Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Prediction of function in daily life following multidisciplinary rehabilitation for individuals with chronic musculoskeletal pain; a prospective study

Prediction of function in daily life following multidisciplinary rehabilitation for individuals... Background: The prevalence of chronic musculoskeletal pain is high, with widespread negative economic, psychological, and social consequences for the individual. It is therefore important to find ways to predict the outcome of rehabilitation programmes in terms of function in daily life. The aims of this study were to investigate the improvements over time from multidisciplinary rehabilitation in terms of pain and function, and analyse the relative impact of individual and psychosocial factors as predictors of function in daily life in individuals with chronic musculoskeletal pain. Methods: A prospective study was conducted among one hundred and forty three (N = 143) musculoskeletal pain patients. Measures of pain, function, and functional health status were obtained at baseline, after 5 weeks of intensive training, at the end of the 57-week rehabilitation programme, and at a 1 year follow-up, using validated self-administrated measures. Linear regression analysis was applied to investigate the relative impact of musculoskeletal pain, individual- , and psychosocial factors in function. Results: The participants studied showed a significant increase in function during the 57 weeks rehabilitation period. There was also a significant increase in function from the end of the rehabilitation period (57th week) to the one year follow-up measures. Pain intensity associated significantly with pain experience over all measurement periods. High levels of pain intensity (β = .42**) and pain experience (β = .37*), and poor psychological capacity (β = -.68*) at baseline, as well as poor physiological capacity (β = -.44**) and high levels of anxiety (β = .48**) and depression (β = .58***) at the end of the rehabilitation program were the most important prognostic factors of variance in functioning over the 4 measurement periods. Conclusion: The data suggest that physical capacity, emotional distress and coping skills should be priority areas in rehabilitation programmes to improve functioning in daily life. Page 1 of 10 (page number not for citation purposes) BMC Musculoskeletal Disorders 2007, 8:65 http://www.biomedcentral.com/1471-2474/8/65 important contributors to disability and adjustment Background Chronic musculoskeletal pain represents an important among people with chronic pain [10,14]. Therefore, the cause of reduced function in daily life, and constitutes a individual's understanding of the symptoms and the significant and increasing medical, social, and economic impact of the symptoms on everyday life might be an challenge in industrialized countries [1,2]. In more than important way of understanding pain and function. 90 % of the musculoskeletal pain cases, no organic reason can explain the pain that for some individuals persists and Chronic widespread pain and poor health functioning are gets worst to the point where it considerably limits func- significantly associated with a number of environmental tion in everyday activities [3-5]. General pain, viewed as a factors [15-17], acting both through and independently of multidimensional phenomenon with varying degrees of disease. This is emphasized by Krokstad and Westin [18] severity, distribution and functional impact, is considered who demonstrate the importance and impact of social-, to be chronic if it lasts for more than three months [3,6,7]. non-medical-, and contextual determinants in disability. Chronification is not only tied to the duration of pain. Factors such as little social support, little social anchorage, Chronic pain is found to be associated with a multitude of or little need of being social are found to significantly secondary stressors such as sleep disruption, unemploy- increase the odds for a person to experience a high level of ment and interpersonal tensions [3,8,9], and psychosocial pain [19,20]. The development of widespread chronic factors are considered to be among the most important pain is also found to be predicted by higher age, drinking variables that influence the total health picture. The influ- alcohol weekly, smoking, traumas in childhood and a ence of individual and psychosocial factors in function is family history of chronic pain. However, optimistic atti- moreover believed to be stronger for people with chronic tudes about how the pain will interfere with daily life, the musculoskeletal pain [6,10]. Pain and function can also individual's social interaction, and the individual's ability be approached in a cultural and historical context, and are to receive assistance are factors that are found to predict viewed as multidimensional phenomena that are influ- pain reduction [9,19,21-24]. Multidisciplinary treat- enced by many factors, such as the effect of previous expe- ments, in general, are found to effectively improve the rience and cultural beliefs, as well as sensory input [7,10]. functioning of chronic musculoskeletal pain patients in In accordance to the International Association for the daily life. Such treatments are more cost-effective than Study of Pain (IASP), the experience of pain is connected alternative pain control treatments (i.e. 'conservative' care to emotions and is defined as 'an unpleasant sensory and and surgery), and achieve equal or greater efficiency emotional experience associated with actual or potential [25,26]. People who have completed treatment typically tissue damage, or described in terms of such damage' report decreased pain intensity, less depression and less [[11], p.108]. A study by Rudy, Lieber, Boston, Gourley pain related anxiety, improved levels of pain coping skills, and Baysal [12] concluded that more than 90 % of the var- and increased function in daily life. iance in performance among disabled individuals with chronic musculoskeletal pain was predicted by psychoso- Chronic pain in the musculoskeletal system and cial factors; self-efficacy, perceived emotional and physi- responses to rehabilitation treatment has often been stud- cal functioning, pain intensity, and pain cognition being ied in terms of clinical factors and objective determinants the most important. This is supported by Geisser, Robin- of the person [9,10,19,22]. Results from several studies son and Miller [10] maintaining that individual and psy- indicate that physical-, psychological-, and socioeco- chosocial factors were deemed to be of great importance nomic variables play a major role in how pain is experi- in the experience of pain. The consequences of pain for a enced, as well as how individuals respond to person's everyday life are therefore not only dependent on rehabilitation treatment for chronic musculoskeletal pain the underlying pathophysiological impairments, but to a conditions [9,10,18-21]. While advanced designs are large extent decided by that person's perception of the dis- appearing more frequently in chronic musculoskeletal ease in their present life situation. Depression, reported to pain research, there is a need for prospective, inception be highly prevalent among people with chronic pain studies so that we can learn more about the nature of the [2,13], can take many forms and vary in the number and risk factors being studied. Longitudinal follow-up studies, severity of symptoms. Even milder symptoms of depres- conducted in a real clinical setting, are therefore still sion have been found to influence the experience of pain. needed. Both somatic and cognitive symptoms of depression are associated with perceived psychosocial functioning The present study uses a biopsychosocial theoretical among people with chronic musculoskeletal pain, even approach and the empirical findings discussed [27-29] in when controlling for pain intensity and other measures order to: 1. Examine improvement in function over time [2,10]. Pain-related anxiety, the belief that pain is a sign of in individuals with chronic musculoskeletal pain partici- damage or harm to the body, and that activities that might pating in a multidisciplinary rehabilitation programme, cause pain should be avoided are also believed to be and 2. Analyse the relative impact of individual and psy- Page 2 of 10 (page number not for citation purposes) BMC Musculoskeletal Disorders 2007, 8:65 http://www.biomedcentral.com/1471-2474/8/65 chosocial factors as predictors of pain intensity, pain expe- were used from an age-matched group (n = 52186, mean rience, and function in daily life in individuals with age = 43/SD = 12.7) from the Nord-Trøndelag Health chronic musculoskeletal pain participating in a multidis- Study (The HUNT 2 Study). The study was approved by ciplinary rehabilitation programme. Norwegian Social Science Data Service (NSD) and the Regional Medical Ethical Committee of Mid-Norway Methods (REK). All patients were volunteers and gave their Subjects informed consent. Confidentiality was emphasized. The study sample consisted of 143 (N = 143) individuals, aged 20–67 (mean age = 45.7/SD = 8.9), with chronic (> Treatment program 3 month) musculoskeletal pain, who participated in a 57- The multidisciplinary rehabilitation programme (see week long multidisciplinary rehabilitation programme at Table 1), based on a biopsychosocial theoretical model a rehabilitation centre in central-Norway. Data were col- [29,30], consisted of a 5-week intensive period, where the lected at four points in time; at the start of the rehabilita- participants attended approximately 6 h/day, 4 days a tion, after 5 weeks of intensive training, at the end of the week, and a follow-up period of 52 weeks, where the par- 57-week rehabilitation period, and at a 1 year follow-up ticipants attended approximately 6 h/day, 1–3 days a after end of the rehabilitation period. All participants (N week. The participants were assigned to the rehabilitation = 143) completed the 57 weeks rehabilitation period, programme by their medical doctor based on interviews, however, the follow-up response 1 year after the partici- observations, and clinical tests. Formulation of individual pants completed the rehabilitation period was 51 % (n = training and exercise programmes is based on the map- 72). The majority of the participants (N = 143) were ping of the participants. All participants had a personal women (74 %), and 79 % of the participants reported to supervisor, and individual counselling is offered during have primary or technical/vocational school for 1–2 years. the training period. In addition, the majority of the participants reported to be unskilled or skilled workers/craftspeople (59 %). In order In cooperation with the National Health Insurance Office, to compare the characteristics of the study sample with Employment office, employer and other Public Health the general population in the same geographic area, data Services, an individual tailored education and coping Table 1: Content of the multidisciplinary rehabilitation programme Period Intervention Duration Period I: Mapping of the participants resources/ • Introduction to the rehabilitation programme 6 h/day, 4 days a week in 5 weeks intensive training period • Mapping physical-, psychological-, and social function • Individual counselling-based on the mapping; preparation of a long-term plan for the rehabilitation process in cooperation with their medical doctor, social security office and the employer. • Individual and group-based training to improve functional capacity: 1. Individual exercise programme with focus on e.g., endurance, strength, mobility, and relaxation techniques, 2. Group-based education/training in different health related subjects e.g., body structure, diet, exercise planning, coping strategies, communication, strategies for conflict negotiations, and social security system 3. Indoor and outdoor activities every day Period II: Follow-up training/rehabilitation Functional capacity training continues 6 h/day, 1–3 days a week in 52 weeks period (individual and group-based, indoor/outdoor activities, education), individual counselling, clarifying function and work ability, prepare a plan for work re-entry in cooperation with the employer, for example. During/after finishing the rehabilitation period In addition to the regular rehabilitation 1 h/1–3 days a week programme (57 weeks), the rehabilitation centre offers exercise groups e.g., endurance groups, water activity groups, and relaxation training groups in the participant's local community. Page 3 of 10 (page number not for citation purposes) BMC Musculoskeletal Disorders 2007, 8:65 http://www.biomedcentral.com/1471-2474/8/65 process was emphasized that sought to increase func- feelings have been reported to correlate well with other tional capacity, decrease affective distress, and educate measures of physical and emotional functioning respec- patients about the positive health process. Although all tively, such as the Barthel Index and the Zung Depression patients did not receive exactly the same standardized Scale [32]. intervention, as would be expected in a randomized con- trolled trial, our aim with this study was to examine indi- Anxiety and depression, used as predictors (independent) vidual effects in function in real clinical settings. of function (functional health status, pain intensity and pain experience), were assessed by using the Hospital Instruments and procedures Anxiety and Depression Scale (HADS) [32,39]. HADS is a Self-reporting measures were administrated individually brief assessment of anxiety and depression, consisting of to the participants at the rehabilitation centre. Data were 14 items divided into two sub-scales for anxiety and collected at baseline, after the 5-week intensive period, depression, in which the patient rates each item on a four- after the 57-week rehabilitation period and at the one year point scale. Individual items are scored from 0–3 to 3-0, follow-up after the participants finished the rehabilitation depending on the direction of the wording of the items. period. The scores of the items represent the degree of distress: none = 0, unbearably = 3. Tests for reliability (test-retest) The Visual Analogue Scale (VAS) [31,32] was employed to of the scale have been satisfactory with a reproducibility of assess variables on pain (worst imaginable pain, how 0.67–0.77 [32,39]. Factor analysis (varimax method) troublesome the pain is), physical capacity (muscle [36], extracted with eigenvalues > 1.00 as a criterion, indi- strength, endurance capacity, energy, mobility, and bal- cated that items could be grouped according to the two ance), psychological capacity (good feeling inside, mood, main constructs. feeling valuable, extroverted/introverted, optimistic/pes- simistic, calm, and balanced), coping (feeling of coping in The participants' self-reporting about education level, daily life, control and influence in daily life), and cogni- type of job, financial matters, social network, sleep distur- tive capacity (concentration, memory, understand/evalu- bance, tiredness, and history of childhood trauma (inde- ate information, and knowledge). The VAS is a line of 10 pendent variables) was supplemented by personal cm on which pain marks are scored in millimetres, repre- interviews. The self-reporting of traumas include experi- senting the continuum of the symptom to be rated. ences such as; bullying, physical-, emotional-, and/or sex- Instructions about how to rate the present pain, how trou- ual abuse. Except for education, categorized in four levels, blesome the pain is and the present function/capacity all information retrieved from the interviews was catego- were given along with the scale. VAS variables were used rized in two levels of categorical variables. The internal as independent variables (predictors of outcome). Moreo- consistency was acceptable in this study and measures ver, pain intensity and pain experience were used as out- such as Cronbach's alpha coefficients were calculated at come measures as well. The use of the VAS is well 0.80–0.85. established in chronic pain populations, and test-retest reliability of the scale has been satisfactory with a repro- Statistical analysis ducibility of 0.75–0.83 [31,33]. The scale has also been Data were analysed using SPSS for Windows (version used in creative ways to further explore the phenomenon 14.0) software. Frequencies, percentages, mean values of pain perception and reporting, in addition to explore and standard deviation were calculated for continuous other health-related phenomena [31,34,35]. Factor analy- and categorical variables. Multivariate tests (single group sis (varimax method) [36], extracted with eigenvalues > repeated measures design) [36] of the significance of the 1.00 as a criterion, indicated that items could be grouped repeated-measures effect (Pillai's Trace) were provided in according to the intended constructs presented above. order to examine the long-term improvements (variance due to passage of time) of the multidisciplinary rehabili- Function in daily life was measured using the Norwegian tation programme in terms of functional health status version of the Functional Health Status measurement (COOP/WONCA), pain intensity (VAS), pain experience COOP/WONCA Charts [37]. The COOP/WONCA charts, (VAS), anxiety (HADS), and depression (HADS). T-tests used as an outcome indicator (dependent), measure six were used to compare the sample (N = 143) with the core aspects of functional status: physical fitness, feelings, HUNT population from the same geographical area, on daily activities, social activities, changes in health and the anxiety and depression variables. For the initial selec- overall health. Each item is rated on a five-point ordinal tion of potential determinants for the outcome measures scale ranging from 1 ('no limitation at all') to 5 ('severely pain intensity, pain experience and functional health sta- limited'). The test-retest reliability of the original Dart- tus (physical fitness, feelings, daily activities, social activi- mouth version and the Norwegian version was found sat- ties and overall health), univariate linear regression isfactory (r = 0.74–0.86) [38]. The charts of function and analysis, done on the baseline, were used with of signifi- Page 4 of 10 (page number not for citation purposes) BMC Musculoskeletal Disorders 2007, 8:65 http://www.biomedcentral.com/1471-2474/8/65 cance level of p < 0.05. Subsequently, all independent var- loskeletal pain sample was equal to the HUNT population iables that showed significant associations with the (Table 2). outcome measures (dependent) were considered for inclusion into the multivariate linear regression models. Functional status These analyses were carried out separately for the defini- Figure 1 shows the mean and standard deviations of pain tion of outcome variables (pain intensity, pain experi- intensity and pain experience (how troublesome the pain ence, functional health status: physical fitness, feelings, is) measured at 3 points in times during the rehabilitation daily activities, social activities and overall health). In period and at a one year follow-up. Pain intensity and order to identify which variables predict change over time pain experience significantly (p < 0.01) decreased from best, all measurements across the rehabilitation period (3 the start of the rehabilitation period to the one year fol- times) were included in steps in the same model with low-up measures at 109 weeks. Table 3 show the long- effects of these variables on the estimated change of the term improvements (trend over time) of the multidiscipli- outcome variables (functional health status, pain inten- nary rehabilitation programme with a significant sity, pain experience) over the 3 measurement periods improvement in cognitive- (p < 0.001), physiological-, (p (T1-T3). The dependent variable at Time 1 was entered < 0.001), and psychological (p < 0.01) capacity, measured first in the model to control for its effect. In addition, all by VAS, in the pain sample during the 57-week rehabilita- measurements across all times were included in steps in tion period. In addition, scores on the Hospital Anxiety the same model with effects of these variables on the esti- and Depression Scale (HADS), seen in Table 3, showed a mated change of the outcome variables (functional health significant (p < 0.01) reduction in both anxiety and status, pain intensity, pain experience) over the 4 meas- depression during the rehabilitation period. Despite this urement periods (T1-T4). To control for the effect of the reduction, the present pain sample still scored signifi- dependent variable at T1, the variable was entered first in cantly (p < 0.001) higher on the anxiety and depression the model. In the final multivariate models only variables variables at all measurement points during the rehabilita- with p-value less than 0.05 were retained. A p-value of less tion period compared to the HUNT population. than 0.05 was considered statistically significant. In Figure 2, measures of function in daily life using the Results COOP/WONCA charts (Functional Health Status) are Response and baseline characterization of the sample presented as mean values. Functional health status signif- All patients (N = 143) included completed the 57-week icantly increased on the variables feelings (p < 0.05), daily rehabilitation programme, which gave a response of 100 activities (p < 0.05), social activities (p < 0.001), and over- % at the end of the rehabilitation period. However, the all health (p < 0.01) from baseline to the end of the 57th response percentage decreased to 51 % (n = 72) at the 1 week of the rehabilitation period in present sample. How- year follow-up questionnaire on pain intensity, pain expe- ever, a comparison of present musculoskeletal pain sam- rience, and functional health status. The non-response ple with a normative randomized sample (N = 2864) group reported mean pain and pain experience measured from the Ullensaker study [38] on the COOP/WONCA by VAS at respectively 75.4 and 68.0 at the end of the reha- charts, demonstrates that the musculoskeletal pain sam- bilitation period. Further the mean measures on func- ple (N = 143) still report significantly lower function (p < tional health status (COOP/WONCA charts) were 0.01) on all core aspects of functional health status at the calculated at: physical fitness; 2.87, feelings; 2.73, daily end of the 57-week rehabilitation period. A relative low activities; 3.08, social activities; 2.23, and overall health; response (51 %) might limit the relevance of the one year 3.08 at the end of the rehabilitation period in the non- follow-up analysis, however the follow-up measures (109 response group. Back, shoulders, and neck were the most weeks) on functional health status showed that the partic- common pain locations in the sample, and 93.8 % of the Table 2: Characteristics of present sample (N = 143) at baseline participants reported pain in more than two locations. As compared to the HUNT population (n = 52 186). seen in Table 2, the majority (68 %) of the sample was married, and the total per cent exposed to traumas in Characteristics N = 143 n = 52 186 childhood in the present pain sample was 37 %. Sixty Married/cohabitant (%) 71.3 71.7 nine per cent reported sleeplessness and 74 % reported Smoking (%) 53.8 36.6 tiredness in everyday daily life. By comparison, the age- Traumas in childhood (%) 37.1 - matched population from the same geographic area Sleeplessness (%) 69.9 31.3 (HUNT 2) (aged 20–67) consisted of 47.5 % men and Tiredness (%) 74.8 45.5 52.5 % women, 70 % reported to have basic or secondary Poor Social network (%) 17.5 17.5 education, and 22.5 % of the HUNT population reported Poor Economy (%) 34.3 15.5 to be unskilled or skilled workers/craftspeople. The por- tion reporting poor social network in the chronic muscu- Not measured in the HUNT population Page 5 of 10 (page number not for citation purposes) BMC Musculoskeletal Disorders 2007, 8:65 http://www.biomedcentral.com/1471-2474/8/65 Table 3: Multivariate tests of the significance for the repeated-measures effect on functional status in the present sample (Pillai's Trace V) Baseline 5 weeks 57 weeks Variables (N = 143). M(SD) M(SD) M(SD) V F df p VAS Physiological capacity 39.9(15.6) 43.6(15.6) 45.8(18.1) .14 11.61 2 000*** Psychological capacity 58.8(16.4) 59.8(16.0) 62.8(18.3) .07 5.97 2 .003** Coping capacity 57.4(15.9) 56.8(16.2) 57.3(19.2) .00 .12 2 .884 Cognitive capacity 47.9(20.1) 50.0(17.5) 54.3(19.1) .13 10.81 2 .000*** HAD Anxiety 8.83(4.29) 8.65(4.46) 7.93(4.53) .06 5.16 2 .007** Depression 6.03(4.16) 5.59(3.96) 5.08(4.30) .07 5.34 2 .006** a b Physiological; muscle strength, endurance capacity, energy, mobility and balance, Psychological; good feeling inside, mood, feeling valuable, extroverted/introverted, optimistic/pessimistic, calm and balanced, Coping; feeling of not coping in daily life, control and influence in daily life, Cognitive; concentration, memory, understand/evaluate information, knowledge. *p < 0.05. **p < 0.01. ***p < 0.001. Univariate linear regression analysis ipants continued to improve their function in daily activ- ities (M = 2.82/SD = .95), feelings (M = 2.55/SD = 1.27), Univariate linear regression analysis, done on the base- and overall health (M = 3.03/SD = .77), compared to the line, showed that a multitude of potential prognostic indi- 57th week measures (daily activities M = 3.10/SD = .95, cators associated significantly with our primary outcome feelings M = 2.71/SD = 1.15, and overall health M = 3.10/ measure functional health status (physical fitness, feel- SD = .85). However, the improvement in function was sig- ings, daily activities, social activities and overall health). nificant only in daily activities (p < 0.05). In addition, the participants reported a decrease in physical fitness and Here, poor physiological capacity (F = 19.92/p < 0.000) social activities one year after they completed the rehabil- and high pain experience (F = 4.06/p < 0.046) signifi- itation period, compared to the 57th week measures. The cantly associated with poor physical fitness, while limita- decrease in physical fitness and social activities was not tion on the outcome variable feelings associated with significant, however. poor financial situation (F = 13.05/p < 0.000), experience of traumas in childhood (F = 11.37/p < 0.001), poor social network (F = 9.85/p < 0.002) and high levels of anx- iety (F = 111.61/p < 0.000) and depression (F = 66.42/p < Pain intensity Pain experience Standard Deviation 5,00 80 M = 77,3 M = 73,6 M = 74,1 M = 74,1 Physical fitness M = 69,8 M = 68,7 M = 67,9 M = 65,6 Feelings 4,00 Daily activities Social activities Overall health 3,00 2,00 Baseline 5 weeks 57 weeks 109 weeks 1,00 Baseline 5 weeks 57 weeks 109 weeks Mean a r Figure 1 ience measured nd standard deviation of by VAS pain intensity and pain expe- Mean and standard deviation of pain intensity and Repea (COOP-WONCA) in the prese ple Figure 2 ted measures mean for Fun nctional Health Status t musculoskeletal pain sam- pain experience measured by VAS. Mean (M) and stand- Repeated measures mean for Functional Health Sta- ard deviation (SD) of pain intensity and pain experience (how tus (COOP-WONCA) in the present musculoskele- troublesome the pain is) measured by Visual Analogue Scale tal pain sample. Mean Functional Health status measured (0–100) at the start of the rehabilitation period, after 5 at the start of the rehabilitation period, after 5 weeks of weeks of intensive training, at the end of the 57-week reha- intensive training, at the end of the 57 weeks rehabilitation bilitation period (N = 143), and at the one year follow-up period (N = 143), and at the one year follow-up measures (n measures (n = 72) in present musculoskeletal pain sample. = 72). 1 = no limitation at all, 5 = severely limited. Page 6 of 10 (page number not for citation purposes) COOP-WONCA Scale BMC Musculoskeletal Disorders 2007, 8:65 http://www.biomedcentral.com/1471-2474/8/65 0.000). Moreover, there was a significant association (β = .15*) and pain experience (β = .35**) (T3), were the between limitation in feelings and poor physiological (F strongest predictors of variance of functioning (functional = 5.77/p < 0.018)-, psychological (F = 68.93/p < 0.000)-, health status measured by COOP/WONCA) over the 3 coping (F = 12.59/p < 0.001)-, and cognitive (F = 13.97/p measurement periods (57 week rehabilitation period) < 0.000) capacity in present musculoskeletal pain sample. (Table 4). Sleeplessness (F = 6.73/p < 0.010), high levels of pain Linear regression analysis (B = Unstandardized Coeffi- intensity (F = 18.26/p < 0.000), pain experience (F = cients, SE = Std. Error, β = Standardized Coefficients 25.27/p < 0.000), anxiety (F = 6.58/p < 0.011) and depres- derived from the final step) was also performed on all sion (F = 6.52/p < 0.012) were in the univariate analysis measurements across all times with effects of the inde- significantly associated with limitation in daily activities. pendent variables on the estimated change of functional Furthermore, limitation in daily activities associated with health status over the 4 measurement periods (not poor physiological (F = 11.90/p < 0.001)-, psychological included in table). Variance in functioning (functional (F = 5.10/p < 0.025) -, coping (F = 6.89/p < 0.010)-, and health status measured by COOP/WONCA) over the 4 cognitive (F = 5.03/p < 0.026) capacity. Limitation in measurement periods (T1-T4) were significantly predicted social activities was significantly associated with poor by experience of traumas in childhood (B(SE) = .50(.24), social network (F = 4.93/p < 0.028), high levels of anxiety β = .29*), high levels of pain intensity (B(SE) = .02(.00), (F = 12.27/p < 0.001) and depression (F = 10.41/p < β = .42**) and pain experience (B(SE) = .02(.00), β = 0.002), and reporting poor physiological (F = 10.41/p < .37*), and poor psychological capacity (B(SE) = .05(.02), 0.002)-, psychological (F = 22.16/p < 0.000) -, coping (F β = -.68*) at baseline (T1). Moreover, poor physiological = 7.79/p < 0.006)-, and cognitive (F = 10.57/p < 0.001) capacity (B(SE) = -.02(.00), β = -.44**) and high levels of capacity in present pain sample. Table 4: Effects of independent variables on the estimated change of functional health status over the 3 measurement periods The univariate analysis, done on the baseline, also 2 2 Variables (N = 143). B (SE) β ∆R R showed a significant association between poor overall health and high age (F = 4.53/p < 0.035), experience of Physical fitness traumas in childhood (F = 7.68/p < 0.006), poor social Step 1: Dependent variable T-1 - - .23 .23 network (F = 10.62/p < 0.001), and high levels of pain Step 2: Independent variables T-1 - - .08 .32 intensity (F = 7.73/p < 0.006). In addition, limitation in Step 3: Independent variables T2-3 - - .10 .42 overall health was significantly associated with high levels Physiological capacity T3 -.07(.00) -.30** - - of anxiety (F = 10.81/p < 0.001) and depression (F = Feelings 24.40/p < 0.000), and poor physiological (F = 24.77/p < Step 1: Dependent variable T-1 - - .20 .20 Step 2: Independent variables T-1 - - .11 .32 0.000)-, psychological (F = 16.19/p < 0.000) -, coping (F Step 3: Independent variables T2-3 - - .40 .72 = 17.04/p < 0.000)-, and cognitive (F = 11.19/p < 0.001) Psychological capacity T3 -.02(.00) -.38** - - capacity in present musculoskeletal pain sample. Anxiety T3 .15(.02) .59*** - - Daily activities Poor physiological capacity was the only variable that sig- Step 1: Dependent variable T-1 - - .15 .15 nificantly associated with high levels of pain intensity (F Step 2: Independent variables T-1 - - .08 .23 Step 3: Independent variables T2-3 - - .28 .52 = 7.88/p < 0.006) and pain experience (F = 11.39/p < Pain experience T3 .01(.00) .35** - - 0.001) in the univariate analysis, done on the baseline, in Social activities the musculoskeletal pain sample. Step 1: Dependent variable T-1 - - .12 .12 Step 2: Independent variables T-1 - - .07 .20 Multivariate linear regression analysis Step 3: Independent variables T2-3 - - .23 .44 Table 4 summarizes the multivariate linear regression Depression T3 .07(.03) .31* - - analysis with effects of the independent variables (only Overall health Step 1: Dependent variable T-1 - - .20 .20 significant variables included in the table) across the reha- Step 2: Independent variables T-1 - - .12 .33 bilitation period (3 times) on the estimated change of the Cognitive capacity -.00(.00) -.17* - - outcome (functional health status) over the 3 measure- Step 3: Independent variables T2-3 - - .30 .64 ment periods (T1-T3). Cognitive capacity (β = -.17*) was Physiological capacity T2 -.01(.00) -.24* - - the only baseline (T1) measure that associated signifi- Physiological capacity T3 -.02(.00) -.45*** - - cantly with functional health status (overall health) in the Pain intensity T3 .00(.00) .15* - - final model (Table 4). Poor physiological (β = -.24*/- .45***) (T2 and T3)- and psychological (β = -.38**) (T3) B = Unstandardized Coefficients, SE = Std. Error, β = Standardized Coefficients derived from the final step. ∆R = change in explanation capacity, high levels of anxiety (β = .59***) and depres- rate in each step. R = proportion of variance explained. *p < 0.05. **p sion (β = .31*) (T3), as well as high levels of pain intensity < 0.01. ***p < 0.001. Page 7 of 10 (page number not for citation purposes) BMC Musculoskeletal Disorders 2007, 8:65 http://www.biomedcentral.com/1471-2474/8/65 anxiety (B(SE) = .13(.04), β = .48**) and depression still report significantly lower function on all core aspects (B(SE) = .16(.04), β = .58***) at the end of the rehabili- of functional health status compared to a normative sam- tation program (T3) were found to significantly predict ple from the Ullensaker study (N = 2864) [38] at all points the variance in functioning (functional health status of measurement. In addition, the significant decrease in measured by COOP/WONCA) over the 4 measurement self-reported physical fitness and social activities from the periods. 57th week of the rehabilitation period to one year after the participants finished the rehabilitation programme Variance in pain intensity over the 3 measurement periods give rise for concern. Lack of physical fitness and partici- (not included in table) was significantly associated with pation in social activities might later on influence several high levels of pain experience (T3) (B(SE) = .54(.07), β = aspects of function in daily life and might not be benefi- .65***) and physiological capacity (T3) (B(SE) = cial to the individuals or to the society. Future studies .19(.09), β = .19*). The association between pain inten- should therefore try to clarify the long-term effect of sity and physiological capacity was however not signifi- multidisciplinary rehabilitation programmes for individ- cant over the 4 measurements periods. High levels of pain uals with chronic musculoskeletal pain in terms of func- β = .48***) was the intensity (T3) (B(SE) = .58(.08), tion in daily life. Non-specific musculoskeletal pain is an strongest predictor of variance in pain experience over the increasing health problem in the Norwegian population. 3 measurement periods. Moreover, the association The increased study of individual rehabilitation in a for- between pain intensity (T3) (B(SE) = .48(.22), β = .36*) mal rehabilitation programme must not reduce focus on and pain experience was significant over all 4 measure- primary prevention programmes at a population level and ment periods. on the social- and economic policy implications of the present findings. Discussion This study showed that a multitude of factors had an effect Several studies [5,10,41] suggest that the impairment of on pain intensity, pain experience, and functional health function in daily life is associated with several psychoso- status over the measurement periods in a Norwegian sam- cial factors. The intent of this study was to study the long- ple, and different variables affected different aspects of term improvements of a multidisciplinary rehabilitation daily life function. The participants were found to signifi- programme, by focusing on interactions and the influence cantly improve several aspects related to function during of a broad range of socio-demographic and psychosocial the rehabilitation period. However, it still might be rele- factors in pain intensity, pain experience, and functional vant to question in what way these changes influence the health status. In order to do that, the predictors of change everyday life of the people in this sample. Ultimately, the in pain intensity, pain experience, and functional health consequences of chronic musculoskeletal pain for every- status over time were studied. The relationship between day function depend not only on pain intensity and pain emotional distress, chronic pain and function in daily life experience, but also on the individual and on each per- has been shown before [13]. In this study experience of son's unique set of earlier experiences, values, and envi- traumas in childhood, emotional distress, high levels of ronmental conditions. This illustrates the complexity of pain intensity and pain experience, and poor physical chronic pain conditions, where the person's perception of capacity, measured at baseline, were significantly predict- pain and function and his/her experiences of what it ing lack of improvement in functional health status over means in their everyday life might be an important way of all measurement periods. In terms of emotional distress, understanding the complexity. Therefore, the relative it is also relevant to notice the relative high percentage (37 influence of psychosocial factors on function may vary a %) of traumas in the present pain sample. The partici- lot depending on the activity the individuals are engaged pants report significantly higher levels of anxiety and in [40]. For the person that receives treatment the impor- depression before, during, and after the treatment period tance of the overall effectiveness of the rehabilitation pro- compared to the normative population from the same gramme is re-establishing function. However, the geographical area (The HUNT Study). Taken together, and programme is also important from a broader perspective. supported by previous studies as well [2,10,13], this illus- The reduction in pain intensity and pain experience along trates the complexity and the relative importance of emo- with improved function in daily life indicate a positive tional distress in chronic musculoskeletal pain effect from the extensive rehabilitation programme. This conditions. is further underlined by the increase in function in daily activities, feelings, and overall health from the 57th week A study by Palermo and Kiska [42] suggested that sleep of the rehabilitation period to one year after the partici- disturbance is closely linked to mood disturbance. How- pants finished the rehabilitation programme. However, it ever, less is known about the complex interrelationship is important to note that although the participants between emotional distress, sleeplessness and function in improved function during the rehabilitation period, they daily life among adults with chronic musculoskeletal pain Page 8 of 10 (page number not for citation purposes) BMC Musculoskeletal Disorders 2007, 8:65 http://www.biomedcentral.com/1471-2474/8/65 conditions. More than 69 % of our sample reported sleep musculoskeletal pain, and in primary prevention at a pop- disturbance in daily life. However, sleeplessness adjusted ulation level. for emotional distress like anxiety and depression, was not found to be useful in the prediction of function. The Conclusion results in this study confirm the physical capacity and cop- This study has evaluated a complexity of factors that have ing aspects in multidisciplinary rehabilitation found in theoretical or empirical relationships to function in daily past research as well [43-45], suggesting that physical life in a sample with persisting musculoskeletal pain. The exercise, behavioral and cognitive-behavioral treatment results of this study highlight important individual per- for chronic pain reduces pain, pain distress, and improves spectives in chronic musculoskeletal pain. These results daily functioning. Moreover, in accordance with a study are important to better understand which variables are by Lame, Peters, Vlaeyen, Kleef and Patijn [46], our study most useful in helping patients re-establish function dur- indicates the relevance of pain experience in predicting ing a rehabilitation programme and they show how to function, and that function in daily life might be associ- address the variables that affect the outcome. In a broader ated with beliefs about pain. perspective, and as seen in relation to the high prevalence of people with chronic musculoskeletal pain conditions, The participants in this study are not randomly sampled; it is also important to pay attention to the underlying they represent all patients participating in the rehabilita- causes of incidence and primary prevention at a popula- tion programme at a given period. The drop-out rate on tion level. long term follow-up might limit the power of the follow- up analysis and results. However, the participants are rep- Competing interests resentative for people with chronic musculoskeletal pain The author(s) declare that they have no competing inter- seeking help at a rehabilitation clinic with respect to age, ests. sex, pain conditions, working ability and sick leave. The sample and the general population from the same geo- Authors' contributions graphical area are almost identical with regard to age dis- ML, SK and GAE designed the study, ML collected the tribution, family situation, social network, and education data, ML analysed and wrote up the manuscript. ML, SK, level. This allows scrutinization of differences between the GAE revised the manuscript. All authors read and sample and the general population without taking the fac- approved the final manuscript. tors mentioned above into consideration as an explana- tory variable. Even with a 100 % response at the end of the Acknowledgements The authors acknowledge the participation of Friskgården, located in Nord- rehabilitation period, a relatively small number of partic- Trøndelag County, that since 1995 has developed a multidisciplinary reha- ipants could lead to a reduction in the power of the anal- bilitation model for individuals on sick leave with complex disease condi- ysis and decrease the possibility of generalization. tions. We also acknowledge the Nord-Trøndelag Health Study (The HUNT Another limitation of the results is the possibility of bias Study) which is a collaboration between HUNT Research Centre, Faculty related to the self-reported data [47]. However, multidi- of Medicine, Norwegian University of Science and Technology (NTNU, mensional rehabilitation, as in present study, represents Verdal), Norwegian Institute of Public Health, and Nord-Trøndelag County an approach that has the potential to effectively focus on Council. function in daily life, not necessarily on rendering the individual symptom free, which might provide poten- References 1. Rustøen T, Wahl AK, Hanestad BR, Lerdal A, Paul S, Miaskowski C: tially greater reliability in the self-assessment of function. Prevalence and characteristics of chronic pain in the general Some might possibly argue that improvement over time is Norwegian population. European Journal of Pain 2004, 8:555-565. not very surprising since pain patients often are selected 2. Ericsson M, Poston WSC, Linder J, Taylor JE, Haddock CK, Foreyt JP: Depression predicts disability in long-term chronic pain close to their worse status. The participants in present patients. Disability and Rehabilitation 2002, 24(6):334-340. study are long-term chronic pain patients, in some cases 3. Anderson IH: The course of non-malignant chronic pain: a 12- reporting pain duration of more than 10 years. Due to the year follow-up of a cohort from the general population. Euro- pean Journal of Pain 2004, 8:47-53. chronicity considerable improvements in function may 4. Law M: Participation in the occupations of everyday life. Amer- therefore not be anticipated. The VAS scale has been used ican Journal of Occupational Therapy 2002, 56:640-649. 5. Truchon M: Determinants of chronic disability related to low in creative ways to explore the phenomenon of pain per- back pain: Towards an integrative biopsychosocial model. ception and reporting in addition to exploring other Disability and Rehabilitation 2001, 23(17):758-767. health related phenomena [32,34,35]. However, the VAS 6. Frischenschlager O, Pucher I: Psychological management of pain. Disability and Rehabilitation 2002, 8:416-422. measures used in present study should be further vali- 7. Honkasalo ML: Chronic pain as a posture towards the world. dated. Despite several shortcomings, the study highlights Scandinavian Journal of Psychology 2000, 41:197-208. 8. Newth S, Delongis A: Individual differences, mood, and coping important perspectives in a real clinical setting that with chronic pain in rheumatoid arthritis: A daily process should be taken into account in rehabilitation of chronic analysis. Psychology and Health 2004, 3:283-305. Page 9 of 10 (page number not for citation purposes) BMC Musculoskeletal Disorders 2007, 8:65 http://www.biomedcentral.com/1471-2474/8/65 9. Keough JL, Fisher TF: Occupational-psychological perceptions for degenerative lumbar spine disorders. Spine 2001, influencing return to work and functional performance of 26:2375-2380. injured workers. Work 2001, 16:101-110. 36. Bryman A, Cramer D: Quantitative Data Analysis with SPSS Release 10 10. Geisser ME, Robinson ME, Miller QL, Bade SM: Psychosocial fac- for Windows. A guide for Social Scientists USA, Routledge; 2001. tors and functional capacity evaluation among persons with 37. Bentsen BG, Natvig B, Winnem M: Assessment of own functional chronic pain. Journal of Occupational Rehabilitation 2003, 4:259-276. capacity. COOP-WONCA Charts in clinical work and 11. Brannon L, Feist J: Health Psychology. An introduction to behaviour and research. Tidsskrift Norske Lægeforening 1997, 117:545-549. health 3rd edition. Brooks/Cole Publishing Company; 1997. 38. Bentsen BG, Natvig B, Winnem M: Questions you didn't ask? 12. Rudy TE, Lieber SJ, Boston JR, Gourley LM, Baysal E: Psychosocial COOP/WONCA Charts in clinical work and research. Family predictors of physical performance in disabled individuals Practice 1999, 2:190-195. with chronic pain. Clinical Journal of Pain 2003, 1:18-30. 39. Mykletun A, Stordal E, Dahl AA: Hospital anxiety and depression 13. Henderson M, Kidd BL, Pearson RM, White PD: Chronic upper (HAD) scale: factor structure, its analyses and internal con- limb pain: an exploration of the biopsychosocial model. Jour- sistency in a large population. British Journal of Psychiatri 2001, nal of Rheumatology 2005, 32(1):118-122. 179:540-544. 14. Vowles KE, Zvolensky MJ, Gross RT, Sperry JA: Pain-related anxi- 40. Jakobsen K: Employment and reconstruction of self. A model ety in the prediction of chronic low-back pain distress. Journal of space for maintenance of identity by occupation. Scandina- of Behavioural Medicine 2004, 1:77-89. vian Journal of Occupational Therapy 2001, 8:40-48. 15. Bergman S: Psychosocial aspects of chronic widespread pain 41. Saastamoinen P, Leino-Arjas P, Laaksonen M, Lahelma E: Socio-eco- and fibromyalgia. Disability and Rehabilitation 2005, 12:675-683. nomic differences in the prevalence of acute, chronic and dis- 16. IJzelenberg W, Burdorf A: Risk factors for musculoskeletal abling chronic pain among ageing employees. Pain 2005, symptoms and ensuing health care use and sick leave. Spine 3:364-371. 2005, 30:1550-1556. 42. Palermo TM, Kiska R: Subjective sleep disturbance in adoles- 17. Lotters F, Burdorf A: Prognostic factors for duration of sickness cents with chronic pain: relationship to daily functioning and absence due to musculoskeletal disorders. Clinical Journal of quality of life. Journal of Pain 2005, 3:201-207. Pain 2006, 22:212-221. 43. McCracken LM, Turk DC: Behavioural and cognitive-behav- 18. Krokstad S, Westin S: Disability in society-medical and non- ioural treatment for chronic pain: outcome, predictors of medical determinants for disability pension in a Norwegian outcome, and treatment. Spine 2002, 22:2564-2573. total county population study. Social Science Medicine 2004, 44. Pedersen BK, Saltin B: Evidence for prescribing exercise as ther- 10:1837-1848. apy in chronic disease. Scandinavian Journal of Medicine & Science in 19. Michaelson P, Sjolander P, Johansson H: Factors predicting pain Sports 2006, 16:3-63. reduction in chronic back and neck pain after multimodal 45. Staal JB, Rainville J, Fritz J, van Mechelen W, Pransky G: Physical treatment. Clinical Journal of Pain 2004, 20(6):447-454. Exercise Intervention to Improve Disability and Return to 20. Katz WA: Musculoskeletal pain and its socioeconomic impli- Work in Low Back Pain; Current Insights and Opportunities cations. Clinical Rheumatology 2002, 21:2-4. for Improvement. Journal of Occupational Rehabilitation 2005, 21. Kaiser PO, Mattsson B, Marklund S, Wimo A: The impact of psy- 15:491-505. chosocial "markers" on the outcome of rehabilitation. Disa- 46. Lame IE, Peters ML, Vlaeyen JW, Kleef M, Patijn J: Quality of life in bility and Rehabilitation 2001, 10:430-435. chronic pain is more associated with beliefs about pain, than 22. Bergmann S, Herrstrom P, Jacobson LT, Petersson IF: Chronic pain intensity. European Journal of Pain 2005, 1:15-24. widespread pain: a three year followup of pain distribution 47. Gatchel RJ: Psychosocial Factors That Can Influence the Self- and risk factors. Journal of Rheumatology 2002, 29(4):818-825. Assessment of Function. Journal of Occupational Rehabilitation 23. Brown J, Berenson K, Cohen P: Documented and self-reported 2004, 14:197-206. child abuse and adult pain in a community sample. Clinical Journal of Pain 2005, 21(5):374-377. Pre-publication history 24. Fillingim RB, Edwards RR: Is self-reported childhood abuse his- The pre-publication history for this paper can be accessed tory associated with pain perception among health young women and men? Clinical Journal of Pain 2005, 21(5):387-397. here: 25. Vowles KE, Gross RT, Sorrell JT: Predicting work status follow- ing interdisciplinary treatment for chronic pain. European Jour- nal of Pain 2004, 4:351-358. http://www.biomedcentral.com/1471-2474/8/65/prepub 26. Turk DC: Clinical effectiveness and cost effectiveness of treat- ments for patients with chronic pain. Clinical Journal of Pain 2002, 18:355-365. 27. Vowles KE, Gross RT: Work-related beliefs about injury and physical capability for work in individuals with chronic pain. Pain 2003, 101:291-298. 28. McCracken LM, Gross RT: Does anxiety affect coping with pain? Clinical Journal of pain 2002, 9:253-259. 29. Engel GL: The need for a new medical model: a challenge for biomedicine. Science 1977, 196:129-136. 30. Lillefjell M, Krokstad S, Espnes GA: Factors predicting work abil- ity following multidisciplinary rehabilitation for chronic mus- Publish with Bio Med Central and every culoskeletal pain. Journal of Occupational Rehabilitation 2006, scientist can read your work free of charge 16:543-555. 31. McDowell I, Newell C: Measuring Health – A Guide to Rating Scales and "BioMed Central will be the most significant development for Questionnaires 2nd edition. New York: Oxford University Press; 1997. disseminating the results of biomedical researc h in our lifetime." 32. Bowling A: Measuring Health. A review of quality of life measurement Sir Paul Nurse, Cancer Research UK scales 2nd edition. Open University Press; 1997. 33. Bergh I: Assessing pain in elderly. Rating scales, and their rela- Your research papers will be: tion to verbal expression of pain and pain relief. In Licentiate available free of charge to the entire biomedical community thesis Gøteborg University, Sweden; 2001. 34. McGeary DD, Mayer TG, Gatchel RJ: High pain ratings predict peer reviewed and published immediately upon acceptance treatment failure in chronic occupational musculoskeletal cited in PubMed and archived on PubMed Central disorders. Journal of Bone and Joint Surgery 2006, 88:317-325. 35. Zanoli G, Stromquist B, Jonsson B: Visual analog scales for inter- yours — you keep the copyright pretation of back and leg pain intensity in patients operated BioMedcentral Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp Page 10 of 10 (page number not for citation purposes) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png BMC Musculoskeletal Disorders Springer Journals

Prediction of function in daily life following multidisciplinary rehabilitation for individuals with chronic musculoskeletal pain; a prospective study

Download PDF
10 pages

Loading next page...
 
/lp/springer-journals/prediction-of-function-in-daily-life-following-multidisciplinary-jHh26FwQWZ

References (67)

Publisher
Springer Journals
Copyright
Copyright © 2007 by Lillefjell et al; licensee BioMed Central Ltd.
Subject
Medicine & Public Health; Orthopedics; Rehabilitation; Rheumatology; Sports Medicine; Internal Medicine; Epidemiology
eISSN
1471-2474
DOI
10.1186/1471-2474-8-65
pmid
17623074
Publisher site
See Article on Publisher Site

Abstract

Background: The prevalence of chronic musculoskeletal pain is high, with widespread negative economic, psychological, and social consequences for the individual. It is therefore important to find ways to predict the outcome of rehabilitation programmes in terms of function in daily life. The aims of this study were to investigate the improvements over time from multidisciplinary rehabilitation in terms of pain and function, and analyse the relative impact of individual and psychosocial factors as predictors of function in daily life in individuals with chronic musculoskeletal pain. Methods: A prospective study was conducted among one hundred and forty three (N = 143) musculoskeletal pain patients. Measures of pain, function, and functional health status were obtained at baseline, after 5 weeks of intensive training, at the end of the 57-week rehabilitation programme, and at a 1 year follow-up, using validated self-administrated measures. Linear regression analysis was applied to investigate the relative impact of musculoskeletal pain, individual- , and psychosocial factors in function. Results: The participants studied showed a significant increase in function during the 57 weeks rehabilitation period. There was also a significant increase in function from the end of the rehabilitation period (57th week) to the one year follow-up measures. Pain intensity associated significantly with pain experience over all measurement periods. High levels of pain intensity (β = .42**) and pain experience (β = .37*), and poor psychological capacity (β = -.68*) at baseline, as well as poor physiological capacity (β = -.44**) and high levels of anxiety (β = .48**) and depression (β = .58***) at the end of the rehabilitation program were the most important prognostic factors of variance in functioning over the 4 measurement periods. Conclusion: The data suggest that physical capacity, emotional distress and coping skills should be priority areas in rehabilitation programmes to improve functioning in daily life. Page 1 of 10 (page number not for citation purposes) BMC Musculoskeletal Disorders 2007, 8:65 http://www.biomedcentral.com/1471-2474/8/65 important contributors to disability and adjustment Background Chronic musculoskeletal pain represents an important among people with chronic pain [10,14]. Therefore, the cause of reduced function in daily life, and constitutes a individual's understanding of the symptoms and the significant and increasing medical, social, and economic impact of the symptoms on everyday life might be an challenge in industrialized countries [1,2]. In more than important way of understanding pain and function. 90 % of the musculoskeletal pain cases, no organic reason can explain the pain that for some individuals persists and Chronic widespread pain and poor health functioning are gets worst to the point where it considerably limits func- significantly associated with a number of environmental tion in everyday activities [3-5]. General pain, viewed as a factors [15-17], acting both through and independently of multidimensional phenomenon with varying degrees of disease. This is emphasized by Krokstad and Westin [18] severity, distribution and functional impact, is considered who demonstrate the importance and impact of social-, to be chronic if it lasts for more than three months [3,6,7]. non-medical-, and contextual determinants in disability. Chronification is not only tied to the duration of pain. Factors such as little social support, little social anchorage, Chronic pain is found to be associated with a multitude of or little need of being social are found to significantly secondary stressors such as sleep disruption, unemploy- increase the odds for a person to experience a high level of ment and interpersonal tensions [3,8,9], and psychosocial pain [19,20]. The development of widespread chronic factors are considered to be among the most important pain is also found to be predicted by higher age, drinking variables that influence the total health picture. The influ- alcohol weekly, smoking, traumas in childhood and a ence of individual and psychosocial factors in function is family history of chronic pain. However, optimistic atti- moreover believed to be stronger for people with chronic tudes about how the pain will interfere with daily life, the musculoskeletal pain [6,10]. Pain and function can also individual's social interaction, and the individual's ability be approached in a cultural and historical context, and are to receive assistance are factors that are found to predict viewed as multidimensional phenomena that are influ- pain reduction [9,19,21-24]. Multidisciplinary treat- enced by many factors, such as the effect of previous expe- ments, in general, are found to effectively improve the rience and cultural beliefs, as well as sensory input [7,10]. functioning of chronic musculoskeletal pain patients in In accordance to the International Association for the daily life. Such treatments are more cost-effective than Study of Pain (IASP), the experience of pain is connected alternative pain control treatments (i.e. 'conservative' care to emotions and is defined as 'an unpleasant sensory and and surgery), and achieve equal or greater efficiency emotional experience associated with actual or potential [25,26]. People who have completed treatment typically tissue damage, or described in terms of such damage' report decreased pain intensity, less depression and less [[11], p.108]. A study by Rudy, Lieber, Boston, Gourley pain related anxiety, improved levels of pain coping skills, and Baysal [12] concluded that more than 90 % of the var- and increased function in daily life. iance in performance among disabled individuals with chronic musculoskeletal pain was predicted by psychoso- Chronic pain in the musculoskeletal system and cial factors; self-efficacy, perceived emotional and physi- responses to rehabilitation treatment has often been stud- cal functioning, pain intensity, and pain cognition being ied in terms of clinical factors and objective determinants the most important. This is supported by Geisser, Robin- of the person [9,10,19,22]. Results from several studies son and Miller [10] maintaining that individual and psy- indicate that physical-, psychological-, and socioeco- chosocial factors were deemed to be of great importance nomic variables play a major role in how pain is experi- in the experience of pain. The consequences of pain for a enced, as well as how individuals respond to person's everyday life are therefore not only dependent on rehabilitation treatment for chronic musculoskeletal pain the underlying pathophysiological impairments, but to a conditions [9,10,18-21]. While advanced designs are large extent decided by that person's perception of the dis- appearing more frequently in chronic musculoskeletal ease in their present life situation. Depression, reported to pain research, there is a need for prospective, inception be highly prevalent among people with chronic pain studies so that we can learn more about the nature of the [2,13], can take many forms and vary in the number and risk factors being studied. Longitudinal follow-up studies, severity of symptoms. Even milder symptoms of depres- conducted in a real clinical setting, are therefore still sion have been found to influence the experience of pain. needed. Both somatic and cognitive symptoms of depression are associated with perceived psychosocial functioning The present study uses a biopsychosocial theoretical among people with chronic musculoskeletal pain, even approach and the empirical findings discussed [27-29] in when controlling for pain intensity and other measures order to: 1. Examine improvement in function over time [2,10]. Pain-related anxiety, the belief that pain is a sign of in individuals with chronic musculoskeletal pain partici- damage or harm to the body, and that activities that might pating in a multidisciplinary rehabilitation programme, cause pain should be avoided are also believed to be and 2. Analyse the relative impact of individual and psy- Page 2 of 10 (page number not for citation purposes) BMC Musculoskeletal Disorders 2007, 8:65 http://www.biomedcentral.com/1471-2474/8/65 chosocial factors as predictors of pain intensity, pain expe- were used from an age-matched group (n = 52186, mean rience, and function in daily life in individuals with age = 43/SD = 12.7) from the Nord-Trøndelag Health chronic musculoskeletal pain participating in a multidis- Study (The HUNT 2 Study). The study was approved by ciplinary rehabilitation programme. Norwegian Social Science Data Service (NSD) and the Regional Medical Ethical Committee of Mid-Norway Methods (REK). All patients were volunteers and gave their Subjects informed consent. Confidentiality was emphasized. The study sample consisted of 143 (N = 143) individuals, aged 20–67 (mean age = 45.7/SD = 8.9), with chronic (> Treatment program 3 month) musculoskeletal pain, who participated in a 57- The multidisciplinary rehabilitation programme (see week long multidisciplinary rehabilitation programme at Table 1), based on a biopsychosocial theoretical model a rehabilitation centre in central-Norway. Data were col- [29,30], consisted of a 5-week intensive period, where the lected at four points in time; at the start of the rehabilita- participants attended approximately 6 h/day, 4 days a tion, after 5 weeks of intensive training, at the end of the week, and a follow-up period of 52 weeks, where the par- 57-week rehabilitation period, and at a 1 year follow-up ticipants attended approximately 6 h/day, 1–3 days a after end of the rehabilitation period. All participants (N week. The participants were assigned to the rehabilitation = 143) completed the 57 weeks rehabilitation period, programme by their medical doctor based on interviews, however, the follow-up response 1 year after the partici- observations, and clinical tests. Formulation of individual pants completed the rehabilitation period was 51 % (n = training and exercise programmes is based on the map- 72). The majority of the participants (N = 143) were ping of the participants. All participants had a personal women (74 %), and 79 % of the participants reported to supervisor, and individual counselling is offered during have primary or technical/vocational school for 1–2 years. the training period. In addition, the majority of the participants reported to be unskilled or skilled workers/craftspeople (59 %). In order In cooperation with the National Health Insurance Office, to compare the characteristics of the study sample with Employment office, employer and other Public Health the general population in the same geographic area, data Services, an individual tailored education and coping Table 1: Content of the multidisciplinary rehabilitation programme Period Intervention Duration Period I: Mapping of the participants resources/ • Introduction to the rehabilitation programme 6 h/day, 4 days a week in 5 weeks intensive training period • Mapping physical-, psychological-, and social function • Individual counselling-based on the mapping; preparation of a long-term plan for the rehabilitation process in cooperation with their medical doctor, social security office and the employer. • Individual and group-based training to improve functional capacity: 1. Individual exercise programme with focus on e.g., endurance, strength, mobility, and relaxation techniques, 2. Group-based education/training in different health related subjects e.g., body structure, diet, exercise planning, coping strategies, communication, strategies for conflict negotiations, and social security system 3. Indoor and outdoor activities every day Period II: Follow-up training/rehabilitation Functional capacity training continues 6 h/day, 1–3 days a week in 52 weeks period (individual and group-based, indoor/outdoor activities, education), individual counselling, clarifying function and work ability, prepare a plan for work re-entry in cooperation with the employer, for example. During/after finishing the rehabilitation period In addition to the regular rehabilitation 1 h/1–3 days a week programme (57 weeks), the rehabilitation centre offers exercise groups e.g., endurance groups, water activity groups, and relaxation training groups in the participant's local community. Page 3 of 10 (page number not for citation purposes) BMC Musculoskeletal Disorders 2007, 8:65 http://www.biomedcentral.com/1471-2474/8/65 process was emphasized that sought to increase func- feelings have been reported to correlate well with other tional capacity, decrease affective distress, and educate measures of physical and emotional functioning respec- patients about the positive health process. Although all tively, such as the Barthel Index and the Zung Depression patients did not receive exactly the same standardized Scale [32]. intervention, as would be expected in a randomized con- trolled trial, our aim with this study was to examine indi- Anxiety and depression, used as predictors (independent) vidual effects in function in real clinical settings. of function (functional health status, pain intensity and pain experience), were assessed by using the Hospital Instruments and procedures Anxiety and Depression Scale (HADS) [32,39]. HADS is a Self-reporting measures were administrated individually brief assessment of anxiety and depression, consisting of to the participants at the rehabilitation centre. Data were 14 items divided into two sub-scales for anxiety and collected at baseline, after the 5-week intensive period, depression, in which the patient rates each item on a four- after the 57-week rehabilitation period and at the one year point scale. Individual items are scored from 0–3 to 3-0, follow-up after the participants finished the rehabilitation depending on the direction of the wording of the items. period. The scores of the items represent the degree of distress: none = 0, unbearably = 3. Tests for reliability (test-retest) The Visual Analogue Scale (VAS) [31,32] was employed to of the scale have been satisfactory with a reproducibility of assess variables on pain (worst imaginable pain, how 0.67–0.77 [32,39]. Factor analysis (varimax method) troublesome the pain is), physical capacity (muscle [36], extracted with eigenvalues > 1.00 as a criterion, indi- strength, endurance capacity, energy, mobility, and bal- cated that items could be grouped according to the two ance), psychological capacity (good feeling inside, mood, main constructs. feeling valuable, extroverted/introverted, optimistic/pes- simistic, calm, and balanced), coping (feeling of coping in The participants' self-reporting about education level, daily life, control and influence in daily life), and cogni- type of job, financial matters, social network, sleep distur- tive capacity (concentration, memory, understand/evalu- bance, tiredness, and history of childhood trauma (inde- ate information, and knowledge). The VAS is a line of 10 pendent variables) was supplemented by personal cm on which pain marks are scored in millimetres, repre- interviews. The self-reporting of traumas include experi- senting the continuum of the symptom to be rated. ences such as; bullying, physical-, emotional-, and/or sex- Instructions about how to rate the present pain, how trou- ual abuse. Except for education, categorized in four levels, blesome the pain is and the present function/capacity all information retrieved from the interviews was catego- were given along with the scale. VAS variables were used rized in two levels of categorical variables. The internal as independent variables (predictors of outcome). Moreo- consistency was acceptable in this study and measures ver, pain intensity and pain experience were used as out- such as Cronbach's alpha coefficients were calculated at come measures as well. The use of the VAS is well 0.80–0.85. established in chronic pain populations, and test-retest reliability of the scale has been satisfactory with a repro- Statistical analysis ducibility of 0.75–0.83 [31,33]. The scale has also been Data were analysed using SPSS for Windows (version used in creative ways to further explore the phenomenon 14.0) software. Frequencies, percentages, mean values of pain perception and reporting, in addition to explore and standard deviation were calculated for continuous other health-related phenomena [31,34,35]. Factor analy- and categorical variables. Multivariate tests (single group sis (varimax method) [36], extracted with eigenvalues > repeated measures design) [36] of the significance of the 1.00 as a criterion, indicated that items could be grouped repeated-measures effect (Pillai's Trace) were provided in according to the intended constructs presented above. order to examine the long-term improvements (variance due to passage of time) of the multidisciplinary rehabili- Function in daily life was measured using the Norwegian tation programme in terms of functional health status version of the Functional Health Status measurement (COOP/WONCA), pain intensity (VAS), pain experience COOP/WONCA Charts [37]. The COOP/WONCA charts, (VAS), anxiety (HADS), and depression (HADS). T-tests used as an outcome indicator (dependent), measure six were used to compare the sample (N = 143) with the core aspects of functional status: physical fitness, feelings, HUNT population from the same geographical area, on daily activities, social activities, changes in health and the anxiety and depression variables. For the initial selec- overall health. Each item is rated on a five-point ordinal tion of potential determinants for the outcome measures scale ranging from 1 ('no limitation at all') to 5 ('severely pain intensity, pain experience and functional health sta- limited'). The test-retest reliability of the original Dart- tus (physical fitness, feelings, daily activities, social activi- mouth version and the Norwegian version was found sat- ties and overall health), univariate linear regression isfactory (r = 0.74–0.86) [38]. The charts of function and analysis, done on the baseline, were used with of signifi- Page 4 of 10 (page number not for citation purposes) BMC Musculoskeletal Disorders 2007, 8:65 http://www.biomedcentral.com/1471-2474/8/65 cance level of p < 0.05. Subsequently, all independent var- loskeletal pain sample was equal to the HUNT population iables that showed significant associations with the (Table 2). outcome measures (dependent) were considered for inclusion into the multivariate linear regression models. Functional status These analyses were carried out separately for the defini- Figure 1 shows the mean and standard deviations of pain tion of outcome variables (pain intensity, pain experi- intensity and pain experience (how troublesome the pain ence, functional health status: physical fitness, feelings, is) measured at 3 points in times during the rehabilitation daily activities, social activities and overall health). In period and at a one year follow-up. Pain intensity and order to identify which variables predict change over time pain experience significantly (p < 0.01) decreased from best, all measurements across the rehabilitation period (3 the start of the rehabilitation period to the one year fol- times) were included in steps in the same model with low-up measures at 109 weeks. Table 3 show the long- effects of these variables on the estimated change of the term improvements (trend over time) of the multidiscipli- outcome variables (functional health status, pain inten- nary rehabilitation programme with a significant sity, pain experience) over the 3 measurement periods improvement in cognitive- (p < 0.001), physiological-, (p (T1-T3). The dependent variable at Time 1 was entered < 0.001), and psychological (p < 0.01) capacity, measured first in the model to control for its effect. In addition, all by VAS, in the pain sample during the 57-week rehabilita- measurements across all times were included in steps in tion period. In addition, scores on the Hospital Anxiety the same model with effects of these variables on the esti- and Depression Scale (HADS), seen in Table 3, showed a mated change of the outcome variables (functional health significant (p < 0.01) reduction in both anxiety and status, pain intensity, pain experience) over the 4 meas- depression during the rehabilitation period. Despite this urement periods (T1-T4). To control for the effect of the reduction, the present pain sample still scored signifi- dependent variable at T1, the variable was entered first in cantly (p < 0.001) higher on the anxiety and depression the model. In the final multivariate models only variables variables at all measurement points during the rehabilita- with p-value less than 0.05 were retained. A p-value of less tion period compared to the HUNT population. than 0.05 was considered statistically significant. In Figure 2, measures of function in daily life using the Results COOP/WONCA charts (Functional Health Status) are Response and baseline characterization of the sample presented as mean values. Functional health status signif- All patients (N = 143) included completed the 57-week icantly increased on the variables feelings (p < 0.05), daily rehabilitation programme, which gave a response of 100 activities (p < 0.05), social activities (p < 0.001), and over- % at the end of the rehabilitation period. However, the all health (p < 0.01) from baseline to the end of the 57th response percentage decreased to 51 % (n = 72) at the 1 week of the rehabilitation period in present sample. How- year follow-up questionnaire on pain intensity, pain expe- ever, a comparison of present musculoskeletal pain sam- rience, and functional health status. The non-response ple with a normative randomized sample (N = 2864) group reported mean pain and pain experience measured from the Ullensaker study [38] on the COOP/WONCA by VAS at respectively 75.4 and 68.0 at the end of the reha- charts, demonstrates that the musculoskeletal pain sam- bilitation period. Further the mean measures on func- ple (N = 143) still report significantly lower function (p < tional health status (COOP/WONCA charts) were 0.01) on all core aspects of functional health status at the calculated at: physical fitness; 2.87, feelings; 2.73, daily end of the 57-week rehabilitation period. A relative low activities; 3.08, social activities; 2.23, and overall health; response (51 %) might limit the relevance of the one year 3.08 at the end of the rehabilitation period in the non- follow-up analysis, however the follow-up measures (109 response group. Back, shoulders, and neck were the most weeks) on functional health status showed that the partic- common pain locations in the sample, and 93.8 % of the Table 2: Characteristics of present sample (N = 143) at baseline participants reported pain in more than two locations. As compared to the HUNT population (n = 52 186). seen in Table 2, the majority (68 %) of the sample was married, and the total per cent exposed to traumas in Characteristics N = 143 n = 52 186 childhood in the present pain sample was 37 %. Sixty Married/cohabitant (%) 71.3 71.7 nine per cent reported sleeplessness and 74 % reported Smoking (%) 53.8 36.6 tiredness in everyday daily life. By comparison, the age- Traumas in childhood (%) 37.1 - matched population from the same geographic area Sleeplessness (%) 69.9 31.3 (HUNT 2) (aged 20–67) consisted of 47.5 % men and Tiredness (%) 74.8 45.5 52.5 % women, 70 % reported to have basic or secondary Poor Social network (%) 17.5 17.5 education, and 22.5 % of the HUNT population reported Poor Economy (%) 34.3 15.5 to be unskilled or skilled workers/craftspeople. The por- tion reporting poor social network in the chronic muscu- Not measured in the HUNT population Page 5 of 10 (page number not for citation purposes) BMC Musculoskeletal Disorders 2007, 8:65 http://www.biomedcentral.com/1471-2474/8/65 Table 3: Multivariate tests of the significance for the repeated-measures effect on functional status in the present sample (Pillai's Trace V) Baseline 5 weeks 57 weeks Variables (N = 143). M(SD) M(SD) M(SD) V F df p VAS Physiological capacity 39.9(15.6) 43.6(15.6) 45.8(18.1) .14 11.61 2 000*** Psychological capacity 58.8(16.4) 59.8(16.0) 62.8(18.3) .07 5.97 2 .003** Coping capacity 57.4(15.9) 56.8(16.2) 57.3(19.2) .00 .12 2 .884 Cognitive capacity 47.9(20.1) 50.0(17.5) 54.3(19.1) .13 10.81 2 .000*** HAD Anxiety 8.83(4.29) 8.65(4.46) 7.93(4.53) .06 5.16 2 .007** Depression 6.03(4.16) 5.59(3.96) 5.08(4.30) .07 5.34 2 .006** a b Physiological; muscle strength, endurance capacity, energy, mobility and balance, Psychological; good feeling inside, mood, feeling valuable, extroverted/introverted, optimistic/pessimistic, calm and balanced, Coping; feeling of not coping in daily life, control and influence in daily life, Cognitive; concentration, memory, understand/evaluate information, knowledge. *p < 0.05. **p < 0.01. ***p < 0.001. Univariate linear regression analysis ipants continued to improve their function in daily activ- ities (M = 2.82/SD = .95), feelings (M = 2.55/SD = 1.27), Univariate linear regression analysis, done on the base- and overall health (M = 3.03/SD = .77), compared to the line, showed that a multitude of potential prognostic indi- 57th week measures (daily activities M = 3.10/SD = .95, cators associated significantly with our primary outcome feelings M = 2.71/SD = 1.15, and overall health M = 3.10/ measure functional health status (physical fitness, feel- SD = .85). However, the improvement in function was sig- ings, daily activities, social activities and overall health). nificant only in daily activities (p < 0.05). In addition, the participants reported a decrease in physical fitness and Here, poor physiological capacity (F = 19.92/p < 0.000) social activities one year after they completed the rehabil- and high pain experience (F = 4.06/p < 0.046) signifi- itation period, compared to the 57th week measures. The cantly associated with poor physical fitness, while limita- decrease in physical fitness and social activities was not tion on the outcome variable feelings associated with significant, however. poor financial situation (F = 13.05/p < 0.000), experience of traumas in childhood (F = 11.37/p < 0.001), poor social network (F = 9.85/p < 0.002) and high levels of anx- iety (F = 111.61/p < 0.000) and depression (F = 66.42/p < Pain intensity Pain experience Standard Deviation 5,00 80 M = 77,3 M = 73,6 M = 74,1 M = 74,1 Physical fitness M = 69,8 M = 68,7 M = 67,9 M = 65,6 Feelings 4,00 Daily activities Social activities Overall health 3,00 2,00 Baseline 5 weeks 57 weeks 109 weeks 1,00 Baseline 5 weeks 57 weeks 109 weeks Mean a r Figure 1 ience measured nd standard deviation of by VAS pain intensity and pain expe- Mean and standard deviation of pain intensity and Repea (COOP-WONCA) in the prese ple Figure 2 ted measures mean for Fun nctional Health Status t musculoskeletal pain sam- pain experience measured by VAS. Mean (M) and stand- Repeated measures mean for Functional Health Sta- ard deviation (SD) of pain intensity and pain experience (how tus (COOP-WONCA) in the present musculoskele- troublesome the pain is) measured by Visual Analogue Scale tal pain sample. Mean Functional Health status measured (0–100) at the start of the rehabilitation period, after 5 at the start of the rehabilitation period, after 5 weeks of weeks of intensive training, at the end of the 57-week reha- intensive training, at the end of the 57 weeks rehabilitation bilitation period (N = 143), and at the one year follow-up period (N = 143), and at the one year follow-up measures (n measures (n = 72) in present musculoskeletal pain sample. = 72). 1 = no limitation at all, 5 = severely limited. Page 6 of 10 (page number not for citation purposes) COOP-WONCA Scale BMC Musculoskeletal Disorders 2007, 8:65 http://www.biomedcentral.com/1471-2474/8/65 0.000). Moreover, there was a significant association (β = .15*) and pain experience (β = .35**) (T3), were the between limitation in feelings and poor physiological (F strongest predictors of variance of functioning (functional = 5.77/p < 0.018)-, psychological (F = 68.93/p < 0.000)-, health status measured by COOP/WONCA) over the 3 coping (F = 12.59/p < 0.001)-, and cognitive (F = 13.97/p measurement periods (57 week rehabilitation period) < 0.000) capacity in present musculoskeletal pain sample. (Table 4). Sleeplessness (F = 6.73/p < 0.010), high levels of pain Linear regression analysis (B = Unstandardized Coeffi- intensity (F = 18.26/p < 0.000), pain experience (F = cients, SE = Std. Error, β = Standardized Coefficients 25.27/p < 0.000), anxiety (F = 6.58/p < 0.011) and depres- derived from the final step) was also performed on all sion (F = 6.52/p < 0.012) were in the univariate analysis measurements across all times with effects of the inde- significantly associated with limitation in daily activities. pendent variables on the estimated change of functional Furthermore, limitation in daily activities associated with health status over the 4 measurement periods (not poor physiological (F = 11.90/p < 0.001)-, psychological included in table). Variance in functioning (functional (F = 5.10/p < 0.025) -, coping (F = 6.89/p < 0.010)-, and health status measured by COOP/WONCA) over the 4 cognitive (F = 5.03/p < 0.026) capacity. Limitation in measurement periods (T1-T4) were significantly predicted social activities was significantly associated with poor by experience of traumas in childhood (B(SE) = .50(.24), social network (F = 4.93/p < 0.028), high levels of anxiety β = .29*), high levels of pain intensity (B(SE) = .02(.00), (F = 12.27/p < 0.001) and depression (F = 10.41/p < β = .42**) and pain experience (B(SE) = .02(.00), β = 0.002), and reporting poor physiological (F = 10.41/p < .37*), and poor psychological capacity (B(SE) = .05(.02), 0.002)-, psychological (F = 22.16/p < 0.000) -, coping (F β = -.68*) at baseline (T1). Moreover, poor physiological = 7.79/p < 0.006)-, and cognitive (F = 10.57/p < 0.001) capacity (B(SE) = -.02(.00), β = -.44**) and high levels of capacity in present pain sample. Table 4: Effects of independent variables on the estimated change of functional health status over the 3 measurement periods The univariate analysis, done on the baseline, also 2 2 Variables (N = 143). B (SE) β ∆R R showed a significant association between poor overall health and high age (F = 4.53/p < 0.035), experience of Physical fitness traumas in childhood (F = 7.68/p < 0.006), poor social Step 1: Dependent variable T-1 - - .23 .23 network (F = 10.62/p < 0.001), and high levels of pain Step 2: Independent variables T-1 - - .08 .32 intensity (F = 7.73/p < 0.006). In addition, limitation in Step 3: Independent variables T2-3 - - .10 .42 overall health was significantly associated with high levels Physiological capacity T3 -.07(.00) -.30** - - of anxiety (F = 10.81/p < 0.001) and depression (F = Feelings 24.40/p < 0.000), and poor physiological (F = 24.77/p < Step 1: Dependent variable T-1 - - .20 .20 Step 2: Independent variables T-1 - - .11 .32 0.000)-, psychological (F = 16.19/p < 0.000) -, coping (F Step 3: Independent variables T2-3 - - .40 .72 = 17.04/p < 0.000)-, and cognitive (F = 11.19/p < 0.001) Psychological capacity T3 -.02(.00) -.38** - - capacity in present musculoskeletal pain sample. Anxiety T3 .15(.02) .59*** - - Daily activities Poor physiological capacity was the only variable that sig- Step 1: Dependent variable T-1 - - .15 .15 nificantly associated with high levels of pain intensity (F Step 2: Independent variables T-1 - - .08 .23 Step 3: Independent variables T2-3 - - .28 .52 = 7.88/p < 0.006) and pain experience (F = 11.39/p < Pain experience T3 .01(.00) .35** - - 0.001) in the univariate analysis, done on the baseline, in Social activities the musculoskeletal pain sample. Step 1: Dependent variable T-1 - - .12 .12 Step 2: Independent variables T-1 - - .07 .20 Multivariate linear regression analysis Step 3: Independent variables T2-3 - - .23 .44 Table 4 summarizes the multivariate linear regression Depression T3 .07(.03) .31* - - analysis with effects of the independent variables (only Overall health Step 1: Dependent variable T-1 - - .20 .20 significant variables included in the table) across the reha- Step 2: Independent variables T-1 - - .12 .33 bilitation period (3 times) on the estimated change of the Cognitive capacity -.00(.00) -.17* - - outcome (functional health status) over the 3 measure- Step 3: Independent variables T2-3 - - .30 .64 ment periods (T1-T3). Cognitive capacity (β = -.17*) was Physiological capacity T2 -.01(.00) -.24* - - the only baseline (T1) measure that associated signifi- Physiological capacity T3 -.02(.00) -.45*** - - cantly with functional health status (overall health) in the Pain intensity T3 .00(.00) .15* - - final model (Table 4). Poor physiological (β = -.24*/- .45***) (T2 and T3)- and psychological (β = -.38**) (T3) B = Unstandardized Coefficients, SE = Std. Error, β = Standardized Coefficients derived from the final step. ∆R = change in explanation capacity, high levels of anxiety (β = .59***) and depres- rate in each step. R = proportion of variance explained. *p < 0.05. **p sion (β = .31*) (T3), as well as high levels of pain intensity < 0.01. ***p < 0.001. Page 7 of 10 (page number not for citation purposes) BMC Musculoskeletal Disorders 2007, 8:65 http://www.biomedcentral.com/1471-2474/8/65 anxiety (B(SE) = .13(.04), β = .48**) and depression still report significantly lower function on all core aspects (B(SE) = .16(.04), β = .58***) at the end of the rehabili- of functional health status compared to a normative sam- tation program (T3) were found to significantly predict ple from the Ullensaker study (N = 2864) [38] at all points the variance in functioning (functional health status of measurement. In addition, the significant decrease in measured by COOP/WONCA) over the 4 measurement self-reported physical fitness and social activities from the periods. 57th week of the rehabilitation period to one year after the participants finished the rehabilitation programme Variance in pain intensity over the 3 measurement periods give rise for concern. Lack of physical fitness and partici- (not included in table) was significantly associated with pation in social activities might later on influence several high levels of pain experience (T3) (B(SE) = .54(.07), β = aspects of function in daily life and might not be benefi- .65***) and physiological capacity (T3) (B(SE) = cial to the individuals or to the society. Future studies .19(.09), β = .19*). The association between pain inten- should therefore try to clarify the long-term effect of sity and physiological capacity was however not signifi- multidisciplinary rehabilitation programmes for individ- cant over the 4 measurements periods. High levels of pain uals with chronic musculoskeletal pain in terms of func- β = .48***) was the intensity (T3) (B(SE) = .58(.08), tion in daily life. Non-specific musculoskeletal pain is an strongest predictor of variance in pain experience over the increasing health problem in the Norwegian population. 3 measurement periods. Moreover, the association The increased study of individual rehabilitation in a for- between pain intensity (T3) (B(SE) = .48(.22), β = .36*) mal rehabilitation programme must not reduce focus on and pain experience was significant over all 4 measure- primary prevention programmes at a population level and ment periods. on the social- and economic policy implications of the present findings. Discussion This study showed that a multitude of factors had an effect Several studies [5,10,41] suggest that the impairment of on pain intensity, pain experience, and functional health function in daily life is associated with several psychoso- status over the measurement periods in a Norwegian sam- cial factors. The intent of this study was to study the long- ple, and different variables affected different aspects of term improvements of a multidisciplinary rehabilitation daily life function. The participants were found to signifi- programme, by focusing on interactions and the influence cantly improve several aspects related to function during of a broad range of socio-demographic and psychosocial the rehabilitation period. However, it still might be rele- factors in pain intensity, pain experience, and functional vant to question in what way these changes influence the health status. In order to do that, the predictors of change everyday life of the people in this sample. Ultimately, the in pain intensity, pain experience, and functional health consequences of chronic musculoskeletal pain for every- status over time were studied. The relationship between day function depend not only on pain intensity and pain emotional distress, chronic pain and function in daily life experience, but also on the individual and on each per- has been shown before [13]. In this study experience of son's unique set of earlier experiences, values, and envi- traumas in childhood, emotional distress, high levels of ronmental conditions. This illustrates the complexity of pain intensity and pain experience, and poor physical chronic pain conditions, where the person's perception of capacity, measured at baseline, were significantly predict- pain and function and his/her experiences of what it ing lack of improvement in functional health status over means in their everyday life might be an important way of all measurement periods. In terms of emotional distress, understanding the complexity. Therefore, the relative it is also relevant to notice the relative high percentage (37 influence of psychosocial factors on function may vary a %) of traumas in the present pain sample. The partici- lot depending on the activity the individuals are engaged pants report significantly higher levels of anxiety and in [40]. For the person that receives treatment the impor- depression before, during, and after the treatment period tance of the overall effectiveness of the rehabilitation pro- compared to the normative population from the same gramme is re-establishing function. However, the geographical area (The HUNT Study). Taken together, and programme is also important from a broader perspective. supported by previous studies as well [2,10,13], this illus- The reduction in pain intensity and pain experience along trates the complexity and the relative importance of emo- with improved function in daily life indicate a positive tional distress in chronic musculoskeletal pain effect from the extensive rehabilitation programme. This conditions. is further underlined by the increase in function in daily activities, feelings, and overall health from the 57th week A study by Palermo and Kiska [42] suggested that sleep of the rehabilitation period to one year after the partici- disturbance is closely linked to mood disturbance. How- pants finished the rehabilitation programme. However, it ever, less is known about the complex interrelationship is important to note that although the participants between emotional distress, sleeplessness and function in improved function during the rehabilitation period, they daily life among adults with chronic musculoskeletal pain Page 8 of 10 (page number not for citation purposes) BMC Musculoskeletal Disorders 2007, 8:65 http://www.biomedcentral.com/1471-2474/8/65 conditions. More than 69 % of our sample reported sleep musculoskeletal pain, and in primary prevention at a pop- disturbance in daily life. However, sleeplessness adjusted ulation level. for emotional distress like anxiety and depression, was not found to be useful in the prediction of function. The Conclusion results in this study confirm the physical capacity and cop- This study has evaluated a complexity of factors that have ing aspects in multidisciplinary rehabilitation found in theoretical or empirical relationships to function in daily past research as well [43-45], suggesting that physical life in a sample with persisting musculoskeletal pain. The exercise, behavioral and cognitive-behavioral treatment results of this study highlight important individual per- for chronic pain reduces pain, pain distress, and improves spectives in chronic musculoskeletal pain. These results daily functioning. Moreover, in accordance with a study are important to better understand which variables are by Lame, Peters, Vlaeyen, Kleef and Patijn [46], our study most useful in helping patients re-establish function dur- indicates the relevance of pain experience in predicting ing a rehabilitation programme and they show how to function, and that function in daily life might be associ- address the variables that affect the outcome. In a broader ated with beliefs about pain. perspective, and as seen in relation to the high prevalence of people with chronic musculoskeletal pain conditions, The participants in this study are not randomly sampled; it is also important to pay attention to the underlying they represent all patients participating in the rehabilita- causes of incidence and primary prevention at a popula- tion programme at a given period. The drop-out rate on tion level. long term follow-up might limit the power of the follow- up analysis and results. However, the participants are rep- Competing interests resentative for people with chronic musculoskeletal pain The author(s) declare that they have no competing inter- seeking help at a rehabilitation clinic with respect to age, ests. sex, pain conditions, working ability and sick leave. The sample and the general population from the same geo- Authors' contributions graphical area are almost identical with regard to age dis- ML, SK and GAE designed the study, ML collected the tribution, family situation, social network, and education data, ML analysed and wrote up the manuscript. ML, SK, level. This allows scrutinization of differences between the GAE revised the manuscript. All authors read and sample and the general population without taking the fac- approved the final manuscript. tors mentioned above into consideration as an explana- tory variable. Even with a 100 % response at the end of the Acknowledgements The authors acknowledge the participation of Friskgården, located in Nord- rehabilitation period, a relatively small number of partic- Trøndelag County, that since 1995 has developed a multidisciplinary reha- ipants could lead to a reduction in the power of the anal- bilitation model for individuals on sick leave with complex disease condi- ysis and decrease the possibility of generalization. tions. We also acknowledge the Nord-Trøndelag Health Study (The HUNT Another limitation of the results is the possibility of bias Study) which is a collaboration between HUNT Research Centre, Faculty related to the self-reported data [47]. However, multidi- of Medicine, Norwegian University of Science and Technology (NTNU, mensional rehabilitation, as in present study, represents Verdal), Norwegian Institute of Public Health, and Nord-Trøndelag County an approach that has the potential to effectively focus on Council. function in daily life, not necessarily on rendering the individual symptom free, which might provide poten- References 1. Rustøen T, Wahl AK, Hanestad BR, Lerdal A, Paul S, Miaskowski C: tially greater reliability in the self-assessment of function. Prevalence and characteristics of chronic pain in the general Some might possibly argue that improvement over time is Norwegian population. European Journal of Pain 2004, 8:555-565. not very surprising since pain patients often are selected 2. Ericsson M, Poston WSC, Linder J, Taylor JE, Haddock CK, Foreyt JP: Depression predicts disability in long-term chronic pain close to their worse status. The participants in present patients. Disability and Rehabilitation 2002, 24(6):334-340. study are long-term chronic pain patients, in some cases 3. Anderson IH: The course of non-malignant chronic pain: a 12- reporting pain duration of more than 10 years. Due to the year follow-up of a cohort from the general population. Euro- pean Journal of Pain 2004, 8:47-53. chronicity considerable improvements in function may 4. Law M: Participation in the occupations of everyday life. Amer- therefore not be anticipated. The VAS scale has been used ican Journal of Occupational Therapy 2002, 56:640-649. 5. Truchon M: Determinants of chronic disability related to low in creative ways to explore the phenomenon of pain per- back pain: Towards an integrative biopsychosocial model. ception and reporting in addition to exploring other Disability and Rehabilitation 2001, 23(17):758-767. health related phenomena [32,34,35]. However, the VAS 6. Frischenschlager O, Pucher I: Psychological management of pain. Disability and Rehabilitation 2002, 8:416-422. measures used in present study should be further vali- 7. Honkasalo ML: Chronic pain as a posture towards the world. dated. Despite several shortcomings, the study highlights Scandinavian Journal of Psychology 2000, 41:197-208. 8. Newth S, Delongis A: Individual differences, mood, and coping important perspectives in a real clinical setting that with chronic pain in rheumatoid arthritis: A daily process should be taken into account in rehabilitation of chronic analysis. Psychology and Health 2004, 3:283-305. Page 9 of 10 (page number not for citation purposes) BMC Musculoskeletal Disorders 2007, 8:65 http://www.biomedcentral.com/1471-2474/8/65 9. Keough JL, Fisher TF: Occupational-psychological perceptions for degenerative lumbar spine disorders. Spine 2001, influencing return to work and functional performance of 26:2375-2380. injured workers. Work 2001, 16:101-110. 36. Bryman A, Cramer D: Quantitative Data Analysis with SPSS Release 10 10. Geisser ME, Robinson ME, Miller QL, Bade SM: Psychosocial fac- for Windows. A guide for Social Scientists USA, Routledge; 2001. tors and functional capacity evaluation among persons with 37. Bentsen BG, Natvig B, Winnem M: Assessment of own functional chronic pain. Journal of Occupational Rehabilitation 2003, 4:259-276. capacity. COOP-WONCA Charts in clinical work and 11. Brannon L, Feist J: Health Psychology. An introduction to behaviour and research. Tidsskrift Norske Lægeforening 1997, 117:545-549. health 3rd edition. Brooks/Cole Publishing Company; 1997. 38. Bentsen BG, Natvig B, Winnem M: Questions you didn't ask? 12. Rudy TE, Lieber SJ, Boston JR, Gourley LM, Baysal E: Psychosocial COOP/WONCA Charts in clinical work and research. Family predictors of physical performance in disabled individuals Practice 1999, 2:190-195. with chronic pain. Clinical Journal of Pain 2003, 1:18-30. 39. Mykletun A, Stordal E, Dahl AA: Hospital anxiety and depression 13. Henderson M, Kidd BL, Pearson RM, White PD: Chronic upper (HAD) scale: factor structure, its analyses and internal con- limb pain: an exploration of the biopsychosocial model. Jour- sistency in a large population. British Journal of Psychiatri 2001, nal of Rheumatology 2005, 32(1):118-122. 179:540-544. 14. Vowles KE, Zvolensky MJ, Gross RT, Sperry JA: Pain-related anxi- 40. Jakobsen K: Employment and reconstruction of self. A model ety in the prediction of chronic low-back pain distress. Journal of space for maintenance of identity by occupation. Scandina- of Behavioural Medicine 2004, 1:77-89. vian Journal of Occupational Therapy 2001, 8:40-48. 15. Bergman S: Psychosocial aspects of chronic widespread pain 41. Saastamoinen P, Leino-Arjas P, Laaksonen M, Lahelma E: Socio-eco- and fibromyalgia. Disability and Rehabilitation 2005, 12:675-683. nomic differences in the prevalence of acute, chronic and dis- 16. IJzelenberg W, Burdorf A: Risk factors for musculoskeletal abling chronic pain among ageing employees. Pain 2005, symptoms and ensuing health care use and sick leave. Spine 3:364-371. 2005, 30:1550-1556. 42. Palermo TM, Kiska R: Subjective sleep disturbance in adoles- 17. Lotters F, Burdorf A: Prognostic factors for duration of sickness cents with chronic pain: relationship to daily functioning and absence due to musculoskeletal disorders. Clinical Journal of quality of life. Journal of Pain 2005, 3:201-207. Pain 2006, 22:212-221. 43. McCracken LM, Turk DC: Behavioural and cognitive-behav- 18. Krokstad S, Westin S: Disability in society-medical and non- ioural treatment for chronic pain: outcome, predictors of medical determinants for disability pension in a Norwegian outcome, and treatment. Spine 2002, 22:2564-2573. total county population study. Social Science Medicine 2004, 44. Pedersen BK, Saltin B: Evidence for prescribing exercise as ther- 10:1837-1848. apy in chronic disease. Scandinavian Journal of Medicine & Science in 19. Michaelson P, Sjolander P, Johansson H: Factors predicting pain Sports 2006, 16:3-63. reduction in chronic back and neck pain after multimodal 45. Staal JB, Rainville J, Fritz J, van Mechelen W, Pransky G: Physical treatment. Clinical Journal of Pain 2004, 20(6):447-454. Exercise Intervention to Improve Disability and Return to 20. Katz WA: Musculoskeletal pain and its socioeconomic impli- Work in Low Back Pain; Current Insights and Opportunities cations. Clinical Rheumatology 2002, 21:2-4. for Improvement. Journal of Occupational Rehabilitation 2005, 21. Kaiser PO, Mattsson B, Marklund S, Wimo A: The impact of psy- 15:491-505. chosocial "markers" on the outcome of rehabilitation. Disa- 46. Lame IE, Peters ML, Vlaeyen JW, Kleef M, Patijn J: Quality of life in bility and Rehabilitation 2001, 10:430-435. chronic pain is more associated with beliefs about pain, than 22. Bergmann S, Herrstrom P, Jacobson LT, Petersson IF: Chronic pain intensity. European Journal of Pain 2005, 1:15-24. widespread pain: a three year followup of pain distribution 47. Gatchel RJ: Psychosocial Factors That Can Influence the Self- and risk factors. Journal of Rheumatology 2002, 29(4):818-825. Assessment of Function. Journal of Occupational Rehabilitation 23. Brown J, Berenson K, Cohen P: Documented and self-reported 2004, 14:197-206. child abuse and adult pain in a community sample. Clinical Journal of Pain 2005, 21(5):374-377. Pre-publication history 24. Fillingim RB, Edwards RR: Is self-reported childhood abuse his- The pre-publication history for this paper can be accessed tory associated with pain perception among health young women and men? Clinical Journal of Pain 2005, 21(5):387-397. here: 25. Vowles KE, Gross RT, Sorrell JT: Predicting work status follow- ing interdisciplinary treatment for chronic pain. European Jour- nal of Pain 2004, 4:351-358. http://www.biomedcentral.com/1471-2474/8/65/prepub 26. Turk DC: Clinical effectiveness and cost effectiveness of treat- ments for patients with chronic pain. Clinical Journal of Pain 2002, 18:355-365. 27. Vowles KE, Gross RT: Work-related beliefs about injury and physical capability for work in individuals with chronic pain. Pain 2003, 101:291-298. 28. McCracken LM, Gross RT: Does anxiety affect coping with pain? Clinical Journal of pain 2002, 9:253-259. 29. Engel GL: The need for a new medical model: a challenge for biomedicine. Science 1977, 196:129-136. 30. Lillefjell M, Krokstad S, Espnes GA: Factors predicting work abil- ity following multidisciplinary rehabilitation for chronic mus- Publish with Bio Med Central and every culoskeletal pain. Journal of Occupational Rehabilitation 2006, scientist can read your work free of charge 16:543-555. 31. McDowell I, Newell C: Measuring Health – A Guide to Rating Scales and "BioMed Central will be the most significant development for Questionnaires 2nd edition. New York: Oxford University Press; 1997. disseminating the results of biomedical researc h in our lifetime." 32. Bowling A: Measuring Health. A review of quality of life measurement Sir Paul Nurse, Cancer Research UK scales 2nd edition. Open University Press; 1997. 33. Bergh I: Assessing pain in elderly. Rating scales, and their rela- Your research papers will be: tion to verbal expression of pain and pain relief. In Licentiate available free of charge to the entire biomedical community thesis Gøteborg University, Sweden; 2001. 34. McGeary DD, Mayer TG, Gatchel RJ: High pain ratings predict peer reviewed and published immediately upon acceptance treatment failure in chronic occupational musculoskeletal cited in PubMed and archived on PubMed Central disorders. Journal of Bone and Joint Surgery 2006, 88:317-325. 35. Zanoli G, Stromquist B, Jonsson B: Visual analog scales for inter- yours — you keep the copyright pretation of back and leg pain intensity in patients operated BioMedcentral Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp Page 10 of 10 (page number not for citation purposes)

Journal

BMC Musculoskeletal Disorders – Springer Journals

Published: Jul 10, 2007

There are no references for this article.