Abstract
Background and aims
Depression is common in older adults with persistent pain. Cognitive-behavioural models of pain propose that the relationship between pain and depression is influenced or mediated by interpretations of events (cognitions), rather than by the event itself. Almost exclusively, the evidence for this position has come from studies of people aged less than 65 years. The role of cognitions in the pain experience of older adults has been unclear due to the limited and conflicting evidence available. The aim of our study was to examine the role of catastrophizing in mediating the relationship between pain intensity and depressed mood in older adults with persistent pain using reliable and valid measures for this population.
Methods
In a two-wave longitudinal design, a sample of 141 patients (89 women, 52 men) 65 years and over with persistent pain participating in an evaluation of a pain self-management programme completed questionnaires measuring usual level of pain intensity (NRS), depressed mood (DASS-21) and the catastrophizing factors of magnification and helplessness (PRSS) at the beginning of the programme and 6 months later. Demographic data and pain history were collected by self-completion questionnaires, which were mailed to patients prior to participating in the programme and returned by post.
Results
Change scores for usual level of pain intensity (NRS), depressed mood (DASS-21) and the magnification and helplessness factors on the PRSS-Catastrophizing were calculated by subtracting the scores at 6 months after completion of the study (Time 2) from the scores at the beginning of the study (Time 1). In the longitudinal analyses of mediation, using a series of regression analyses, change scores for both factors (magnification, helplessness) of the measure of catastrophizing (PRSS) totally and significantly mediated the relationship between change scores for pain intensity and depressed mood. The significant relative magnitude of beta between pain intensity and depression reduced from 0.22 to 0.13 and became non-significant after introducing magnification as a mediating variable, whilst the significant relative magnitude of beta reduced from 0.22 to 0.12 and also became non-significant after introducing helplessness as a mediating variable.
Conclusions
These results support a cognitive-behavioural model and highlight the importance of cognitive factors, in this case catastrophizing, in the relationship between pain intensity and depressed mood in older adults with persistent pain.
Implications
These findings also have important clinical implications for the treatment of older adults with persistent pain. They highlight the importance of targeting interventions to reduce the influence of catastrophizing as a prerequisite for reducing depressive symptoms in this growing population.
1 Introduction
Many older adults with persistent pain also suffer from depression [2,26], with higher levels of pain associated with increased severity of depression [5,26,27,44]. While depression is relatively common in older adults with persistent pain, it seems less frequent or less severe in comparison to younger adults with persistent pain [6,29,46]. No explanations for this difference have been established, but one possibility is that older people are more stoic, especially as their pain severity levels are similar to those reported by younger adults [45]. This might mean older adults catastrophize less about their pain, or there is a different relationship between catastrophizing and pain in older people. Cognitive-behavioural mediation models propose that behaviour and emotions are influenced or mediated by interpretations of events (cognitions), rather than by the event itself [39]. Catastrophizing about pain has been broadly defined as an exaggerated negative appraisal of pain and the pain experience [37]. The relationship between pain and depression has been shown in younger adults to be mediated by cognitive factors [31], in particular catastrophizing [17,28,30,35,41]. Also, change in catastrophizing has been associated with change in depression in patients receiving a pain management intervention [32,43], including in older adults [8]. However the role of cognitive factors in older adults has been unclear due to the limited and conflicting evidence available. Some studies have supported a role for catastrophizing [7,21], while one study found no support for the role of cognitive factors and instead found a direct association between pain and depression [38]. A recent study found that catastrophizing mediated the relationship between pain intensity and depressed mood in older adults [47]. This study helped clarify the mediating role of cognitive factors, in this case catastrophizing, in the paindepression relationship in older adults. However, these findings were based on a cross-sectional analysis. A significant limitation of cross-sectional analysis is that analysis is confined to a specific point in time. Longitudinal analysis not only allows the examination of differences among individuals but importantly adds the examination of change within individuals [24]. This is important when examining depression in older adults as depression is associated with temporal factors such as cognitive decline [10]. The mediating role of catastrophizing in the pain-depression relationship in older adults over time has not been examined. The analysis of a mediation model with longitudinal data will strengthen the existing evidence of the mediating role of catastrophizing in the relationship between pain intensity and depressed mood in older adults with persistent pain.
This study is a two-wave design examining whether changes in catastrophizing over 6 months mediate the relationship between changes in pain intensity and depressed mood over the same period. Given the previous finding of the significant mediating role of catastrophizing in the relationship between pain intensity and depressed mood at a single point in time, we hypothesized that the significant role of catastrophizing in mediating this relationship at a single point in time would hold over a 6 month period.
2 Method
2.1 Participants
Participants included patients aged 65 years and over with persistent non-cancer pain for more than 6 months who were part of a larger study evaluating a pain self-management programme [25]. Recruitment of participants commenced in June 2006 and ended in June 2009. Participants were assessed for cognitive deficits and were only included if they showed evidence of normal range shortterm memory functioning as determined by a score of 22 or greater on the Rowland Universal Dementia Assessment Scale (RUDAS) [34]. For a full description of the participants and the inclusion and exclusion criteria refer to the above study [25].
2.2 Procedures
Demographic information and pain history was collected by self-completion questionnaires, which were mailed to patients and returned by post. Patients completed a set of self-report measures, including the PRSS-Catastrophizing, DASS-21 and NRS at the start of the study (Time 1). These measures will be described in the following section. Patients were also requested to complete the same set of self-report measures 6 months later (Time 2). None of the measures were modified in any way (e.g. font size), although participants were permitted to use spectacles, etc. to account for any sensory loss.
2.3 Measures
2.3.1 Pain-Related Self Statements Scale-Catastrophizing subscale (PRSS-Catastrophizing)
The Pain-Related Self Statements (PRSS) [12,13] scale was developed to assess the frequency of use of cognitions by people with persistent pain that either assist or hinder their attempts to cope with severe pain. The measure consists of two subscales: active coping and catastrophizing. This study only used the catastrophizing subscale (PRSS-Catastrophizing). The catastrophizing subscale contains nine items and for each item individuals are asked to rate on a six-point scale, with 0 = “Almost never” and 5 = “Almost always”, how often they think in such a way when they experience severe pain. The total score for all items is divided by nine to obtain a mean item score. A higher score indicates a greater frequency of catastrophizing. The PRSS-Catastrophizing was validated by the authors [12] using predominantly younger adults with persistent pain, and was shown to have excellent internal consistency (α = 0.92) and construct and discriminant validity. The psychometric properties of the PRSS-Catastrophizing subscale have also been established in a sample of older adults with persistent pain [47]. This study found that the PRSS-Catastrophizing consisted of 2 factors labelled magnification and helplessness. These factors were used in the analysis.
2.3.2 Depression Anxiety Stress Scales (short version) (DASS-21)
The Depression Anxiety Stress Scales (DASS) [22] was developed to assess the severity of the core symptoms of depression, anxiety and stress. The DASS-21 is a 21-item short version of the DASS with three scales of 7 items labelled: Depression, Anxiety and Stress. It has been shown to have the same factor structure as the 42item version [1]. The psychometric properties of the DASS-21 have been established in a sample of older adults with persistent pain [46]. Only the Depression scale was used for this study and scores are doubled so they are comparable to scores for the full 42-item version [22], which range between 0 and 42, with higher scores indicating more severe levels of depression.
2.3.3 Numerical Rating Scale (NRS)
A Numerical Rating Scale was used to measure the usual level of pain in the preceding week. Patients were asked to rate the intensity of their pain on a 0-10 (11-point) scale, where 0 indicates “no pain” and 10 indicates “worst pain imaginable”, by circling a number on the scale. The psychometric properties of the NRS have been established in a sample of older adults with persistent pain [45].
2.4 Statistical analyses
Change scores for usual level of pain (NRS), depression (DASS-21) and the magnification and helplessness factors on the PRSS-Catastrophizing were calculated by subtracting the scores at 6 months after completion of the study (Time 2) from the scores at the beginning of the study (Time 1).
Pearson’s product moment correlations were calculated to examine the relationships among the change score variables. Mediation was tested for each factor from the PRSS-Catastrophizing according to the procedure detailed by Baron and Kenny [3]. They describe 4 conditions which must be met for a variable to function as a mediator. Firstly, the independent variable (in this study, change in usual level of pain) must significantly predict the mediator variable (in this study, change in PRSS-Catastrophizing factors). Secondly, the independent variable (change in usual level of pain) must significantly predict the dependent variable (in this study, change in DASS-21-Depression). Thirdly, the mediator variable (change in PRSS-Catastrophizing factors) must significantly predict the dependent variable (change in DASS-Depression-21). Finally, while controlling for the mediator variable (change in PRSS-Catastrophizing factors), the previously significant relationship between the independent variable (change in usual level of pain) and the dependent variable (change in DASS-Depression-21) must be reduced or eliminated. A series of regression analyses were used to assess whether the conditions for mediation were met. A test of the significance of the indirect or mediation effect of the independent variable on the dependent variable via the mediator was calculated using Sobel’s test [33].
Analyses were performed using the Statistical Package for the Social Sciences (SPSS) version 17.0 (SPSS, Chicago, IL).
2.5 Ethics approval
The use of the de-identified dataset for this study was approved by the Hospital’s Ethics Committee.
3 Results
3.1 Sample characteristics
There were 141 patients who completed the measures at Time 1 and 112 patients who completed the measures at Time 2. There were 29 patients who completed the measures at Time 1 but declined to complete the measures 6 months later at Time 2. An independent samples t-test revealed that patients who declined to complete the Time 2 measures were significantly older (M = 76.38, SD = 6.65) than the group who provided data for both Time 1 and 2 (M = 73.29, SD = 5.75), t(139) = 2.50, p = 0.014. However, there were no significant differences on pain duration or any of the Time 1 measures.
The demographic and pain-related characteristics of the sample are presented in Tables 1 and 2. The mean age was 73.92 years and ranged from 65 to 87 years. Almost two thirds (63.1%) of the participants were female. The majority of patients (54.6%) were either married or in a de facto relationship, however, a significant proportion (17.7%) was widowed. Approximately 65% of patients reported 9 or more years of education. Approximately half of the patients (51.8%) were born in Australia and were retired from work (53.2%).
Demographic characteristics.
| Mean | Range | |
|---|---|---|
| Mean | Range | |
| Age (years) | 73.92 | 65–87 |
| Gender | n | (%) |
| Male | 52 | (36.9) |
| Female | 89 | (63.1) |
| Marital status [a] | ||
| Married/de facto | 77 | (54.6) |
| Separated/Divorced | 11 | (7.8) |
| Widowed | 25 | (17.7) |
| Single | 7 | (5.0) |
| Educational status [a] | ||
| Post-high school qualification | 56 | (39.7) |
| Completed secondary schooling | 15 | (10.6) |
| Between 9 and 11 years ofeducation | 22 | (15.6) |
| Less than 9 years of education | 5 | (3.5) |
| Other | 18 | (12.8) |
| Birthplace [a] | ||
| Australia | 73 | (51.8) |
| Other countries | 44 | (30.9) |
| Current work status [a] | ||
| Full-time/part-time/voluntary work | 10 | (7.1) |
| Home duties | 6 | (4.3) |
| Unemployed due to pain | 5 | (3.5) |
| Retired | 75 | (53.2) |
Pain-related characteristics.
| Pain duration[a] | ||
| Median (months) | 72.0 | |
| Mode of onset of pain [a] | N | (%) |
| Accident at work | 9 | (6.4) |
| At work but not involving an accident | 1 | (0.7) |
| Accident at home | 8 | (5.7) |
| Motor vehicle accident | 6 | (4.3) |
| After surgery | 20 | (14.2) |
| After illness | 7 | (5.0) |
| Pain just began, no obvious reason | 49 | (34.8) |
| Other reasons | 24 | (16.9) |
| Pain site[a] | N | (%) |
| Head, face, mouth | 9 | (6.4) |
| Neck | 3 | (2.1) |
| Upper shoulders and upper limbs | 1 | (0.7) |
| Lower back, lower spine, sacrum | 23 | (16.3) |
| Lower back and lower limbs | 22 | (15.6) |
| Legs | 9 | (6.4) |
| Pelvic region | 1 | (0.7) |
| 2 or more major pain sites (generalized pain) | 57 | (40.4) |
| Measures at Time1, 2 | ||
| NRS-Usual level pain preceding week/Time 1/(0–10) [b] | 5.39 | (2.09) |
| DASS-21-Depression/Time 1/(0–42) [b] | 10.79 | (9.76) |
| PRSS-Catastrophizing-Magnification/Time 1/(0–5) [b] | 2.24 | (1.37) |
| PRSS-Catastrophizing-Helplessness/Time 1/(0–5) [b] | 2.58 | (1.03) |
| NRS-Usual level pain preceding week/Time 2/(0–10) [b] | 5.10 | (2.24) |
| DASS-21-Depression/Time 2/(0–42) [b] | 10.78 | (9.85) |
| PRSS-Catastrophizing-Magnification/Time 2/(0–5) [b] | 2.07 | (1.51) |
| PRSS-Catastrophizing-Helplessness/Time 2/(0–5) [b] | 2.33 | (1.11) |
The median pain duration was 6 years. A considerable proportion of the patients (34.8%) reported that their pain started for no obvious reason, although a sizeable proportion (14.2%) reported that the pain began following surgery. A large proportion of patients reported generalized pain, that is, pain in two or more major sites (40.4%). Other pain sites reported included: head, face and mouth (6.4%); neck (2.1%); upper shoulders and upper limbs (0.7%); lower back (16.3%); lower back and lower limbs (15.6%); legs (6.4%) and pelvic region (0.7%). Multimorbidity levels were substantial and consistent with community norms for this age group [9]. Approximately 70% of patients reported one or more ongoing diseases or health conditions besides their persisting pain. These included joint diseases, cardiovascular conditions, visual impairments, gastrointestinal conditions, and endocrine, skin, and bladder disorders. The mean (SD) number of comorbidities in this sample was 3.05 (2.16).
In order to gauge the severity of depressive symptoms in this sample, categories were derived using the means and standard deviations of DASS-21 scores from a population with similar demographic characteristics [16]. Consequently, these definitions are age adjusted and differ from those listed in the DASS manual [22]. “Normal” was defined as up to but not including the mean of the scale (D:0–8). “Moderate” severity was defined by the mean of the scale up to but not including one standard deviation above the mean (D:9–16). “Severe” was defined by the mean plus one standard deviation up to but not including two standard deviations above the mean (D:17–24), whilst “Extremely severe” was defined as two standard deviations above the mean up to the maximum score (D:25–42). This placed the mean depression scores for the participants in this study in the moderate range at both Times 1 and 2.
3.2 Pearson’s correlations among the variables used in the mediation analysis
A bivariate correlation matrix for the variables used in the mediation analysis (i.e. changes in usual level of pain, magnification, helplessness and depression) is presented in Table 3 and supports the predicted relationships amongst these variables.
Pearson correlations among the variables used in the mediation analysis
| Δ Usual level pain preceding week | Δ Magnification | Δ Helplessness | Δ Depression | |
|---|---|---|---|---|
| Δ Usual level pain preceding week | - | |||
| 0.33[**] | – | |||
| Δ Magnification | ||||
| Δ Helplessness | 0.34[**] | 0.63[**] | – | |
| Δ Depression | 0.22[*] | 0.28[**] | 0.31[**] | - |
-
Δ Change score = Time 1 Time 2
3.3 Testing the mediation of change in catastrophizing on the relationship between change in pain intensity and change in depression
A summary of the series of regression analyses testing for mediation is shown in Table 4.
Series of regression analyses testing mediation.
| Measure | Predictors | B (95% CI) | SE | β | Adj. R2% | Sobel’stest | |||
|---|---|---|---|---|---|---|---|---|---|
|
|
|||||||||
| Indirect effect (95%CI) | Teststatistic | p value | |||||||
| Δ F1-Magnification | |||||||||
| Step1 | |||||||||
| Δ PRSS-Magnification | Δ Usual level pain | 0.22 | (0.10–0.33) | 0.06 | 0.33 | [**] | 9.9 | ||
| Step 2 | |||||||||
| Δ DASS-21-Depression | Δ Usual level pain | 0.42 | (0.06–0.77) | 0.18 | 0.22 | [*] | 3.8 | ||
| Step 3 | |||||||||
| Δ DASS-21-Depression | Δ Usual level pain | 0.26 (-0.12 to 0.64) | 0.19 | 0.13 | ns | 7.7 | 0.16 (-0.39 to 0.70) | 2.03 | 0.04 |
| Δ Magnification | 0.71 (0.13–1.28) | 0.29 | 0.24 | [*] | |||||
| Δ F2-Helplessness | |||||||||
| Step1 | |||||||||
| Δ PRSS-Helplessness | Δ Usual level pain | 0.22 (0.10–0.33) | 0.06 | 0.34 | [**] | 11.0 | |||
| Step 2 | |||||||||
| Δ DASS-21 Depression | Δ Usual level pain | 0.42 (0.06–0.77) | 0.18 | 0.22 | [*] | 3.8 | |||
| Step 3 | |||||||||
| Δ DASS-21-Depression | Δ Usual level pain | 0.23 (-0.14 to 0.61) | 0.19 | 0.12 | ns | 9.1 | 0.18 (-0.38 to 0.75) | 2.21 | 0.03 |
| Δ Helplessness | 0.83 (0.23–1.42) | 0.30 | 0.27 | [**] | |||||
-
Δ Change score = Time 1 Time 2. ns, non-significant
The preconditions for mediation were met for each factor. The independent variable of change in usual level of pain significantly predicted the mediator variables of change in magnification and helplessness. Also, change in usual level of pain significantly predicted the dependent variable of change in depression. Finally, the mediator variables of change in magnification and helplessness significantly predicted the dependent variable of change in depression. A final regression analysis for each factor where change in usual level of pain was entered into the regression with each factor showed that the magnitude of prediction of change in usual level of pain was reduced and became non-significant. For change in magnification the magnitude of beta for change in pain intensity reduced from 0.22 to 0.13. For change in helplessness the magnitude of beta for pain intensity reduced from 0.22 to 0.12. These relationships are shown in Figs. 1 and 2.
Change over 6 months in magnification mediates the relationship between change in pain intensity and change in depressed mood (*p<0.05; **p<0.01; ns, non-significant; fi, standardized coefficients).
Change over 6 months in helplessness mediates the relationship between change in pain intensity and change in depressed mood (*p<0.05; **p<0.01; ns, non-significant; fi, standardized coefficients).
Therefore it can be said that change in usual level of pain has indirect effects on change in depression. The indirect effects of change in usual level of pain on change in depression through change in magnification and helplessness factors were both significant according to Sobel’s test [33].
4 Discussion
The aim of this study was to examine the mediating role of catastrophizing over time in the relationship between pain intensity and depressed mood in older adults with persistent pain. This study showed that change over 6 months in the two factors of the catastrophizing subscale of the PRSS (magnification, helplessness) totally and significantly mediated the relationship between the change in pain intensity and the change in depression score on the DASS-21 over the same time period. The significant relative magnitude of beta between pain intensity and depression reduced from 0.22 to 0.13 and became non-significant after introducing magnification as a mediating variable, whilst the significant relative magnitude of beta reduced from 0.22 to 0.12 and also became non significant after introducing helplessness as a mediating variable. This finding using longitudinal data is consistent with a previous finding [47], which used cross-sectional data from a different sample to examine the mediating role of catastrophizing. In that study, which also used the PRSS as a measure of catastrophizing, the relationship between pain intensity and depression was significantly mediated by magnification and helplessness. The relative magnitude of beta between pain intensity and depression reduced from 0.28 to 0.11 after introducing magnification as a mediating variable, whilst the relative magnitude of beta reduced from 0.28 to 0.17 when helplessness was introduced as the mediating variable. Together, the findings from this study using longitudinal data and the previous findings using cross-sectional data demonstrate that cognitive factors are important mediators of the pain experience in older adults with persistent pain and contradicts a previous finding that found a strong and direct relationship between pain intensity and depression in this population [38]. Our findings are also consistent with previous findings of a significant association between change in catastrophizing and change in depression in patients receiving a pain management intervention [8,32,43].
These results have important implications for the assessment and treatment of older adults with persistent pain. It is recognized that the population in most western societies is rapidly ageing and additionally it is known that the prevalence of persistent pain increases with age [4,11,42]. Therefore, it is imperative that effective treatments are developed for this population. This study has shown that cognitive factors, in this case catastrophizing, are important mediators in the pain experience of older adults. Firstly, these findings suggest that the assessment of persistent pain in older adults should also routinely include the assessment of catastrophizing, using measures such as the PRSS, shown to be reliable and valid when used in this population. Also given the findings from previous research of age differences in the frequency and severity of depression in pain patients, future research should investigate age differences in the mediating role of catastrophizing in the relationship between pain intensity and depression. Additionally, future research should focus on uncovering other cognitive factors which may also mediate the relationship between pain intensity and depression in this population. The relatively modest strength of the relationships between pain, the mediating factors and depression suggests there remains significant unexplained variance in the model. There is evidence for the role of other cognitive factors in the pain experience of older adults. Self-perceived interference from pain and pain locus of control have been found to be important predictors of depression in older adults with persistent pain [15]. Additionally, high self-efficacy was found to be associated with better adjustment to persistent pain, including lower pain-related disability and depressive symptoms [40].
These findings also suggest that treatments for older adults with pain should focus on reducing the frequency of catastrophizing to reduce the severity of depression in addition to focusing on reducing pain intensity. Treatments such as cognitive behaviour therapy (CBT), which focus on the role of cognitive factors, are underutilized in this population [18]. However, to date there have been mixed findings regarding the effectiveness of CBT for this population [8,23].
There are a number of limitations of the study. A significant limitation of our study is that despite using longitudinal data, the analysis used is only able to establish relationships between the variables and is unable to establish a directional or causal effect. Future research should focus on evaluating treatments that aim at reducing catastrophizing in this population to uncover the causal nature of catastrophizing in the pain-depression relationship in older adults. Second, there was a lack of change in the mean scores of the variables used in the mediation analysis. However, the analysis of the longitudinal data in this study focused on within subject change from Time 1 to Time 2. Third, participants with comorbid conditions associated with pain and depression were not excluded from the study. However, multimorbidity is common in this age group [9] and around 70% of participants reported one or more ongoing diseases or health conditions in addition to persisting pain. Excluding 70% of participants would have resulted in a sample with limited power. Finally, some studies have argued that catastrophizing scales measure cognitive symptoms of depression and not distinct pain-related cognitions, which highlights the difficulty of assessing catastrophizing as a mediating variable in the relationship between pain and depression [35]. However, a later study by the same principal author argued that while catastrophizing shares variance with depression in predicting pain-related disability, it also adds unique variance [36]. Additionally, other studies have supported the argument against the redundancy between catastrophizing and depression [14,19,20].
This study also has a number of strengths. This is the first study to examine the mediating role of catastrophizing using longitudinal data in a two wave design. The use of longitudinal data allows the examination of change in these variables and controls for the influence of extraneous variables when variables are measured only at a single time point. A further strength of this study is that participants were screened for cognitive deficits.
In summary these results show that change in magnification and helplessness totally and significantly mediate the relationship between change in pain intensity and change in depressed mood in older adults with persistent pain. These results highlight the importance of targeting interventions to reduce the influence of these mediating factors to effectively reduce depression in this growing population.
Highlights
The cognitive-behavioural model proposes a role for cognitions in the pain experience.
The role of cognitions in the pain experience of older adults has been unclear.
The role of catastrophizing in the pain experience of older adults was examined.
Catastrophizing mediated the relationship between pain intensity and depressed mood.
Reducing catastrophizing in older adults is an important treatment target.
DOI of refers to article: http://dx.doi.org/10.1016/j.sjpain.2016.02.002.
-
Disclosures: This study was supported by a grant from the Australian Health Ministers Advisory Council (Grant: AHMAC PDR 2005/08).
-
Conflict of interest: There are no conflicts of interest to declare.
References
[1] Antony MM, Bieling PJ, Cox BJ, Enns MW, Swinson RP. Psychometric properties of the 42-item and 21-item versions of the Depression Anxiety Stress Scales in clinical groups and a community sample. Psychol Assess 1998;10: 176–81.Search in Google Scholar
[2] Bair MJ, Robinson RL, Katon W, Kroenke K. Depression and pain comorbidity: a literature review. Arch lntern Med 2003;163:2433–45.Search in Google Scholar
[3] Baron RM, Kenny DA. The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. J Pers Soc Psychol 1986;51:1173–82.Search in Google Scholar
[4] Blyth FM, March LM, Brnabic AJ, Jorm LR, Williamson M, Cousins MJ. Chronic pain in Australia: a prevalence study. Pain 2001;89:127–34.Search in Google Scholar
[5] Casten RJ, Parmelee PA, Kleban MH, Lawton MP, Katz IR. The relationships among anxiety, depression, and pain in a geriatric institutionalized sample. Pain 1995;61:271–6.Search in Google Scholar
[6] Corran TM, Farrell MJ, Helme RD, Gibson SJ. The classification of patients with chronic pain: age as a contributing factor. Clin J Pain 1997;13:207–14.Search in Google Scholar
[7] Corran TM, Gibson SJ, Farrell MJ, Helme RD. Comparison of chronic pain experience between young and elderly patients. In: Gebhart GF, Hammond DL, Jensen TS, editors. Proceedings of the 7th world congress on pain, progress in pain research and management. Seattle: lASP Press;1994. p. 895–906.Search in Google Scholar
[8] Darchuk KM, Townsend CO, Rome JD, Bruce BK, Hooten WM. Longitudinal treatment outcomes for geriatric patients with chronic non-cancer pain at an interdisciplinary pain rehabilitation program. Pain Med 2010;11:1352–64.Search in Google Scholar
[9] Dominick CH, Blyth FM, Nicholas MK. Unpacking the burden: understanding the relationships between chronic pain and comorbidity in the general population. Pain 2012;153:293–304.Search in Google Scholar
[10] Dotson VM, Resnick SM, Zonderman AB. Differential association of concurrent, baseline, and average depressive symptoms with cognitive decline in older adults. Am J Geriatr Psychiatry 2008;16:318–30.Search in Google Scholar
[11] Eriksen J, Jensen MK, Sjogren P, Ekholm O, Rasmussen NK. Epidemiology of chronic non-malignant pain in Denmark. Pain 2003;106:221–8.Search in Google Scholar
[12] Flor H, Behle DJ, Birbaumer N. Assessment of pain-related cognitions in chronic pain patients. Behav Res Ther 1993;31:63–73.Search in Google Scholar
[13] Flor H, Turk DC. Chronic back pain and rheumatoid arthritis: predicting pain and disability from cognitive variables. J Behav Med 1988;11:251–65.Search in Google Scholar
[14] Geisser ME, Robinson ME, Keefe FJ, Weiner ML. Catastrophizing, depression and the sensory, affective and evaluative aspects of chronic pain. Pain 1994;59:79–83.Search in Google Scholar
[15] Gibson SJ, Helme RD. Cognitive factors and the experience of pain and suffering in older persons. Pain 2000;85:375–83.Search in Google Scholar
[16] Gloster AT, Rhoades HM, Novy D, KlotscheJ, Senior A, Kunik M, Wilson N, Stanley MA. Psychometric properties of the Depression Anxiety and Stress Scale-21 in older primary care patients. J Affect Disord 2008;110:248–59.Search in Google Scholar
[17] Jensen MP, Ehde DM, Hoffman AJ, Patterson DR, Czerniecki JM, Robinson LR. Cognitions, coping and social environment predict adjustment to phantom limb pain. Pain 2002;95:133–42.Search in Google Scholar
[18] Kee WG, Middaugh SJ, Redpath S, Hargadon R. Age as a factor in admission to chronic pain rehabilitation. Clin J Pain 1998;14:121–8.Search in Google Scholar
[19] Keefe FJ, Brown GK, Wallston KA, Caldwell DS. Coping with rheumatoid arthritis pain: catastrophizing as a maladaptive strategy. Pain 1989;37:51–6.Search in Google Scholar
[20] Keefe FJ, Lefebvre JC, Egert JR, Affleck G, Sullivan MJ, Caldwell DS. The relationship of gender to pain, pain behavior and disability in osteoarthritis patients: the role of catastrophizing. Pain 2000;87:325–34.Search in Google Scholar
[21] Lopez-Lopez A, Montorio I, Izal M, Velasco L.The role of psychological variables in explaining depression in older people with chronic pain. Aging Ment Health 2008;12:735–45.Search in Google Scholar
[22] Lovibond SH, Lovibond PF. Manual forthe depression anxiety stress scales. 2nd ed. Sydney: Psychology Foundation; 1995.Search in Google Scholar
[23] Lunde LH, NordhuslH, Pallesen S. The effectiveness of cognitive and behavioural treatment of chronic pain in the elderly: a quantitative review. J Clin Psychol Med Settings 2009;16:254–62.Search in Google Scholar
[24] MacKinnon DP. Introduction to statistical mediation analysis. New York, NY: Taylor & Francis Group/Lawrence Erlbaum Associates; 2008.Search in Google Scholar
[25] Nicholas MK, Asghari A, Blyth FM, Wood BM, Murray R, McCabe R, Brnabic A, Beeston L, Corbett M, Sherrington C, Overton S. Self-management intervention for chronic pain in older adults: a randomised controlled trial. Pain 2013;154:824–35.Search in Google Scholar
[26] Onder G, Landi F, Gambassi G, Liperoti R, Soldato M, Catananti C, Finne-Soveri H, Katona C, Carpenter l, Bernabei R. Association between pain and depression among older adults in Europe: results from the aged in home care (AdHOC) project: a cross-sectional study. J Clin Psychiatry 2005;66:982–8.Search in Google Scholar
[27] Parmelee PA, Katz IR, Lawton MP. The relation of pain to depression among institutionalized aged. J Gerontol 1991;46:15–21.Search in Google Scholar
[28] Richardson EJ, Ness TJ, Doleys DM, Banos JH, Cianfrini L, Richards JS. Depressive symptoms and pain evaluations among persons with chronic pain: catastrophizing, but not pain acceptance, shows significant effects. Pain 2009;147:147–52.Search in Google Scholar
[29] Riley III JL, Wade JB, Robinson ME, Price DD. The stages of pain processing across the adult lifespan. J Pain 2000;1:162–70.Search in Google Scholar
[30] Roth RS, Lowery JC, Hamill JB. Assessing persistent pain and its relation to affective distress, depressive symptoms, and pain catastrophizing in patients with chronic wounds: a pilot study. Am J Phys Med Rehabil 2004;83:827–34.Search in Google Scholar
[31] Rudy TE, Kerns RD, Turk DC. Chronic pain and depression: toward a cognitivebehavioral mediation model. Pain 1988;35:129–40.Search in Google Scholar
[32] Smeets RJ, Vlaeyen JW, Kester AD, Knottnerus JA. Reduction of pain catastrophizing mediates the outcome of both physical and cognitive-behavioral treatment in chronic low back pain. J Pain 2006;7:261–71.Search in Google Scholar
[33] Sobel ME. Asymptotic confidence intervals for indirect effects in structural equation models. In: Leinhart S, editor. Sociological methodology 1982. San Francisco: Jossey-Bass; 1982. p. 290–312.Search in Google Scholar
[34] Storey JE, Rowland JT, Conforti DA, Dickson HG. The Rowland Universal Dementia Assessment Scale (RUDAS): a multicultural cognitive assessment scale. Int Psychogeriatr2004;16:13–31.Search in Google Scholar
[35] Sullivan MJ, D’Eon JL. Relation between catastrophizing and depression in chronic pain patients. J Abnorm Psychol 1990;99:260–3.Search in Google Scholar
[36] Sullivan MJ, Stanish W, Waite H, Sullivan M, Tripp DA. Catastrophizing, pain, and disability in patient with soft-tissue injuries. Pain 1998;77:253–60.Search in Google Scholar
[37] Sullivan MJ, Thorn B, Haythornthwaite JA, Keefe F, Martin M, Bradley LA, Lefebvre JC. Theoretical perspectives on the relation between catastrophizing and pain. Clin J Pain 2001;17:52–64.Search in Google Scholar
[38] Turk DC, Okifuji A, Scharff L. Chronic pain and depression: role of perceived impact and perceived control in different age cohorts. Pain 1995;61:93–101.Search in Google Scholar
[39] Turk DC, Rudy TE. Cognitive factors and persistent pain: a glimpse into Pandora’s box. Cognit Ther Res 1992;16:99–122.Search in Google Scholar
[40] Turner JA, Ersek M, Kemp C. Self-efficacy for managing pain is associated with disability, depression, and pain coping among retirement community residents with chronic pain. J Pain 2005;6:471–9.Search in Google Scholar
[41] Turner JA, Jensen MP, Romano JM. Do beliefs, coping, and catastrophizing independently predict functioning in patients with chronic pain? Pain 2000;85:115–25.Search in Google Scholar
[42] Verhaak PF, Kerssens JJ, Dekker J, Sorbi MJ, Bensing JM. Prevalence of chronic benign pain disorder among adults: a review of the literature. Pain 1998;77:231–9.Search in Google Scholar
[43] Wideman TH, Adams H, Sullivan MJL. A prospective sequential analysis of the fear-avoidance model of pain. Pain 2009;145:45–51.Search in Google Scholar
[44] Williamson GM, Schulz R. Pain, activityrestriction, and symptoms of depression among community-residing elderly adults. J Gerontol 1992;47:367–72.Search in Google Scholar
[45] Wood BM, Nicholas MK, Blyth F, Asghari A, Gibson S. Assessing pain in older people with persistent pain: the NRS is valid but only provides part of the picture. J Pain 2010;11:1259–66.Search in Google Scholar
[46] Wood BM, Nicholas MK, Blyth F, Asghari A, Gibson S. The utility of the short version of the Depression Anxiety Stress Scales (DASS-21) in elderly patients with persistent pain: does age make a difference? Pain Med 2010;11:1780–90.Search in Google Scholar
[47] Wood BM, Nicholas MK, Blyth F, Asghari A, Gibson S. Catastrophizing mediates the relationship between pain intensity and depressed mood in older adults with persistent pain. J Pain 2013;14:149–57.Search in Google Scholar
© 2016 Scandinavian Association for the Study of Pain
