Introduction
Pulmonary rehabilitation (PR), an exercise and education intervention, is strongly recommended for people with chronic respiratory disease (CRD) in international guidelines (Bolton et al., Reference Bolton, Bevan-Smith, Blakey, Crowe, Elkin, Garrod, Greening, Heslop, Hull, Man, Morgan, Proud, Roberts, Sewell, Singh, Walker and Walmsley2013; Alison et al., Reference Alison, Mckeough, Johnston, Mcnamara, Spencer, Jenkins, Hill, Mcdonald, Frith, Cafarella, Brooke, Cameron-Tucker, Candy, Cecins, Chan, Dale, Dowman, Granger, Halloran, Jung, Lee, Leung, Matulick, Osadnik, Roberts, Walsh, Wootton and Holland2017; Rochester et al., Reference Rochester, Alison, Carlin, Jenkins, Cox, Bauldoff, Bhatt, Bourbeau, Burtin, Camp, Cascino, Koppel, Garvey, Goldstein, Harris, Houchen-Wolloff, Limberg, Lindenauer, Moy, Ryerson, Singh, Steiner, Tappan, Yohannes and Holland2023). It leads to improvements in exercise capacity, breathlessness, and health-related quality of life (Mccarthy et al., Reference Mccarthy, Casey, Devane, Murphy, Murphy and Lacasse2015; Dowman et al., Reference Dowman, Hill, May and Holland2021; Lee et al., Reference Lee, Gordon and Osadnik2021; Yang et al., Reference Yang, Gao, Wang, Deng and Gao2022; He et al., Reference He, Wang, Feng, Li and Xie2023), as well as a reduction in hospital readmissions (Moore et al., Reference Moore, Palmer, Newson, Majeed, Quint and Soljak2016). Despite its effectiveness, approximately 5% of people who would benefit from referral ever access a program, and fewer still complete (Rochester et al., Reference Rochester, Vogiatzis, Holland, Lareau, Marciniuk, Puhan, Spruit, Masefield, Casaburi, Clini, Crouch, Garcia-Aymerich, Garvey, Goldstein, Hill, Morgan, Nici, Pitta, Ries, Singh, Troosters, Wijkstra, Yawn and Zuwallack2015; Spitzer et al., Reference Spitzer, Stefan, Priya, Pack, Pekow, Lagu, Pinto-Plata, Zuwallack and Lindenauer2019). Physiotherapists and Accredited Exercise Physiologists (AEPs) are suitably qualified health professionals to conduct a PR program. Although suitably trained, many physiotherapists and AEPs report no exposure to PR in their training or on clinical placement (Johnston et al., Reference Johnston, Maxwell and Alison2011; Rochester et al., Reference Rochester, Vogiatzis, Holland, Lareau, Marciniuk, Puhan, Spruit, Masefield, Casaburi, Clini, Crouch, Garcia-Aymerich, Garvey, Goldstein, Hill, Morgan, Nici, Pitta, Ries, Singh, Troosters, Wijkstra, Yawn and Zuwallack2015). The majority of PR programs are provided within the hospital setting, with 93% of Australian PR programs run by hospital staff (Johnston et al., Reference Johnston, Maxwell and Alison2011). Conversely, 60% of physiotherapists and 57% of AEPs work in private practices (PPs) (Australian Physiotherapy Association, 2023; Exercise & Sports Science Australia, 2023). Highlighting a large potential workforce that could deliver PR in primary care in the PP setting with appropriate upskilling.
There are emerging models of PR such as home-based PR and telerehabilitation which have been shown to be as effective as traditional center-based PR (Holland et al., Reference Holland, Mahal, Hill, Lee, Burge, Cox, Moore, Nicolson, Ohalloran, Lahham, Gillies and Mcdonald2017; Cox et al., Reference Cox, Dal Corso, Hansen, Mcdonald, Hill, Zanaboni, Alison, O’halloran, Macdonald and Holland2021; Stafinski et al., Reference Stafinski, Nagase, Avdagovska, Stickland and Menon2022; Machado et al., Reference Machado, Burtin and Spruit2024). However, in practice these models are still staffed by the existing hospital-based PR workforce. Many of these services operate with extensive waitlists (Lung Foundation Australia, 2023) and constraints of public hospital workforce challenges and funding cuts (Philip, Reference Philip2015). In Australia, there have been calls to broaden the utilization of allied health in primary care (Australian Government Department of Health, 2023) and the introduction of funding schemes to support the delivery of PR in the primary care setting (Lung Foundation Australia, 2023; Australian Commission on Safety and Quality in Health Care, 2024). With appropriate upskilling of the existing primary care physiotherapy and AEP workforce, there is scope for PR to be delivered in this setting, complementing, existing services delivered by hospital-based staff and increasing reach.
Patients often report travel, transport, and timing as barriers to attendance and completion of PR (Keating et al., Reference Keating, Lee and Holland2011; Cox et al., Reference Cox, Oliveira, Lahham and Holland2017). Hospital-based programs are provided in work hours and often operate in shared gym spaces so timing may be relatively inflexible. Comparatively, PPs are often open after hours or on weekends, may have their own gym space, and are typically embedded within local communities, offering geographic proximity to patients thereby reducing the travel burden. Therefore, offer both a workforce and infrastructure to provide PR, potentially increasing access.
Limited data exist on upskilling healthcare professionals to provide PR. A 2-day PR training program for healthcare professionals working in rural communities lead to significant improvements in knowledge and confidence and facilitated the establishment of new PR programs (Johnston et al., Reference Johnston, Maxwell, Boyle, Maguire and Alison2013). A systematic review of education interventions for healthcare professionals managing COPD found only one study targeting either physiotherapists or AEPs (Cross et al., Reference Cross, Thomas, Liang, Abramson, George and Zairina2022). To our knowledge, no studies have been conducted specifically targeting private practitioners. A recent survey of Australian physiotherapists and AEP working in private practice reported that clinicians were interested in upskilling to provide PR in their practices and that this aligned with their business model (Walsh et al., Reference Walsh, Dennis, Alison, Thom, Herrmann, Fisher, McKeough and Dale2025a). This study aimed to investigate the efficacy of a PR training program for PP physiotherapists and AEPs to develop the knowledge, skills, and confidence necessary to conduct PR.
Methods
Participants
This study is part of a larger feasibility trial, and the detailed methods are available in the published study protocol (Walsh et al., Reference Walsh, Mckeough, Dale, Alison and Dennis2025b). The study was conducted in two primary health networks (PHNs) in Sydney, Australia. Participants were recruited from March 2023 to November 2024 through PHNs and professional societies. Participants were registered physiotherapists or AEPs, working in PP where there was: at least two staff members, adequate space for group exercise, a 10-meter straight track for a field walking test, and accessible for people with CRD. Physiotherapists or AEPs with post-graduate PR experience or qualifications were excluded.
Interventions
Participants were given access to the Lung Foundation Australia (LFA) PR Online Training Program (Lung Foundation Australia), a 6–8 hour self-directed online program to increase knowledge and skills of clinicians to deliver evidence-based PR. Modules include: Establishing a pulmonary rehabilitation program; The respiratory system and COPD; Comprehensive patient assessment; Spirometry; Assessing exercise capacity; Assessing quality of life; Exercise training; Patient education; Program evaluation and troubleshooting. Participants then attended a 4-hour small group practical face-to-face module, delivered by experts in PR, which focused on practical aspects of PR including performing exercise capacity tests and exercise prescription.
Participants were required to pass a predefined competency threshold of 80% (i.e., 9 out of 11) for an objective knowledge test embedded within the LFA PR Online Training Program: an 11-item multiple-choice questionnaire based on a clinical case vignette, with items covering the key domains of PR assessment and delivery. A copy of the questionnaire is available as supplementary material in the published protocol (Walsh et al., Reference Walsh, Mckeough, Dale, Alison and Dennis2025b). The 80% competency threshold was pre-specified in the study protocol. Competency for the practical skills was to be ‘proficient’ in practical skills of the six-minute walk test, incremental shuttle walk test, and 1-minute sit-to-stand test. After completing the workshop, each participant was directly observed by an assessor with expertise in PR while performing each skill. Proficiency was determined against a pre-defined competency checklist covering correct setup, standardized instructions, safety monitoring and test termination criteria. The self-rated confidence, knowledge, and skills questionnaire assessed the content domains across both the LFA online training modules and the practical skills workshop.
Outcomes
The primary outcomes were i) change in participant knowledge in the knowledge test (scored 0–11) and ii) change in participant self-rated confidence, knowledge, and skills using a 5-point Likert scale questionnaire before and after the training course. Secondary outcomes included i) proportion of participants meeting 80% in the knowledge test and proficient in practical skills assessment, and ii) maintenance of participant knowledge at 3 months post-training. Participants also completed a survey of their satisfaction with the training using a 5-point Likert scale.
Statistical methods
Fifteen participants were needed to detect a mean (SD) change of 4.6 (2.5) points change in participant knowledge (α = 0.05, β = 0.8), accounting for 20% attrition (Johnston et al., Reference Johnston, Maxwell, Boyle, Maguire and Alison2013). This calculation assumed a normal distribution, which may not hold for a small cohort; it was used as an approximate planning estimate rather than a formal guarantee of power. Data were summarized using descriptive statistics. Likert scales were analyzed as ordinal scales, with 5-point scales assigned values from −2 to +2, except satisfaction scores which were assigned values 1–5. Change data were analyzed over all timepoints using the Friedman test, with pairwise comparisons using the Wilcoxon signed-rank test and Bonferroni correction applied, with alpha p < 0.05. Analysis was conducted on participants with available data at these timepoints. Hedges’ g was calculated for the primary knowledge outcome to quantify effect size. All continuous data are reported as median (interquartile range).
Results
Invitations were sent to 332 email addresses; 32 replies (10%) were received. Of those, 12 were excluded (8 = out of area, 4 = duplicate contact from same PP), leaving 22 eligible clinicians. 17 clinicians (8 physiotherapists, 9 AEPs), 11 (65%) males, completed the training program and post-training measures, 13 (76%) completed the 3-month follow-up. Median (IQR) years working in their profession was 3.5 (1.75, 9) years and median (IQR) number of clinical staff working in the practices was 4 (4,6).
The median (IQR) knowledge score pre-training was 7 (6,8) points, compared to 10 (9,11) points at post-training (z = −3.43, p < 0.001)). The effect size for the primary knowledge outcome was large (Hedges’ g = 1.52, 95% CI 0.81 to 2.21). Post training, 82% of participants met the competency threshold for the knowledge test (i.e., score ≥ 80%), compared to only 18% pre-training. Almost all (94%) of participants met the pre-defined practical competencies. Practical skills competency was assessed post-training only so improvement in this domain cannot be quantified.
At 3-month follow-up, the median (IQR) knowledge score was 9 (7.75, 10), which was lower than post-training (z = −2.05, p = 0.04). The Friedman test showed a significant difference across the three timepoints (X 2 = 20.42, p < 0.001), with pairwise comparisons (Bonferroni-correction α = 0.02) showing significant improvement pre-training to post-training (p < 0.001) and pre-training to 3 months (p = 0.01) but a non-significant decline post-training to 3 months (p = 0.04). Self-rated confidence, knowledge, and skills pre- and post-training are displayed in Figure 1. There was a statistically significant change in median scores for each component pre- to post-training. The median (IQR) satisfaction score for the online training was 5 (4, 5) and for the practical module was 4 (4, 5) (out of 5).
Likert scales of participant’s self-rated: a) confidence pre-training b) confidence post-training c) knowledge and skills pre-training d) knowledge and skills post-training. Median (IQR) scores of participants’ responses to 5-point Likert scales with values assigned −2 to +2. IRQ, interquartile range.
Discussion
In this study, a short training program was associated with improvements in PR knowledge, skills, and confidence among primary care physiotherapists and AEPs, and these findings provide early-stage evidence to support workforce development as a strategy for expanding access to PR. Baseline knowledge was below the competency threshold, aligning with reports on lack of exposure to education and training in this area of practice (Johnston et al., Reference Johnston, Maxwell and Alison2011; Rochester et al., Reference Rochester, Vogiatzis, Holland, Lareau, Marciniuk, Puhan, Spruit, Masefield, Casaburi, Clini, Crouch, Garcia-Aymerich, Garvey, Goldstein, Hill, Morgan, Nici, Pitta, Ries, Singh, Troosters, Wijkstra, Yawn and Zuwallack2015; Rochester et al., Reference Rochester, Langer and Singh2021). This is unsurprising given that only 2% of Australian physiotherapists report they work in cardiorespiratory, compared to 37% working in musculoskeletal (Australian Physiotherapy Association, 2023).
Assessment and interpretation of spirometry was the area of practice that participants rated themselves as having the least amount of knowledge, skills, and confidence at baseline. Although improvements were observed across the three domains post-training, it remained the area of practice with the lowest score. The LFA training module on spirometry focuses on interpretation of existing spirometry results (e.g., understanding FEV1 and FVC values to identify obstructive or restrictive patterns) rather than on the performance of spirometry testing itself. Primary care physiotherapists and AEPs are unlikely to have access to a spirometer and perform spirometry so the relevant skill is interpreting any spirometry results provided by the referring doctor. Therefore, the scope of skills required to perform PR for these clinicians may need to be adapted, compared to those of hospital-based clinicians where spirometry is a key component of the pre-program respiratory assessment, used to confirm diagnosis and classify disease severity, rather than an ongoing measurement within the PR exercise program itself. Importantly, this adaptation relates to the assessment phase rather than exercise training delivery, which can be conducted effectively with minimal equipment and without compromising clinical outcomes (Cheng et al., Reference Cheng, Mckeough, McNamara and Alison2023). However, there may be implications for referral pathways, with patients requiring spirometry to be performed prior to referral to PR.
Among the self-rated domains, exercise capacity assessment, exercise prescription, and patient education showed the most clinically meaningful shifts pre- to post-training and are the competencies most directly required for day-to-day PR delivery in private practice. Sustained competency in these domains at 3 months is encouraging for the prospects of translation to practice, while acknowledging that actual clinical application in a patient-facing program is required to confirm whether training-acquired competency translates to safe and effective delivery.
Poor accessibility to PR has long been recognized since most PR programs are only available through secondary care hospital-funded programs in Australia (Johnston et al., Reference Johnston, Maxwell and Alison2011). Primary care provides an alternative setting and workforce to increase provision of PR beyond hospital-based programs. It is also recognized that primary care allied health professionals do not utilize their full skill set (Dennis et al., Reference Dennis, Ball, Harris and Refshauge2021). Private practices provide a potential alternative PR workforce and infrastructure (Australian Physiotherapy Association, 2023; Exercise & Sports Science Australia, 2023). However, given the lack of appropriately trained clinicians (Rochester et al., Reference Rochester, Langer and Singh2021), education programs to build knowledge and skills of primary care physiotherapists and AEPs in PR may improve access to PR. There are other barriers to the provision of PR in the primary care setting. In Australia for example, there is no funding model for primary care practitioners to provide PR, meaning as little as 6% of PR programs are conducted in PPs nationally (Lung Foundation Australia, 2022). However, the fact that a short training course can improve knowledge and skills of physiotherapists and AEPs to provide PR, without the need for formal post-graduate training, may mean that PR can be delivered in the primary care setting in a scalable and feasible way, increasing the provision of PR which is a greatly underutilized intervention.
There are some limitations to the study. The single-arm, pre-post design without a control group precludes causal inference, and self-selection bias among motivated early-adopters cannot be excluded. The 10% response rate from initial invitations, combined with eligibility exclusions, resulted in a final sample of 17 clinicians from two Sydney PHNs, limiting generalization to the broader private practice physiotherapy and AEP workforce, so replication across geographically diverse settings is important. It should be noted that clinicians recruited to this study had to commit to both training and a feasibility RCT, an offer of standalone training may have achieved higher uptake. A recent national survey of 245 PP physiotherapists and AEPs in Australia reported that 91% would be interested in providing PR in their practice, especially if funding was available, suggesting that there would likely be uptake of training (Walsh et al., Reference Walsh, Dennis, Alison, Thom, Herrmann, Fisher, McKeough and Dale2025a). Formal psychometric validation data are not available for the knowledge test, and the 80% competency threshold, while consistent with medical education literature, was not derived through formal procedures.
For widespread workforce scale-up, cost, time, and perceived relevance of the training program to everyday private practice may all influence uptake. For participants enrolled in the study, training was provided at no cost; however, the LFA online training program is commercially available at AUD $175. Encouragingly, in a parallel national workforce survey conducted as part of this research program, 90% of 221 private practice clinicians indicated they would pay the AUD $175 training fee if it were required to access a Medicare funding rebate for PR delivery, suggesting that financial motivation aligned with sustainable funding models may be a powerful lever for engagement. The translation of skills and knowledge from the training program also should be tested in a practical way, with participants going on to conduct a PR program in the practices in which they work. This study is currently underway, and the protocol has been published (Walsh et al., Reference Walsh, Mckeough, Dale, Alison and Dennis2025b).
In conclusion, this study showed that a short training program improved knowledge, skills and confidence in primary care physiotherapists and AEPs. This study reports the training component of the PURE PRIME study (Walsh et al., Reference Walsh, Mckeough, Dale, Alison and Dennis2025b). While these findings demonstrate that primary care clinicians can acquire foundational PR knowledge and skills through a brief training program, the translation of this capacity into sustained, patient-facing PR delivery requires evaluation of the implementation component, which is currently underway.
Data availability statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Author contributions
All authors: concept and design of the study. JAW: data collection. JAW: data analysis. All authors: drafted the manuscript. All authors reviewed the final draft of the manuscript.
Funding statement
This study is funded by Lung Foundation Australia/Better Breathing Foundation PhD Scholarship in Chronic Respiratory Disease, awarded to JAW.
Competing interests
ZJM is the managing director of the Better Breathing Foundation. JAW is supported by a Lung Foundation Australia/Better Breathing Foundation PhD Scholarship in Chronic Respiratory Disease. MTD, JAA and SMD have no relevant disclosures.
Ethical standards
Ethical approval from Sydney Local Health District (RPAH Zone) Human Research Ethics Committee (2022/ETH02324). Australian and New Zealand Clinical Trials Registry number: ACTRN12622001501730.
Open access
