Implementing an integrated psychological and physical intervention in routine physiotherapy practice for patients with musculoskeletal road traffic injury: protocol for a hybrid implementation-effectiveness type III cluster randomised controlled trial

Abstract

Background

Up to 50% of individuals with musculoskeletal road traffic injury (RTI) develop chronic pain, resulting in substantial individual and societal burden. Integrated psychological and physical care, such as StressModex, improves patient outcomes compared to physical treatment alone. However, StressModex is not routinely implemented in physiotherapy practice, due to limited training access and physiotherapists’ lack of confidence in delivering psychological care. To address this gap, we developed a blended learning implementation strategy—Physiotherapist bIopsyChosocial On-line Training (PICOT)—guided by the integrated-Promoting Action on Research Implementation in Health Services (i-PARIHS) framework. The aims of this trial are to compare: (1) the effectiveness of PICOT versus in-person training on the reach of StressModex in routine community private physiotherapy practice; (2) the effectiveness of PICOT versus in-person training on adoption, implementation fidelity, sustainability, and maintenance of StressModex; (3) the effectiveness of PICOT versus in-person training on patient health outcomes; and (4) the cost-effectiveness of PICOT versus in-person training. Trial outcomes are informed by the RE-AIM framework.

Methods

This is a hybrid type III implementation-effectiveness, cluster randomised, superiority trial with embedded economic and qualitative process evaluations. Thirty primary care physiotherapy clinics across Australia will be randomly assigned to either the PICOT or traditional 2-day in-person training. PICOT includes a 6-week online program, 6 weeks (once/week) of real-time online group training with individualised feedback, then 3 clinical supervision on-line sessions (once per fortnight). All on-line sessions are co-facilitated by a clinical psychologist and expert physiotherapist. Following training, physiotherapists will deliver StressModex to eligible patients (≥ 18 years, ≤ 12 weeks of musculoskeletal spinal pain post RTI, and at risk of poor recovery). The primary implementation outcome is reach, defined as the proportion of eligible patients treated with StressModex over 8 months. Secondary outcomes include adoption (training participation and initial uptake), implementation (dose, fidelity, and sustainability of delivery), patient health outcomes (collected at Time1, 8 weeks, 6-, and 12 months), and cost-effectiveness.

Discussion

This trial will provide critical evidence on scalable training models for embedding integrated psychological and physical care into physiotherapy practice. Findings will inform strategies to improve the implementation and sustainment of evidence-based interventions for musculoskeletal RTIs.

Trial registration

ACTRN12624001268538. Registered on 18 October 2024. https://www.anzctr.org.au/Trial/Registration/TrialReview.aspxid=388006&showOriginal=true&isReview=true

Contribution to the literature.

• Although integrated psychological and physical interventions are effective for musculoskeletal pain, their uptake in physiotherapy practice is limited—due to lack of training and clinician confidence. Research on scalable training strategies to support implementation is scarce.

• Guided by the i-PARIHS framework, we developed a blended learning training strategy (PICOT: Physiotherapist biopsychosocial On-line Training), incorporating online modules, facilitated sessions, and practice feedback to build physiotherapists’ capability to deliver integrated care.

• This study is a pragmatic, multi-site, Type III hybrid implementation-effectiveness trial comparing PICOT to traditional in-person training for implementation of StressModex, a physiotherapist-delivered integrated psychological and physical intervention for patients with musculoskeletal road traffic injuries at risk of poor recovery.

• The trial will generate critical evidence on the effectiveness of an implementation strategy to support reach of StressModex by physiotherapists to patients, with implications for future training models, clinical practice, and health policy.

Background

Road traffic injuries (RTIs) affect 20 to 50 million people globally [1], with 75% being non-catastrophic musculoskeletal injuries, such as whiplash affecting the neck and back [2]. These injuries result in substantial economic and personal burden in many countries [2–4]. Up to 50% of people with acute musculoskeletal RTI do not recover well but report chronic pain, associated disability and posttraumatic stress symptoms, anxiety and depression [5, 6].

Clinical guidelines for acute and subacute musculoskeletal RTIs recommend advice to return to usual activity and exercise as first-line treatment [7], but randomised controlled trials (RCTs) show that these treatments have only small effects compared to advice only [8, 9]. Our 2019 RCT found greater effect sizes when patients at risk of poor recovery were identified and treatment targeting psychological distress was provided [10]. In this earlier trial, physiotherapists were trained by psychologists to deliver Stressmodex, a 6-week integrated intervention that targeted stress symptoms (stress inoculation training) and integrated with guideline-recommended exercise. The stress inoculation component included six modules: 1) education about the effects of stress on pain and recovery, 2) relaxation strategies including relaxed deep breathing and progressive muscle relaxation, 3) problem-solving techniques, 4) positive coping self-statements, 5) practice using these skills in stressful situations, and 6) relapse prevention and planning. The exercise component was individually prescribed by the physiotherapist based on the presentation of each patient. It could include range of movement, strengthening and sensori-motor control exercise as deemed appropriate. The integrated treatment (StressModex) resulted in clinically relevant improvements in pain related disability compared to exercise alone immediately post treatment (SMD: −10; 95% confidence interval (CI):−15.5 to −0.9 points on 0–100 Neck Disability Index); at 6 months (− 7.8; − 13.8 to − 1.8) and at 12 months (− 10.1; − 16.3 to − 4.0) post randomisation [10]. Table 1 outlines the treatments delivered in the SressModex treatment arm of our previous trial.

Table 1.

Outline of the StressModex intervention [10]

Week	StressModex
1	Introduction to stress inoculation training, theories of stress and pain, abdominal breathing exercises. Physiotherapy exercise.
2	Muscle relaxation training (body scan exercise). Physiotherapy exercise.
3	Problem solving for stressful situations. Physiotherapy exercise.
4	Use of positive coping statements. Physiotherapy exercise.
5	Applying stress inoculation training to the real world. Physiotherapy exercise.
6	Coping skills maintenance. Physiotherapy exercise.

Physiotherapists integrating psychological care with physical therapy is not new and systematic reviews have found consistent moderate to high quality evidence that physiotherapist-delivered integrated psychological and physical care is effective compared to education and/or exercise alone and also cost-effective for the management of musculoskeletal pain [11, 12]. The reviews conclude that further effectiveness trials are not needed. Rather, the priority is to determine the best ways to implement models of care that integrate psychological and physical care in routine physiotherapy practice [12]. However, there are barriers to implementation. Physiotherapists receive only limited education and training in delivering psychological care [13] and consistently report a lack of skills and confidence to incorporate psychological strategies into their practice [14]. Despite this, physiotherapists report a need for more training in the delivery of psychological care for patients with musculoskeletal pain and distress [15]. Training is essential to support effective translation of these treatments into current clinical care.

RCTs of physiotherapist-delivered integrated psychological and physical therapy usually train physiotherapists in-person over various time-periods. In our StressModex RCT, in-person training was over two days [10]. In-person training is difficult to scale, limiting widespread implementation and sustainable change in clinical practice. To address these barriers and guided by the integrated-Promoting Action on Research Implementation in Health Services (i-PARIHS) framework, we have developed a novel blended learning strategy (PICOT: Physiotherapist bIopsyChosocial On-line Training) to facilitate rapid and widespread effective implementation of StressModex in community physiotherapy practice. Blended learning models that include e-learning, face-to-face training and support can overcome geographic limitations and provide greater access to training, especially for clinicians in regional and remote areas [16] and have been shown to be superior to in-person workshops for educating and training health professionals [16, 17].

This paper describes the protocol for a hybrid type III effectiveness-implementation trial by comparing outcomes of two training strategies to promote implementation of the evidence-based StressModex intervention in routine community physiotherapy practice.

Trial Aims

The primary aim of this trial is to compare the effectiveness of two implementation training strategies (PICOT vs. in-person training) on the reach of StressModex. Reach in this context is defined as the proportion of eligible patients with musculoskeletal RTI at risk of poor outcome who are treated with StressModex by participating physiotherapists.

The secondary aims are as follows:

1. To compare the two implementation training strategies on the dose and fidelity of StressModex delivery by physiotherapists, and the sustainability of StressModex provision over time

2. To compare the two implementation training strategies on physiotherapist training outcomes, including commencement of, and attendance at, training sessions, fidelity of training delivery, and sustainability of training.

3. To compare the effectiveness of the two implementation training strategies on patient health outcomes.

4. To compare the cost-effectiveness of the two implementation training strategies.

5. To explore determinants influencing the delivery of StressModex by physiotherapists.

Hypothesis

It is hypothesised that reach—the proportion of eligible patients with musculoskeletal RTI and at risk of poor outcome who are treated with StressModex—will be greater among physiotherapists who are randomised to the blended learning training (PICOT) compared to those who are randomised to traditional in-person training.

Methods

Study design

This is a type III hybrid implementation-effectiveness cluster randomised, multisite superiority trial with two parallel groups (Fig. 1). Economic evaluation and process evaluation are embedded in the trial. The trial results will be reported in accordance with the Consolidated Standards of Reporting Trials (CONSORT): extension to cluster randomised controlled trials [18] and the Standards for Reporting Implementation Studies (StaRI) Statement [19]. The trial has been approved by The University of Queensland Human Research Ethics Committee A (2024/HE000426) and registered prospectively on the Australian and New Zealand Clinical Trials Registry (ACTRN12624001268538).

Fig. 1.

Trial Design

Setting

The units of randomisation for this trial are community primary care physiotherapy clinics in metropolitan and regional areas across Australia.

Participant groups

Primary care physiotherapy clinics

Primary care physiotherapy clinics are eligible if they employ at least three physiotherapists, see ≥ 12 patients/year with musculoskeletal RTI, and consent for their clinic to be randomised in the trial. Eligible clinics must have at least one physiotherapist willing to participate, but all physiotherapists from the clinic can participate.

Physiotherapy clinics will be recruited from the research team’s network of physiotherapists and through social media (e.g., X, LinkedIn). If necessary, we will recruit additional physiotherapy clinics through advertising in the Australian Physiotherapy Association newsletter. Interested clinics will complete an on-line and secure Excel spreadsheet with details for eligibility screening, including their name, address, monthly caseload of patients with musculoskeletal RTIs, and number of physiotherapists employed.

Each physiotherapy clinic will nominate a trial delegate and a champion physiotherapist. The clinic trial delegate will liaise with the research team regarding administrative aspects of the trial. The champion physiotherapist will be a senior physiotherapist who will champion the trial at their clinic site.

Each participating physiotherapy clinic will receive a maximum $8,000 over the course of the trial to compensate costs associated with participation in the trial including administrative costs, training time, screening and consenting of patient participants. Administration of funds will be aligned with key project milestones (e.g., training attendance).

Patients

Consecutive patients meeting eligibility criteria at participating physiotherapy clinics will be invited by their treating physiotherapist to participate in the trial study data collection. Patients will be eligible if they have musculoskeletal spinal (neck and/or back) pain of ≤ 12 weeks duration following RTI, screened at medium to high risk of poor recovery using a validated tool appropriate for the clinical presentation (e.g., WhipPredict [20] or the Örebro Musculoskeletal Pain Screening Questionnaire [21], are aged 18 years or older, and are proficient in written and spoken English. Patients will be excluded if they have, or are suspected to have, a spinal fracture or other serious or suspected serious spinal pathology (e.g., spinal infection, cauda equina, progressive neurological signs/symptoms warranting urgent medical attention, cervical myelopathy, cancer-related spinal pain), or diagnosed systematic inflammatory arthritis (e.g., rheumatoid arthritis) or neurological conditions (e.g., brain injury, multiple sclerosis, stroke).

Recruitment of patients will commence after completion of physiotherapist training (6-week online or 2-day in-person). Physiotherapists will identify eligible patients during initial consultations and conduct eligibility screening as part of their routine assessments. If patients are eligible, physiotherapists will provide information about the trial, a participant information sheet, and obtain informed consent for data collection, including self-reported questionnaires, access to physiotherapy clinical records, and video recording of their physiotherapy sessions. Patient participants in both arms of the trial will be provided the same information, consent forms and questionnaires. They will not be informed of the specific aims of the trial, nor the group their physiotherapy clinic was assigned to, but that we are interested in health outcomes after road traffic injury and that their physiotherapist has undergone some training.

The following processes will be undertaken to mitigate potential selection bias: 1) the physiotherapists will be trained in and provided with a clear checklist of patient eligibility criteria, 2) consecutive patients will be screened for eligibility, 3) each clinic will maintain a screening log of potentially eligible patients attending the clinic including reasons for inclusion or exclusion, 4) These logs will be audited by the trial research staff every month, to monitor adherence and identify any patterns suggestive of selection bias, and 5) Recruitment data (e.g., demographics, injury type, risk-stratification subgroup, pain intensity) will be reviewed monthly to ensure balance across arms and sites. Feedback and mitigation activities with sites will be employed if deemed necessary.

Implementation framework

PICOT was explicitly designed using the i-PARIHS framework which identifies facilitation as the active ingredient essential for successful implementation. i-PARIHS conceptualises implementation success as a function of the characteristics of the Innovation (StressModex), the intended Recipients (community physiotherapists), and the Context (local clinic and broader healthcare system). Within the Innovation domain, the evidence strength, complexity, and compatibility of StressModex with existing physiotherapy practices were key considerations in designing PICOT. In the Recipients domain, PICOT targets physiotherapists’ motivation, values, goals, knowledge, and self-efficacy related to delivering integrated psychological and physical care. The Context domain accounts for organisational readiness, leadership support, and resource availability across participating clinics.

In this trial, external facilitation will be provided through an interactive experiential learning model, combining online modules, real-time virtual workshops, coaching, and practice feedback to address individual and organisational barriers to adopting StressModex. Facilitation strategies will include educational materials, case studies, video demonstrations, feedback and self-monitoring tools, and peer discussion forums. Local facilitation will be embedded by appointing senior physiotherapists at each clinic as local champions to support peer engagement, encourage practice change, and sustain implementation efforts beyond the training period. Each component of PICOT is mapped to the core constructs of the i-PARIHS framework (see Table 2).

Table 2.

Relationship between i-PARIHS Constructs and the Research Plan and PICOT blended learning Implementation Strategy

i-PARIHS Construct		Relationship to research plan/implementation strategies
Innovation	Underlying knowledge	Evidence supporting the delivery of integrated psychological and physical interventions by physiotherapists for people with musculoskeletal pain including after RTI is strong [11, 12] Blended learning models that include e-learning and face-to-face training and support have been shown to be superior to in-person training of health care providers [16, 17]
	Compatibility (degree of fit with existing practices and values)	The StressModex intervention integrated psychological strategies with guideline recommended physiotherapy care which is usual practice for physiotherapy. No additional time/clinical session was required to deliver StressModex over and above usual clinical session times1. Physiotherapists report a lack of confidence and skills in delivering psychological care with limited training opportunities available [13]. They value being upskilled in the delivery of psychological strategies [15].
	Accessibility and usability	We have piloted the PICOT strategy with physiotherapists locally in Brisbane, nationally and internationally. Results show an increase in physiotherapists confidence and skills.
	Relative advantage over current practice.	Current practice in postnominal physiotherapist training is in-person. The PICOT strategy has advantages in terms of reach and accessibility of physiotherapist training.
	Observable results	We will measure a range of implementation outcomes.
Recipients	Motivation, values, and beliefs	Physiotherapists express a desire to improve their skills in delivering psychological care to ensure their patients have the best outcomes [15].
	Skills and knowledge	In our piloting of the PICOT strategy, we have demonstrated that physiotherapists’ skills and knowledge can be improved.
	Time, resources, support	Some time and resources will be required to be allocated by the participating physiotherapist clinics and physiotherapy champions.
	Local opinion leaders, Policy Maker	Physiotherapist champions will be appointed at each clinic. These will be senior physiotherapist opinion leaders. The training of physiotherapists will be facilitated by psychology and physiotherapy opinion leaders (RE, MS, DB, TA). The trial has the support of policy-making organisations including government regulators, road traffic injury insurers, peak professional and consumer organisations.
	Existing networks	The research team have existing relationships with the clinics involved in the trial.
Context	Local level	Physiotherapist champions provide local leadership and support. There is sufficient time to deliver StressModex in practice. Evaluation and feedback provided to physiotherapists on their delivery of StressModex during training
	Organisational level	Physiotherapist clinics are supportive of PICOT. The over-arching organisation that many of the participating clinics belong is supportive of their involvement. The over-arching organisation as a history of innovation and change and has a network of learning opportunities.
	External health system level	Policy makers and treatment funds (insurers) are partners in the trial and are supportive of PICOT. PICOT fits within their regulatory frameworks. The peak professional organisation (Australian Physiotherapy Association) is a partner in the trial.

PICOT: Physiotherapists bIopsyChosocial On-line Training; i-PARIHS: Integrated-Promoting Action on Research Implementation in Health Services

Implementation strategies

The implementation strategies are the two training strategies: blended online and clinician-supported learning (PICOT) versus traditional in-person training. PICOT will be delivered over 6-weeks, while in-person training will be delivered over 2 days (as per training model adopted in the Stressmodex RCT [10]), scheduled toward the latter period of the PICOT period to ensure similar time between training and implementation in practice for both groups of physiotherapists. The total time involvement of each implementation strategy is approximately 12 h per physiotherapist. Both groups of physiotherapists will be trained by an experienced clinical psychologist and expert musculoskeletal or pain physiotherapist skilled in biopsychosocial management of patients with musculoskeletal pain and who have teaching experience using various learning methods. Both implementation strategies will include discussion about scope of physiotherapist practice including when to refer patients to a psychologist if indicated.

The PICOT implementation strategy

PICOT is a blended learning approach that includes

1) online self-paced modules, 2) real time online training sessions co-facilitated by a clinical psychologist and expert physiotherapist, and 3) practice feedback. Blended learning, a combination of traditional face-to-face learning and asynchronous or synchronous online learning [22] has shown to have a consistent positive effect for health professional learners compared to traditional learning approaches (e.g., face-to-face seminars) [23, 24].

Online self-paced modules

Physiotherapists will complete 6 online self-paced modules weekly over 6 weeks, with each module taking 20–30 min to complete. Each module will be released 7 days prior to the on-line training sessions, with physiotherapists asked to complete them during the week prior to the on-line training sessions. The modules reflect the content of the original StressModex treatment manual (see supplementary materials of the original trial) [10] and include: (1) pain education and the influence of stress on pain, (2) stress-management strategies, e.g., relaxed breathing, muscle relaxation, (3) problem solving skills, (4) managing unhelpful thoughts, (5) generalisation of skills, and (6) relapse prevention. Several aspects of the original treatment manual were further developed and refined (e.g., considerations when referring to a psychologist, cognitive-behavioural links between thoughts, feelings and behaviours as part of pain education).

All modules are integrated with guideline recommended exercise [25]. The modules will be presented using engaging strategies that includes quizzes, videos of experts delivering the intervention, videos of patient and physiotherapist describing their experiences with StressModex, Q&A videos, and downloadable written resources.

Real time online training sessions

Weekly group online training sessions (maximum duration 1 h) will be co-facilitated by the clinical psychologist and expert musculoskeletal or pain physiotherapist. These sessions are designed to provide physiotherapists with a formal process of clinical support, reflection, and learning that aims to enhance their skills, capabilities, and confidence in delivering StressModex in practice. Sessions will include skills practice using role plays, case studies, group discussion and reflection.

Individual practice reflection and feedback

After each training session, physiotherapists will be asked to demonstrate one of the cognitive behavioural skills (e.g., problem solving) and how they would integrate it into physiotherapy care. Physiotherapists will videotape a randomly selected practice session with a colleague and provide written reflection on their practice. A psychologist and expert musculoskeletal or pain physiotherapist will review the videotapes and provide written feedback to the physiotherapists.

Real time online clinical support sessions

Following the 6-week program, online facilitated clinical support sessions (each 1 h) will be held offered fortnightly for a further 6 weeks with the same facilitators to support and enhance physiotherapist skills and confidence as they implement StressModex into practice.

In-person training implementation strategy

Physiotherapists in the clinics randomised to usual training will be offered 2 days of in-person training in StressModex, led by the same facilitators of the online training (an experienced clinical psychologist and expert musculoskeletal or pain physiotherapist), as per our original StressModex trial [10] and is consistent with other RCTs of physiotherapist-delivered integrated physical and psychological treatments [12]. The same 6 modules will be covered during the 2-day workshop, using the StressModex treatment manual [10]. Small group skills role plays of all modules, reflection and discussion will be included. The 2-day training workshops will be conducted in Brisbane, Australia.

Trial outcomes and assessments of effectiveness

The selection of primary and secondary outcomes is guided by the RE-AIM framework (Reach, Effectiveness, Adoption, Implementation, Maintenance) [26]. The primary outcome is reach, defined as the proportion of eligible patients treated with StressModex over 8 months.

Secondary outcomes cover the remaining RE-AIM domains: effectiveness (patient health outcomes), adoption (physiotherapist engagement with training), implementation (dose, fidelity, and sustainability of StressModex delivery), maintenance (ongoing delivery of StressModex and sustained training practices), and cost-effectiveness (economic evaluation of the two training strategies). A mapping of the trial aims to the RE-AIM domains is presented in Table 3.

Table 3.

Data collection framework for RE-AIM outcomes

RE-AIM Domain	Outcome Measure	Data source	Collection Timepoints
Reach	Proportion of eligible patients treated with StressModex (≥4 of 6 sessions within 8 weeks).	Blinded audit of anonymised physiotherapy clinical records.	Over the 8-month period following completion of training.
Effectiveness	Patient-reported health outcomes: pain, function, quality of life, psychological symptoms, therapeutic alliance.	Validated questionnaires (BPI, EQ-5D-5L, DASS-21, PCL-5, Pain Self-Efficacy, Session Rating Scale).	Baseline, 8-weeks, 6- and 12- months post baseline
Adoption	(1) % of physiotherapists who attempt StressModex delivery. (2) Physiotherapist intention to use StressModex. (3) % of physiotherapists who complete training. For PICOT: proportion of physiotherapist-participants who complete all on-line modules/provide audiotapes/attend all zoom group sessions; proportion of clinics with a champion. (4) Number of training sessions attended	(1) Physiotherapy record audit (blinded) (2) A single-item questionnaire adapted from the Implementation Evaluation Questions52: ‘I intend to use the StressModex program when appropriate to treat my patients with musculoskeletal road traffic injury’ (“1” = “strongly disagree” to “5”= “strongly agree”). (3–4) Study specific log	(1) Over the 8-month period following training. (2) Immediately after completion of training period. (3–4) Immediately after completion of training period.
Implementation	(1) Dose: Number of StressModex sessions delivered (2) Fidelity of delivery:  (i) Adherence to protocol  (ii) Physiotherapist confidence  (iii) Physiotherapist competence (3) Fidelity of training delivery.	Physiotherapy record audit (blinded) 2(i). Record audit and 50% of video recordings rated against checklist (blinded). 2(ii–iii). Self-rated questionnaires; expert-rated Therapist Quality Scale (TQS) for 50% of videos. Study specific training log	Over the 8-month period following training. Over the 8-month period following training. During training delivery
Maintenance	(1) Proportion of patients treated with StressModex, dose and fidelity of StressModex provided to patients. (2) Training sustainability	Physiotherapy record audit (blinded) Percentage of physiotherapists who are interested in training additional physiotherapists	8- to12- month period following the completion of training. 12 months following completion of training.

To complement the quantitative outcomes, qualitative interviews will explore physiotherapists’ perceptions of barriers and enablers to implementation. This qualitative component is designed to enrich interpretation of RE-AIM outcomes, particularly those relating to context and maintenance.

Primary implementation outcome

The primary implementation outcome is reach, defined as the proportion of eligible patients who are treated with StressModex over the 8-months following completion of training. Reach will be measured by blinded audit of anonymised clinical records from physiotherapy sessions. Patients will be considered to have received StessModex if they receive at least 4 of the 6 StressModex sessions over 8-weeks (StressModex is a 6-week intervention, but we will allow an additional 2-weeks for potential interruptions such as illness, holidays, or rescheduling of appointments).

Secondary implementation outcomes

Secondary implementation outcomes will evaluate adoption, implementation, and maintenance. These include:

• Adoption: physiotherapist enrolment in and attendance at the training program.

• Implementation: dose and fidelity of StressModex delivery (number of sessions delivered, adherence to protocol content), as well as fidelity of training delivery.

• Maintenance: sustained use of StressModex by physiotherapists post-intervention and ongoing engagement with training content.

Operational definitions and data sources for each outcome are presented in Table 3.

Secondary outcomes – patient reported outcomes

RE-AIM measures of effectiveness will include patient reported outcomes measured at baseline (in their physiotherapy clinic after recruitment to the study and before commencing treatment), and at 8-weeks, 6- and 12-months following baseline assessment (via online questionnaires). Patient health outcomes are selected according to core outcome set recommendations for whiplash-associated disorders [27] and Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials (IMMPACT) recommendations for chronic pain clinical trials [28].

Pain-related interference measured by the 7-item pain interference subscale of the Brief Pain Inventory (BPI) [29] consisting of 7-items to assess the extent to which pain interferes with various aspects of their lives, including general activity, walking ability, work, sleep, enjoyment of life, mood, and relationships, on an 11-point numerical rating scale. The total score ranges from 0 to 70, with higher scores indicating more pain interference.

Pain severity measured by the 4-item pain severity subscale of the BPI [29]. Participants rate their current pain intensity from 0 (“no pain”) to 10 (“pain as bad as you can imagine”), as well as the average, least, and worst pain intensity in the past week. Both the single item score and a mean severity score (average of the 4 items) will be used for analysis.

Perceived recovery measured using an 11-point (−5 to + 5) global rating of change scale [27], where 0 represents no change in a patient’s overall condition since baseline, negative values denote worsening, and positive values denote improvement.

Pain self-efficacy measured using the Pain Self-Efficacy Questionnaire [30], which assesses an individual’s confidence to perform specific tasks despite the presence of pain. It consists of 10-items with responses on a 7-point Likert-type scale, ranging from 0 (not at all confident) to 6 (completely confident). The total score ranges from 0 to 60, with higher scores indicating stronger pain self-efficacy beliefs.

Depression, anxiety and stress symptoms measured by the Depression, Anxiety and Stress Scale 21 (DASS-21) [31]. Each subscale consists of 7-items with participants rating the extent to which each statement (item) applied to them in the previous week on a scale of 0 (never), 1 (sometimes), 2 (often), and 3 (almost always). The scores for each subscale are summed and then doubled. Scores range from 0 to 42 for each subscale, with higher scores indicating more higher levels of depression, anxiety and stress.

Health-related quality of life measured using the EQ-5D-5L, a multilevel preference-based measure for assessing health-related quality of life [32]. It consists of five dimensions, which include mobility, self-care, usual activities, pain/discomfort, and anxiety/depression. Each dimension has five response options that are coded from 1 to 5 on an ordinal scale. The sixth item consists of a visual analogue scale (VAS) and asks people to rate their current overall health on a scale from 0 (the worst health state they can imagine) to 100 (the best health state they can imagine).

Posttraumatic stress symptoms measured using the Posttraumatic Stress Disorder (PTSD) Checklist for the Diagnostic and Statistical Manual of Mental Disorders, fifth edition (DSM–5; PCL–5) [33]. The PCL-5 comprises 20-items that assess symptoms commonly experienced by individuals who have experienced a traumatic event. Individuals rate each item on a scale from 0 (not at all) to 4 (extremely), indicating how much they have been bothered by each symptom in the past month. Higher total scores on the PCL-5 suggest a greater likelihood of meeting criteria for PTSD.

Therapeutic alliance measured using the Session Rating Scale [34] at the end of the last patient treatment session. It provides a quantitative measure of a patient’s experience and satisfaction with a particular session within the therapy. The Session Rating Scale includes four items, each represented on a 10 cm visual analogue scale (patients rate their experience on a continuum from "not at all" to "very much"), covering the following categories: the therapist-client relationship, the relevance of therapy goals and topics, the therapist-client approach or method fit, and the overall session experience. The Session Rating Scale is scored by summing the marks made to the nearest metric with the maximum achievable score being 40-points.

Costs

The costs associated with each implementation training strategy will be estimated, including expenses related to educational materials, the time of psychologists and expert physiotherapists delivering the training, and the time commitment of clinic champions. For the in-person training arm, travel and accommodation costs, as well as the opportunity costs of physiotherapists’ time, will also be included. The costs of delivering StressModex will be calculated based on the number of treatment sessions each patient receives, as determined through clinical record audits, and will reflect the service rates reported by participating clinics.

Process evaluation

Implementation determinants will be assessed using a mixed-methods approach. Quantitative data will include physiotherapist and patient demographics, contextual factors (e.g., clinic characteristics, treatment funding models, referral patterns), and trial-related factors (e.g., randomisation, training delivery). These variables will be considered in relation to implementation success.

Physiotherapists’ attitudes and beliefs about pain will be measured using the Pain Attitudes and Beliefs Scale for Physiotherapists (PABS-PT) [35]. Confidence and competence in delivering each StressModex module will be assessed through custom questionnaires (confidence: 1 = not confident at all to 5 = extremely confident; competence: 1 = beginner to 5 = specialist level). Self-reported difficulty and perceived effort related to participation in the training strategies will also be collected (0 = not difficult to 10 = very difficult). These will be measured immediately after completion of training.

Qualitative data will be collected via two virtual focus groups (one per training arm), held on Zoom following the patient recruitment phase. Each group will include up to 10 physiotherapists, purposively sampled from those who express interest. Discussions will include both participating and champion physiotherapists and last up to 1.5 h. The interview guide will be developed using the i-PARIHS framework, which emphasises the interaction between the innovation, recipients, context, and facilitation. Questions will explore physiotherapists’ perceptions of the feasibility and acceptability of the training strategies and StressModex delivery, as well as barriers, enablers, and suggestions for adaptation to enhance future implementation. This approach ensures that key implementation constructs are captured. Qualitative findings will be reported in accordance with the Consolidated Criteria for Reporting Qualitative Research (COREQ) [36].

Blinding

Blinding physiotherapist participants and research team members involved in organising and delivering the training is not feasible due to the nature of the implementation strategies. Patients will remain blinded to the superiority hypothesis of the trial. Blinded assessors will undertake the physiotherapy clinical record audit and evaluate the videotapes of physiotherapists delivering StressModex. The statistician will also be blinded to the implementation training strategy the physiotherapy clinics are allocated to.

Randomisation

Randomisation will be conducted at the physiotherapy clinic level. A trial statistician, who is not involved in data collection or analysis, will generate the allocation sequence using a web-based randomisation service. Clinics will be stratified by state and setting (urban/regional) to ensure balance across these factors. Within each stratum, clinics will be paired based on the estimated monthly volume of musculoskeletal RTI patient encounters. One clinic from each pair will be randomised to the PICOT group and the other to the in-person training group, in a 1:1 ratio.

Sample size

The sample size is calculated based on superiority hypothesis and the primary outcome (proportion of patients treated with StressModex). Given a ~ 30% adoption rate of physiotherapy interventions following previous in-person training [37], we assume this proportion will increase to 40% with trial participation and hypothesise that PICOT will increase this proportion by an additional 20%. Without accounting for clustering, we would need data from 186 individuals to detect a 20% between-group difference with 80% power (alpha = 0.05). To account for the cluster design, we estimate we will recruit and retain an average of 17 patients per cluster and conservatively estimate an ICC = 0.10 (chosen based on previous reporting of ICCs of the order of 0.10 for process variables in primary care [35]), giving a design effect of 2.6 and a total sample size of 484 (from 30 clinics). These numbers will also provide 90% power to detect a clinically significant between-group difference in a key patient reported outcome of one unit of pain intensity (on the BPI) assuming standard deviation = 2.7, alpha = 0.05 and 10% non-compliance.

Data management and analysis

Data management

We will collect data using electronic and paper questionnaires, anonymised physiotherapy medical record audits, video recordings, and trial-specific logs. We will use the Research Electronic Data Capture (REDCap) system (Vanderbilt USA), hosted on a secure university server will be used capture all electronic questionnaire data. Patients will complete Time 1 questionnaires at the clinic as part of usual clinic practices, which the clinic delegate will scan and send to the research team for entry into the secure REDCap database. Follow-up questionnaires for patients and all questionnaires for physiotherapists will be completed online with email links sent from REDCap to participant email addresses. The clinic delegate will also scan and send physiotherapy clinical records to the research team for entry into REDCap. Data quality will be ensured through range checks, and trial data collected will be regularly monitored for omissions and errors.

Data analysis

Intention-to-treat analyses will be conducted. For the primary implementation outcome, reach, which will be analysed at the individual participant level, we will build a mixed effects logistic regression model with implementation training strategy and time (8/12 months) included as the main fixed effects, with a strategy-by-time interaction term, and stratification factors (state and setting) included as covariables. Site will be included as a random effect to account for the probable non-independence of observations from within each site. The between-strategy difference at 8 months will be presented as odds ratio with 95% CI.

Secondary implementation outcomes and patient effectiveness outcomes will be analysed using appropriate mixed-effects linear (continuous outcomes), logistic (binary outcomes) or Poisson (count outcomes) regression models as per the primary outcome. For all patient outcomes, Time 1 values will be included as model covariables. A separate Statistical Analysis Plan will be formulated and published prior to the trial database lock.

Cost-effectiveness analysis will be conducted from a health system perspective. The costs and effects of the two implementation training strategies will be compared and reported as an incremental cost effectiveness ratio. All resources utilised in implementation (e.g., training material, personnel time) will be identified, measured, and valued using market prices and industry wages. Effects will be expressed as 1) the proportion of eligible patients who receive StressModex, and 2) quality-adjusted life-years gained, calculated using the area under the curve of the utility scores derived from EQ-5D-5L over time (patient Time 1, 8-weeks, 6- and 12-months). Non-parametric bootstrap will be used to characterise the uncertainty in the results.

Quantitative and qualitative data will be synthesised using mixed methods approaches [36] to understand implementation determinants. Changes in physiotherapists pain attitudes and beliefs, self-reported confidence and competence in delivering StressModex, as well as satisfaction, difficulty and effort involved with training participation will be presented as mean (SD) for each item. Univariate correlations will explore the relationships between these determinants and reach, dose delivered and fidelity of StressModex provided by the physiotherapists to help understand key elements of the implementation strategies for future training of physiotherapists.

Audio-recordings of focus group interviews will be transcribed, and transcripts will be uploaded into qualitative software NVivo 12 (QSR International, Australia). Transcripts will be initially coded using a priori constructs consistent with our conceptual model, relevant implementation outcomes (i.e., acceptability, adoption, fidelity, feasibility), and i-PARIHS, which will be outlined in a codebook with definitions and examples. This will be supplemented by inductive coding to capture additional relevant information.

Discussion

Research has consistently shown that integrating psychological strategies as part of physiotherapy treatment is effective for patients with musculoskeletal spinal pain including after RTI [11, 12]. However, without adequate or accessible training models to upskill physiotherapists in integrated physiotherapy and psychological treatments, successful and widespread implementation is unlikely [12]. This trial aims to address these challenges by comparing the effectiveness of a blended learning versus in-person training model for broad implementation of StressModex, a physiotherapist-delivered integrated psychological and physical intervention for people with musculoskeletal RTI at risk of poor recovery [10]. By using a hybrid type III cluster design, this trial will provide real-world insights into the most effective strategies for training physiotherapists who work in private community clinics, where most patients with musculoskeletal RTI seek care in Australia.

This trial fills a critical gap in training physiotherapists to confidently and competently deliver person-centred care for patients with musculoskeletal pain conditions that consider a biopsychosocial model of pain. Physiotherapists consistently report a lack of skills and confidence in delivering psychological interventions [13]. Yet guidance on effective training methods for physiotherapists is limited. In 2020, we piloted an early iteration of PICOT with 18 physiotherapists. Results (unpublished data) showed that physiotherapists’ scores on self-reported intent to use StressModex in practice (3.1/5 to 4.5/5), level of confidence (3/5 to 4.5/5) and knowledge (28/60 to 47/60) improved pre-post 6-weeks training. The current trial will substantiate whether training physiotherapists with an accessible blended learning package results in a greater number of patients receiving an evidence-based integrated treatment (StressModex) compared to traditional learning methods. The findings have implications for shaping cognitive behavioural training models for health professionals of the future, support person-centred care and reduce the burden of musculoskeletal pain. Additionally, our quantitative and qualitative process evaluation will inform broad and rapid scale-up beyond this trial, contributing to greater scalability, reach, and improved patient health outcomes.

There are some potential risks associated with this trial. Physiotherapy clinic and patient recruitment is often unpredictable in real-world clinical settings. This trial will only recruit physiotherapy clinics that see sufficient numbers of patients with musculoskeletal RTIs. To further mitigate this risk, we will provide these clinics with an executive summary of the trial, which they can use to help recruit eligible patients from local General Practices and Emergency Departments. The research team has developed an engagement strategy which includes regular email newsletters updating physiotherapists about trial progress.

Ensuring adherence to the trial protocol may be a challenge due to the trial’s complex design, differences in clinic environments and workplace cultures, the busy and demanding nature of clinical work, and the need for ongoing motivation and engagement from physiotherapists throughout the trial. To mitigate this, we will develop a clear and detailed manual of trial procedures for research staff and participating physiotherapists and provide training in these procedures prior to commencement. The Trial Coordinator will maintain regular contact with the clinics, offering ongoing support and addressing any questions or concerns. The physiotherapy clinics will receive financial support to facilitate the trial processes. Additionally, trial newsletters will keep clinics updated on progress, fostering engagement and retention. The Trial Coordinator will also monitor adherence to the trial protocol and conduct audits of anonymised physiotherapy medical records to ensure compliance with documentation standards and address any identified issues with participating clinics. The clinic delegate will facilitate documentation sharing between their clinic and the research team, and the nominated clinic champion will have a key role in coordinating and engaging physiotherapists for the study duration. There is also a risk of selection bias, given that the same physiotherapists who are trained to deliver interventions in the trial will also be identifying, screening and consenting patients for data collection, and we have taken steps to mitigate this.

Conclusion

This trial will address gaps in knowledge about the best model of training physiotherapists to deliver evidence-based integrated psychological and physical care for patients with musculoskeletal RTI. Enhanced understanding of the most effective training model will inform rapid and broad scale-up, facilitating better reach of these integrated interventions like StressModex to patients, including those who may be living in rural and remote areas. Focus on continued professional development training models and research methods that examine implementation outcomes and treatment effectiveness in real-world clinical settings could ultimately improve patient health and quality of life after RTI.

Abbreviations

RTIs

Road traffic injuries

PICOT

Physiotherapists bIopsyChosocial On-line Training

RCT

Randomised controlled trial

STARI

Standards for Reporting Implementation Studies

i-PARIHS

Integrated-Promoting Action on Research Implementation in Health Services framework

PTSD

Posttraumatic Stress Disorder

Authors’ contributions

MS and RE conceived the study. RF, JLo, NEF, JT, HS, HT, KP, SF, JLy, SO, TR, RW, KE, TA, MN, DB contributed to the study design and protocol. MS and RE designed the PICOT implementation package with input from RF, JL, CP, YX, AC, CE and EH. HT designed the economic evaluation. RW designed the statistical analysis. MS and YX drafted the manuscript. All authors provided critical review and approved the final manuscript version.

Funding

This trial was funded by The Medical Research Future Fund (MRFF) – 2022 Clinician Researchers: Nurses, Midwives and Allied Health (APP 2024146).

Medical Research Future Fund,2024146,Michele Sterling

Data availability

Not applicable.

Declarations

Ethics approval and consent to participate

This project is approved by The University of Queensland Human Research Ethics Committee A on 14 June 2024 (2024/HE000426).

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

MS is funded through an Australian National Health and Medical Research Council (NHMRC) Investigator Grant (ID: 2017405). MS receives unrestricted funding from the Motor Accident Insurance Commission Qld.

NEF is funded through an Australian National Health and Medical Research Council (NHMRC) Investigator Grant (ID: 2018182).