- Email: [email protected]
The Effect of Neuroticism on the Recall of Persistent Low-Back Pain and Perceived Activity Interference
The Journal of Pain, Vol 14, No 9 (September), 2013: pp 948-956 Available online at www.jpain.org and www.sciencedirect.com
The Effect of Neuroticism on the Recall of Persistent Low-Back Pain and Perceived Activity Interference John C. Lefebvre* and Francis J. Keefey *Department of Psychology, Wofford College, Spartanburg, South Carolina. y Department of Psychiatry and Behavioral Sciences, Duke University Medical Center, Durham, North Carolina.
Abstract: The assessment of persistent pain often relies on recalling and then summarizing the entire pain experience using a single rating. Newer methodologies, such as the Original Pain Recall Assessment, ask people to recall the pain they experienced over a specific period of time by tracing a single line in a graph to represent their pain levels. One advantage of this approach is that one can compare recalled levels of pain with actual daily diary pain ratings. This methodology was used to investigate the effects of neuroticism on the recall of levels and patterns of persistent pain. The study involved 70 participants who completed a measure of neuroticism, depressive symptoms, and up to 15 daily diaries that asked for ratings of pain intensity, pain unpleasantness, and activity interference due to pain. Following completion of the daily diary period, the participants were asked to recall the entire diary period using the Original Pain Recall Assessment methodology. The analyses revealed that higher levels of neuroticism were related to significantly better recall of the variability of pain unpleasantness over time. Furthermore, individuals who reported higher levels of depressive symptoms were less accurate in the recall of pain in general. Perspective: Memory for pain is crucial in the assessment of pain, with little research devoted to the study of this topic. The current study demonstrates that people higher on neuroticism had better recall of pain unpleasantness, and people with higher levels of depressive symptoms had poorer recall of pain in general. ª 2013 by the American Pain Society Key words: Memory for pain, neuroticism, depression, low-back pain.
M
emory is a reconstructive process and is subject to the effects of a number of factors.18 However, the underlying assumption of pain assessment is that people are able to accurately recall past painful experiences over a given period of time. Early studies in the field focused predominantly on accuracy and showed that some participants were more accurate and others less so.15,16,21 These studies often assumed a small difference between scores,7,16 or a significant correlation16 as an indication of accuracy in the recall of pain. More recent studies focus on factors that might influence the accuracy of the recall of pain. These studies have suggested that factors such as pain ratings at both Received September 17, 2012; Revised February 21, 2013; Accepted March 14, 2013. The study was not supported by any grant funding sources. This study is the original work of the authors and neither author has any conflicts of interest or financial interest in the research. Address reprint requests to John C. Lefebvre, PhD, Department of Psychology, Wofford College, 429 N Church St, Spartanburg, SC 29303. E-mail: [email protected] 1526-5900/$36.00 ª 2013 by the American Pain Society http://dx.doi.org/10.1016/j.jpain.2013.03.006
948
the ‘‘peak and end’’ of recording periods27 and personality26 influence the recall of pain. A limitation of these earlier studies was the use of a single recall rating and not recognizing that an individual could also recall patterns of change in their pain experiences over time. The current study uses a novel approach to the study of memory for pain. It asks participants to recall the pattern of their pain by tracing a single continuous line in a graph to represent their daily pain levels over a 15-day period. Values from this line are then compared to daily ratings of pain. This methodology is important because it allows for the analysis of different aspects of memory for daily pain. It also permits one to analyze the accuracy of daily pain recall (ie, absolute differences between initially reported and recalled pain ratings) and to assess the recall of variability in the daily pain experience (ie, patterns of change in pain ratings).14 Finally, this study examined the recall of 3 aspects of daily pain— pain intensity, pain unpleasantness, and perceived activity interference due to pain—rather than focusing on recall of pain intensity only. Neuroticism is a fundamental aspect of personality characterized as a general tendency to experience
Lefebvre and Keefe 6
negative affect. A number of studies suggest that higher levels of neuroticism are associated with reports of higher levels of acute experimental pain and persistent pain.1,8 In addition, studies show that higher levels of neuroticism are related to higher reports of somatic symptoms and to a similar bias in the recall of these symptoms.12,13 Because neuroticism is associated with consistent biases in both the report and recall of symptoms, it should lead to people being less accurate in their recall of pain. Interestingly, to date, only 1 study of persons with persistent pain has examined this relationship. It found no significant relationship between neuroticism and the recall of pain intensity and pain unpleasantness.26 However, this study relied on single measures of pain intensity and unpleasantness. The current study is one of a small number of studies that assess the variability of recalled pain over a series of days.14,19 In studying the influence of neuroticism on the recall of pain, it is also important to consider the effects of depression.6,10,17,20 Depression is common in persons with persistent pain and is known to influence the reports of pain.1,29 Given this relationship, and depression’s effect on recall in general,5 this study measured and controlled for the effect of the severity of depressive symptoms on the recall of daily pain intensity, pain unpleasantness, and pain interference. The current study hypothesized that after controlling for the severity of depressive symptoms, 1) neuroticism would be related to poorer accuracy between recalled and actual levels of pain and pain interference, and 2) neuroticism would be related to lower levels of agreement between recalled and actual day-to-day variability of pain and pain interference.
Methods Participants Participants for the study were recruited using advertisements and direct solicitations from clinics at the Duke University Medical Center as well as from the local community. All participants in the study reported experiencing low-back pain for at least 6 months prior to participation in the study. All participants were at least 18 years of age and were currently experiencing low-back pain on a daily basis. Of the 96 people who agreed to participate in the study, 10 participants failed to complete the study by failing to return for the experimental session. Participants with moderate to high levels of depression (Beck Depression Inventory 1 [BDI-1] score of 15 or greater2,9) were screened out of the study (n = 16) because moderate to high levels of depression have been found to have a significant influence on memory.5 Data in the present study were thus derived from 70 participants (40 women and 30 men). The sample was primarily middle aged (average age = 43.89 years, SD = 12.5, range = 18–71) and had experienced low-back pain for a significant period of time (close to 10 years on average). The sample was predominantly Caucasian
The Journal of Pain
949
(85.7%) and married (65.7%). The majority of the participants had not undergone surgery for their low-back pain (75.7%) and were not receiving disability payments (88.6%). The study was approved by the institutional review board of Duke University and all participants provided informed consent to participate in the study. Participants were compensated up to $16.25 for participation.
Procedures The study was divided into 3 phases: an orientation phase, a 15-day diary phase, and a laboratory recall phase. During the orientation phase, information on demographic variables was collected and participants completed baseline measures of pain interference, neuroticism, and depression. During the diary phase, participants were asked to complete a structured daily diary at the end of each of 15 days. The diary consisted of standardized measures of pain intensity, pain unpleasantness, and perceived activity interference due to pain. During the recall phase, the participants came to a laboratory setting and were asked to recall their pain intensity, pain unpleasantness, and activity interference due to pain using 3 separate Original Pain Recall Assessment (OPRA) forms (described below). At this time, the participants also provided ratings of their current pain intensity and pain unpleasantness using a mechanical visual analog scale (VAS) whose endpoints were identical to those used in the daily diary.23
Measures Diary-Based Measures Daily pain intensity and pain unpleasantness were measured using 100-mm VASs. Daily pain intensity was measured using a VAS whose endpoints were ‘‘no pain sensation’’ and ‘‘most intense pain sensation imaginable.’’ Daily pain unpleasantness used a VAS whose endpoints were ‘‘not at all unpleasant’’ and ‘‘most unpleasant imaginable.’’ Price et al24 have demonstrated that VASs can reliably differentiate the sensory-intensity and the affective-unpleasantness dimensions of pain. Daily activity interference due to low-back pain was assessed by asking participants to rate the relative degree of interference on a 100-mm VAS anchored by ‘‘no interference’’ and ‘‘could not do daily activities.’’ This question has been shown to be reliable when used as part of a protocol to grade the severity of chronic pain.33 A key requirement of diary studies is that participants record each day’s data at the end of that day.1,31 To reinforce this expectation, participants were asked to mail each day’s diary back the following morning. As an added incentive to return the diaries on time, each participant was paid $.25 for each completed diary returned within a 48-hour period, and an extra $1.25 if all 15 diaries were returned. Of the 1050 possible diaries to be completed (15 days by 70 participants), 972 daily diaries were completed and returned, representing a
950
The Journal of Pain
93.6% return rate. Missing diary data led to omission of recalled values for those days in subsequent analyses. Following the 15-day diary period, each participant was asked to return for a recall session on the day immediately following the completion of the final daily diary. Several participants came in the day after completing their final diary (31%), with 79% of participants completing the recall assessment within 7 days of their final diary. Given that there were variations in the length of time between diary completion and the recall session, this variable was used as a control variable in all analyses.
Original Pain Recall Assessment Form During the laboratory recall phase, participants were asked to recall the level of pain intensity, pain unpleasantness, and the amount of activity interference caused by low-back pain over the course of the 15-day diary period. Each participant completed separate OPRA forms, 1 each for pain intensity, pain unpleasantness, and pain interference. On the OPRA, participants are asked to draw a single continuous line from left to right that represented the pattern of their pain level over the entire diary period. To illustrate this approach, Fig 1A provides a tracing obtained from a study participant who was asked to recall his or her pain intensity on the OPRA. The OPRA ratings procedure was repeated for an assessment of recall of pain unpleasantness and for an assessment of recall of pain activity interference. MacDonald et al19 have demonstrated that the OPRA is a reliable and valid method of assessing memory for pain in a persistent pain sample. The lines representing recall of pain intensity, pain unpleasantness, and pain interference were scored using a template. The scoring template consisted of a series of
Neuroticism, Depression, and the Recall of Pain 15 vertical 100-mm VAS meridians, spaced 10 mm apart. Each meridian represented 1 of the 15 days of the diary period. The scoring template was placed over the recalled pattern using a see-through acetate template. Fig 1B demonstrates the same tracing in Fig 1A with the scoring template superimposed. Recalled pain level for each day was defined as the distance in millimeters from the lower endpoint of the VAS (bottom of the box) to the point where the recalled pattern crossed the corresponding meridian.
Questionnaire Measures Neuroticism Subscale of the NEO Personality Inventory–Revised. Neuroticism was measured by the 48 items that make up the neuroticism subscale of the NEO Personality Inventory–Revised (NEO-PI-R6). Participants were asked to indicate the extent to which they agreed with each of the items using a 5-point scale (strongly disagree, disagree, neutral, agree, strongly agree). A global neuroticism score was obtained by summing scores on the 6 facets of neuroticism including depression, anxiety, hostility, self-consciousness, impulsiveness, and vulnerability. The global score for neuroticism has been found to have good internal consistency, with a reported Cronbach alpha of .92.6 Beck Depression Inventory 1. BDI-1 is a 21-item scale that covers the symptoms and attitudes associated with depression.3 Each item contains a graded series of 4 self-evaluative statements, representing a behavioral manifestation of depression. The statements are weighted (0–3) to reflect the range of severity of the symptom from neutral to maximally severe. Each participant was asked to select 1 statement per item to describe how he/she felt over the preceding week. The weights of the selected statements were summed to provide a depression score, with higher
Figure 1. (A) OPRA with an example of a diary recall tracing for pain intensity. (B) OPRA with an example of a diary recall tracing for pain intensity with superimposed scoring template. (C) An example of the resulting recalled and diary values for pain intensity used in the analysis of recall of pain level and recall of pain variability.
Lefebvre and Keefe scores representing higher levels of depression. The scale has been shown to have good internal and split-half reliability.3 Usual Activity Interference Due to Pain. To assess their usual level of activity interference due to pain, each participant was asked to rate the degree to which pain interfered with their daily activities over the previous 6 months on an 11-point scale (0 = no interference, 10 = unable to carry on any activities).33 This is the same scale that was used for the diaries except that the response scale was not a VAS but a verbal numeric scale.
Demographic and Medical History Measures Each participant completed a form that provided his or her age, gender, ethnic group, marital status, and duration of disease. In addition, participants were asked if they had undergone surgery to alleviate low-back pain, if they were currently receiving disability payments, and to describe their current treatment regimen for low-back pain.
Data Analysis Data analysis for the current study progressed from univariate statistics that assessed the mean level of all of the variables used in subsequent analyses. In addition, the data analyses also tried to match the analyses that are common in previous research on the recall of pain. In the univariate section, the study assessed if there was a significant difference between the diary and recalled values for pain intensity, pain unpleasantness, and pain interference. In terms of bivariate analyses, separate series of Pearson product-moment correlations were conducted to assess the relationship between the diary and recalled values of pain, the relationship between neuroticism and depression to the diary and recalled values of pain, and the interrelationship between the independent, covariate, and dependent variables. Past research into the recall of pain has primarily focused on a single measure, usually the difference between a single measure of ‘‘actual’’ pain and the associated single measure of recalled pain. However, individuals may also recall and report on the general pattern of the pain (ie, whether it fluctuates over time). The OPRA provides a means of assessing a wider variety of dimensions to the recall of pain. One benefit of the OPRA methodology is that it provides a set of matched pairs for each day of the diary period. Thus, one could measure recall by subtracting 1 value from its matched pair. Unlike traditional measures, the number of assessment points for both recalled and daily pain ratings provides greater stability in these mean values. The methodology also provides a means of determining if the individual is accurate in recalling the day-to-day variations over time by comparing the agreement between the matched pairs over the 15-day diary period. An example of the resulting matched recall and diary pairs is provided in Fig 1C. Based on the benefits of the OPRA, the current study had 2 principal
The Journal of Pain
951
dependent measures for all analyses: 1) recall of pain level and 2) recall of pain variability. The recall of pain level measure provided an estimate of the absolute difference between the recalled and actual pain measures. For each of the 3 dependent variables (daily pain intensity, daily pain unpleasantness, and daily pain interference), the recall of pain level measure was based on the absolute difference between the recalled and diary ratings. To calculate the measure, a series of daily difference scores was calculated by subtracting each available day’s diary ratings of pain and interference from the corresponding recalled levels of pain and interference. The absolute value of these difference scores was then created. The average of these absolute scores was then used to form a single measure. Higher scores on the recall of pain level measure suggest greater inaccuracy in the recall of pain and activity interference. The recall of pain variability measure provided an estimate of the degree of agreement between the recalled values of pain and interference and the corresponding diary ratings. Given the nature of the OPRA methodology, the multiple recalled and diary ratings provide a means of assessing the agreement of the recalled and daily patterns across time. The recall of pain variability measure was calculated using the intraclass correlation (ICC) between the recalled and diary ratings of pain and interference for each participant over time. The ICC was calculated based on a 2-way mixed consistency model using the average values in order to provide greater stability.19 As shown in Fig 1C, the closer the 2 patterns agree with each other, the higher the ICC and the better the recall of pain variability. Finally, 2 regression analyses were conducted to assess the degree to which neuroticism was related to diary values and recalled values of pain level for pain intensity, pain unpleasantness, and pain interference. Because of the importance of depressive symptoms to the report of pain, both regression analyses controlled for the effect of depressive symptoms as well as covariates that are typically assumed to be involved in memory for pain such as the number of days between the end of the diary period and the evaluation session (lapse) and the duration of pain. In addition, for each model, the current level of pain intensity or pain unpleasantness was added to the model given the robust finding that the current level of pain is related to the recall of pain.7 Usual level of activity interference due to pain, assessed at the orientation phase, was included as a covariate in the analysis predicting the recall of pain interference as a control for baseline interference.
Results Univariate Descriptive Analyses Table 1 presents the descriptive statistics for the dependent and questionnaire measures. As can
952
Neuroticism, Depression, and the Recall of Pain
The Journal of Pain
be seen from Table 1, participants reported low to moderate levels of pain intensity and unpleasantness during both the diary phase and the recall phase. The traditional approach to the assessment of accuracy of the recall of pain has been to assess if there is a significant difference in the levels of pain recalled and reported in the diary. If we were to adopt this approach, there was a significant difference between the average level of pain intensity reported in the diaries and those recalled on the OPRA (t(69) = 2.35, P = .02, d = .28), but no significant differences between the average diary reports and recalled reports for pain unpleasantness (t(69) = .92, P = .36, d = .11) or activity interference (t(69) = 1.01, P = .32, d = .12). Thus, according to this more traditional approach, the sample was relatively accurate in terms of recalling pain unpleasantness and activity interference, but was inaccurate in terms of recalling pain intensity. On average they recalled their pain intensity to be significantly lower than was reported in the daily diary. In terms of the recall of pain variability measure, the results show that the ICCs between recalled and initial diary pain ratings were positive but weak. Despite the fact that overall level of agreement between recalled and initial diary pain ratings was low, considerable variability was noted, with some individuals being more accurate in recalling the pattern of their pain experience over time than others.
Descriptive Findings for Dependent and Questionnaire Measures
Table 1.
MEAN
SD
RANGE
Diary values (aggregated over 15 days) Pain intensity 35.45 20.58 3.5–81.5 Pain unpleasantness 34.37 20.57 .6–83.3 Pain interference 29.52 21.37 .0–91.5 Recalled values (aggregated over 15 days) Pain intensity 32.18 21.12 2.2–79.3 Pain unpleasantness 33.11 23.42 1.6–89.2 Pain interference 27.99 24.66 .2–91.7 OPRA analyses Recall of pain level (absolute difference between pairs) Pain intensity 15.09 7.69 1.5–47.7 Pain unpleasantness 15.54 8.12 1.4–40.7 Pain interference 14.80 10.08 .2–56.9 Recall of pain variability (ICC) Pain intensity .21 .47 1.00–.95 Pain unpleasantness .19 .36 .81–.96 Pain interference .19 .35 .48–.97 Questionnaire measures Neuroticism 115.90 18.43 76.0–161.0 Depressive symptoms 6.89 3.72 .0–14.0 Usual activity interference 4.60 2.49 .0–9.0 Duration of pain (months) 106.26 121.31 6.0–468.0 Current pain intensity 23.23 21.16 .0–72.0 Current pain unpleasantness 21.77 22.16 .0–85.0 Days since completion of diary 5.34 5.76 1.0–30.0 period (Lapse)
Bivariate Analyses: The Relationship of Neuroticism and Depression to Mean Values of Recalled and Diary Ratings of Pain Intensity, Pain Unpleasantness, and Pain Interference Levels A series of correlation analyses was conducted to assess the relationship between the mean recall levels of pain intensity, pain unpleasantness, and pain interference and the mean diary values for each of these variables. The results showed a significant correlation between the mean recalled and diary levels of pain intensity (r = .84, P < .001), pain unpleasantness (r = .88, P < .001), and activity interference (r = .86, P < .001). As with the difference scores, more traditional studies would have concluded that the significant correlations suggest that, as a whole, these individuals were accurate in their recall in all 3 aspects of their pain experience. Correlation analyses were conducted to assess the relationship between neuroticism and mean recalled levels of pain intensity, pain unpleasantness, and pain interference. The analyses showed that neuroticism was not significantly correlated with any of the mean values of the recalled pain measures. A series of correlation analyses was also conducted to assess the relationship between neuroticism and mean diary ratings of pain intensity, pain unpleasantness, and pain interference. Once again, neuroticism was not significantly correlated with any of the mean values of the daily diary pain measures. Similar correlational analyses were conducted to examine the relationship of the severity of depressive symptoms to the recall of pain. These analyses showed that depressive symptoms were significantly correlated with mean recalled pain intensity (r = .34, P = .004), mean recalled pain unpleasantness (r = .35, P = .003), and mean recalled pain interference (r = .38, P = .001). Correlational analyses were also conducted to assess the relationship of the severity of depressive symptoms to the report of daily pain. These analyses found that the severity of depressive symptoms was significantly correlated with mean diary ratings of pain intensity (r = .33, P = .006), mean diary ratings of pain unpleasantness (r = .35, P = .003), and mean diary ratings of pain interference (r = .38, P = .001). Finally, a series of correlational analyses was conducted to assess the interrelationship among the variables used in the analyses. The results showed that neuroticism was significantly correlated with the degree of depressive symptoms (r = .47, P = .001). In terms of the diary values, pain intensity was significantly related to pain unpleasantness (r = .96, P = .001) and activity interference (r = .90, P = .001). Daily pain unpleasantness was significantly related to daily pain interference (r = .93, P = .001). A similar pattern was found for the recalled values, with pain intensity significantly related to pain unpleasantness (r = .94, P = .001) and pain interference (r = .84, P = .001). Finally, recalled values for pain unpleasantness were significantly correlated to recalled values of pain interference (r = .85, P = .001). Taken together, in terms of the recall of pain, these results suggest that neuroticism is not significantly
Lefebvre and Keefe
The Journal of Pain
correlated with the average level of both diary and recalled values of pain intensity, pain unpleasantness, and pain interference derived from the OPRA. The results did indicate that participants who reported higher levels of depressive symptoms were significantly more likely to recall having higher levels of pain intensity, pain unpleasantness, and pain interference, as well as report higher levels of daily pain intensity, pain unpleasantness, and pain interference.
Multivariate Analyses: The Relationship of Neuroticism to Recall of Pain Level The results of the regression analyses are presented in Table 2. As can be seen, neuroticism was not significantly correlated with the accuracy of recall of pain intensity, pain unpleasantness, or pain interference on the OPRA. Interestingly, the regression analyses did find a significant relationship between baseline severity of depressive symptoms with all 3 recall of pain level measures on the OPRA. Patients with greater severity of depressive symptoms showed poorer recall accuracy (greater absolute difference) when comparing the diary and recalled values on all 3 measures. A second regression analysis was conducted to assess the degree to which neuroticism was related to the recall of pain variability measure for pain intensity, pain unpleasantness, and pain interference. These analyses included the same set of variables used in the previous regression analyses. The results of the regression analysis are presented in Table 3. As can be seen, neuroticism was significantly related to the concordance between diary and recalled values in pain unpleasantness. Patients who scored higher on neuroticism showed greater concordance in diary and recalled fluctuations in pain unpleasantness. Neuroticism was not related to the
Results of Regression Analyses Relating Neuroticism and Evaluation Measures to the Recall of Pain Level Measures
Table 2.
R2 VARIABLE
TOTAL
DF
F RATIO b ON 2 FOR R ENTRY SIGNIFICANCE
Pain intensity .13 (5, 64) 1.97 Lapse (days since last diary) .17 Duration of pain .17 Current pain intensity .06 Depression .26 Neuroticism .13 Pain unpleasantness .11 (5, 64) 1.57 Lapse (days since last diary) .14 Duration of pain .07 Current pain unpleasantness .03 Depression .32 Neuroticism .08 Pain interference .32 (5, 64) 7.54 Lapse (days since last diary) .40 Duration .13 Usual activity interference .15 Depression .27 Neuroticism .10
.15 .15 .61 .04 .34 .28 .55 .88 .03 .58 .001 .21 .13 .04 .41
953
concordance in day-to-day variations in pain intensity or pain interference.
Discussion The present study investigated the effects of neuroticism and depression on the degree to which participants suffering from low-back pain could accurately recall their daily experience of pain intensity, pain unpleasantness, and perceived activity interference. The study had 2 major findings. First, patients scoring higher on neuroticism were significantly more accurate in their recall of the variability of pain unpleasantness over time. Second, patients who reported more severe depressive symptoms were less accurate in terms of recalling pain intensity, pain unpleasantness, and perceived activity interference. One of the most interesting findings of this study was the significant relationship between neuroticism and recall of day-to-day variations in pain unpleasantness. The literature suggests that higher levels of neuroticism are associated with a bias in report and recalling symptoms as being more severe. The current study found that neuroticism was related to the agreement in patterns between recalled pain unpleasantness and actual pain unpleasantness but was not related to a difference score between recalled and diary levels of pain unpleasantness. Thus, it may be that people with persistent pain who are higher in neuroticism may not be accurate in their recall of pain unpleasantness if assessed with a single value, but may be better at recalling variability of pain unpleasantness over time. This suggests that neuroticism may be related to noticing relative changes in pain level rather than absolute pain levels.26 One could assess how changes in pain are recalled by people higher on neuroticism compared to the recall of a stable level of pain. One could also assess
Results of Regression Analyses Relating Neuroticism and Evaluation Measures to the Recall of Pain Variability Measures
Table 3.
R2 VARIABLE
TOTAL
DF
F RATIO b ON 2 FOR R ENTRY SIGNIFICANCE
Pain intensity .10 (5, 64) 1.37 Lapse (days since last diary) .27 Duration of pain .10 Current pain intensity .13 Depression .10 Neuroticism .01 Pain unpleasantness .14 (5, 64) 2.01 Lapse (days since last diary) .07 Duration of pain .10 Current pain unpleasantness .17 Depression .02 Neuroticism .27 Pain interference .02 (5, 64) .30 Lapse (days since last diary) .06 Duration of pain .07 Usual activity interference .10 Depression .02 Neuroticism .08
.03 .42 .30 .49 .93 .53 .40 .16 .89 .05 .63 .57 .44 .91 .60
954
The Journal of Pain
how fluctuations in pain lead to the increase or decrease in the use of specific coping strategies. Future empirical studies might be able to manipulate pain level and then assess the recall of pain among those higher and lower on neuroticism. Because it provides a series of scores for the recall of pain, the OPRA provides a closer approximation between the recall and diary assessment methods. In prior research studies, diary techniques were used to assess the daily level of pain with a great temporal resolution. However, the recall of pain did not match the resolution of the assessment technique. Rather, recall is assessed by having the participant complete a single VAS to represent his or her usual pain intensity over the diary period. Thus, there is a qualitative and quantitative difference between how pain is assessed and recalled. The OPRA also provides a means of assessing accuracy in the recall of pain using the methodology of traditional studies: difference between mean actual and recalled pain scores and correlation between actual and recalled pain scores. Although these results may appear revealing, they may distort the reality of the relationship by simplifying a very complex process into a comparison between single values across individuals. These approaches also fail to determine what consistent patterns exist within the data that led to these values. Newer approaches in the study of the recall of pain have focused less on determining accuracy and more on assessing what influences the variability in the recall of pain between individuals. To this end, the OPRA methodology goes beyond the traditional approaches by providing more information about how the relationship of daily pain ratings and recalled pain ratings varies over time. It is interesting to note that the results of the current study found the relationship between neuroticism to be stronger for the recall of pain unpleasantness. These results of this study are in line with the study by Harkins and colleagues,8 which found that neuroticism was significantly related to the report of acute and clinical pain unpleasantness and not pain intensity. In another study, Wade et al34 found that neuroticism was again only related to pain unpleasantness and not pain intensity among a group of chronic pain patients. Taken together, the results of the current study when combined with the results of the previous studies suggest that neuroticism appears to favor a focus on the emotional aspects of pain, in terms of both the report of pain unpleasantness and the recall of the unpleasantness of this pain. The precise mechanisms by which neuroticism affects memory for chronic pain remain uncertain. One view proposes that individuals higher on neuroticism are more somatically hypervigilant, constantly searching for early warning signs of disease or increased pain.35 This would suggest that people who are higher on neuroticism would be more focused on their pain and devote more attention resources to monitor their pain. Thus, one possibility is that higher levels of neuroticism are involved in slowing the process of forgetting details about the chronic pain experience. Forgetting has been
Neuroticism, Depression, and the Recall of Pain hypothesized to be the loss of detail about stimulus attributes over time.28 Perhaps the somatic hypervigilance that is characteristic of neuroticism increases the salience of pain attributes and thus slows the forgetting process, leading to better recall. An interesting, though expected, result was the relationship of depressive symptoms to the recall of pain. There is an extensive literature on the effects of depression on memory.4,5,32 However, the specific relationship between depression and memory for pain has not been fully researched. In fact, most of the research available discusses the impact of depression on the recall of personal attributes and cognitive bias among patients with persistent pain.22 The current study is novel in that it has demonstrated that the severity of depressive symptoms does have an effect on the accuracy of the recall of pain but not the recall of pain variability. Even after controlling for severity of depressive symptoms through a limit imposed on the BDI-1 and the statistical control of the variable, the relationship is clear: those individuals who show higher levels of depressive symptoms are more inaccurate in the recall of pain. The severity of depressive symptoms was included as a covariate in order to isolate its effect on memory. Further research is needed to assess the degree to which clinical depression is related to the recall of pain. There are some limitations to this study. First, the study relies on self-report data both for the recall of pain and interference and also for the to-be-remembered stimuli. In memory research, the participants are exposed to a standard set of stimuli and then the recall of this material can be easily quantified at a later time. Given the subjective nature of pain, this approach is difficult but not impossible. A study by Rainville and colleagues25 demonstrates that methodology such as the matching to sample paradigm can be used in the study of memory for pain. This methodology, commonly used in the study of memory, allows for the assessment of memory without the use of verbal descriptions. This approach is important because pain is a nonverbal memory and the recall of verbal or numeric pain ratings may interfere with the recall of pain sensations. Second, in this study the daily diary data on pain were collected using pencil-and-paper diaries. The methodology was to have the participants complete the diary at the end of the day and return the diary by mail. This approach does not guarantee that the person completed the diary on time, and some diaries were not returned at all. Newer technology using portable electronic assessments offer the possibility of collecting data in a more systematic and consistent fashion.26,30 These methods may also afford better temporal resolution in the assessment of the recall of pain by placing the assessment and recall tasks in closer proximity. This approach could assess if certain days (ie, higher levels of pain) are better recalled than others. Third, the results show a high level of variability in the recall of pain. The degree of variability may be due to individual differences, or it could be that the OPRA methodology requires more refinement. Fourth, the current study did not assess and control for the effects
Lefebvre and Keefe of anxiety on memory for pain. Previous research suggests that anxiety related to dental procedures may affect memory for pain.11 Future research may need to explore the relationship of pain anxiety and worry about pain to memory for pain. Fifth, although neuroticism was found to be related to the recall of pain unpleasantness, it was not related to diary reports of pain intensity or pain unpleasantness. Neuroticism is usually found to be associated with reports of pain. As suggested above, neuroticism may be more related to changes in pain rather than single reports of usual pain. Finally, the study failed to assess the medications that the participants were taking at the time of the study. Some participants may have been on medications that could affect their memory in general. The current study offers a vista on a new and potentially exciting field of research, not only in terms of the memory for pain but also in the recall of other
References 1. Affleck G, Tennen H, Urrows S, Higgins P: Neuroticism and the pain-mood relation in rheumatoid arthritis: Insights from a prospective daily study. J Consult Clin Psychol 60: 119-126, 1992
The Journal of Pain
955
somatic sensations. The vast majority of the research conducted in order to study the mechanisms of memory and pain is focused on verbal memory. Methodologies such as the one presented in this study may be useful in better understanding how memory functions in the recall of somatic sensations.
Acknowledgments The authors of the study would like to thank David Caldwell, MD, Bruno Urban, MD, William Spillane, MD, John Goreki, MD, Lloyd Hey, MD, Diane Scott, MD, and William Richardson, MD, for their assistance in recruiting patients to participate in this study. The authors would also like to thank 3 anonymous reviewers for their insightful comments on a previous draft of this manuscript.
14. Lefebvre JC, Keefe FJ: Memory for pain: The relationship of catastrophizing to the recall of daily rheumatoid arthritis pain. Clin J Pain 18:56-63, 2002 € testam KG: A clinical comparison of two 15. Linton SJ, Go pain scales: Correlation, remembering chronic pain, and a measure of compliance. Pain 17:57-65, 1983
2. Beck AT, Beck RW: Screening depressed patients in family practice: A rapid technic. Postgrad Med 52:81-85, 1972
16. Linton SJ, Melin L: The accuracy of remembering chronic pain. Pain 13:281-285, 1982
3. Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J: An inventory for measuring depression. Arch Gen Psychiatry 4: 561-571, 1961
17. Lloyd GG, Lishman WA: Effect of depression on the speed of recall of pleasant and unpleasant experiences. Psychol Med 5:173-180, 1975
4. Blaney PH: Affect and memory: A review. Psychol Bull 99: 229-246, 1986
18. Loftus EF: Our changeable memories: Legal and practical implications. Nat Rev Neurosci 4:231-234, 2003
5. Burt DB, Zeembar MJ, Niederehe G: Depression and memory impairment: A meta-analysis of the association, its pattern, and specificity. Psychol Bull 117:285-305, 1995
€ jmark M: Recon19. MacDonald S, Linton SJ, Jansson-Fro structing the pain on the Original Pain Recall Assessment form (OPRA). J Pain 10:809-817, 2009
6. Costa PT, McCrae RR: NEO PI-R: Professional manual. Odessa, FL, Psychological Assessment Resources, Inc, 1992
20. Martin M: Neuroticism as predisposition towards depression: A cognitive mechanism. Pers Individ Diff 6: 353-365, 1985
7. Eich E, Reeves JL, Jaeger B, Graff-Radford SB: Memory for pain: Relation between past and present pain intensity. Pain 23:375-379, 1985
21. Morley SJ: Reliability and accuracy of memory for pain. Pain 57:132, 1994
8. Harkins SW, Price DD, Braith J: Effects of extroversion and neuroticism on experimental pain, clinical pain, and illness behavior. Pain 36:209-218, 1989 9. Haslam N, Beck AT: Categorization of major depression in an outpatient sample. J Nerv Ment Dis 181:725-731, 1993 10. Kendler KS, Kuhn J, Prescott CA: The interrelationship of neuroticism, sex, and stressful life events in the prediction of episodes of major depression. Am J Psychiatry 161:631-636, 2004 11. Kent G: Memory for dental pain. Pain 21:187-194, 1985 12. Larsen RJ: Neuroticism and selective encoding and recall of symptoms: Evidence from a combined concurrentretrospective study. J Pers Soc Psychol 62:480-488, 1992 13. Larsen RJ, Kasimatis M: Day-to-day physical symptoms: Individual differences in the occurrence, duration, and emotional concomitants of minor daily illnesses. J Pers 59: 387-423, 1991
22. Pincus T, Morley SJ: Cognitive-processing bias in chronic pain: A review and integration. Psychol Bull 127:599-617, 2001 23. Price DD, Bush FM, Long S, Harkins SW: A comparison of pain measurement characteristics of mechanical visual analog and simple numerical rating scales. Pain 56: 217-226, 1994 24. Price DD, McGrath PA, Rafii A, Buckingham B: The validation of visual analog scales as ratio scale measures for chronic and experimental pain. Pain 17:45-56, 1983 25. Rainville P, Doucet JC, Fortin MC, Duncan GH: Rapid deterioration of pain sensory-discriminative information in short-term memory. Pain 110:605-615, 2004 26. Raselli C, Broderick JE: The association of depression and neuroticism with pain reports: A comparison of momentary and recalled pain assessment. J Psychosom Res 62:313-320, 2007 27. Redelmeier DA, Kahneman D: Patients’ memories of painful medical treatments: Real-time and retrospective
956
The Journal of Pain
evaluations of two minimally invasive procedures. Pain 66: 3-8, 1996 28. Riccio DC, Rabinowitz VC, Axelrod S: Memory: When is less more. Am Psychol 49:917-926, 1994 29. Romano JM, Turner JA: Chronic pain and depression: Does the evidence support a relationship? Psychol Bull 97: 18-34, 1985
Neuroticism, Depression, and the Recall of Pain 32. Veiele HOF: A preliminary profile of neuropsychological deficits associated with major depression. J Clin Exp Neuropsychol 19:587-603, 1997 33. Von Korff M, Ormel J, Keefe FJ, Dworkin SF: Grading the severity of chronic pain. Pain 50:133-149, 1992
30. Stone A, Shiffman S: Ecological Momentary Assessment (EMA) in behavioral medicine. Ann Behav Med 16:199-202, 1994
34. Wade JB, Dougherty LM, Hart RP, Rafii A, Price DD: A canonical correlation analysis of the influence of neuroticism and extraversion on chronic pain, suffering, and pain behavior. Pain 51:67-73, 1992
31. Stone A, Kessler R, Haythornthwaite J: Methodological considerations in the measurement of daily events and experiences. J Person 59:575-607, 1991
35. Watson D, Pennebaker JW: Health complaints, stress, and distress: Exploring the central role of negative affectivity. Psychol Rev 96:234-254, 1989
The Effect of Neuroticism on the Recall of Persistent Low-Back Pain and Perceived Activity Interference John C. Lefebvre* and Francis J. Keefey *Department of Psychology, Wofford College, Spartanburg, South Carolina. y Department of Psychiatry and Behavioral Sciences, Duke University Medical Center, Durham, North Carolina.
Abstract: The assessment of persistent pain often relies on recalling and then summarizing the entire pain experience using a single rating. Newer methodologies, such as the Original Pain Recall Assessment, ask people to recall the pain they experienced over a specific period of time by tracing a single line in a graph to represent their pain levels. One advantage of this approach is that one can compare recalled levels of pain with actual daily diary pain ratings. This methodology was used to investigate the effects of neuroticism on the recall of levels and patterns of persistent pain. The study involved 70 participants who completed a measure of neuroticism, depressive symptoms, and up to 15 daily diaries that asked for ratings of pain intensity, pain unpleasantness, and activity interference due to pain. Following completion of the daily diary period, the participants were asked to recall the entire diary period using the Original Pain Recall Assessment methodology. The analyses revealed that higher levels of neuroticism were related to significantly better recall of the variability of pain unpleasantness over time. Furthermore, individuals who reported higher levels of depressive symptoms were less accurate in the recall of pain in general. Perspective: Memory for pain is crucial in the assessment of pain, with little research devoted to the study of this topic. The current study demonstrates that people higher on neuroticism had better recall of pain unpleasantness, and people with higher levels of depressive symptoms had poorer recall of pain in general. ª 2013 by the American Pain Society Key words: Memory for pain, neuroticism, depression, low-back pain.
M
emory is a reconstructive process and is subject to the effects of a number of factors.18 However, the underlying assumption of pain assessment is that people are able to accurately recall past painful experiences over a given period of time. Early studies in the field focused predominantly on accuracy and showed that some participants were more accurate and others less so.15,16,21 These studies often assumed a small difference between scores,7,16 or a significant correlation16 as an indication of accuracy in the recall of pain. More recent studies focus on factors that might influence the accuracy of the recall of pain. These studies have suggested that factors such as pain ratings at both Received September 17, 2012; Revised February 21, 2013; Accepted March 14, 2013. The study was not supported by any grant funding sources. This study is the original work of the authors and neither author has any conflicts of interest or financial interest in the research. Address reprint requests to John C. Lefebvre, PhD, Department of Psychology, Wofford College, 429 N Church St, Spartanburg, SC 29303. E-mail: [email protected] 1526-5900/$36.00 ª 2013 by the American Pain Society http://dx.doi.org/10.1016/j.jpain.2013.03.006
948
the ‘‘peak and end’’ of recording periods27 and personality26 influence the recall of pain. A limitation of these earlier studies was the use of a single recall rating and not recognizing that an individual could also recall patterns of change in their pain experiences over time. The current study uses a novel approach to the study of memory for pain. It asks participants to recall the pattern of their pain by tracing a single continuous line in a graph to represent their daily pain levels over a 15-day period. Values from this line are then compared to daily ratings of pain. This methodology is important because it allows for the analysis of different aspects of memory for daily pain. It also permits one to analyze the accuracy of daily pain recall (ie, absolute differences between initially reported and recalled pain ratings) and to assess the recall of variability in the daily pain experience (ie, patterns of change in pain ratings).14 Finally, this study examined the recall of 3 aspects of daily pain— pain intensity, pain unpleasantness, and perceived activity interference due to pain—rather than focusing on recall of pain intensity only. Neuroticism is a fundamental aspect of personality characterized as a general tendency to experience
Lefebvre and Keefe 6
negative affect. A number of studies suggest that higher levels of neuroticism are associated with reports of higher levels of acute experimental pain and persistent pain.1,8 In addition, studies show that higher levels of neuroticism are related to higher reports of somatic symptoms and to a similar bias in the recall of these symptoms.12,13 Because neuroticism is associated with consistent biases in both the report and recall of symptoms, it should lead to people being less accurate in their recall of pain. Interestingly, to date, only 1 study of persons with persistent pain has examined this relationship. It found no significant relationship between neuroticism and the recall of pain intensity and pain unpleasantness.26 However, this study relied on single measures of pain intensity and unpleasantness. The current study is one of a small number of studies that assess the variability of recalled pain over a series of days.14,19 In studying the influence of neuroticism on the recall of pain, it is also important to consider the effects of depression.6,10,17,20 Depression is common in persons with persistent pain and is known to influence the reports of pain.1,29 Given this relationship, and depression’s effect on recall in general,5 this study measured and controlled for the effect of the severity of depressive symptoms on the recall of daily pain intensity, pain unpleasantness, and pain interference. The current study hypothesized that after controlling for the severity of depressive symptoms, 1) neuroticism would be related to poorer accuracy between recalled and actual levels of pain and pain interference, and 2) neuroticism would be related to lower levels of agreement between recalled and actual day-to-day variability of pain and pain interference.
Methods Participants Participants for the study were recruited using advertisements and direct solicitations from clinics at the Duke University Medical Center as well as from the local community. All participants in the study reported experiencing low-back pain for at least 6 months prior to participation in the study. All participants were at least 18 years of age and were currently experiencing low-back pain on a daily basis. Of the 96 people who agreed to participate in the study, 10 participants failed to complete the study by failing to return for the experimental session. Participants with moderate to high levels of depression (Beck Depression Inventory 1 [BDI-1] score of 15 or greater2,9) were screened out of the study (n = 16) because moderate to high levels of depression have been found to have a significant influence on memory.5 Data in the present study were thus derived from 70 participants (40 women and 30 men). The sample was primarily middle aged (average age = 43.89 years, SD = 12.5, range = 18–71) and had experienced low-back pain for a significant period of time (close to 10 years on average). The sample was predominantly Caucasian
The Journal of Pain
949
(85.7%) and married (65.7%). The majority of the participants had not undergone surgery for their low-back pain (75.7%) and were not receiving disability payments (88.6%). The study was approved by the institutional review board of Duke University and all participants provided informed consent to participate in the study. Participants were compensated up to $16.25 for participation.
Procedures The study was divided into 3 phases: an orientation phase, a 15-day diary phase, and a laboratory recall phase. During the orientation phase, information on demographic variables was collected and participants completed baseline measures of pain interference, neuroticism, and depression. During the diary phase, participants were asked to complete a structured daily diary at the end of each of 15 days. The diary consisted of standardized measures of pain intensity, pain unpleasantness, and perceived activity interference due to pain. During the recall phase, the participants came to a laboratory setting and were asked to recall their pain intensity, pain unpleasantness, and activity interference due to pain using 3 separate Original Pain Recall Assessment (OPRA) forms (described below). At this time, the participants also provided ratings of their current pain intensity and pain unpleasantness using a mechanical visual analog scale (VAS) whose endpoints were identical to those used in the daily diary.23
Measures Diary-Based Measures Daily pain intensity and pain unpleasantness were measured using 100-mm VASs. Daily pain intensity was measured using a VAS whose endpoints were ‘‘no pain sensation’’ and ‘‘most intense pain sensation imaginable.’’ Daily pain unpleasantness used a VAS whose endpoints were ‘‘not at all unpleasant’’ and ‘‘most unpleasant imaginable.’’ Price et al24 have demonstrated that VASs can reliably differentiate the sensory-intensity and the affective-unpleasantness dimensions of pain. Daily activity interference due to low-back pain was assessed by asking participants to rate the relative degree of interference on a 100-mm VAS anchored by ‘‘no interference’’ and ‘‘could not do daily activities.’’ This question has been shown to be reliable when used as part of a protocol to grade the severity of chronic pain.33 A key requirement of diary studies is that participants record each day’s data at the end of that day.1,31 To reinforce this expectation, participants were asked to mail each day’s diary back the following morning. As an added incentive to return the diaries on time, each participant was paid $.25 for each completed diary returned within a 48-hour period, and an extra $1.25 if all 15 diaries were returned. Of the 1050 possible diaries to be completed (15 days by 70 participants), 972 daily diaries were completed and returned, representing a
950
The Journal of Pain
93.6% return rate. Missing diary data led to omission of recalled values for those days in subsequent analyses. Following the 15-day diary period, each participant was asked to return for a recall session on the day immediately following the completion of the final daily diary. Several participants came in the day after completing their final diary (31%), with 79% of participants completing the recall assessment within 7 days of their final diary. Given that there were variations in the length of time between diary completion and the recall session, this variable was used as a control variable in all analyses.
Original Pain Recall Assessment Form During the laboratory recall phase, participants were asked to recall the level of pain intensity, pain unpleasantness, and the amount of activity interference caused by low-back pain over the course of the 15-day diary period. Each participant completed separate OPRA forms, 1 each for pain intensity, pain unpleasantness, and pain interference. On the OPRA, participants are asked to draw a single continuous line from left to right that represented the pattern of their pain level over the entire diary period. To illustrate this approach, Fig 1A provides a tracing obtained from a study participant who was asked to recall his or her pain intensity on the OPRA. The OPRA ratings procedure was repeated for an assessment of recall of pain unpleasantness and for an assessment of recall of pain activity interference. MacDonald et al19 have demonstrated that the OPRA is a reliable and valid method of assessing memory for pain in a persistent pain sample. The lines representing recall of pain intensity, pain unpleasantness, and pain interference were scored using a template. The scoring template consisted of a series of
Neuroticism, Depression, and the Recall of Pain 15 vertical 100-mm VAS meridians, spaced 10 mm apart. Each meridian represented 1 of the 15 days of the diary period. The scoring template was placed over the recalled pattern using a see-through acetate template. Fig 1B demonstrates the same tracing in Fig 1A with the scoring template superimposed. Recalled pain level for each day was defined as the distance in millimeters from the lower endpoint of the VAS (bottom of the box) to the point where the recalled pattern crossed the corresponding meridian.
Questionnaire Measures Neuroticism Subscale of the NEO Personality Inventory–Revised. Neuroticism was measured by the 48 items that make up the neuroticism subscale of the NEO Personality Inventory–Revised (NEO-PI-R6). Participants were asked to indicate the extent to which they agreed with each of the items using a 5-point scale (strongly disagree, disagree, neutral, agree, strongly agree). A global neuroticism score was obtained by summing scores on the 6 facets of neuroticism including depression, anxiety, hostility, self-consciousness, impulsiveness, and vulnerability. The global score for neuroticism has been found to have good internal consistency, with a reported Cronbach alpha of .92.6 Beck Depression Inventory 1. BDI-1 is a 21-item scale that covers the symptoms and attitudes associated with depression.3 Each item contains a graded series of 4 self-evaluative statements, representing a behavioral manifestation of depression. The statements are weighted (0–3) to reflect the range of severity of the symptom from neutral to maximally severe. Each participant was asked to select 1 statement per item to describe how he/she felt over the preceding week. The weights of the selected statements were summed to provide a depression score, with higher
Figure 1. (A) OPRA with an example of a diary recall tracing for pain intensity. (B) OPRA with an example of a diary recall tracing for pain intensity with superimposed scoring template. (C) An example of the resulting recalled and diary values for pain intensity used in the analysis of recall of pain level and recall of pain variability.
Lefebvre and Keefe scores representing higher levels of depression. The scale has been shown to have good internal and split-half reliability.3 Usual Activity Interference Due to Pain. To assess their usual level of activity interference due to pain, each participant was asked to rate the degree to which pain interfered with their daily activities over the previous 6 months on an 11-point scale (0 = no interference, 10 = unable to carry on any activities).33 This is the same scale that was used for the diaries except that the response scale was not a VAS but a verbal numeric scale.
Demographic and Medical History Measures Each participant completed a form that provided his or her age, gender, ethnic group, marital status, and duration of disease. In addition, participants were asked if they had undergone surgery to alleviate low-back pain, if they were currently receiving disability payments, and to describe their current treatment regimen for low-back pain.
Data Analysis Data analysis for the current study progressed from univariate statistics that assessed the mean level of all of the variables used in subsequent analyses. In addition, the data analyses also tried to match the analyses that are common in previous research on the recall of pain. In the univariate section, the study assessed if there was a significant difference between the diary and recalled values for pain intensity, pain unpleasantness, and pain interference. In terms of bivariate analyses, separate series of Pearson product-moment correlations were conducted to assess the relationship between the diary and recalled values of pain, the relationship between neuroticism and depression to the diary and recalled values of pain, and the interrelationship between the independent, covariate, and dependent variables. Past research into the recall of pain has primarily focused on a single measure, usually the difference between a single measure of ‘‘actual’’ pain and the associated single measure of recalled pain. However, individuals may also recall and report on the general pattern of the pain (ie, whether it fluctuates over time). The OPRA provides a means of assessing a wider variety of dimensions to the recall of pain. One benefit of the OPRA methodology is that it provides a set of matched pairs for each day of the diary period. Thus, one could measure recall by subtracting 1 value from its matched pair. Unlike traditional measures, the number of assessment points for both recalled and daily pain ratings provides greater stability in these mean values. The methodology also provides a means of determining if the individual is accurate in recalling the day-to-day variations over time by comparing the agreement between the matched pairs over the 15-day diary period. An example of the resulting matched recall and diary pairs is provided in Fig 1C. Based on the benefits of the OPRA, the current study had 2 principal
The Journal of Pain
951
dependent measures for all analyses: 1) recall of pain level and 2) recall of pain variability. The recall of pain level measure provided an estimate of the absolute difference between the recalled and actual pain measures. For each of the 3 dependent variables (daily pain intensity, daily pain unpleasantness, and daily pain interference), the recall of pain level measure was based on the absolute difference between the recalled and diary ratings. To calculate the measure, a series of daily difference scores was calculated by subtracting each available day’s diary ratings of pain and interference from the corresponding recalled levels of pain and interference. The absolute value of these difference scores was then created. The average of these absolute scores was then used to form a single measure. Higher scores on the recall of pain level measure suggest greater inaccuracy in the recall of pain and activity interference. The recall of pain variability measure provided an estimate of the degree of agreement between the recalled values of pain and interference and the corresponding diary ratings. Given the nature of the OPRA methodology, the multiple recalled and diary ratings provide a means of assessing the agreement of the recalled and daily patterns across time. The recall of pain variability measure was calculated using the intraclass correlation (ICC) between the recalled and diary ratings of pain and interference for each participant over time. The ICC was calculated based on a 2-way mixed consistency model using the average values in order to provide greater stability.19 As shown in Fig 1C, the closer the 2 patterns agree with each other, the higher the ICC and the better the recall of pain variability. Finally, 2 regression analyses were conducted to assess the degree to which neuroticism was related to diary values and recalled values of pain level for pain intensity, pain unpleasantness, and pain interference. Because of the importance of depressive symptoms to the report of pain, both regression analyses controlled for the effect of depressive symptoms as well as covariates that are typically assumed to be involved in memory for pain such as the number of days between the end of the diary period and the evaluation session (lapse) and the duration of pain. In addition, for each model, the current level of pain intensity or pain unpleasantness was added to the model given the robust finding that the current level of pain is related to the recall of pain.7 Usual level of activity interference due to pain, assessed at the orientation phase, was included as a covariate in the analysis predicting the recall of pain interference as a control for baseline interference.
Results Univariate Descriptive Analyses Table 1 presents the descriptive statistics for the dependent and questionnaire measures. As can
952
Neuroticism, Depression, and the Recall of Pain
The Journal of Pain
be seen from Table 1, participants reported low to moderate levels of pain intensity and unpleasantness during both the diary phase and the recall phase. The traditional approach to the assessment of accuracy of the recall of pain has been to assess if there is a significant difference in the levels of pain recalled and reported in the diary. If we were to adopt this approach, there was a significant difference between the average level of pain intensity reported in the diaries and those recalled on the OPRA (t(69) = 2.35, P = .02, d = .28), but no significant differences between the average diary reports and recalled reports for pain unpleasantness (t(69) = .92, P = .36, d = .11) or activity interference (t(69) = 1.01, P = .32, d = .12). Thus, according to this more traditional approach, the sample was relatively accurate in terms of recalling pain unpleasantness and activity interference, but was inaccurate in terms of recalling pain intensity. On average they recalled their pain intensity to be significantly lower than was reported in the daily diary. In terms of the recall of pain variability measure, the results show that the ICCs between recalled and initial diary pain ratings were positive but weak. Despite the fact that overall level of agreement between recalled and initial diary pain ratings was low, considerable variability was noted, with some individuals being more accurate in recalling the pattern of their pain experience over time than others.
Descriptive Findings for Dependent and Questionnaire Measures
Table 1.
MEAN
SD
RANGE
Diary values (aggregated over 15 days) Pain intensity 35.45 20.58 3.5–81.5 Pain unpleasantness 34.37 20.57 .6–83.3 Pain interference 29.52 21.37 .0–91.5 Recalled values (aggregated over 15 days) Pain intensity 32.18 21.12 2.2–79.3 Pain unpleasantness 33.11 23.42 1.6–89.2 Pain interference 27.99 24.66 .2–91.7 OPRA analyses Recall of pain level (absolute difference between pairs) Pain intensity 15.09 7.69 1.5–47.7 Pain unpleasantness 15.54 8.12 1.4–40.7 Pain interference 14.80 10.08 .2–56.9 Recall of pain variability (ICC) Pain intensity .21 .47 1.00–.95 Pain unpleasantness .19 .36 .81–.96 Pain interference .19 .35 .48–.97 Questionnaire measures Neuroticism 115.90 18.43 76.0–161.0 Depressive symptoms 6.89 3.72 .0–14.0 Usual activity interference 4.60 2.49 .0–9.0 Duration of pain (months) 106.26 121.31 6.0–468.0 Current pain intensity 23.23 21.16 .0–72.0 Current pain unpleasantness 21.77 22.16 .0–85.0 Days since completion of diary 5.34 5.76 1.0–30.0 period (Lapse)
Bivariate Analyses: The Relationship of Neuroticism and Depression to Mean Values of Recalled and Diary Ratings of Pain Intensity, Pain Unpleasantness, and Pain Interference Levels A series of correlation analyses was conducted to assess the relationship between the mean recall levels of pain intensity, pain unpleasantness, and pain interference and the mean diary values for each of these variables. The results showed a significant correlation between the mean recalled and diary levels of pain intensity (r = .84, P < .001), pain unpleasantness (r = .88, P < .001), and activity interference (r = .86, P < .001). As with the difference scores, more traditional studies would have concluded that the significant correlations suggest that, as a whole, these individuals were accurate in their recall in all 3 aspects of their pain experience. Correlation analyses were conducted to assess the relationship between neuroticism and mean recalled levels of pain intensity, pain unpleasantness, and pain interference. The analyses showed that neuroticism was not significantly correlated with any of the mean values of the recalled pain measures. A series of correlation analyses was also conducted to assess the relationship between neuroticism and mean diary ratings of pain intensity, pain unpleasantness, and pain interference. Once again, neuroticism was not significantly correlated with any of the mean values of the daily diary pain measures. Similar correlational analyses were conducted to examine the relationship of the severity of depressive symptoms to the recall of pain. These analyses showed that depressive symptoms were significantly correlated with mean recalled pain intensity (r = .34, P = .004), mean recalled pain unpleasantness (r = .35, P = .003), and mean recalled pain interference (r = .38, P = .001). Correlational analyses were also conducted to assess the relationship of the severity of depressive symptoms to the report of daily pain. These analyses found that the severity of depressive symptoms was significantly correlated with mean diary ratings of pain intensity (r = .33, P = .006), mean diary ratings of pain unpleasantness (r = .35, P = .003), and mean diary ratings of pain interference (r = .38, P = .001). Finally, a series of correlational analyses was conducted to assess the interrelationship among the variables used in the analyses. The results showed that neuroticism was significantly correlated with the degree of depressive symptoms (r = .47, P = .001). In terms of the diary values, pain intensity was significantly related to pain unpleasantness (r = .96, P = .001) and activity interference (r = .90, P = .001). Daily pain unpleasantness was significantly related to daily pain interference (r = .93, P = .001). A similar pattern was found for the recalled values, with pain intensity significantly related to pain unpleasantness (r = .94, P = .001) and pain interference (r = .84, P = .001). Finally, recalled values for pain unpleasantness were significantly correlated to recalled values of pain interference (r = .85, P = .001). Taken together, in terms of the recall of pain, these results suggest that neuroticism is not significantly
Lefebvre and Keefe
The Journal of Pain
correlated with the average level of both diary and recalled values of pain intensity, pain unpleasantness, and pain interference derived from the OPRA. The results did indicate that participants who reported higher levels of depressive symptoms were significantly more likely to recall having higher levels of pain intensity, pain unpleasantness, and pain interference, as well as report higher levels of daily pain intensity, pain unpleasantness, and pain interference.
Multivariate Analyses: The Relationship of Neuroticism to Recall of Pain Level The results of the regression analyses are presented in Table 2. As can be seen, neuroticism was not significantly correlated with the accuracy of recall of pain intensity, pain unpleasantness, or pain interference on the OPRA. Interestingly, the regression analyses did find a significant relationship between baseline severity of depressive symptoms with all 3 recall of pain level measures on the OPRA. Patients with greater severity of depressive symptoms showed poorer recall accuracy (greater absolute difference) when comparing the diary and recalled values on all 3 measures. A second regression analysis was conducted to assess the degree to which neuroticism was related to the recall of pain variability measure for pain intensity, pain unpleasantness, and pain interference. These analyses included the same set of variables used in the previous regression analyses. The results of the regression analysis are presented in Table 3. As can be seen, neuroticism was significantly related to the concordance between diary and recalled values in pain unpleasantness. Patients who scored higher on neuroticism showed greater concordance in diary and recalled fluctuations in pain unpleasantness. Neuroticism was not related to the
Results of Regression Analyses Relating Neuroticism and Evaluation Measures to the Recall of Pain Level Measures
Table 2.
R2 VARIABLE
TOTAL
DF
F RATIO b ON 2 FOR R ENTRY SIGNIFICANCE
Pain intensity .13 (5, 64) 1.97 Lapse (days since last diary) .17 Duration of pain .17 Current pain intensity .06 Depression .26 Neuroticism .13 Pain unpleasantness .11 (5, 64) 1.57 Lapse (days since last diary) .14 Duration of pain .07 Current pain unpleasantness .03 Depression .32 Neuroticism .08 Pain interference .32 (5, 64) 7.54 Lapse (days since last diary) .40 Duration .13 Usual activity interference .15 Depression .27 Neuroticism .10
.15 .15 .61 .04 .34 .28 .55 .88 .03 .58 .001 .21 .13 .04 .41
953
concordance in day-to-day variations in pain intensity or pain interference.
Discussion The present study investigated the effects of neuroticism and depression on the degree to which participants suffering from low-back pain could accurately recall their daily experience of pain intensity, pain unpleasantness, and perceived activity interference. The study had 2 major findings. First, patients scoring higher on neuroticism were significantly more accurate in their recall of the variability of pain unpleasantness over time. Second, patients who reported more severe depressive symptoms were less accurate in terms of recalling pain intensity, pain unpleasantness, and perceived activity interference. One of the most interesting findings of this study was the significant relationship between neuroticism and recall of day-to-day variations in pain unpleasantness. The literature suggests that higher levels of neuroticism are associated with a bias in report and recalling symptoms as being more severe. The current study found that neuroticism was related to the agreement in patterns between recalled pain unpleasantness and actual pain unpleasantness but was not related to a difference score between recalled and diary levels of pain unpleasantness. Thus, it may be that people with persistent pain who are higher in neuroticism may not be accurate in their recall of pain unpleasantness if assessed with a single value, but may be better at recalling variability of pain unpleasantness over time. This suggests that neuroticism may be related to noticing relative changes in pain level rather than absolute pain levels.26 One could assess how changes in pain are recalled by people higher on neuroticism compared to the recall of a stable level of pain. One could also assess
Results of Regression Analyses Relating Neuroticism and Evaluation Measures to the Recall of Pain Variability Measures
Table 3.
R2 VARIABLE
TOTAL
DF
F RATIO b ON 2 FOR R ENTRY SIGNIFICANCE
Pain intensity .10 (5, 64) 1.37 Lapse (days since last diary) .27 Duration of pain .10 Current pain intensity .13 Depression .10 Neuroticism .01 Pain unpleasantness .14 (5, 64) 2.01 Lapse (days since last diary) .07 Duration of pain .10 Current pain unpleasantness .17 Depression .02 Neuroticism .27 Pain interference .02 (5, 64) .30 Lapse (days since last diary) .06 Duration of pain .07 Usual activity interference .10 Depression .02 Neuroticism .08
.03 .42 .30 .49 .93 .53 .40 .16 .89 .05 .63 .57 .44 .91 .60
954
The Journal of Pain
how fluctuations in pain lead to the increase or decrease in the use of specific coping strategies. Future empirical studies might be able to manipulate pain level and then assess the recall of pain among those higher and lower on neuroticism. Because it provides a series of scores for the recall of pain, the OPRA provides a closer approximation between the recall and diary assessment methods. In prior research studies, diary techniques were used to assess the daily level of pain with a great temporal resolution. However, the recall of pain did not match the resolution of the assessment technique. Rather, recall is assessed by having the participant complete a single VAS to represent his or her usual pain intensity over the diary period. Thus, there is a qualitative and quantitative difference between how pain is assessed and recalled. The OPRA also provides a means of assessing accuracy in the recall of pain using the methodology of traditional studies: difference between mean actual and recalled pain scores and correlation between actual and recalled pain scores. Although these results may appear revealing, they may distort the reality of the relationship by simplifying a very complex process into a comparison between single values across individuals. These approaches also fail to determine what consistent patterns exist within the data that led to these values. Newer approaches in the study of the recall of pain have focused less on determining accuracy and more on assessing what influences the variability in the recall of pain between individuals. To this end, the OPRA methodology goes beyond the traditional approaches by providing more information about how the relationship of daily pain ratings and recalled pain ratings varies over time. It is interesting to note that the results of the current study found the relationship between neuroticism to be stronger for the recall of pain unpleasantness. These results of this study are in line with the study by Harkins and colleagues,8 which found that neuroticism was significantly related to the report of acute and clinical pain unpleasantness and not pain intensity. In another study, Wade et al34 found that neuroticism was again only related to pain unpleasantness and not pain intensity among a group of chronic pain patients. Taken together, the results of the current study when combined with the results of the previous studies suggest that neuroticism appears to favor a focus on the emotional aspects of pain, in terms of both the report of pain unpleasantness and the recall of the unpleasantness of this pain. The precise mechanisms by which neuroticism affects memory for chronic pain remain uncertain. One view proposes that individuals higher on neuroticism are more somatically hypervigilant, constantly searching for early warning signs of disease or increased pain.35 This would suggest that people who are higher on neuroticism would be more focused on their pain and devote more attention resources to monitor their pain. Thus, one possibility is that higher levels of neuroticism are involved in slowing the process of forgetting details about the chronic pain experience. Forgetting has been
Neuroticism, Depression, and the Recall of Pain hypothesized to be the loss of detail about stimulus attributes over time.28 Perhaps the somatic hypervigilance that is characteristic of neuroticism increases the salience of pain attributes and thus slows the forgetting process, leading to better recall. An interesting, though expected, result was the relationship of depressive symptoms to the recall of pain. There is an extensive literature on the effects of depression on memory.4,5,32 However, the specific relationship between depression and memory for pain has not been fully researched. In fact, most of the research available discusses the impact of depression on the recall of personal attributes and cognitive bias among patients with persistent pain.22 The current study is novel in that it has demonstrated that the severity of depressive symptoms does have an effect on the accuracy of the recall of pain but not the recall of pain variability. Even after controlling for severity of depressive symptoms through a limit imposed on the BDI-1 and the statistical control of the variable, the relationship is clear: those individuals who show higher levels of depressive symptoms are more inaccurate in the recall of pain. The severity of depressive symptoms was included as a covariate in order to isolate its effect on memory. Further research is needed to assess the degree to which clinical depression is related to the recall of pain. There are some limitations to this study. First, the study relies on self-report data both for the recall of pain and interference and also for the to-be-remembered stimuli. In memory research, the participants are exposed to a standard set of stimuli and then the recall of this material can be easily quantified at a later time. Given the subjective nature of pain, this approach is difficult but not impossible. A study by Rainville and colleagues25 demonstrates that methodology such as the matching to sample paradigm can be used in the study of memory for pain. This methodology, commonly used in the study of memory, allows for the assessment of memory without the use of verbal descriptions. This approach is important because pain is a nonverbal memory and the recall of verbal or numeric pain ratings may interfere with the recall of pain sensations. Second, in this study the daily diary data on pain were collected using pencil-and-paper diaries. The methodology was to have the participants complete the diary at the end of the day and return the diary by mail. This approach does not guarantee that the person completed the diary on time, and some diaries were not returned at all. Newer technology using portable electronic assessments offer the possibility of collecting data in a more systematic and consistent fashion.26,30 These methods may also afford better temporal resolution in the assessment of the recall of pain by placing the assessment and recall tasks in closer proximity. This approach could assess if certain days (ie, higher levels of pain) are better recalled than others. Third, the results show a high level of variability in the recall of pain. The degree of variability may be due to individual differences, or it could be that the OPRA methodology requires more refinement. Fourth, the current study did not assess and control for the effects
Lefebvre and Keefe of anxiety on memory for pain. Previous research suggests that anxiety related to dental procedures may affect memory for pain.11 Future research may need to explore the relationship of pain anxiety and worry about pain to memory for pain. Fifth, although neuroticism was found to be related to the recall of pain unpleasantness, it was not related to diary reports of pain intensity or pain unpleasantness. Neuroticism is usually found to be associated with reports of pain. As suggested above, neuroticism may be more related to changes in pain rather than single reports of usual pain. Finally, the study failed to assess the medications that the participants were taking at the time of the study. Some participants may have been on medications that could affect their memory in general. The current study offers a vista on a new and potentially exciting field of research, not only in terms of the memory for pain but also in the recall of other
References 1. Affleck G, Tennen H, Urrows S, Higgins P: Neuroticism and the pain-mood relation in rheumatoid arthritis: Insights from a prospective daily study. J Consult Clin Psychol 60: 119-126, 1992
The Journal of Pain
955
somatic sensations. The vast majority of the research conducted in order to study the mechanisms of memory and pain is focused on verbal memory. Methodologies such as the one presented in this study may be useful in better understanding how memory functions in the recall of somatic sensations.
Acknowledgments The authors of the study would like to thank David Caldwell, MD, Bruno Urban, MD, William Spillane, MD, John Goreki, MD, Lloyd Hey, MD, Diane Scott, MD, and William Richardson, MD, for their assistance in recruiting patients to participate in this study. The authors would also like to thank 3 anonymous reviewers for their insightful comments on a previous draft of this manuscript.
14. Lefebvre JC, Keefe FJ: Memory for pain: The relationship of catastrophizing to the recall of daily rheumatoid arthritis pain. Clin J Pain 18:56-63, 2002 € testam KG: A clinical comparison of two 15. Linton SJ, Go pain scales: Correlation, remembering chronic pain, and a measure of compliance. Pain 17:57-65, 1983
2. Beck AT, Beck RW: Screening depressed patients in family practice: A rapid technic. Postgrad Med 52:81-85, 1972
16. Linton SJ, Melin L: The accuracy of remembering chronic pain. Pain 13:281-285, 1982
3. Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J: An inventory for measuring depression. Arch Gen Psychiatry 4: 561-571, 1961
17. Lloyd GG, Lishman WA: Effect of depression on the speed of recall of pleasant and unpleasant experiences. Psychol Med 5:173-180, 1975
4. Blaney PH: Affect and memory: A review. Psychol Bull 99: 229-246, 1986
18. Loftus EF: Our changeable memories: Legal and practical implications. Nat Rev Neurosci 4:231-234, 2003
5. Burt DB, Zeembar MJ, Niederehe G: Depression and memory impairment: A meta-analysis of the association, its pattern, and specificity. Psychol Bull 117:285-305, 1995
€ jmark M: Recon19. MacDonald S, Linton SJ, Jansson-Fro structing the pain on the Original Pain Recall Assessment form (OPRA). J Pain 10:809-817, 2009
6. Costa PT, McCrae RR: NEO PI-R: Professional manual. Odessa, FL, Psychological Assessment Resources, Inc, 1992
20. Martin M: Neuroticism as predisposition towards depression: A cognitive mechanism. Pers Individ Diff 6: 353-365, 1985
7. Eich E, Reeves JL, Jaeger B, Graff-Radford SB: Memory for pain: Relation between past and present pain intensity. Pain 23:375-379, 1985
21. Morley SJ: Reliability and accuracy of memory for pain. Pain 57:132, 1994
8. Harkins SW, Price DD, Braith J: Effects of extroversion and neuroticism on experimental pain, clinical pain, and illness behavior. Pain 36:209-218, 1989 9. Haslam N, Beck AT: Categorization of major depression in an outpatient sample. J Nerv Ment Dis 181:725-731, 1993 10. Kendler KS, Kuhn J, Prescott CA: The interrelationship of neuroticism, sex, and stressful life events in the prediction of episodes of major depression. Am J Psychiatry 161:631-636, 2004 11. Kent G: Memory for dental pain. Pain 21:187-194, 1985 12. Larsen RJ: Neuroticism and selective encoding and recall of symptoms: Evidence from a combined concurrentretrospective study. J Pers Soc Psychol 62:480-488, 1992 13. Larsen RJ, Kasimatis M: Day-to-day physical symptoms: Individual differences in the occurrence, duration, and emotional concomitants of minor daily illnesses. J Pers 59: 387-423, 1991
22. Pincus T, Morley SJ: Cognitive-processing bias in chronic pain: A review and integration. Psychol Bull 127:599-617, 2001 23. Price DD, Bush FM, Long S, Harkins SW: A comparison of pain measurement characteristics of mechanical visual analog and simple numerical rating scales. Pain 56: 217-226, 1994 24. Price DD, McGrath PA, Rafii A, Buckingham B: The validation of visual analog scales as ratio scale measures for chronic and experimental pain. Pain 17:45-56, 1983 25. Rainville P, Doucet JC, Fortin MC, Duncan GH: Rapid deterioration of pain sensory-discriminative information in short-term memory. Pain 110:605-615, 2004 26. Raselli C, Broderick JE: The association of depression and neuroticism with pain reports: A comparison of momentary and recalled pain assessment. J Psychosom Res 62:313-320, 2007 27. Redelmeier DA, Kahneman D: Patients’ memories of painful medical treatments: Real-time and retrospective
956
The Journal of Pain
evaluations of two minimally invasive procedures. Pain 66: 3-8, 1996 28. Riccio DC, Rabinowitz VC, Axelrod S: Memory: When is less more. Am Psychol 49:917-926, 1994 29. Romano JM, Turner JA: Chronic pain and depression: Does the evidence support a relationship? Psychol Bull 97: 18-34, 1985
Neuroticism, Depression, and the Recall of Pain 32. Veiele HOF: A preliminary profile of neuropsychological deficits associated with major depression. J Clin Exp Neuropsychol 19:587-603, 1997 33. Von Korff M, Ormel J, Keefe FJ, Dworkin SF: Grading the severity of chronic pain. Pain 50:133-149, 1992
30. Stone A, Shiffman S: Ecological Momentary Assessment (EMA) in behavioral medicine. Ann Behav Med 16:199-202, 1994
34. Wade JB, Dougherty LM, Hart RP, Rafii A, Price DD: A canonical correlation analysis of the influence of neuroticism and extraversion on chronic pain, suffering, and pain behavior. Pain 51:67-73, 1992
31. Stone A, Kessler R, Haythornthwaite J: Methodological considerations in the measurement of daily events and experiences. J Person 59:575-607, 1991
35. Watson D, Pennebaker JW: Health complaints, stress, and distress: Exploring the central role of negative affectivity. Psychol Rev 96:234-254, 1989