Demographic and clinical factors associated with psychological wellbeing in people with chronic, non-specific musculoskeletal pain engaged in multimodal rehabilitation: –a cross-sectional study with a correlational design

Introduction

Musculoskeletal disorders are common and pose a significant problem among the working population. Three out of every five workers in the EU have reported musculoskeletal complaints in the back, upper limbs, and/or lower limbs [1]. In Sweden, musculoskeletal disorders constitute the second highest number of approved occupational diseases [2], causing individual suffering as well as societal costs. During 2020, 15–18 % of all sick leave was due to problems in the musculoskeletal system.

Such disorders not only result in musculoskeletal pain but also affect mobility, activities of daily life, and emotions. Recommended treatment is therefore multimodal in teams with different specialties. Evaluation has shown multimodal treatment for low-back problems to be more effective than conventional pain treatments, such as only physiotherapy or medication in reducing pain and activity limitations and increasing return to work [3]. Conversely, multimodal rehabilitation compared to conventional treatment has not proven cost-effective from a societal perspective. Costs increased due to lower employment rates, decreased income, and increased number of days on sick leave, as the treatment tended to entail initial sick leave which then became permanent [4]. However, a recent study report cost-effectiveness in terms of increased quality of life and decreased sickness absence at one-year follow-up [5].

In the evaluation of symptoms and rehabilitation, pain intensity, physical functioning, emotional functioning, and participant ratings of improvement and satisfaction with treatment have been recommended [6]. Evaluation of a patient’s symptoms and the results of rehabilitation thus often focus on pain combined with other aspects. However, the choice of which components to evaluate differs between instruments [7], [8], [9] and the correspondence between the instruments and the symptoms the patients experience is often low [10].

For those affected by musculoskeletal disorders wellbeing and enjoyment of life have been appraised as the most important health aspects [11] and more influential related to care-seeking [12] than pain reduction. A theoretical model of psychological well-being encompassing 6 distinct dimensions (Autonomy, Environmental Mastery, Personal Growth, Positive Relations with Others, Purpose in Life, Self-Acceptance) has been presented by Ryff and Keyes [13]. Incorporating psychological wellbeing as a measure of a patient’s health and investigating the relationship between psychological wellbeing and relevant symptoms of non-specific musculoskeletal disorders may show the relevance of different aspects of wellbeing in evaluating symptoms and rehabilitation. This insight may contribute valuable knowledge as finding a cure and alleviating symptoms of non-specific musculoskeletal disorders can be challenging. Consequently, our research question was:

1. To what extent are measures of general, physical and mental health; functional limitations; pain intensity; mood and sleep associated with psychological wellbeing among working-aged men and women who participate in multimodal rehabilitation for non-specific musculoskeletal disorders?

Materials and methods

Results

Discussion

Our research question was to what extent is psychological wellbeing associated with general health, physical and mental health, pain intensity, limitations in daily life, depression and sleep? The results showed decreased odds of psychological wellbeing for persons who rated high on depression, results that remained significant after adjusting for sex and age. Being a woman increased the odds of high psychological wellbeing. None of the other independent variables studied was significantly associated with high vs. low psychological wellbeing.

Our choice of instruments to use in the analyses, both in the present and in previous studies [23, 24], relied on the bio-psycho-social model [25, 26], which most often is used in analyses of patient problems and decisions on treatment approach [27]. Another basis for our decision came from the IMMPACT group, which recommends measurement of pain, physical and emotional functioning, and the patient’s rating of global improvement and satisfaction with treatment [6, 19]. A further important theoretical model that was considered is the ICF model for the classification of functioning, disability and health [28]. Thus, we included measures for pain, physical functioning, psychological functioning and general health.

As a point of comparison, a variety of different measures were found in the studies included in a recent review of treatment studies [27]. The SF-36 (or the 12-item version) was most often used to measure health-related quality of life, both as a total score and for the subscales. The ICF domain of Body Function was most often measured in terms of pain intensity and depression, and occasionally sleep. The ICF domain Activity and participation was most often measured in terms of pain-related physical and social-functional limitations. The questionnaires for these two ICF domains were not specified in the report.

Consensus on what to measure in evaluating treatment would make comparisons easier. There are, however, a multitude of instruments in use. Prior reviews on the content and quality of questionnaires used to evaluate musculoskeletal symptoms [7], [8], [9] suggest that questionnaires aimed at evaluating the same condition are often not comparable due to differing items. Questionnaires often cover only some of the symptoms patients experience and different aspects of those experiences [10]. A consensus on measures would increase the possibility of drawing reliable conclusions from the results of different interventions in rehabilitation. Such consensus is needed for both content and quality of the chosen instruments.

Depression scales are most often used to measure emotional functioning. We used a measure of positive emotional functioning and assessed if there was any specific relation between wellness and symptoms often present in musculoskeletal disorders. The reason for this choice is that emotional wellbeing and enjoyment of life have been appraised as the most important health aspects, in addition to pain reduction, for those affected by musculoskeletal disorders [11]. When it comes to care-seeking, wellbeing has also proven to be more influential than specific somatic symptoms [12].

In a previous study of women with musculoskeletal disorders [23] we attempted to construct a short-form questionnaire. The results indicated the relevance of including wellbeing as a measure of health. Even though successfully used in this prior research, the questionnaire may not be the most suitable for measuring wellbeing, leaving open the question of a possible better choice. Other commonly used questionnaires seem to measure general health or quality of life rather than wellbeing as such. The results in our follow-up study to test the questionnaire in a mixed-sex sample [24] did not indicate the same fit as for the women-only sample, raising the question of the difference being gender related.

The composition and characteristics of a sample in a study must be considered. A recent review reported multimodal and interdisciplinary treatments to be comparable to other interventions, additionally indicating that they may be more effective for people with long-term pain [25]. However, these results showed moderate reliability and there was uncertainty regarding sample comparability with those intended for multimodal rehabilitation for long-term pain in Swedish health care. One criterion for inclusion in multimodal rehabilitation in Sweden is that the participant does not have any co-morbidity or other condition that could hinder participation in the rehabilitation program. Despite this criterion, the degree of co-morbidity in the sample was surprisingly high. Twenty participants reported 4 other physician-diagnosed diseases, and only 13 of the participants reported no diagnosed co-morbidity at all. Co-morbidity to this extent may affect the results of rehabilitation as well as the participants’ wellbeing. Previous studies have shown that people with long-term pain more often have other medical conditions such as cardiovascular disease [29], as was also shown in the present study. This information is important because studies [30, 31] have found that the risk of a person with severe long-term pain dying from some other disease is higher than for a person without, or with mild, long-term pain. Nevertheless, in the present study, neither pain nor limitations in daily life turned out to be significant aspects related to psychological wellbeing, while depression did.

Early treatment is important to avoid severe long-term pain; however, finding and including participants at an early stage of the disorder has proven to be difficult, probably because people tend to delay seeking care until the disorder severely affects their ability to work, interfering with their ordinary daily life.

A strength of the present study is that we took previous recommendations under consideration in our choice of outcome measures. Furthermore, a health-promoting perspective is included by adding a measure of psychological wellbeing in an effort to grasp the entire picture.

However, the sample in the present study was small, even though we made efforts to increase participation by engaging the rehabilitation teams in data collection, extending the sampling period, and including additional rehabilitation groups. In the present study we asked all the rehabilitation teams in the county if they wanted to participate, but for various reasons the participation was less than expected. This was mainly due to the fact that some teams did not have a continuous staffing cover, which meant that their activities were limited. Another reason was that relatively few patients in the rehabilitation groups chose to participate. The patients were already asked by their rehabilitation teams to fill out questionnaires to assess the results of the rehabilitation. When they were also asked if they wanted to answer the questionnaire in this study, it may have been perceived as too demanding. Due to the low number of participants only a limited number of variables could be included in the logistic regression analyses, which in turn limited statistical power. A higher number of participants, and thereby a higher statistical power, could have yielded more reliable results. Furthermore, we cannot generalize the results since the participants were not randomly selected. However, the sample was representative of the national population engaged in multimodal rehabilitation regarding age and gender [32].

Conclusions

The present study aimed to investigate which, if any, of the included variables were associated with psychological wellbeing, possibly pointing to alternative directions in rehabilitation. Depression turned out to be significantly related to psychological wellbeing, contrary to pain and limitations in daily life. If further studies with larger, random samples can confirm these results, this knowledge may be important both in clinical settings and in future research.