Persistent low back pain (LBP) is among the most prevalent, disabling, and expensive conditions to treat [1]. The 2023 Global Burden of Disease study estimates that 619 million people lived with LBP in 2020, with numbers projected to rise to 843 million by 2050—nearly 39% of the associated disability can be attributed to occupational ergonomic exposures, smoking, and high BMI [2]. It can affect an individual’s ability to work, maintain relationships, and function in normal day-to-day life [1]. Recent European data confirm LBP’s labor-market impact: in a nationwide Swedish cohort of more than 830,000 workers, any episode of LBP-related sickness absence tripled the likelihood of subsequent disability retirement (hazard ratio 3.2) [3]. The UK evidence from the Whitehall II data similarly reports that recurrent back pain is associated with higher odds of exiting from paid work for health reasons, particularly in those in lower and middle grade jobs [3].

As the incidence of LBP is becoming more prevalent globally, with the steepest relative increases projected for low- and middle-income countries [4], where resources may be limited, it is important that accessible and effective interventions are available. The World Health Organization’s 2023 Digital Health Strategy cites musculoskeletal conditions as a priority area for scalable eHealth solutions in resource-constrained settings. Since most LBP is not considered to be related to a specific identifiable spinal pathology, digital educational self-management programs are appealing as they encourage and facilitate wider participation. Digital delivery, including web-based and mobile formats, expands reach, allows asynchronous access, and can reduce per-capita costs by leveraging economies of scale [5].

Digital educational self-management programs have demonstrated effectiveness in a range of chronic disorders such as cardiovascular disease [4], chronic obstructive pulmonary disease [6], and diabetes [5]. For LBP specifically, a 2025 systematic review of digital interventions for employed adults with LBP identified small-to-moderate improvements in pain, disability, and work ability across 18 trials, although study quality was mixed and attrition was high [7]. The following are the findings of 2 recent randomized controlled trials: the UK SupportBack 2 trial (n=1201) found that an internet-delivered program with or without brief clinician support was safe, cost-effective, and produced modest reductions in disability at 12 months [8]; the Scandinavian selfBACK trial (n=346) reported a statistically significant reduction in Roland-Morris disability scores that persisted to 9 months among workers with physically demanding jobs [9]. However, digital interventions to date appear to largely focus on individuals seeking care within health systems, limiting accessibility. An option for improving this and scalability is by using a “MOOC” (massive open online course) format. MOOCs are a type of digital course designed using principles of educational technology. For employees with persistent LBP, a MOOC offers flexible, asynchronous learning that can be fitted around working hours and accessed from any location with an internet connection. While MOOCs have shown promise during COVID in helping health workers through the provision of health education [9], the feasibility of using this format to facilitate an educational self-management program for working individuals with back pain has not been investigated.

The MyRelief MOOC was developed specifically to support individuals in employment with LBP. The materials were derived from a synthesis of consistent recommendations across 6 clinical guidelines [10-15]. The key focus of the MyRelief program was to provide a summary of evidence-based information in user-friendly formats.

The aim of this study was to determine the feasibility and acceptability of the MyRelief MOOC.

The MyRelief study protocol, providing the rationale and description of the study, has been published elsewhere [16]. This paper was prepared in accordance with the Consolidated Standards of Reporting Trials (CONSORT) guidelines for reporting of a pilot or feasibility trial, and a CONSORT checklist has been included as Checklist 1.

The MyRelief study was a multicenter, multinational, uncontrolled feasibility study in Italy, Portugal, Sweden, and the United Kingdom. A fifth partner, Lithuania, contributed as a technology specialist but was not directly involved in the feasibility study. A separate paper was published [17] focusing on the technological aspects of the study. Ethics approval was obtained from the relevant ethics committee in each country. This study was funded by the European Commission’s Erasmus+ Program.

Recruitment of adult workers with nonspecific, persistent LBP, present for at least 3 months, took place at the 4 study sites in Italy, Portugal, Sweden, and the United Kingdom between 2020 and 2021. Recruitment took place via several methods depending on each country's ethical approval, for example, via waiting list for outpatient rehabilitation services (Italy), email circulated to all staff and students at University (Sweden and the United Kingdom) or patients at a pain clinic (Portugal) or Social Media (Italy, Sweden, the United Kingdom, and Portugal). For the latter approach, a Facebook page was created with information about the project. Links to this page were posted on Instagram, Twitter, and other relevant forums. Individuals with unexplained symptoms, underlying pathology, or specific LBP were excluded, as were those who were not interested in using a digital intervention delivered via computer, laptop, tablet, or mobile. A full list of inclusion/exclusion criteria is in our study protocol [16].

Regardless of the method of recruitment, potential participants accessed the study information via a link provided in the recruitment material. If interested, they completed a web-based inclusion/exclusion questionnaire. Those meeting the inclusion criteria were given access to a digital consent form. After completing the consent form, they received access to the MyRelief study website to complete the baseline assessment and access the digital education program.

All participants were also asked to enter their email address and phone number to enable reminders and contacts with those interested in participating in the qualitative interview.

Various strategies to enhance recruitment were adopted in the different countries. In Sweden and the United Kingdom, interested participants responded via email to a member of the research team who checked eligibility and gave an overview of the study, answering any queries, prior to providing the study web link. The researcher arranged email, telephone, or video call support, depending on participant preference, to support them to get started with the system and checked in to resolve any technical issues. In Italy, multiple routes (email, telephone, and in-person recruitment meetings) were used to recruit participants, who were staff at the research hospital (nurses), local university (employees and administrative staff), and the general public. In Portugal, participants were recruited by direct contact with private clinics dealing with the issue.

The MyRelief digital education program was designed to be accessed on a computer or a mobile device. The MyRelief content was informed by focus groups with adult workers with persistent LBP, LBP guidelines, and the clinical expertise of the research team. It comprised 8 modules which covered the following topics: understanding LBP; physical activity and exercise in relation to LBP; psychological factors; sleep/nutrition; management of LBP in the workplace; communication with health care providers; and other issues related to LBP (eg, medication, scans, and alternative therapies). Each module was designed to be short enough to complete during a work break, that is, <20 minutes, and was made up of short written and video content, as well as knowledge quizzes. Participants received 4 weeks’ access to the digital education program prior to being invited to complete the postintervention assessment. Thus, completion of 2 online modules per week was encouraged to facilitate completion of the course within the timeframe.

The feasibility of the study was assessed as the ability to recruit participants and retain them in the study, and completion of functional disability and quality of life measures. Recruitment was assessed as the proportion of the recruitment target met during the timeframe. A record was kept of the number of individuals who entered the system, completed the screening questions, and provided consent. Retention was assessed by calculating the proportion of participants who completed outcome assessments at both baseline and postintervention timepoints. Completion of outcome measures at baseline and postintervention timepoints was recorded.

Participant views on usability and satisfaction with the system were assessed postintervention using the System Usability Scale (SUS), which is an industry standard measure with validity to measure a variety of aspects of usability, such as the need for support, training, and complexity [18].

Qualitative feedback was assessed through semistructured telephone interviews with a subgroup of study participants (Italy) and comments provided through email correspondence between participants and the study teams (Sweden and the United Kingdom). Semistructured interview questions included feedback on participants’ experience using the digital education program, likes and dislikes, perceived usefulness, problems experienced, and suggestions for improvement.

Participant characteristics (demographics, history of LBP, type of occupation, etc) were collected at baseline via the study website. Type of occupation was defined as sedentary (mostly sitting, eg, office worker, driver, and professional), light manual (mostly standing or walking, eg, retail worker, health care worker, and cook), heavy manual (heavy labor, eg, farming, construction, and cleaning), unsure, or other.

The following outcome measures were collected at baseline and postintervention via the study website:

1. Functional disability was measured using the Oswestry disability questionnaire (ODQ) v2.1a, a valid and reliable measure of pain and physical function in people with LBP [19,20]. A minimum important change of between 10 and 12 points over time, or an improvement from baseline of between 20% and 30% for an individual, has been recommended [21].

2. Health-related quality of life was measured using the EuroQol questionnaire (EQ-5D-3L), which has been recommended for use in LBP research [22]. A minimum important change of 0.03 points for the categorical scales of the EQ-5D and 10.5 points for the EQ-5D Visual Analog Scale has been suggested [22].

The following outcomes were measured at postintervention only:

3. Change in understanding and coping with their LBP was assessed via the study website using the Patient Enablement Index (adapted to reflect enablement after engaging with the system), which is considered “the gold standard” for measuring enablement [23,24]. This measure is calculated as a sum score ranging between –6 to 10 points. A score greater than 6 points is suggested to reflect “high” enablement [25].

4. SUS: Scores above 70 can be interpreted as good usability, while scores below 50 indicate unacceptably low usability according to Bangor et al [18].

5. Semistructured interview (see above).

Statistical analysis was performed using Microsoft Excel. Participant characteristics and feasibility outcomes for recruitment and retention were presented as frequency and percentage (%), mean and SD, or median and IQR based on the distribution of the data.

The following feasibility criteria were applied: observed recruitment rate falls short of 70% of that anticipated, overall drop out of over 35%, no apparent change in the outcomes with confidence intervals that include large negative values, feedback from participants that they were unable to complete (or lack of engagement with) MyRelief, feedback from participants that suggests that the MyRelief content or format was not acceptable or usable.

Clinical and usability outcomes were reported descriptively as means and 95% CIs; significance tests were not performed in this feasibility study.

A number of pragmatic changes were implemented during study delivery. Initially, study recruitment was entirely online via a web link. Engagement was poor with this automated recruitment method, so additional methods were implemented to support recruitment. Changes related to researcher-supported recruitment are described above in the “Participants and Recruitment” section. In addition to these changes, participants also received emails (Sweden, the United Kingdom, and Portugal) or telephone reminders (Italy) to complete the follow-up questionnaire.

The study protocol outlined plans to collect qualitative data from participants related to usability and acceptability at all study sites; although 3 follow-up telephone interviews were successfully conducted in Italy, Sweden, and the United Kingdom, informal participant feedback was collated via telephone and email correspondence due to challenges with recruitment.

Ethics approval for this study has been obtained from the relevant ethics committee at each intervention site (Swedish Ethical Review Authority Dnr 2020-03536). The study was classified as human research involving minimal risk. The trial has been registered on ClinicalTrials.gov (NCT03575091). Regardless of the method of recruitment, potential participants accessed the study information via a link provided in the recruitment material. If interested, they completed a web-based inclusion/exclusion questionnaire. Those meeting the inclusion criteria were given access to a digital consent form. The consent covered both primary data collection and potential secondary analysis of deidentified data. Participants were informed of their right to withdraw from the study at any time without any consequences. All data were deidentified prior to analysis, and all materials are presented to ensure complete anonymity of the study participants. Participants were assigned unique study codes, and no personally identifiable information was stored with the research data. All digital files were stored on secure, password-protected servers in accordance with institutional data protection guidelines. Participants did not receive any compensation for their contribution to the study.

A total of 40 participants (19 male and 21 female; mean age 57 years) were recruited. Baseline demographics are reported in Table 1. Mean pain duration was 11.5 years and differed between countries from 7 years (Portugal) to 16.3 years (Sweden), and among individuals, it ranged from 1 year to 38 years. A total of 28 (70%) of participants reported a sedentary occupation. Among the participants in Portugal and Italy, these levels were lower, 40% and 54%, respectively (Table 1).

This study explored the feasibility, usability, and acceptability of a digital education program for people living and working with persistent LBP. Recruitment of study participants reached the a priori feasibility criteria (>70% recruitment). However, retention was challenging, and the a priori feasibility criteria were not met (<35% dropouts). The quality of outcome measure completion was good. The SUS indicates that the MyRelief website met minimal industry standards, but qualitative feedback confirmed some challenges with usability in terms of access to and navigation within the website. Although mean functional disability was moderate, actually, half of our sample only had mild functional disability and low pain intensity. A clinically important increase between baseline and follow-up was found for EQ-5D Weighted Health Index, suggesting a meaningful improvement in patients’ perceived health status. This aligns with the idea that participation in patient education sessions may play a positive role in enhancing quality of life. While the observed improvement is promising, further exploration is needed to determine whether this alone is sufficient to sustain long-term improvements in quality of life.

The SUS score of 72.4 suggests that our digital education program meets acceptable usability standards and can be considered to offer good usability [18], indicating that MyRelief digital education has the prerequisites required for further development. In comparison to other digital interventions to support and facilitate self-management of LBP, previous mean scores have ranged from 64.7 to 70.5 [17,35]. However, comparability is limited as previous studies have mostly focused on the usability of mobile applications. Nonetheless, it is promising that the current digital program was perceived as usable by MyRelief study participants. The current platform did not support detailed use analytics, which limited our ability to assess which modules were most engaging or where participants may have disengaged. Future versions should include integrated tracking of module completion, time spent, and quiz interaction to better understand user patterns and adapt content accordingly. There was a lack of qualitative feedback on usability in our study, but the feedback that was collected primarily highlighted issues around accessibility to the website. It has been reported that a clear and intuitive navigation system can significantly enhance accessibility, ensuring easy access to important information [36]. Improved understanding regarding accessibility and usability preferences from the end users’ perspective is required in order to improve the utility of the digital education format for health-related programs.

Also, the acceptability of the MyRelief digital education program was assessed through qualitative feedback collected via semistructured interviews and email correspondence. Participants generally appreciated the relevance of the content and its focus on self-management, with several highlighting the usefulness of practical examples. However, feedback also revealed that some participants found the material too basic in relation to their existing knowledge or needs. This suggests that although the education program was acceptable to many, others desired greater depth or personalization.

While the overall design and delivery format were well received, some users expressed preferences for shorter modules or more interactive elements. These comments provide important insights for future refinement of the education program. Acceptability could potentially be improved by tailoring content to different levels of prior knowledge or by offering optional advanced sections. Although qualitative feedback was limited in volume, it was important in identifying strengths and specific areas for improvement.

We conducted this study in accordance with CONSORT guidelines and followed a prospectively registered protocol [16]. A key strength and a novelty of this trial is the transnational approach in developing evidence-based material that could facilitate health education, providing an infrastructure that can be upscaled to reach more learners, leveraging digital tools and materials. However, the transnational development process also revealed that the cultural context of health differed between countries. A lot of the workload within the project was attributed to developing materials that required multiple translations and heavy editing to ensure the advice and information were appropriate for and sensitive to different cultural contexts. To maximize the utility of interventions delivered using a digital education format, cultural contexts of health need to be considered. The multi-country nature of this study revealed valuable insights into the importance of cultural and contextual adaptation. Although no formal subgroup comparisons were made, observations from the project teams indicated differences in how participants across sites perceived the program’s content and usability. Some adjustments—such as modifying terminology, examples, and health care system references—were made during translation to improve relevance. Nonetheless, variations in language availability and local health literacy may have influenced both engagement and acceptability, highlighting the need for cocreation [37-39] and iterative user testing in future implementations. Although we conducted initial focus groups with individuals with LBP, further iterative review opportunities with patient representatives and clinicians were not feasible within the project timeline. For future studies, more focus should be dedicated to structured cocreation and the involvement of health care providers in this process, including repeated feedback loops from patients in each participating country. In addition, the diversity of learners’ own goals, motivations, and interests needs to be satisfied through the design, as the learning experience will benefit from personalization [40]. Another key consideration for future development is the importance of using an administratively simple and flexible course system. The current platform lacked integrated tracking features, which limited our ability to monitor progress and engagement. To minimize challenges with technology, interdisciplinary collaboration with developers and clinician academics is crucial. For digital education programs to be scalable and sustainable, they must also be supported by systems that allow easy management of participants, automated data collection (eg, module completion, quiz performance), and user support. A streamlined and user-friendly backend interface is essential for both researchers and educators to efficiently deliver content, monitor engagement, and provide timely feedback.

This study was feasible in terms of recruitment but not retention. About half of our sample had low levels of functional disability, and in the future, recruitment via health care providers may be more appropriate, and recruitment strategies should focus on efforts to recruit individuals from heavy manual and repetitive task jobs, as they are more representative of the wider persistent LBP population. Our usability scores suggest our website meets minimal industry standards. Completion of outcomes was good, apart from ODQ. Future studies should explore culturally adapted versions of the program, developed in collaboration with patients and health care providers. Earlier and more frequent end user input may help refine language and relevance, thereby improving both recruitment and retention.