Efficacy of YOga in people with chronic non-specific low BACK pain and poor PROgnosis (YOBACK-PRO): protocol for a randomised clinical trial

Abstract

Introduction

Yoga has been shown to improve pain and function compared with no exercise in people with chronic low back pain (LBP), but treatment effects are small. Given that yoga is a mind–body intervention that addresses physical as well as psychological factors, it may be more effective for patients with chronic LBP who are at high risk of poor prognosis. The study aims to investigate the efficacy of a 12-week yoga programme combined with education in reducing pain and disability for individuals with chronic LBP at high risk of poor prognosis at short (12 weeks) and intermediate (24 weeks) terms, compared with a control group receiving education only.

Methods and analysis

A randomised controlled trial will include 110 adults with chronic non-specific LBP reporting an average pain intensity of 3 points or more on a 0–10 scale over the past week and classified as high risk of poor prognosis (ie, scoring 50 points or above) on the Orebro Musculoskeletal Pain Questionnaire short-form. Participants in the control group will receive an educational booklet and attend three face-to-face lectures over a 3-month period. In the intervention group, in addition to the booklet and lectures, participants will attend group yoga sessions twice a week for 12 weeks, totalling 24 yoga sessions. The primary outcome is disability assessed at 12 weeks, measured using the Roland-Morris Disability Questionnaire.

Ethics and dissemination

The study was approved by the Human Research Ethics Committee of Universidade Federal de Minas Gerais (Protocol number CAAE: 57028022.0.0000.5149). Findings will be disseminated to trial participants, clinicians and the broader public and scientific community.

Trial registration number

NCT05953155.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• Prospectively registered randomised controlled trial with prespecified primary and secondary outcomes.

• Targeted, clearly defined high-risk subgroup (adults with chronic low back pain (LBP) at high risk of poor prognosis), reducing heterogeneity and testing a plausible mind–body intervention of strong clinical relevance.

• Standardised yoga protocol (24 sessions over 12 weeks), refined via a feasibility study and delivered by trained clinicians to support reproducibility.

• Blinding of participants and therapists was not feasible given the nature of the intervention and comparator.

• Generalisability is limited to adults aged 18–60 years with chronic LBP at high risk of poor prognosis.

Introduction

Low back pain (LBP) is the leading cause of disability worldwide,¹ with a point prevalence of 18% and a 1 year prevalence of 38%.² While most cases of acute LBP show significant recovery within 6 weeks,³ nearly 40% progress to chronic symptoms.⁴ Chronic LBP imposes a substantial economic burden globally, contributing to both direct and indirect societal costs.^{5 6} International clinical practice guidelines consistently recommend non-pharmacological interventions,⁷ but treatment effects are, at best, small, and the certainty of evidence is not rated as high.^{8 9} The recent WHO guideline,¹⁰ for instance, conditionally recommends (ie, based on low-certainty evidence) structured and standardised education, structured exercise therapies and cognitive behavioural therapy. This conditional recommendation means that although these interventions are recommended in most cases, they may not be suitable for every patient.¹⁰

LBP is a complex condition influenced by multiple factors, including biophysical, psychological and social factors, all of which contribute to its chronicity. Given this multifactorial nature, targeting a single factor in isolation may be insufficient, as it fails to address all underlying mechanisms.¹¹ Another common criticism in this field is that the small effects observed for most guideline-endorsed interventions may be due to the inclusion of a highly heterogeneous group of participants in LBP trials, making it difficult for any single intervention to demonstrate large effects.¹² Among trial participants, for instance, some may be more affected by psychological factors, such as depressive symptoms¹³ and fear avoidance beliefs,¹⁴ while others may not. As a result, interventions that benefit one group may be ineffective for another. This may explain why the interventions in the WHO guideline were conditionally recommended in favour, meaning that they may not be suitable for every patient. Hence, trials testing interventions that target multiple contributing factors and are tailored to a more homogeneous sample might result in large treatment effects.

Yoga is a mind–body intervention that integrates movement-based exercises, breath control and meditation. It can influence both physical factors, including flexibility, strength and movement confidence, and psychological factors, such as depressive symptoms, anxiety and catastrophising.15,17 While the exercise component of yoga encourages patients to be more active and confident in moving, the breath control and meditation components promote relaxation, alleviate stress and reduce pain sensitivity.¹⁸ When practised in groups, yoga has been reported to increase social connectedness by fostering a sense of community among participants, which may be particularly valuable for individuals with depression symptoms or at high risk of social isolation.^{19 20} Evidence from a recent Cochrane review shows that yoga is probably better than no exercise for pain and function, but the treatment effects are small.²¹ We would argue that larger effects may be found when yoga is performed in a group setting and targeted to patients with chronic symptoms and high risk of poor prognosis (ie, who are feeling depressed, anxious and/or fearful of moving and exercising), as it addresses not only the physical factors but also the psychological and social factors involved in chronic LBP. Such patients can be clinically identified using screening tools such as the Örebro Musculoskeletal Pain Questionnaire (OMPQ).²² The OMPQ was originally developed to predict the risk of chronic LBP in patients with acute symptoms, and more recently, the OMPQ short-form has been considered a promising tool to be used as a prognostic tool in people with chronic LBP.²³

The integration of yoga into physiotherapy clinical practice is now a reality.²⁴ A survey of physiotherapists (n=333) in the USA has shown that more than two-thirds (71%) of the participants used yoga to target functional outcomes, nearly half (46%) used yoga more than four times during the previous month and one-third (36%) reported recommending yoga as part of their client’s home exercise programme.²⁵ However, despite the increased use of yoga as a treatment modality by physical therapists, it is still unclear if yoga combined with education can promote clinical improvements in a more homogenous clinical population compared with a control group receiving education only. Therefore, this randomised controlled trial will investigate the efficacy of a 12-week yoga programme combined with education in reducing disability for individuals with chronic LBP at high risk of poor prognosis at short (12 weeks) and intermediate (24 weeks) terms, compared with education only. Secondary outcomes for this trial include pain intensity, depressive symptoms, pain self-efficacy and quality of life.

Methods

Study design

This protocol of a randomised controlled trial follows the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) statement. The SPIRIT checklist is available as online supplemental material 1. This trial has been prospectively registered at ClinicalTrials.gov (NCT05953155) on 12 July 2023. The study’s design is described in figure 1. This is a protocol for an efficacy (explanatory) trial, designed to isolate the added benefit of yoga when delivered alongside a standardised education component. Using education as the comparator enables estimation of the incremental effect attributable to yoga. While credible education should not be seen as a stand-alone treatment for chronic LBP, it provides a practical and credible comparator for an efficacy trial.

Figure 1. Study design showing participant timeline, including schedule of enrolment, interventions and time-points of assessment. (X, indicates when data will be collected; black circles, education lectures; double-headed arrow line, 2×/week yoga classes).

Study settings and eligibility criteria

Participants will be recruited from the community (local media and social media) and from waiting lists at physiotherapy outpatient clinics in Belo Horizonte, Brazil. Individuals interested in participating in the study will be screened for eligibility and assessed by a research assistant. The intervention and control interventions will be delivered at the participating outpatient clinics on separate days to avoid cross-group contamination. Participants will be included if they meet the following criteria: aged between 18 and 60 years, experiencing chronic non-specific LBP (ie, pain between the lower ribs and the gluteal folds lasting more than 3 months, with or without leg pain), reporting an average pain intensity of 3 points or more on the Numeric Pain Rating Scale (NPRS) over the past week and classified as high risk of poor prognosis (ie, scoring 50 points or above) on the OMPQ short-form. Exclusion criteria include diagnosed serious spinal pathology (ie, tumours, fractures and inflammatory disease), prior spinal surgery, pregnancy or childbirth within the past 3 months, medical conditions preventing physical activity (eg, severe cardiovascular, renal or neurological diseases) or musculoskeletal conditions limiting participation in a yoga programme. We will also exclude individuals with radicular pain (ie, defined as pain radiating down below the knee following a dermatomal distribution) and radiculopathy (ie, clinically defined as a positive straight leg raise test or the presence of at least two concordant neurological signs: weakness, reflex changes or sensation loss in the same spinal nerve distribution). Patients with suspicion of radiculopathy will be assessed in person by a trained researcher in a separate consultation room. Participants will also be excluded if they are currently receiving other physical activity-based interventions (eg, Pilates, physical therapy, strength training and aquatic exercises) for their LBP or have insufficient Portuguese literacy to complete study questionnaires. Participants will be advised to follow their allocated intervention. We will not restrict or control any concurrent treatments or physical activity undertaken after randomisation in either group for the duration of the trial.

Randomisation

Trained research assistants will explain the study’s objectives, procedures, risks and benefits to all potential participants. Participants will be given adequate time to ask questions before being asked to provide written informed consent by signing the consent form. This process will be documented for each participant. Eligible participants will provide informed consent and undergo baseline data collection. Participants will be randomly assigned either to the control group or the intervention group. Block randomisation with variable block sizes of 2, 4 and 6 will be used, with the sequence generated by a computer. An external investigator not involved in data collection will generate the randomisation sequence and prepare sequentially numbered, opaque and sealed envelopes for allocation. The envelopes will be opened by the research assistant after baseline assessment to assign participants to their groups. Due to the nature of the intervention, participants and therapists cannot be blinded. As all outcomes are self-reported by participants, outcome assessors are similarly not blinded to group allocation.

Control group

Participants in the control group will receive an educational booklet and attend three monthly face-to-face group lectures, each designed to stand alone so that no specific sequence is required. The first lecture introduces the structure and function of the spine and general aspects of LBP, discusses how to interpret imaging findings, distinguishes types of LBP, addresses fear of movement and emphasises optimism and patient empowerment in pain management. The second lecture focuses on pain education, covering the role of pain, how the pain system operates, factors that modulate pain, common myths and simple strategies for short-term relief. The third lecture examines factors that influence pain more broadly, encourages changes in physical and mental habits, outlines the benefits of exercise and provides practical guidance on initiating physical activity, concluding with an integrated synthesis of the key concepts covered across the programme.

Yoga intervention group

In addition to the booklet and monthly lectures, participants will attend group yoga sessions twice a week for 12 weeks, totalling 24 yoga sessions. Each session will last 1 hour. The yoga intervention tested in this trial was adapted to incorporate principles of gradual activity through the selection and progression of postures, duration of practice and verbal guidance, reinforcing the concept that pain during movement does not necessarily indicate tissue damage. A feasibility study with eight participants who met the eligibility criteria for this trial was conducted to assess the acceptability of the programme. In this study, the programme was offered twice a week for 6 weeks, with exercises and postures adapted and modified to meet the participants’ needs. Based on the findings of this feasibility study, we developed a manual for clinicians to ensure consistency in yoga prescription. The clinicians’ manual is available in online supplemental material 2. The main components of yoga sessions include movement-based exercises, breathing techniques, relaxation and meditation. The exercises gradually increase in complexity throughout the 12-week intervention. Sessions will be conducted by three certified yoga trainers, two of whom are also physical therapists and one a nurse. Participants will be asked to be barefoot and wear comfortable clothing. Yoga mats will be used for ground exercises, and pillows may be provided to assist with the postures. Participants will be allowed to continue any pain medications previously prescribed. Table 1 details the intervention following the Template for Intervention Description and Replication. Participants may discontinue the intervention in cases of adverse events, persistent pain flare-ups or personal request.

Table 1. Template for Intervention Description and Replication (TIDieR) checklist: important points of the intervention.

1. Brief name	Yoga programme for individuals with chronic LBP and high risk of poor prognosis
2. Why	Chronic LBP has been described as a recurrent condition that affects quality of life and involves high public health costs. Current clinical guidelines highlight the importance of performing physical activity as the first line of treatment and have also suggested a multidisciplinary approach to encompass the biopsychosocial aspects that affect this condition. Yoga, an integrative practice, is a mind–body therapy that could thus address both the physical and biopsychosocial aspects. Through the integration of movements, breathing techniques and meditations with the purpose of acting on flexibility and strength, it enables more confidence in moving and thus becoming more active, in addition to being able to promote relaxation, stress relief and reduce sensitivity to pain. This intervention might be even more effective if targeting a more homogeneous group with chronic symptoms and a high risk of poor prognosis, in which biopsychosocial factors are already interfering with their health condition.
3. What: materials	An educational booklet with guidance on back pain and self-management of LBP will be given to the control and intervention groups. Mats and pillows will be used to practise yoga.
4. What: procedures	The two groups will participate in face-to-face lectures, with guidance on back pain and strategies for pain self-management. The intervention group, in addition to educational meetings, will receive a yoga programme with 24 sessions.
5. Who provided	The professionals who will offer the yoga programme have been trained in yoga and participated in a feasibility study to adapt the postures and exercises for individuals with LBP.
6. How	Face-to-face lectures and yoga practices will be held in groups, each lasting approximately 1 hour. Each yoga session will be organised to include four components: postures and exercises, breathing techniques, relaxation strategies and meditation.
7. Where	The yoga programme will be offered in physiotherapy clinics and health centres.
8. When and how much	A total of 3 face-to-face lectures and 24 yoga sessions will be offered. Face-to-face lectures will last approximately 1 hour. Yoga sessions will be offered twice a week for 12 weeks. Each session will be 60 min long.
9. Tailoring	Practitioners will be guided to respect their limits during the session. If participants report too much discomfort while maintaining the postures, they will be instructed to stop, and adaptations will be provided by the teachers.
10. Modifications	Modifications in yoga postures and exercises will be provided by teachers to meet the capabilities and individualities of each practitioner.

LBP, low back pain.

Data collection

For this trial, only participants classified as having a high risk of poor prognosis will be recruited. Risk assessment will be conducted using the OMPQ short version as part of the eligibility criteria.²² This questionnaire evaluates the risk of persistent pain due to the presence of psychosocial factors through 10 questions, each scored from 0 to 10. The total score ranges from 0 to 100 points, with scores above 50 indicating a high risk of poor prognosis.²²

At baseline, we will collect information on demographic data (age, weight, height and sex), comorbidities (eg, diabetes, hypertension), education level (primary, secondary or higher), symptom duration, ongoing treatments and current medications.

The primary outcome is disability assessed at 12 weeks, measured using the Roland-Morris Disability Questionnaire (RMDQ), which evaluates activity limitations due to LBP. The RMDQ consists of 24 yes/no questions, with higher scores indicating greater disability.

Secondary outcomes include disability assessed at 24 weeks and pain intensity, quality of life, depressive symptoms, pain self-efficacy and impression of recovery measured at 12 weeks (ie, postintervention) and 24 weeks. Average pain intensity over the past week will be assessed using the NPRS, which ranges from 0 to 10, with higher scores indicating greater pain intensity. Quality of life will be assessed using the EuroQol-Visual Analogue Scale, which records the participants’ self-rated health on a vertical visual analogue scale. The endpoints are 0 (‘the worst health you can imagine’) and 100 (‘the best health you can imagine’).²⁶ Depressive symptoms will be assessed using the Center for Epidemiological Studies-Depression (CES-D) scale, which evaluates the frequency of depressive symptoms over the past week. The CES-D consists of 20 questions, each rated on a 4-point Likert scale, with a total score ranging from 0 to 60 points. A score above 16 indicates the presence of depressive symptoms.²⁷ Data monitoring will be performed every 4 months by an independent researcher not involved in data collection. The Pain Self-Efficacy Questionnaire (PSEQ) will be used to assess participants’ confidence in performing various activities despite pain.²⁸ Pain self-efficacy refers to an individual’s belief in their ability to successfully perform tasks or behaviours to achieve a desired outcome. PSEQ consists of ten items, each scored from 0 to 6, with a total score ranging from 0 (‘very low confidence’) to 60 (‘very high confidence’). The global perceived effect scale (GPES) assesses the patients’ overall impression of recovery, comparing their condition at the beginning of the trial to their most recent health status. Participants will be asked, ‘compared with the start of the treatment, how would you describe your back pain nowadays?’.²⁹ GPES is an 11-point numerical scale, ranging from −5 (vastly worse) to +5 (completely recovered).

We will monitor adherence to yoga classes in the intervention group using an attendance spreadsheet, recording the frequency of session attendance. Participants missing two consecutive sessions will be contacted to re-engage. We will report mean attendance to the yoga classes and the proportion of participants attending 100% and 70% of the sessions. Adverse events will be tracked at the end of the study using two questions: ‘Have you experienced any medical event or exacerbation of back pain-related symptoms in the past 3 months following baseline evaluation (eg, dizziness, syncope or symptom worsening)?’. If participants answer yes, they will need to provide further details of the incident. Adverse events will be classified as serious or non-serious, depending on whether the incident requires hospitalisation and is associated with the trial interventions. No interim analyses or stopping rules are planned due to the short intervention period and low-risk nature of the study. Data will be collected and entered into Research Electronic Data Capture by outcome assessors. To ensure data quality, a second researcher will verify a random 10% of the entered data against the original forms. Range checks for data values will be built in the database to minimise data entry errors. The database will be stored on a secure university server and backed up regularly. Participants will be contacted via phone up to five times to minimise loss to follow-up. Figure 1 lists the outcomes to be collected and the time-point of assessments.

Sample size

The sample size for this trial was calculated based on a previous study.³⁰ A total of 110 participants (55 per group) will be required to detect a between-group difference of 2.5 points on the RMDQ, assuming a SD of 4.3 points, 80% statistical power and 0.05 significance level, accounting for a 15% dropout rate.

Patient and public involvement

There was no formal patient or public involvement in the study’s conception and design. Patients provided inputs to refine the yoga intervention during the feasibility study before the start of the trial.

Data analysis plan

Data will be analysed by a statistician blinded to group allocation. Descriptive analysis will be conducted to report baseline data. Data normality will be verified through the Kolmogorov-Smirnov test. Normally distributed data will be reported as a mean (SD), while data with non-normally distributed data will be reported as a median (IQR). Dichotomous and categorical variables will be reported as frequency (percentage). Flow of participants throughout the study will be reported using the template in figure 2.

Figure 2. Flow of participants during the study.

The effects on primary and secondary outcomes at 12- and 24-week postrandomisation will be calculated using fixed linear models, adjusted for baseline outcome. Multiple imputation techniques will be used in case of missing data.

Statistical analysis will be conducted following the intention-to-treat principle, which means that participants will be analysed in the group they are initially randomised to, regardless of whether they received the assigned intervention. Primary and secondary outcomes will be analysed using separate linear regression models for analysis of covariance. In these models, the groups will be included as independent variables, the outcome score at the time-point of assessment (ie, postintervention or at 24 weeks) will be included as a dependent variable, and baseline outcomes will be included as covariates. Between-group comparisons will be reported through the mean difference and their respective 95% CIs. All analyses will be performed using the IBM SPSS software V.20.0 (IBM Corporation, Somers, NY, USA).

Ethics and dissemination

The protocol for this study has been approved by the ethics committee of the Universidade Federal de Minas Gerais and the Institute of Public Servants Welfare of the Minas Gerais State (Protocol number CAAE: 57028022.0.0000.5149). All collected data will be de-identified to protect participant confidentiality. Personal information will be stored separately from research data in a secure password-protected database accessible only to key research personnel. The results of the trial will be disseminated to trial participants, clinicians and the broader public and scientific community.

Discussion

Globally, there has been a growing interest in complementary therapeutic methods, such as yoga for back pain, leading to an increase in scientific research in this field. Healthcare systems in upper-income countries (eg, the United Kingdom)³¹ and the USA and middle-income countries (eg, Brazil)³² have recognised their potential benefits. In Brazil, yoga has been incorporated into the Brazilian National Health System as part of the National Policy for Integrative and Complementary Practices, established in 2006.³² However, before policy makers can consider implementation of yoga in the healthcare system, it is essential to first determine whether there is a specific subset of patients with chronic LBP that show large clinical improvements. If the results in our efficacy trial are promising, we expect future studies to test the effectiveness of yoga in more pragmatic trials.

This interest is supported by evidence from a systematic review that has shown favourable results for yoga when compared with no exercise.²¹ However, the effect of treatment is small and based on low-certainty evidence. Our trial aims to address limitations from previous trials by focusing on a more homogeneous group of patients who are more likely to respond to a body–mind intervention. We will use OMPQ to screen and select patients at higher risk of poor prognosis, often characterised by psychosocial factors, such as depressive symptoms, fear of movement and stress. Although OMPQ has been developed as a multicomponent intervention, yoga has the potential to address both physical and psychosocial factors.

The results of this trial will be disseminated regardless of the outcome. We plan to submit the findings for publication in a peer-reviewed journal. A lay summary of the results will also be made available to trial participants and to the public.

Any important modifications to this protocol will be submitted as a formal amendment to the approving ethics committee (Universidade Federal de Minas Gerais, CAAE: 57028022.0.0000.5149) and updated in the ClinicalTrials.gov registry (NCT05953155).

Study limitations

The findings will be generalisable to adults with back pain aged 18–60 years. Consequently, results should not be extrapolated to adolescents or older adults without caution and replication. Because of the nature of the intervention, blinding of participants and treating clinicians is not feasible. All primary and secondary outcomes are patient-reported. As such, independent blinded outcome assessment is not possible and detection bias cannot be completely excluded.

Conclusion

Yoga is a mind–body intervention that addresses physical as well as psychological factors; it may be more effective for patients with chronic LBP who are at high risk of poor prognosis. The results of this efficacy trial should be seen as the initial step in the implementation pathway. If yoga shows promising results, future effectiveness trials are needed before policymakers decide whether yoga should be offered to patients with chronic LBP.