Pre-consultation biopsychosocial data from patients admitted for management at pain centers in Norway

1 Introduction

Chronic non-malignant pain (CNMP) is a complex and multifaceted health problem. The condition is associated with a variety of physical, psychological and social factors often resulting in a significant negative impact on the patient’s quality of life [1], [2], [3], [4], [5], [6], [7]. Due to this complexity, a bio-psycho-social approach has been recommended for multidisciplinary pain clinics treating patients with severe CNMP [3], [8], [9].

To plan individual patient pathways at the multidisciplinary pain clinics, it has been recommended to assess potential biological, psychological and social barriers to pain management with validated questionnaires [10]. Several studies have explored how self-reported, pre-consultation measures might identify patterns relevant for treatment [3], [4], [11], [12], [13], [14], [15], [16]. A Danish study, assessing patient reported data from 1176 patients, reported that lower age was associated with higher anxiety and depression scores [13] while full and part time employment was associated with lower levels of pain intensity, discomfort, anxiety and depression, and higher quality of life levels [13]. A US study showed that higher levels of anger and depression were associated with lower physical function as well as decreased social satisfaction [12] and revealed a strong association between low physical function and psychological well-being [4]. Characteristics, associated with symptoms, may thus be important information for the health professionals being involved when targeting a multi-facet treatment for the individual patient.

The overarching aim of this study was to examine data from of patients referred to three large multidisciplinary pain centers in Norway and to explore if there are differences between the patient samples. The hospitals in Norway are owned by the government, and the responsibility for provision of specialized services to the population are organized within four geographically defined health regions. The population sizes in 2017 was 2.9 million in the southern region, 1.1 million in the western region, 0.7 million in the central region to 0.5 million in the northern region. Within each region there is one university hospital with a tertiary, multidisciplinary pain center: Oslo University Hospital (OUS), Haukeland University Hospital (HUS), St. Olavs Hospital, Trondheim University Hospital (STO), and University Hospital North Norway (UNN). Annually, these centers receive between 4 and 5,000 referrals from general practitioners or hospital wards and treat around three thousand patients. The rejection rates at in 2017 were as follows: OUH: 28%, HUS: 31%, STO: 39% and UNN: 34%. During the last years STO, HUS and UNN have, as part of the assessment routine, asked the patients to answer the same set of questionnaires providing demographic information, assessment of symptoms and health related quality of life, before the first consultation. This tool is called the “Patient response package”.

The specific aims of this study were (1) to describe characteristics of a sample of outpatients referred and accepted for treatment/management to three of the regional multidisciplinary pain centers in Norway, (2) to examine patient differences between the centers and (3) to study associations between symptom scores (insomnia, fatigue, depression, anxiety) and patient characteristics as well as between health related quality of life (HRQoL) and patient characteristics.

2 Methods

3 Results

A total of 1563 patients answered the Patient response package between 01.01.2017 and 15.06.2018, of these 43% at STO, 32% at HUS, and 24% at UNN. Descriptive statistics of patients are shown in Table 1, both according to center and the overall sample.

4 Discussion

This cross-sectional study showed that the samples of chronic non-malignant pain patients, examined at three tertiary, regional multidisciplinary pain centers in Norway present with many years of pain, mean age below 50 years of age, a high female proportion, a low employment rate and low physical activity rate compared to the general population [30], [31]. Also, a large proportion (61%) reported to have been exposed to serious life event(s). Health related quality scores were lower than those reported for patient groups like rheumatoid arthritis, chronically ill hemodialysis patients and stroke patients, but were similar to those for patients with chronic low back pain, liver transplant and chronic obstructive pulmonary disease [32], and corroborate to a large extent with patient characteristics described in corresponding Scandinavian and other international studies on pain [4], [13], [15], [33], [34], [35], [36].

Our patient samples reported high levels of fatigue, moderate/severe insomnia and symptoms of mental distress based on clinical cut-off scores, in accordance with several studies [3], [4], [36], [37], [38]. We found some differences between the centers. Patient at STO reported a higher score of mental distress than UNN, however, health-related quality of life was reported to be higher at STO than HUS and UNN. It is reason to believe that the observed differences between the centers to a large extent are associated with regional differences which might lead to a different selection of patients. Although the pain centers share the same population responsibility for provision of treatment, there are differences in personnel capacity of physicians, psychologists, physiotherapists and nurses. The division of tasks between the pain center and other hospital wards may also differ. However, despite the observed differences, the similarities are more striking than the differences.

When comparing our result to findings from a recent study evaluating the largest pain center in Norway (OUS), situated in the southern region [36], those patients were in average 10-years older than those at STO. This difference might partly be due to that center’s responsibility for the Norwegian National Advisory Unit on Neuropathic Pain. Thus, the center gets referred more patients with neuropathic pain [36]. STO is on the other hand, responsible for the Norwegian National Advisory Unit on Complex Symptom Disorders and might therefore get referred a different selection of patients.

Non-worker, no physical activity, lack of social network, reports of exposure to serious life event(s) and widespread pain were all characteristics repeatedly associated with clinically high levels of the three symptom scores: fatigue, insomnia and mental distress. Our results corroborate with previous results. For instance, Jensen et al., 2016, provided a description of 1176 patients referred to a Danish multidisciplinary pain center and found that non-work participation was significantly correlated to mental distress and to low health-related quality of life. Patients on sick leave had the poorest outcomes on these measures [13].

The odds of reporting clinically high level of fatigue were twice as high for patients with chronic widespread pain compared with patients without widespread pain, and the numbers are similar for moderate/severe insomnia and symptomatic mental distress. Insomnia and fatigue are both found to be important statistical predictors of the onset of chronic widespread pain [37]. The interaction between insomnia and mental distress is also well-established [37], but the latter relationship is not evaluated here.

Physical inactivity among CNMP patients is well known, but the direction of the association is unclear and several hypotheses have been suggested [39]. A common suggestion is that pain creates fear of movement which is assumed to play an important role in the chronification process [40], [41], [42]. The patient adapts to an avoidance behavior which in the long run might cause disability and mental distress [37], [40], [41]. From this hypothesis, physical activity has become a frequent rehabilitation modality and has shown physical improvement amongst CNMP patients [43]. In this study, we found clear associations between physical inactivity and clinically high levels of all the three symptom scores. This supports a potential interaction between physical inactivity in CNMP patients which needs to be further investigated.

Several studies have shown a relationship between social support and pain-related disability [9], [44], [45], [46], and more perceived social support is associated with better clinical outcomes [9], [44]. For patients with other chronic diseases the benefits of social support are found to be significant, whereas social isolation and loneliness may have detrimental effect on mental and physical health [46]. Our findings showed that the proportion of patients, reporting clinically high level of mental distress, was significantly lower among those reporting sufficient social support compared with those reporting lack of social support.

A large proportion of the patients reported to have been exposed to serious life event (61%), and this was associated with clinically higher values for all three symptom scores. It is well known that a number of physiological and psychological processes become enacted when a person experiences acute stress or trauma which again might trigger dysregulation of the physiologic stress system and development of stress-related conditions [47]. Several studies have suggested a potential link between traumatic experiences and subsequent development of chronic pain [9], [47], [48]. A large metanalysis revealed that individuals reporting exposure to previous trauma, regardless of what type, was 2.7 times more likely to have a CNMP [47]. The link between early traumatic experiences and development of chronic pain, however, is still debated. Other researchers have questioned the validity of self-reported data from retrospective cross-sectional studies due to the risk of memory bias [49]. In a prospective cohort study with documented cases of child abuse or neglect, Raphael et al. found the odds of reporting one or more unexplained pain symptoms not associated with childhood victimization [49]. However, a more recent and large, prospective cohort study by the Macfarlane group [48] contrastingly found a clear association between adverse childhood events, defined as physical trauma and poor social/psychological environment, and adult CNMP. The discussion has also raised the question about how childhood traumas are explored [50]. A report of previous exposure to serious life event may or may not be a recall of a traumatic event and should, if reported, be further explored. Collecting data by a brief “yes or no” questionnaire, as in our study, has indeed not the same validity as careful examination by trained interviewers or other objective documentation.

Although the patient characteristics might carry important information for further assessment and planning of a multimodal treatment, the direction of the relationship between these characteristics and symptoms are unclear and warrants further investigation. It is likely that most of these associations are multidirectional, and that there are interactions between the different characteristics and between the different symptoms. In the light of the complex nature of CNMP and acceptance of a biopsychosocial perspective, it is important to assess whether some specific characteristics carry clues useful for further investigations and most importantly, for appropriate interventions.

The present study has several limitations. A cross-sectional study is not designed to define any cause-effect relations between the variables. All data are based on patient reported measures and under and over reporting of symptoms are well known challenges. These are measuring errors that can influence the observed means and proportions. In addition, there might be underlying systematic differences between the centers which we have not been able to control for resulting in sample bias. Example of such could be systematic referral biases by the general practitioners (GP). If the GPs systematically choose not to refer the active working patient to the pain center our results regarding associations of work situation with fatigue, insomnia and mental distress can reflect systematically selected sample.

Methodological limitations are the possibility of some item overlap between the different symptoms score and wrongful associations. Although statistical measures have been taken to avoid type 1 errors, some of the observed associations could still be due to chance.

The strength of our study is the completeness of data and that the study is done in a regular clinical setting.

5 Conclusions

This present study on the samples of chronic, nonmalignant pain patients referred to three regional multidisciplinary pain centers in Norway shows (1) patient characteristics of many years of pain, mean age of 42.9 years, high proportion women, low employment rate, low physical activity rate and a large proportion reporting to have been exposed to serious life event(s); (2) differences in the patient samples between the three centers which probably reflect structural differences, and (3) patient characteristics such as no work participation, no physical activity, lack of social network, previous exposure to serious life event and chronic widespread pain are all statistically associated with high clinical levels of symptoms such as insomnia, fatigue and mental distress, and patient characteristics such as no work participation, no physical activity, lack of social network, and previous exposure to serious life event(s) are associated with low HRQL.