 9.2 Behaving confused or disoriented 9.2 Depression was assessed with the Beck Depression Inventory (BDI 8 ) a well-known and well-validated 21-item measure of common symptoms of depression. Pain-related anxiety was assessed with the Pain Anxiety Symptoms Scale (PASS 9,10 ). The PASS is a 40-item instrument assessing cognitive and physiological symptoms, avoidance, and fearful thinking associated with pain. Cognitive complaints and sleep quality were assessed with the Sickness Impact Profile (SIP 11 ). The SIP is a 136-item behaviorally-based checklist assessing the effects of illness on 12 categories of daily functioning. Patients are asked to endorse statements that describe them today in relation to their state of health. For purposes of this study we focused on the 10-item Alertness Behavior subscale and an item addressing sleeping problems ( I sleep less at night, for example, wake up too early, don t fall asleep for a long time, awaken frequently ). The Alertness Behavior subscale is a face valid inventory of potential cognitive problems in daily life including confusion, slowed reaction time, poor concentration, difficulty reasoning, forgetfulness, and inattention. Table 3 Predictors of Cognitive Complaints r P(2-tailed) Age.04 ns Education Gender (0 female, 1 male) Pain chronicity.05 ns Involved in litigation.04 ns Pain in head or neck.06 ns Using a narcotic analgesic Using a tranquilizer.08 ns Using an antidepressant Sleep disruption Pain severity (100 mm VAS) Pain-related anxiety Depression % Results The most frequently reported cognitive complaints included forgetfulness (23.4%), minor accidents (23.1%), difficulty finishing tasks (20.5%), and difficulty with attention (18.7%). Fifty-four percent of patients reported at least one problem with cognitive functioning. The frequency of patients endorsing each item of the Alertness Behavior subscale is shown in Table 2. Thirteen variables were examined to test their correlations with the total score for cognitive complaints (see Table 3). Age, chronicity of pain, involvement in litigation, the presence of pain in the head or neck, and using tranquilizing medication were not significantly associated with cognitive complaints. With P 0.05 as the significance criteria, education, gender, narcotic use, and sleep disruption showed small correlations with cognitive complaints. Those who are less educated, males, on narcotic analgesics, and suffering with sleeping problems reported more complaints. Using antidepressant medication, pain severity, pain-related anxiety, and depression showed moderate positive correlations with cognitive complaints. A Bonferroni correction would require a P ( 0.05/13) for significance in this correlation matrix. With that P-value as the criteria for significance, only using antidepressants, pain severity, pain-related anxiety, and depression were significantly associated with cognitive complaints. A multiple regression analysis was used to identify the set of variables that account for the most unique variance in cognitive complaints (see Table 4). All variables with significant correlations with cognitive complaints at P 0.05 were tested for entry (P to enter 005, P to remove 0.10) in the regression equation. The final set of selected variables included depression, antidepressant use, and pain-related anxi- Table 4 Results from Stepwise (Statistical) Multiple Regression Analysis Predicting Total Cognitive Complaints Predictor Beta R 2 P Total R 2 Depression Antidepressant use Pain-related anxiety Variables tested but not in equation: gender, education, narcotic use, sleep, and pain severity.

4 Vol. 21 No. 5 May 2001 Impaired Cognitive Functioning in Chronic Pain Patients 395 ety. These three variables accounted for 36.0% of the variance in cognitive complaints. Gender, education, opioid use, sleep disruption, and pain severity were not retained as significant, unique predictors. The instrument used in this study to assess depression, the BDI, includes one item concerning problems with decision-making. Since that is a cognitive complaint we examined its contribution to the correlation and regression results. The bivariate correlation between cognitive complaints and the BDI score minus this item was r 0.52 (compared to r 0.54 with item included). The R 2 for this recalculated score when it was entered in the regression analyses was 0.27 (compared to R ). Based on these results the contribution of this item to results was considered negligible. Discussion Perceived cognitive problems are common in persons with chronic pain. Of this sample of 275 patients, 54% endorsed at least one cognitive complaint. These results are consistent with a body of literature indicating that perceived cognitive dysfunction is a non-specific symptom complex common to many groups, including mental health outpatients, 12 general medical outpatients, 13,14 and personal injury claimants. 13,14 Depression and pain-related anxiety were significantly and uniquely related to complaints of cognitive problems in the present sample. Greater emotional distress was associated with a greater perception of cognitive dysfunction. These results are similar to results from studies of physical complaints in persons with chronic pain, showing they too are strongly associated with emotional distress. 6 They suggest that the many threatening, discouraging, and stressful circumstances faced by those with chronic pain contribute to the perception of impaired cognitive functioning, as well as complaints of collateral physical symptoms. These circumstances may be (a) directly pain-related; (b) related to general health, family problems, or financial problems; or (c) linked to some other social or emotional influences in the clinic or work environments. The finding that antidepressant use and sleep disruption also predict cognitive complaints certainly is consistent with the interpretation that emotional distress and related disruption of routines may be at the root of the problem. It is interesting to note that although there was a small to moderate association between pain severity and perceived cognitive problems (r 0.24), pain severity did not contribute unique variance in the prediction model. This result is somewhat at odds with the findings showing that pain relief from opioid analgesics is associated with improved cognitive performance. 7 During clinic visits patients often indicate that their memory and concentration abilities are hindered only when pain is severe. Our results suggest that in terms of perceived cognitive functioning, emotional responses provoked by pain may be more important than the pain itself. Therefore, reductions in pain severity may not result in perceived improvement in cognitive functioning unless it is accompanied by improvement in emotional functioning. Our results are also somewhat at odds with results from Sjogren et al. 15 Their results showed that patients with chronic pain on long-term opioid therapy showed reduced performance in vigilance and attention, psychomotor speed, and working memory tasks compared to pain-free controls on no opioids. Our results showed a non-significant relationship between cognitive complaints and opioid use. Their results considering the role of depression were mixed. They showed that depression predicted psychomotor speed but not the other performances assessed. The most obvious explanation for these differences is that our measure of cognitive complaints did not focus on specific domains as did their direct performance measures. Clearly, relations between cognitive impairment, opioid use, and depression in patients with chronic pain deserve further study. This study focused on cognitive complaints. As indicated earlier, patients who make these complaints may or may not also experience directly observable cognitive impairment, as measured by neuropsychological tests. These are separate but potentially related issues. Both deserve attention. For instance, patients who perceive themselves to be impaired may not feel secure engaging in normal activity and may withdraw. Once they voice complaints of cognitive impairment, they may be dissatisfied with treatment if these complaints are not ad-

5 396 McCracken and Iverson Vol. 21 No. 5 May 2001 dressed. They may blame their cognitive problems on medication and may then refuse to comply with appropriate prescriptions, even if they are mistaken. The methods of this study present some limitations. First, we did not directly measure cognitive performance. It is unclear how self-reported complaints would correlate with the actual performances of patients in this study. Predictors of complaints cannot be assumed to equally predict performances. Second, despite our exclusion of persons with known histories of head trauma or stroke, some patients in our study may suffer with identifiable brain impairment. Thus, our results cannot be assumed to apply purely to persons with no biologically based cognitive dysfunction. Finally, our measure of cognitive complaints included multiple domains of cognitive functioning but does not include a measure of perceived severity. Predictors of focused cognitive complaints reported in severe terms may differ from predictors of complaints of multiple cognitive problems. These results are preliminary. However, they may have implications for clinical evaluation and management of persons with chronic pain in the future. Given the high frequency of complaints of impaired cognitive functioning, this realm of functioning deserves routine assessment. When complaints are encountered, a careful history can examine their clinical significance and precipitating circumstances, and generate a list of causal influences that may underlie them. Neuropsychological evaluation may be in order. However, vague complaints of cognitive deficits that (a) have not produced any significant adverse consequences, (b) seem to vary over time in relation to psychological influences, and (c) occur without a history of head trauma or other risk factors suggesting underlying brain impairment may not need extensive testing and evaluation. Recognizing and treating the distress and suffering of some patients may reduce complaints of cognitive problems. References 1. Schnurr RF, MacDonald MR. Memory complaints in chronic pain. Clin J Pain 1995;11: Iverson GL, McCracken LM. Postconcussive symptoms in persons with chronic pain. Brain Injury 1997;11: American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 4th Edition. Washington, DC: American Psychiatric Association, Anderson MJ, Kaplan MS, Felsenthal GF. Brain injury obscured by chronic pain: a preliminary report. Arch Phys Med Rehabil 1990;71: Andary MT, Crewe N, Ganzel SK, et al. Traumatic brain injury/chronic pain syndrome: a case comparison study. Clin J Pain 1997;13: McCracken LM, Faber SD, Janeck AS. Painrelated anxiety predicts non-specific physical complaints in persons with chronic pain. Behav Res Ther 1998;36: Haythornthwaite JA, Menefee LA, Quartrano- Piacenti AL, Pappagallo M. Outcome of chronic opioid therapy for non-cancer pain. J Pain Symptom Manage 1998;15: Beck AT, Rush AJ, Shaw BF, Emery G. Cognitive therapy of depression. New York: Guilford, McCracken LM, Zayfert C, Gross RT. The Pain Anxiety Symptoms Scale: development and validation of a scale to measure fear of pain. Pain 1992;50: McCracken LM, Gross RT. The Pain Anxiety Symptoms Scale (PASS) and the assessment of emotional responses to pain. In: VandeCreek L, Knapp S, Jackson TL, eds. Innovations in clinical practice: a sourcebook, vol. 14. Sarasota, FL: Professional Resources Press, 1995: Bergner M, Bobbit RA, Carter WB, Gilson BS. The Sickness Impact Profile: development and final revision of a health status measure. Med Care 1981; 29: Fox DD, Lees-Haley PR, Earnest K, Dolezal- Wood S. Post-concussive symptoms: base rates and etiology in psychiatric patients. Clin Neuropsychologist 1995;9: Dunn JT, Lees-Haley PR, Brown RS, et al. Neurotoxic complaint base rates of personal injury claimants: implications for neuropsychological assessment. J Clin Psychol 1995;51: Lees-Haley PR, Brown RS. Neuropsychological complaint base rates of 170 personal injury claimants. Arch Clin Neuropsychol 1993;8: Sjogren P, Thomsen AB, Olsen AK. Impaired neuropsychological performance in chronic nonmalignant pain patients receiving long-term oral opioid therapy. J Pain Symptom Manage 2000;19: