A total of 110 participants diagnosed with FM (99% women) were recruited from the Central Sensitivity Syndromes Specialized Unit (CSSSU) at the Vall d’Hebron University Hospital (Spain). Table 1 presents the sociodemographic characteristics of the study participants. No significant differences were found between the groups at baseline in terms of the general measures or clinical outcomes.

To be eligible for participation, individuals had to meet the following criteria: (a) a diagnosis of FM according to the 2010/2011 American College of Rheumatology (ACR) criteria,12 (b) 18 years of age or older, and (c) fluent in Spanish. Individuals with terminal illnesses or scheduled treatments that could disrupt their participation in the study were excluded (n=19). Recruitment for the study was conducted from February to March 2022, and the interventions were conducted from April to July 2022. The flow of participants through the study is shown in Fig. 1.

Fig. 1.

Flow of participants through the trial.

This was a 12-week, randomized controlled trial in which recruited participants were randomly assigned to online FIBROWALK (n=39) or blended FIBROWALK (n=71). The blended study arm was divided into two subgroups: four outdoor sessions at the park (n=35) and four outdoor sessions at the beach (n=36).

All procedures followed the ethical standards established in the 1964 Declaration of Helsinki and its subsequent revisions. The FIBROWALK study protocol was approved by the the Ethical Committee of Clinical Investigation of Vall d’Hebron Hospital (code: PR(AG)249/2020) and registered on ClinicalTrials.gov (NCT05395832). This study adhered to the guidelines specified by the Consolidated Standards of Reporting Trials (CONSORT).13

Participants diagnosed with FM from the Central Sensitivity Syndromes Specialised Unit (CSSSU) at the Vall d’Hebron University Hospital (Spain) were initially screened by the main researcher (M.S.) and informed of the study. After obtaining informed consent, the participants were assigned an alphanumeric code and randomized into either the online FIBROWALK group or the blended FIBROWALK group using SPSS version 26. The participants were assessed at baseline and upon completion of the intervention (12 weeks after the beginning) through an online survey platform (REDCap). Participants in both arms continued their usual treatment, mainly pharmacological treatment tailored to each patient's symptom profile.

Adherence to the online sessions in both formats was evaluated using a weekly content test, and adherence to the outdoor sessions was evaluated using attendance logs.

Online FIBROWALK: This program comprised 12 weekly sessions, each lasting 60min. The participants were sent via e-mail a weekly link to a 60-min video. Each video module integrated diverse components of the program, including Pain Neuroscience Education (PNE), therapeutic physical exercise, Self-management Patient Education, CBT techniques, with an emphasis on cognitive restructuring and mindfulness training. The content and structure of the program are described in detail elsewhere.10 Additionally, the program integrated homework tasks aimed at enhancing patients’ endurance and resilience through consistent challenges. To ensure that participants engaged in the program, a brief content test consisting of (<10 questions) about the session was sent via email every week.

Blended FIBROWALK: The contents are identical to those delivered in the online FIBROWALK, but sessions corresponding to weeks 2, 6, 8, and 11 were performed face-to-face in addition to online sessions. Participants freely chose to attend outdoor sessions. These sessions were carried out in a park located near the referral hospital (“Parc del Cargol”) for one of the subgroups and in a beach (“Platja del Bogatell”) for the other.

Sociodemographic and clinical variables: Data on gender, age, marital status, cohabitation, educational level, and employment status were collected using an ad hoc survey. Concerning clinical variables, we asked participants about their years with an FM diagnosis, possession of a disability certificate, height, weight, and comorbidity with chronic fatigue syndrome.

The Revised Fibromyalgia Impact Questionnaire (FIQR)14 was used to measure functional impairment. This instrument assesses three dimensions: physical dysfunction, impact of FM, and severity of symptoms. The total maximum score can be calculated by adding the three subscale scores, where a higher score indicates greater impairment (range of scores: 0–100). The Spanish version was used15; in our sample, it showed good internal consistency (α pre=.95; α post=.95).

The Hospital Anxiety and Depression Scale (HADS)16 was used to measure anxiety and depressive symptoms. It comprises 14 items answered using a four-point Likert scale. An overall psychological distress score can be calculated by adding all item scores, with higher values indicating greater anxiety-depressive symptomatology (range of scores: 0–42). A Spanish version was used.17,18 It showed adequate internal consistency in our sample (α pre=.85/.87; α post=.89/.89).

The physical functioning subscale of the Short Form-36 Health Survey (SF-PF-36)19 was used to measure physical function. This dimension consists of ten items rated on a three-point Likert scale. The total score can be calculated by adding all item scores. Higher scores indicate higher physical functioning (range: 0–100). The Spanish version was used in this study.20 It showed adequate internal consistency in our sample (α pre=.86; α post=.88).

The Tampa Scale for Kinesiophobia (TSK)21 was used to measure fear of movement and pain. This scale consists of 11 items rated on a four-point Likert scale. The total score can be calculated by adding all item scores. Higher scores indicate greater fear of movement (range of scores: 11–44). The Spanish version was used.22 It showed adequate internal consistency in our sample (α pre=.86; α post=.89).

The Credibility/Expectancy Questionnaire(CEQ)23 was used to assess treatment expectancy and opinions of the interventions. It consists of six items on an 11-point Likert scale (from 0 to 10), with the pre-test part focusing on therapy credibility and expectations, and the post-test part gathering patient opinions after treatment completion. Higher scores indicate a better perception of the treatment (range of scores: 0–10). An ad-hoc Spanish version was used.7

Data analyses were performed using SPSS v26. Descriptive statistics for all variables were calculated and reported using means and standard deviations for continuous data, and frequencies and percentages for categorical data. Student's t-tests and χ2-tests were conducted to further analyze any potential differences in sociodemographic and clinical characteristics between the online and blended FIBROWALK groups. In addition, the continuous variables were checked for normality (kurtosis ranging from −2 to +2 and skewness from −7 to +7, Kolmogorov–Smirnov's statistic with p>0.05, histograms, and Q–Q plots), with no major violation noted of this assumption for parametric tests. Moreover, within-paired-samples t-tests were conducted to evaluate the impact of the interventions in each group on the FIQR, HADS, SF-PF, and TSK scores.

A series of one-way between-groups analyses of covariance (ANCOVA) was conducted to compare the effectiveness of both interventions. The independent variable was the type of intervention (online FIBROWALK, blended FIBROWALK), and the dependent variables consisted of total scores on the FIQR, HADS, SF-PF-36, and TSK administered after the intervention was completed. Participants’ scores on the pre-intervention administration of these measures were used as covariates in each analysis. Cohen's d was calculated as measure of effect size using the pooled initial SD to measure differences in the initial and subsequent mean values and to correct for the estimated population (d=0.20, small effect size; 0.50, moderate; and 0.80, large).24 For this analysis, we employed a ‘completers approach’, which provided insight into the effectiveness of patients who reached the study endpoint. In addition, a ‘per-protocol analysis’ was performed including only those who completed the study and attended all online sessions in both interventions and at least two outdoor sessions of the blended format.

A Chi-Square Test for Independence was conducted to explore the relationship between the type of intervention and the proportion of ‘responders’. A ≥20% reduction in the FIQR total score was considered a criterion of a clinically relevant change.14 Student's t-tests and χ2-tests were conducted to further analyze any potential differences in sociodemographic and clinical characteristics between responders and non-responders within each arm.

Eleven participants (10%) dropped out due to personal or family circumstances at the beginning of the intervention period. Dropout rates did not vary significantly across treatment arms (χ2=2.54, p=.111): one participant (2.56%) in the online group and ten (14.01%) in the blended group. There were no statistically significant differences in baseline characteristics between dropouts and completers in both arms (p>.05 in all cases). Participants who completed both interventions (38 in online FIBROWALK and 61 in blended FIBROWALK) attended all online sessions. In the blended format, 19 participants (26.76%) attended at least two sessions.

Table 2 presents the participants’ scores on expectancy and opinion items for both the interventions. Overall, participants had high expectations of online FIBROWALK (CEQ total mean score=8; SD=1.40) and blended FIBROWALK (CEQ total mean score=7.68; SD=1.66), with no significant between-group differences (p=.331). An inspection of the items scores indicated that the online and blended FIBROWALK were expected to be logical, satisfactory, useful, recommendable, useful for treating other problems (M ranging between 7.71 and 8.50 for online FIBROWALK; M ranging between 7.51 and 8.11 for blended FIBROWALK), and non-aversive (M=3.08 for online FIBROWALK; M=3.23 for blended FIBROWALK), with no significant between-group differences (p>.05 in all cases). The per-protocol analysis showed similar results (see Table S1 in the Supplemental Material).

Overall, after the interventions were completed, participants had a good opinion of online FIBROWALK (CEQ total mean score=8.06; SD=1.44) and blended FIBROWALK (CEQ total mean score=7.49; SD=1.95), with no significant between-group differences (p=.120). Concretely, the CEQ item scores indicated that both the online and blended FIBROWALK were considered logical, satisfactory, useful, recommendable, useful for treating other problems (M ranging between 7.16 and 8.87 for online FIBROWALK; M ranging between 7.07 and 8.39 for blended FIBROWALK), and non-aversive (M=2.29 for online FIBROWALK; M=4.03 for blended FIBROWALK), with no significant differences between group scores (p>.05 in all cases), except for the aversiveness item (p=.013). The per-protocol analysis reflected that the blended FIBROWALK group perceived it as more logical, and participants were more satisfied than those in the online group. In addition, the aversiveness item score was lower and there was no statistically significant between-group difference (p=.368) (see Table S1 in the Supplemental Material for more details).

Overall, paired-samples t-tests showed that there were statistically significant improvements in the FIQR, HADS, SF-PF, and TSK scores from baseline to post-treatment in both conditions with small-to-large effect sizes (p<.05 in all cases, Cohen's d ranged from .19 to 1.43; see Table S2 in the online Supplemental Materials for more details). However, the online FIBROWALK group did not exhibit a significantly lower average score on the HADS from baseline to post-test (p=.052, d=.19).

Table 3 shows the means and SDs of the patient-reported outcome measures at baseline and post-treatment in the online and blended FIBROWALK groups (with completers). A statistically significant between-group difference with a moderate effect size was found in the HADS scores (p=.042; d=.60), indicating that the blended FIBROWALK group had a lower level of psychological distress than the online group at the end of the intervention period. The per-protocol analysis reflected the same results (see Table S3 in the online Supplemental Materials for more details).

Among the participants, 33 (33.33%) of the total sample achieved responder status (i.e. reduction of ≥20% in FIQR scores), of which 13 (34.21%) belonged to the online group and 20 (32.79%) to the blended group. The rate of responders did not vary significantly across treatment arms (χ2=0.02, p=.884).

Tables S4 and S5 in the Supplemental Material show that there was a statistically significant difference in years of illness between responders (M=9.08; SD=7.34) and non-responders (M=16.64; SD=12.20) of the online FIBROWALK group (p=.048) as well as a statistically significant difference in BMI between responders (M=30.80; SD=7.70) and non-responders (M=27.07; SD=6.02) of the blended FIBROWALK group (p=.043). In contrast, their sociodemographic and clinical characteristics and baseline FIQR, HADS, SF-PF, and TSK scores were not significantly different (p>.05 in all cases).