Multidisciplinary treatments of fibromyalgia (FM) have demonstrated efficacy. Nevertheless, they have been criticized for not maintaining their benefits and for not being studied for specific populations. Our objectives were to determine the efficacy of a multidisciplinary treatment for FM adapted for patients with low educational levels and to determine the maintenance of its therapeutic benefits during a long-term followup period.
Inclusion criteria consisted of female sex, a diagnosis of FM (using American College of Rheumatology criteria), age between 18 and 60 years, and between 3 and 8 years of schooling. Patients were randomly assigned to 1 of the 2 treatment conditions: conventional pharmacologic treatment or multidisciplinary treatment. Outcome measures were functionality, sleep disturbances, pain intensity, catastrophizing, and psychological distress. Analysis was by intent-to-treat and missing data were replaced following the baseline observation carried forward method.
One hundred fifty-five participants were recruited. No statistically significant differences regarding pretreatment measures were found between the 2 experimental groups. Overall statistics comparison showed a significant difference between the 2 groups in all of the variables studied (P < 0.0001). Mixed linear model analysis demonstrated the superiority of the multidisciplinary treatment in all of the studied variables at posttreatment. The differences were maintained at 12-month followup in sleep disturbances (P < 0.0001), catastrophizing (P < 0.0001), and psychological distress (P < 0.01).
Multidisciplinary treatment adapted for individuals with low educational levels is effective in reducing key symptoms of FM. Some improvements were maintained 1 year after completing the multidisciplinary treatment.
Fibromyalgia (FM) is a chronic syndrome characterized by generalized musculoskeletal pain and multiple symptoms, including fatigue, sleep alterations, cognitive dysfunction, and psychological stress (1, 2). Various studies from the US and Western Europe have indicated that FM has an estimated prevalence between 2% (3) and 2.9% (4).
The debate over the efficacies of various pharmacologic and nonpharmacologic treatments for FM has led to the publication of evidence-based clinical guidelines, specifically the guidelines of the American Pain Society (2005), the Association of the Scientific Medical Societies in Germany (2008), and the European League Against Rheumatism (2007). These guidelines assign a higher grade of evidence to a multidisciplinary treatment for FM, but they disagree regarding the various components that the treatment should integrate (5). However, despite the divergence, the guidelines do agree that a multidisciplinary treatment for FM should include the following: medication management, education, physical exercise, and cognitive–behavioral therapy (CBT) (5).
The combination of pharmacologic treatment, education, physical exercise, and CBT has demonstrated efficacy in reducing some of the key symptoms of FM, such as pain, fatigue, and depression, and it has also demonstrated efficacy in the improvement of self-sufficiency and physical aptitude (6, 7). However, multidisciplinary treatments for FM have been criticized for not maintaining their long-term therapeutic benefits and the lack of studies directed toward specific populations (6, 7). In fact, of the small number of studies that have explored the efficacy of a multidisciplinary therapy for the treatment of FM (8–16), only 3 had a followup of more than 6 months, and none were directed toward a specific subgroup of patients (10, 13, 15, 16). Therefore, as previously noted by Hauser et al (7), there is a need for improved studies exploring the long-term effects of a multicomponent therapy for FM.
Although the relationship between educational level and FM remains unclear (4), several studies have found an association between low educational level and FM (17–23). Furthermore, some studies have linked the effectiveness of CBT and the educational level in the treatment of chronic diseases (24), complicated grief (25), health-related clinical problems (26), and FM (27). However, the patient's educational level has not usually been considered as a relevant variable in studies that assess the efficacy of multidisciplinary treatments for FM. In fact, in those programs that have considered the participants' educational level in demographic data, the percentage of participants with a low educational level (primary care studies) was lower than the percentage of patients with a high educational level (university studies) (8, 11, 13–15). Therefore, what seems to be clear from all of these studies is that educational level is an important variable to consider in multidisciplinary treatment for FM. Furthermore, this is especially relevant if we consider that 66.3% of the patients in the Spanish primary public health care system have an educational level that is less than or equivalent to eighth grade (28).
The objectives of our study were to determine the efficacy of a multidisciplinary treatment for FM adapted for patients with low educational levels, and to determine the maintenance of its therapeutic benefits during a long-term followup period. A disorder-specific instrument assessing the consequences of FM and a measure of sleep disturbances were chosen as the primary outcome measures. We hypothesized that the multidisciplinary group would demonstrate the greatest improvements on primary outcomes. Also, we hypothesized that the improvements on primary outcomes would be maintained during followup.
Significance & Innovations
A tailored multidisciplinary treatment of fibromyalgia composed of medication, cognitive–behavioral therapy, and physical therapy (hydrokinesiotherapy and kinesiotherapy) has been demonstrated to be effective in women with a low educational level.
Improvements in key symptoms of fibromyalgia such as sleep disturbances, functionality, psychological distress, and catastrophizing were maintained at 1-year followup.
To assess the importance of improvements in pain intensity and functionality, clinically significant change was evaluated. A cutoff of 30% was accepted for pain intensity (numerical rating scale) and a cutoff of 14% was accepted for functionality (Fibromyalgia Impact Questionnaire total score).
PATIENTS AND METHODS
Study design and participants
This study was performed in accordance with the standards of the Clinical Research Ethics Committee at the Hospital Universitari de Tarragona Joan XXIII (Tarragona John XXIII University Hospital), and all of the participants signed an informed consent document. To be included in the study, the participants had to have met the following inclusion criteria: female sex, a diagnosis of FM based on the diagnostic criteria of the American College of Rheumatology (29), age between 18 and 60 years, and between 3 and 8 years of schooling. The exclusion criteria were as follows: another severe chronic pain pathology (e.g., sciatica or complex regional pain syndrome), having been diagnosed with inflammatory rheumatic disease, being physically unable to perform the exercises, an open wound, a skin disease, being under psychiatric and/or psychological treatment within the past 3 years, significant suicidal ideation, cognitive or sensorial deterioration that impedes an adequate followup to the treatment, or a pending legal resolution for disability.
Patients in the study were recruited from consultation with a rheumatologist. Consecutive participants who met all of the inclusion criteria and agreed to participate in the study were assigned to 1 of the 2 treatment conditions (conventional pharmacologic treatment or multidisciplinary treatment). Patients were randomly assigned in a 1:1 ratio in blocks of 32 according to a computer-generated random number table. Of the 460 individuals who were evaluated, 174 were included in the study. Of these 174 participants, 19 were included in the pilot study. Finally, the study sample consisted of 155 participants with a mean ± SD age of 48.9 ± 7 years (range 26–60 years) and a mean ± SD pain duration of 11.6 ± 9 years (range 0.5–49 years). A total of 98% of the sample was white (Figure 1).
The study was performed in 2 phases. In the first phase, the program was designed and applied to a small group of participants. At this initial stage, the therapists (clinical psychologist and physiotherapist) were observed by another clinical psychologist (AC) in CBT and by another physiotherapist (IM) in physical therapy when applying the treatment in order to identify what aspects could be improved. The results of this pilot study were assessed and used to improve the content and procedures of the sessions to be included in the final manualized treatment protocol. Finally, in the second phase, the definitive program was applied and the rating measurements were performed. The CBT program was delivered by a clinical psychologist specifically trained for this program. The physical therapy program was delivered by a physiotherapist also specifically trained for this program.
Before the start of either of the treatments, demographic and clinical data were collected and the measurement instruments were applied. In the week immediately after the completion of the treatment, the same measurement instruments were reapplied and the posttreatment score was obtained. The same instruments were reapplied after 3, 6, and 12 months. Obviously, the participants and therapists (clinical psychologist and physical therapist) could not be blinded to treatment conditions. Instead, the evaluators (physician and psychologist) were blinded to different treatment conditions.
The participants in the control group received a conventional pharmacologic treatment that essentially included analgesics, antidepressants (tricyclics, selective serotonin reuptake inhibitors, and dual reuptake inhibitors), benzodiazepine, and nonbenzodiazepine hypnotics (2, 30). The participants in the multidisciplinary group received the same conventional pharmacologic treatment, CBT, and physical therapy. Once baseline measures were taken, the pharmacologic treatment was started. Just after finishing treatment and at 3-, 6-, and 12-month followup, both groups of participants were individually attended again by the physician and the initial drug treatment was adjusted when necessary. Drug treatment was adjusted in both groups (control and multidisciplinary) as recommended by guidelines (2, 30). In the multidisciplinary program, the CBT and physical therapy were performed in a group format (8 patients per group). The participants underwent a total of 24 sessions at a frequency of 2 days per week. When they attended the treatment sessions, the participants underwent 1 hour of CBT and 1 hour of physical therapy.
CBT protocols developed for the current study were adapted from Bradley (1996) (31), Espie (2006) (32), Thorn (2004) (33), and Waters et al (2004) (34). The adaptation consisted of adjusting contents to the low educational level of the sample. In particular, the vocabulary used was simplified at most and concepts were explained through everyday examples and graphic bullets. The therapist sought to maximize patient rapport, patient adherence, group cohesion, and group discussion related to the session topics. The participants were given a patient guide to follow or discuss during sessions and read between sessions. Also, home tasks were planned and revised every session. The CBT program included the following: information about FM, theory of pain perception, cognitive restructuring skills training, CBT for primary insomnia, assertiveness training, goal setting, activity pacing and pleasant activity scheduling training, life values, and relapse prevention (see panel 1 in Supplementary Appendix A, available in the online version of this article at http://onlinelibrary.wiley.com/doi/10.1002/acr.21818/abstract).
The physical therapy treatment emphasized aerobic capacity, muscular strengthening, and flexibility and alternated with sessions of hydrokinesiotherapy and kinesiotherapy in a gymnasium. All of the sessions included overall aerobic work, coordination exercises, and flexibility exercises. The difficulty of the exercises was individually tailored and progressively increased through the use of resistance media and a slow execution velocity. During the physical therapy sessions, the participants practiced Schultz autogenic training (35). The sessions of hydrokinesiotherapy were conducted in a heated pool at 30°C. Each session started with global aerobic exercises combined with diaphragmatic breathing. Afterward, exercises to coordinate the upper and lower extremities followed. Finally, the session ended with relaxation exercises and gentle stretching of the thorax muscle groups and extremities (36). Each session of kinesiotherapy in a gymnasium started with breath awareness and the work of pelvic floor muscles. Afterward, exercises that reinforce lumbar stabilization and lumbar–pelvic dissociation followed. Finally, each session ended with training of the deep cervical muscles. Due to the variability of the physical condition of patients with FM, the intensity of the exercises was adapted to each participant (37). Physical therapy was supplemented with an exercise routine between sessions and a scheduled daily march to facilitate the incorporation of the regular exercise into daily life (see panels 2 and 3 in Supplementary Appendix A, available in the online version of this article at http://onlinelibrary.wiley.com/doi/10.1002/acr.21818/abstract). Furthermore, all of the participants were given an audio CD to practice Schultz autogenic training at home.
The patient indicated the maximum, minimum, and usual intensities of pain experienced in the last week using a numerical rating scale, with values ranging from 0–10. From these 3 independent scores, a combined retrospective measure was obtained (38). The Hospital Anxiety and Depression Scale (HADS) (39), which evaluates the presence of anxiety and depression, was used as an indicator of psychological distress (40). We used the catastrophizing subscale of the Coping Strategies Questionnaire (41) to measure catastrophizing related to pain. The Fibromyalgia Impact Questionnaire (FIQ) (42) was used to assess the impact of FM on a patient's life. We used the Dartmouth COOP/WONCA Functional Health Assessment Charts (43) to assess health-related quality of life. The total score was considered as a measure of quality of life. The Medical Outcomes Study Sleep Scale (44) was used to measure sleep disturbances.
Demographic and pretreatment outcome variables of the 2 treatment groups were compared using chi-square tests for categorical variables and t-test analyses for continuous variables. An intent-to-treat analysis including subjects who dropped out of the trial was performed. Participants with missing data were included in the analysis, and missing final outcome variable values were replaced following the baseline observation carried forward method (45) to diminish the different attrition rates between the groups and obtain conservative results. Overall statistical comparisons in each group were performed using an analysis of variance repeated-measures test, and post hoc comparisons were corrected by the Bonferroni method. Mixed linear model analysis was performed to determine the effects of the reported treatments on changes in outcome measures. The analysis focused on the fixed effects, considering whether there was an interaction between time and group and a random effect for subject. Significant interactions were broken down by exploring pairwise comparisons and corrected by the Bonferroni method.
Finally, the percentage of patients who experienced a clinically significant change was also evaluated, a cutoff of 30% was accepted for pain intensity (46), and a cutoff of 14% was accepted for the FIQ total score (47). To compare these percentages between the 2 groups, chi-square tests were performed and odds ratios (ORs) were calculated. All of the analyses were performed with the SPSS statistical package, version 20. Conventional alpha and power levels (0.05 and 0.8, respectively) were considered.
No statistically significant differences were found between the 2 experimental groups on age, body mass index, marital status, or work status, as well as on the number of tender points, pain duration, and drug therapy (Table 1). Also, no differences were found between both groups at baseline on measures of pain intensity, catastrophizing, psychological distress, FIQ total score, and sleep index problems (Table 2). However, significant differences were found in baseline measures of health-related quality of life. For this reason, this variable was eliminated from the statistical analysis.
Table 1. Total participants at baseline (pilot study excluded)*
Control group (n = 74)
Multidisciplinary group (n = 81)
Continuous variables are the mean ± SD and qualitative variables are the number (percentage) of subjects. BMI = body mass index.
Age, mean ± SD years
48.8 ± 7.2
49.0 ± 6.8
BMI, mean ± SD kg/m2
28.8 ± 5.8
27.6 ± 4.8
Marital status, no. (%)
Work status, no. (%)
Pain duration, mean ± SD years
10.8 ± 7.9
12.5 ± 9.9
Number of tender points, mean ± SD
15.6 ± 2.2
15.7 ± 2.3
Medications, no. (%)
Acetaminophen + tramadol
Dual reuptake inhibitors
Selective serotonin reuptake inhibitors
Benzodiazepine and nonbenzodiazepine hypnotics
Table 2. Pretreatment, posttreatment, 3-month followup, 6-month followup, and 12-month followup outcome measures of participants in each treatment condition*
Control group (n = 74)
Multidisciplinary group (n = 81)
Values are the mean ± SD. T-test P values were adjusted by the Bonferroni method. Values of significance in the respective line correspond to: baseline to posttreatment, baseline to 3-month followup, baseline to 6-month followup, and baseline to 12-month followup. Intent-to-treat analysis was performed and missing final outcome variable values were replaced following the baseline observation carried forward method. FIQ = Fibromyalgia Impact Questionnaire.
The percentage of participants with a HADS depression score ≥9 was 60.9% in the control group and 54.3% in the multidisciplinary group. Concerning the percentage of participants with a HADS depression score ≥11, the percentage was 51.4% in the control group and 35.8% in the multidisciplinary group. No significant differences were found between the 2 groups in these percentages. Finally, participants in the multidisciplinary group attended a mean ± SD of 22.3 ± 1.8 of the CBT sessions and a mean ± SD of 21.6 ± 2.2 of both types of physical therapy sessions.
Overall, statistics comparison showed a significant difference between multidisciplinary treatment and conventional pharmacologic treatment in pain intensity (t = −4.626, P < 0.0001), catastrophizing (t = −6.910, P < 0.0001), psychological distress (t = −7.751, P < 0.0001), FIQ total score (t = −7.708, P < 0.0001), and sleep index problems (t = 11.503, P < 0.0001).
Mixed linear model analysis showed a significant interaction of group × time in pain intensity (F = 2.965, P < 0.05), catastrophizing (F = 5.947, P < 0.0001), psychological distress (F = 10.526, P < 0.0001), FIQ total score (F = 13.362, P < 0.0001), and sleep index problems (F = 16.269, P < 0.0001).
When outcomes were compared with baseline in each of the 2 groups, post hoc analyses adjusted with the Bonferroni method showed a significant effect of multidisciplinary treatment on all of the variables studied (Table 2).
When treatments were compared between each other, estimated outcome means were used to perform a post hoc power calculation, given a sample size. Statistical power among treatments for each period of time was superior to 80% in all cases, except for baseline comparisons. Mixed-model analyses showed significant differences between multidisciplinary treatment and conventional pharmacologic treatment in FIQ total score (P < 0.0001), sleep index problems (P < 0.0001), pain intensity (P < 0.0001), catastrophizing (P < 0.0001), and psychological distress (P < 0.0001) posttreatment. At 3-month followup, the differences between both groups were significant in FIQ total score (P < 0.0001), sleep index problems (P < 0.0001), catastrophizing (P < 0.0001), and psychological distress (P < 0.0001). At 6-month followup, the differences between the 2 groups were significant in FIQ total score (P < 0.01), sleep index problems (P < 0.0001), catastrophizing (P < 0.0001), and psychological distress (P < 0.0001). Finally, at 12-month followup, the differences between multidisciplinary treatment and conventional pharmacologic treatment were significant in sleep index problems (P < 0.0001), catastrophizing (P < 0.0001), and psychological distress (P < 0.01) (Figures 2–4).
Finally, the percentage of participants who met the standard criteria for the minimum clinically significant difference in pain intensity and FIQ total score was calculated. The percentages of patients with the minimum clinically significant difference in pain intensity at posttreatment and at 3, 6, and 12 months of followup were 22.2%, 13.6%, 16.0%, and 8.6%, respectively, in the multidisciplinary group, and 6.7%, 10.8%, 5.4%, and 0%, respectively, in the control group. Also, the percentages of patients with the minimum clinically significant difference in FIQ total score were 64.2%, 48.1%, 42.0%, and 27.2%, respectively, in the multidisciplinary group, and 24.3%, 23.0%, 18.9%, and 4%, respectively, in the control group. Significant differences were found between the multidisciplinary treatment and control groups in pain intensity posttreatment (χ2 = 7.319, P < 0.01; OR 3.943 [95% confidence interval (95% CI) 1.382–11.245]), at 6-month followup (χ2 = 4.487, P < 0.05; OR 3.346 [95% CI 1.039–10.772]), and at 12-month followup (χ2 = 6.698, P < 0.05; OR 0.500 [95% CI 0.426–0.587]). With respect to the FIQ total score, significant differences between the multidisciplinary and control treatments were found at posttreatment (χ2 = 24.825, P < 0.0001; OR 5.579 [95% CI 2.773–11.221]), at 3-month followup (χ2 = 10.621, P < 0.01; OR 3.113 [95% CI 1.554–6.240]), at 6-month followup (χ2 = 9.616, P < 0.01; OR 3.100 [95% CI 1.494–6.434]), and at 12-month followup (χ2 = 15.263, P < 0.0001; OR 8.825 [95% CI 2.516–30.947]).
The main findings of this study are that the patients who underwent the multidisciplinary treatment for FM that was adapted for individuals with low educational levels improved more compared to the patients who followed the conventional pharmacologic treatment, and some improvements in key symptoms of FM, such as sleep disturbances, poor functionality, psychological distress, and catastrophizing, were maintained 1 year after completing the multidisciplinary treatment.
The results confirmed our original hypothesis. Functionality and sleep problems improved significantly with the multidisciplinary treatment for FM and clinical changes were maintained at long-term followup. Regarding sleep disturbances, the improvement was maintained at 12-month followup. One-third of the patients in the multidisciplinary group showed significant changes in functionality at followup when clinically significant changes were considered rather than significance from the mixed linear model analysis. Catastrophizing and psychological distress also showed improvements, and those were maintained in time. By contrast, when considering pain intensity, changes were not maintained at followup. Nevertheless, when the groups were compared considering the clinically significant change, multidisciplinary treatment was superior at the 6-month followup and at the 12-month followup. Finally, as it was pointed out in the results, we did not analyze health-related quality of life due to the differences found at baseline.
When our program was designed, adjustment of the contents and the form of presentation to the educational level of the participants was considered a priority. Nonetheless, the maintenance of the essential elements that make up a multidisciplinary treatment of FM was another key criterion to design the program (1, 7), with the purpose of being able to generalize our results and to compare them with other studies.
Compared to other works, this study has provided more evidence for the beneficial effects of a multidisciplinary treatment for FM. Specifically, our work expands the existing evidence for the effects of a multidisciplinary therapy on reduced pain intensity and mood improvement in patients with FM (7). It also provides evidence for the effects of a multidisciplinary treatment on other important aspects of FM such as poor functionality and catastrophizing.
This study has also provided data regarding the long-term maintenance of the therapeutic benefits. This refutes one of the most common criticisms of the efficacy of a multidisciplinary treatment for FM (i.e., that the therapeutic benefits are only briefly maintained) (7). However, in our treatment program, the improvements in key symptoms of FM (psychological distress, catastrophizing, and sleep problems) were maintained after 1 year of followup. Moreover, given the percentage of participants who obtained a clinically significant improvement in pain intensity (46) and functionality (47), our study provides more evidence for the long-term maintenance of the therapeutic benefits of combining the drug treatment with CBT and physical therapy.
Although the multidisciplinary treatments for FM that have been performed to date have not demonstrated an efficacious reduction in the number of sleep disturbances (even in the short term) (7), the program designed in this study achieved a decrease in the number of sleep disturbances in the short term and maintained these benefits in the long term. This fact could have considerable clinical repercussions, given that a decrease in the number of sleep problems is related to the level of functionality in patients with chronic pain (48). This improvement may be due to various factors. One factor is the patients' introduction to relaxation as one of the treatment components. In fact, relaxation has been shown to increase the effects of CBT in the treatment of insomnia in patients with FM (49). Another factor is the introduction of specific aspects of CBT for chronic insomnia, such as sleep restriction and stimulus control, which have demonstrated their utility in patients with FM (50).
Aside from its purely clinical aspects, the adaptation of our program for patients with low educational levels had other notable characteristics. On the one hand, it did not involve a significant increase in the treatment duration. In fact, the treatments included in the meta-analysis by Haüser et al (7) lasted between 18 and 46 hours, which is similar to the duration of our program (48 hours). Conversely, our program could be applied in a community setting, which can facilitate its accessibility to patients. Another significant aspect of our work is the maintenance of some improvements at long-term followup. CBT teaches patients about interrelationships among thoughts, feelings, behavior, disability, and pain. Also, CBT provides training in a variety of cognitive and behavioral techniques that can help our patients preserve the skills acquired. Nevertheless, the introduction of specific contents such as life values and relapse prevention would help in the maintenance of changes.
Our experiment has some limitations that must be considered. When comparing the multidisciplinary treatment group to the control group, we did not consider the possible effect of contact, and some of the results could have been due to the different degrees of time and attention that were dedicated to each of the groups. Unfortunately, we did not include an attention control group, and for this reason cannot determine the independent effect of this variable. We also could not determine the differential effects of each of the separate components of the multidisciplinary treatment or the effects of some predictors of treatment outcomes as self-efficacy or treatment expectancies. Finally, because we limited our study sample to women with low educational levels, the potential for generalizing our results to other population groups is partially limited, as stated above. However, this limitation is nuanced by the fact that our primary objective was, precisely, to determine the efficacy of a multidisciplinary treatment for FM in a population with these specific characteristics. Further experimental research should determine the additional value of the adaptation for patients with a low educational level, comparing a tailored program with another one not specifically tailored for patients with these characteristics.
In conclusion, a multidisciplinary treatment program for FM that combines pharmacologic treatment, education, physical therapy, and CBT (the contents of which have been adapted for patients with low educational levels and applied in a group format in a nonhospital environment) has demonstrated efficacy in the treatment of the key symptoms of FM and the long-term maintenance of these improvements.
All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Castel had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study conception and design. Castel, Fontova, Montull, Miralles, Cascón-Pereira, Salvat, Monterde, Añez, Rull.
Acquisition of data. Periñán, Poveda, Aragonés, Castro, Padrol, Sala.
Analysis and interpretation of data. Castel, Fontova, Hernández, Castro.
We would like to thank Marta Salvat, Patricia Zaldívar, and Eva Luque, scholarship holders in our unit, for their efficient collaboration in the project. We would also like to thank all of the participants for their collaboration and patience, without whom we would not have been able to accomplish this study.