Effectiveness of a group intervention using pain neuroscience education and exercise in women with fibromyalgia: a pragmatic controlled study in primary care

Fibromyalgia (FM) is a complex condition that includes persistent multi-focal pain and hyperalgesia, together with tiredness, sleep and cognitive alterations and other effects [1], with a potentially-serious impact in the quality of life of those who suffer from this condition [2, 3].

Over the last 20 years, the definition of FM has been extended, despite the lack of specific diagnostic tests or pathognomonic findings [4]. This has hampered its treatment which, until now, has consisted in mitigating its symptoms with drugs [5] or even using invasive procedures [6], since no specific treatment exists [4, 7, 8]. The latest recommendations of the EULAR (European League Against Rheumatism) for treating patients with FM indicate that exercise is the only recommended therapeutic option that is strongly supported by evidence. If this fails, other individualised alternatives should be explored such as psychological and pharmacological treatments (for severe pain or sleep alterations) and/or multi-modal rehabilitation programs (in the case of severe disability) [8].

Although they have achieved modest results, several studies have evaluated the efficacy of various non-pharmacological therapies in the treatment of chronic pain syndromes, such as cognitive-behavioural therapy, physical exercise, mindfulness, multidisciplinary approaches, amygdala retraining [9‐12] and psychoeducational interventions [13].

In recent years, abundant evidence has been obtained regarding the implication in FM of an abnormal cerebral process primarily mediated by the Central Nervous System (CNS) that forms part of the so-called Central Sensitization Syndromes [4, 14]. Aided by advances in neuro-imaging [15], this concept of pain experience, which differs from nociception, defends the leading role of the brain in the intensity and perception of pain, leaving the relationship with body tissues in second place [16]. In this setting, we find a permanently “hyperalert” nervous system that amplifies sensory stimuli, accompanied by a comprehensive autonomic and/or neuro-endocrine motor cortex. This approach has led to the development of Pain Neuroscience Education (PNE), which proposes that understanding the neurobiological mechanisms involved in the experience of pain reduces the cerebral perception of threat [16]. Achieving a change in the cognition of, and the attitudes related to, the pain process would also moderate the activation of motor, sympathetic and neuro-endocrine protection mechanisms [17].

Very positive effects have been described as regards the application of this therapeutic approach to chronic pain [18‐22] and migraine [23, 24], with clinical and functional improvements in patients, including FM sufferers, both as the main therapy [25, 26] and as part of a multidisciplinary approach [27, 28]. Even in the current pandemic due to SARS-CoV-2 infection, some authors such as A. Goicoechea or P. Garner have proposed its use in patients with long COVID-19 due to the similarities of this entity with FM and other central sensitisation syndromes [29, 30]. Nevertheless, there is little data regarding the applicability of this method in real clinical practice in a primary care (PC) setting or of its medium- and long-term effectiveness. For this reason, we undertook a pragmatic study to explore, prospectively, the effectiveness of a double PNE intervention, at group and individual levels, coupled with conscious movement techniques, as compared with treatment as usual (TAU), in PC.

Between January 2018 and October 2019, 4 group courses were organized for a total of 45 patients who met FM criteria (IG). In parallel with this, data were collected from 19 women who did not receive the intervention (CG). Of these 64 women, 53 answered the initial and final questionnaires (the latter 1 year after the initial interview), 35 from the IG and 18 from the CG (Fig. 1).

The baseline characteristics of the 53 women included in the analysis are shown in Table 1. The average age of the women in the CG was greater than that in the IG (64 vs 57 years). In general, the IG included a greater proportion of single women, women with secondary/higher education, with paid work, with a history of significant pain during infancy and with a higher score for severity of symptoms than did the CG.

Table 2 shows the variations observed in the different variables between the baseline scores and the scores after 1 year, both within each group and between the IG and the CG. In the intervention group, a reduction (improvement) was observed in the scores of all variables at the one-year follow-up. In the control group, a worsening (increase in score) in symptom seriousness (SS), impact of FM on daily life (FIQ) and functional capacity (HAQ) was noted. Moreover, the improvement observed in the scores of all variables (comparing the differences between the scores obtained for each variable in IG and CG) was greater in the IG than in the CG, with all differences reaching statistical significance (p < 0.05), after adjusting for age and baseline scores. In addition, the effect sizes of the differences between the intervention and the control groups were large for pain and FIQ total score, and medium for the remaining dimensions (Table 2).

As shown in Table 3, 17 (49%, CI: 33–64%) of the women in the IG no longer fulfilled the FM criteria 1 year later, according to the ACR 2010 classification.

Regarding the evaluation of the impact of fibromyalgia, more women in the IG showed a reduction ≥20% and ≥ 50% between the baseline FIQ score and the final score. Also, at the end of the study, 2 (11%, CI: 3–33%) and 11 (31%, CI 19.48%) of patients achieved a FIQ score < 39 (mild functional impairment) in the CG and the IG, respectively, although this difference was not statistically significant (Table 3).

Our study shows that a group intervention based on PNE, preceded by an active-listening interview and accompanied by exercises of increasing intensity, can improve, among other factors, the severity of symptoms, the functional capacity and the impact of pain in the daily life of women with FM. This improvement was sustained for 1 year.

Up to now, few studies have been published on patients with FM carried out in PC to evaluate PNE-based interventions. Our results corroborate those obtained in an uncontrolled before-after study where a similar PNE methodology was used, but without the addition of exercise [25]. As in that study, we used a modified form of classical PNE, incorporating the hypothesis of learned cerebral evaluation error. This hypothesis implies an original approach since it implicates an unconscious learning process in the aetiopathogenesis of this disease, removing blame from the subject. It focusses on the unconscious transmission of knowledge in order to unlearn the painful experience associated with the learned cerebral evaluation error, which is reinforced by the current alarmist culture, model copying and instructions from experts [24].

The results obtained in our study are comparable to the best results reported in published controlled studies that used various interventions based on PNE [26, 34, 39]. Van Ittersum et al. [26] concluded that supplying written information on the physiology of pain, followed by a motivational telephone call to resolve doubts, was not effective in patients with FM in terms of modifying the impact of pain in daily life, of improving feelings related to pain (catastrophizing) nor in perceptions regarding the condition. In the study by van Oosterwijck et al. [39], an intervention consisting of 2 individual PNE sessions, each of 30 min, achieved a reduced degree of anxiety in the short term and longer-term improvements in vitality, functionality, mental health and in the general perception of health; however, the pressure pain thresholds remained unchanged. The EFFIGACT study, the only one carried out in PC, showed that the group that received the PNE intervention (8 sessions of 2.5 h each with groups of 10 to 15 patients) experienced a greater increase in their overall functional state than the group treated with drugs or the CG; they also reported improvements in the pain catastrophizing scale, acceptance of pain, subjective pain, quality of life, and anxiety and depression, with medium-sized effects in a majority of cases.

These differences may be due not only to the particular design of each study, but also to the type and duration of the PNE sessions. In a recently-published study that evaluated the effects of different dosages of PNE in 3 centres specializing in FM in Spain [40], it was found that higher dosages (6 sessions, each of 45 min) produced a larger improvement in pain severity at three-month follow-up than other dosages of PNE (2 sessions of 45 min or 6 sessions of 15 min) and biomedical education (2 sessions of 45 min); however, PNE (regardless of its duration) was not superior to biomedical education in the central nociceptive processing, disability, or psychological variables in patients with FM.

In addition, in all our sessions, the physiotherapists explained how to carry out different exercises, and the fifth session was dedicated exclusively to exercises. It should be emphasised that the physiotherapist is the most qualified professional to carry out the exercise programme of the intervention, so it is strongly recommended that these professionals are the ones who teach and direct the exercises. In this sense, the results of a study carried out in a hospital environment have just been published, following the recommendations of the EULAR [8], which establishes education and physical exercise (carried out by physiotherapists) as the initial management for FM. This study concluded that a multidisciplinary intervention (including PNE and exercise) achieved moderate to large effect at six-weeks, when compared with the control group (TAU), for functional impairment, anxiety, kinesiophobia, perceived competence, and positive reappraisal in patients with FM [41].

Another difference in our intervention, compared to that used in the aforementioned studies, is that before starting the group sessions, each patient was interviewed individually, using active listening to aid in the verbalization of their experience of the disease. During these interviews, we could perceive certain characteristics that could be associated a priori with poorer results in the variables analyzed, such as the expectation of receiving an indemnity (for inability to work or some degree of disability), social/family or labour-related advantages derived from suffering this disease, or a feeling of dissatisfaction with the way in which the health system (including some professionals) had treated them. These aspects deserve to be investigated in ad hoc studies using qualitative research techniques such as focal groups.

As described above, the sessions were delivered in groups of 9–13 subjects. Although theoretically this intervention could be applied to a single subject, it is much more cost-effective to use it in a group of subjects, as proposed in most studies, in the same way that we offer other services such as maternal education, pelvic exercises to improve urinary health, etc. in PC. It is known that there are some limitations in some PC sites to implement group interventions, so it is necessary to make an effort to ask health authorities and PC managers to promote and allocate the necessary resources where they do not exist to offer these activities to the population.

In order to be recruited for our study, patients had to meet the ACR 2010 diagnostic criteria for FM [1]. As in the case reported by Barrenengoa-Cuadra et al. [25], it is noteworthy that in our study almost half of the patients in the intervention group no longer met the criteria for FM 1 year later. As the authors of that study also remarked, we have not found this evaluation criterion in the literature that we have reviewed, so it would be interesting to evaluate its usefulness in following up these patients.

Among the strengths of this study are that it is a pragmatic, controlled trial that evaluates an intervention carried out by a professional multidisciplinary team (physiotherapists, nurses and family doctors) in PC, integrated in the normal activities of this level of care, without requiring additional resources. Although the ideal design for demonstrating the benefits of an intervention is the classical randomised clinical trial, it has the limitations of using strict selection criteria with very clear and complex rules for implementing the intervention, which makes it difficult to replicate in clinical practice. As commented above, some randomised clinical trials have been published exploring the efficacy of the PNE approach. Therefore, we decided to apply this method in a standard clinical setting to demonstrate that it is feasible and compatible with the routine care of other patients. In this sense, we decided to offer the intervention to all potential patients, as we do with other interventions/treatments, and let them decide whether they wanted to be included in the intervention or control group, depending on their current availability to attend the sessions. The selection criteria used in our study were not excessively strict, so the results obtained could be extrapolated to most of patients with FM in PC (high external validity).

Although male patients were not excluded, none were included in the study because no males with FM visited the Physiotherapy Unit during the recruitment period: however, in the studies of patients with FM that we have consulted, it is unusual to find a proportion of male patients greater than 5–10%, which is in line with the known prevalence of this condition. The low drop-out is also noteworthy, despite the sessions being carried out every week during almost 2 months, and with a one-year follow-up period (in most of the clinical trials published, the follow-up was 6 months at most).

All data on the variables studied were self-reported, so, although this was not a blinded trial, it would be difficult for the researchers to have influenced the results. Given the nature of the intervention, it would not be possible for the educators to be unaware of the patients’ groups.

The relatively small size of the sample (especially in the case of the control group) did not allow for subgroup analysis to determine which subgroups of patients could benefit most from this intervention, nor to detect small differences between the groups. Nevertheless, given the magnitude of the differences observed between the two populations in most of the variables studied, it is not likely that the true differences differ greatly from those observed. In addition, the 95% CI is given for all results, so that readers can draw their own conclusions in each case. On the other hand, the distribution of the patients between the intervention and the control groups was determined by their own preferences, which allows us to suppose that there was greater motivation in the intervention group, and this may have influenced the magnitude of the results. However, this is a characteristic of most studies that investigate cognitive-behavioural therapies and is difficult to control for.

As for the non-existence of a control group that received no treatment, which would have made it possible to compare the effect of the intervention with the natural history of FM, it did not seem ethical to us, in a pathology associated with a high degree of physical and psychological suffering, asking patients to abandon any therapy they were receiving. However, given that all the patients continued their usual follow-up by professionals from the same Health Area, it is not expected that there would be relevant differences in the basic treatment received by the patients in both groups: therefore, the differences observed can reasonably be attributed to the intervention studied, especially bearing in mind that they are in line with those described in other studies.

The results obtained show that this intervention based on PNE and exercise is feasible and effective in patients with fibromyalgia in real practice in PC, possibly the ideal level of care for the management of this type of patients. Group therapy can increase the degree of adherence and the positive responses of peers and can promote and reinforce the assimilation of the changes required in order to unlearn the painful experience associated with the learned cerebral evaluation error. This hypothesis deserves to be tested in an ad hoc study that could also study the efficiency of this group intervention compared with TAU in terms of direct costs (number of consultations in PC, hospitals and emergency departments, medical drugs consumption, etc.) and of indirect costs (days off work, disabilities, etc.).