Journal of Physiotherapy 69 (2023) 128 j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j p hy s Appraisal Appraisal of Clinical Practice Guideline: Clinical Practice Guideline for hamstring strain injury in athletes Date of latest update: March 2022. Date of next update: 2026 or earlier if needed. Patient group: Athletes Commentary: This guideline clearly describes the aims and objectives of the guideline, population of with hamstring strain injury. Intended audience: Orthopaedic and sports physiotherapists. Additional interest, search methods, selection criteria, process for developing and refining recommendations versions: N/A. Funded by: NHMRC Investigator Grant (APP1194105), The Academy of Orthopaedic Physical (including end-user involvement), and evidence that contributes to each recommendation.1 The Therapy (AOPT) and American Academy of Sports Physical Therapy (AASPT). Expert working group: guideline also describes how conflicts of interest were managed in formulating the Physiotherapists and a doctor. Consultation with: AOPT members, consumer/patient representatives, recommendations, which is a major strength. Some weaknesses of the guideline include lack of external stakeholders, claims reviewers, medical coding experts, academics, clinical educators, physician consideration of benefits and harms of recommended care, and limited information on facilitators specialists, researchers and clinical practice guideline methodologists. Approved by: AOPT Guideline and barriers to implementation and resource implications (eg, costs of implementing guideline- Editors and Clinical Practice Guideline Managing Editor. Location: https://doi.org/10.2519/jospt.2022.0301 based assessment and treatment). Future research should focus on better defining performance char- acteristics as risk factors (eg, hamstring weakness) and improving methods used to assess them, better Description: Recommendations were developed by a series of author groups, with feedback from re- understanding the clinical course of hamstring strain injury (and how prognostic factors modify this) to view panels. There were 14 recommendations, 11 of which were based on moderate (Level B) or strong reduce re-injury risk, and determining the most effective type and dose of injury prevention programs. (Level A) evidence. Re-injury risk and return to play recommendations: use history of hamstring strain injury in return to play progression (it is a risk factor for re-injury); use caution in return to play de- Provenance: Invited. Not peer reviewed. cisions for those who did not complete an appropriately progressed and comprehensive impairement- based functional exercise program including eccentric training; and use hamstring strength, pain in- Joshua R Zadro tensity at the time of injury, days from injury to pain-free walking, and area of tenderness to estimate The University of Sydney, Australia return-to-play time. For diagnosis/classification, diagnose as hamstring strain injury when someone presents with sudden-onsent posterior thigh pain during activity, pain when the hamstring is length- Reference ened or contracted, tenderness with palpation and loss of function. Recommendations for examination: quantify knee flexor strength following hamstring strain injury using a handheld or isokinetic dyna- 1. Martin RL, et al. J Orthop Sports Phys Ther. 2022;52:Cpg1–44. mometer; assess hamstring length using an inclinometer; document objective measures of ability to walk, run and sprint to monitor progress; and use the Functional Assessment Scale for Acute Hamstring https://doi.org/10.1016/j.jphys.2023.02.010 Injuries before and after interventions. For injury prevention, include the Nordic hamstring exercise in a prevention program, along with warm-up, stretching, stability training, strengthening and functional movements. Recommendations for management: use eccentric training, stretching, strengthening, stabilisation and progressive running to reduce return to play time; and use progressive agility and trunk stabilisation, stretching, strengthening, and functional exercises to reduce re-injury risk. Appraisal of Clinical Practice Guideline: Exercise intensity assessment and prescription in cardiovascular rehabilitation and beyond: why and how. A position statement from the Secondary Prevention and Rehabilitation Section of the European Association of Preventive Cardiology Date of latest update: January 2021. Date of next update: Not stated. Patient group: Patients with Commentary: The strengths of this guideline include clear guideline objectives, description of the cardiovascular disease (CVD). Intended audience: Cardiologists, physiotherapists, clinical exercise population, search methods, link between recommendations and evidence, and management op- physiologists, nurses and specialists who deal with exercise for CVD patients. Additional versions: tions; consideration of benefits and harms; and specific recommendations.1 Some weaknesses Eur J Prev Cardiol. 2013;20:442–467. Funded by: NHMRC Investigator Grant (APP1194105). Expert include limited information on how: studies were selected; consensus was reached; the working group: Individuals from relevant cardiac rehabilitation professional groups (European As- recommendations were formulated; the views of end users/external reviewers informed the sociation of Preventive Cardiology, American Association of Cardiovascular and Pulmonary Rehabil- recommendations; and funding or conflicts of interest potentially influenced guideline itation, Canadian Association of Cardiac Rehabilitation). Consultation with: Not stated. Approved by: development. There was also no systematic review and meta-analysis of the available evidence Expert opinion. Location: https://doi.org/10.1093/eurjpc/zwab007 and no information about dissemination of the recommendations (eg, resource implications, barriers, facilitators). Future aerobic exercise training research should investigate the impact of different ex- Description: The following recommendations on aerobic and resistance exercise prescription align ercise intensities on CVD risk factors and other health indicators under isocaloric conditions, and the with current cardiac rehabilitation position papers or guidelines and exercise prescription for pa- long-term safety of unsupervised high-intensity exercise. Future resistance exercise research should tients with CVD. For aerobic exercise: aim to optimise total energy expenditure rather than one investigate the benefits of different types of resistance exercise as an adjunct to aerobic exercise, the specific training characteristic (eg, intensity); use intensity to design programs that are time-efficient haemodynamic impact of high-intensity versus low-intensity exercise, a standardised approach for and induce short-term clinical benefits; higher intensities increase the relative risk of cardiovascular exercise progression, the impact of functional resistance training in prolonged rehabilitation, and the events in patients with coronary artery disease and heart failure, but the absolute risk remains low; acceptability of self-selected versus imposed exercise intensities. intensity should be decided upon using a shared decision-making approach; commonly used peak indices (eg, VO2peak, HRpeak) should be applied with great caution; assessment of the first venti- Provenance: Invited. Not peer reviewed. latory threshold, VT1, and VT2 during a cardiopulmonary exercise test is a reliable measure of ex- ercise intensity in most CVD patients; the talk test and Borg Rated Percieved Exertion scale should Joshua R Zadro only be considered as adjuncts to objective intensity determination; progress duration of the exercise The University of Sydney, Australia session before progressing intensity; and re-assess VT1 and VT2 as fitness increases to inform in- tensity. For resistance exercise: high-intensity dynamic resistance training (versus low-intensity) Reference leads to greater improvements in muscle strength in patients with elevated CVD risk and is equally safe when properly executed; intensity should be decided upon using a shared decision- 1. Hansen D, et al. Eur J Prev Cardiol. 2022;29:230–245. making approach; to determine intensity, use a , 10 RM test to set the initial weight and the Omnibus Resistance Exercise Scale to track perceived intensity during sessions; and ongoing strength https://doi.org/10.1016/j.jphys.2023.02.011 improvements should be facilitated by progressing repetitions, intensity and rest periods (in that order). 1836-9553/© 2023 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/4.0/).
Journal of Physiotherapy 69 (2023) 128 j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j p hy s Appraisal Appraisal of Clinical Practice Guideline: Clinical Practice Guideline for hamstring strain injury in athletes Date of latest update: March 2022. Date of next update: 2026 or earlier if needed. Patient group: Athletes Commentary: This guideline clearly describes the aims and objectives of the guideline, population of with hamstring strain injury. Intended audience: Orthopaedic and sports physiotherapists. Additional interest, search methods, selection criteria, process for developing and refining recommendations versions: N/A. Funded by: NHMRC Investigator Grant (APP1194105), The Academy of Orthopaedic Physical (including end-user involvement), and evidence that contributes to each recommendation.1 The Therapy (AOPT) and American Academy of Sports Physical Therapy (AASPT). Expert working group: guideline also describes how conflicts of interest were managed in formulating the Physiotherapists and a doctor. Consultation with: AOPT members, consumer/patient representatives, recommendations, which is a major strength. Some weaknesses of the guideline include lack of external stakeholders, claims reviewers, medical coding experts, academics, clinical educators, physician consideration of benefits and harms of recommended care, and limited information on facilitators specialists, researchers and clinical practice guideline methodologists. Approved by: AOPT Guideline and barriers to implementation and resource implications (eg, costs of implementing guideline- Editors and Clinical Practice Guideline Managing Editor. Location: https://doi.org/10.2519/jospt.2022.0301 based assessment and treatment). Future research should focus on better defining performance char- acteristics as risk factors (eg, hamstring weakness) and improving methods used to assess them, better Description: Recommendations were developed by a series of author groups, with feedback from re- understanding the clinical course of hamstring strain injury (and how prognostic factors modify this) to view panels. There were 14 recommendations, 11 of which were based on moderate (Level B) or strong reduce re-injury risk, and determining the most effective type and dose of injury prevention programs. (Level A) evidence. Re-injury risk and return to play recommendations: use history of hamstring strain injury in return to play progression (it is a risk factor for re-injury); use caution in return to play de- Provenance: Invited. Not peer reviewed. cisions for those who did not complete an appropriately progressed and comprehensive impairement- based functional exercise program including eccentric training; and use hamstring strength, pain in- Joshua R Zadro tensity at the time of injury, days from injury to pain-free walking, and area of tenderness to estimate The University of Sydney, Australia return-to-play time. For diagnosis/classification, diagnose as hamstring strain injury when someone presents with sudden-onsent posterior thigh pain during activity, pain when the hamstring is length- Reference ened or contracted, tenderness with palpation and loss of function. Recommendations for examination: quantify knee flexor strength following hamstring strain injury using a handheld or isokinetic dyna- 1. Martin RL, et al. J Orthop Sports Phys Ther. 2022;52:Cpg1–44. mometer; assess hamstring length using an inclinometer; document objective measures of ability to walk, run and sprint to monitor progress; and use the Functional Assessment Scale for Acute Hamstring https://doi.org/10.1016/j.jphys.2023.02.010 Injuries before and after interventions. For injury prevention, include the Nordic hamstring exercise in a prevention program, along with warm-up, stretching, stability training, strengthening and functional movements. Recommendations for management: use eccentric training, stretching, strengthening, stabilisation and progressive running to reduce return to play time; and use progressive agility and trunk stabilisation, stretching, strengthening, and functional exercises to reduce re-injury risk. Appraisal of Clinical Practice Guideline: Exercise intensity assessment and prescription in cardiovascular rehabilitation and beyond: why and how. A position statement from the Secondary Prevention and Rehabilitation Section of the European Association of Preventive Cardiology Date of latest update: January 2021. Date of next update: Not stated. Patient group: Patients with Commentary: The strengths of this guideline include clear guideline objectives, description of the cardiovascular disease (CVD). Intended audience: Cardiologists, physiotherapists, clinical exercise population, search methods, link between recommendations and evidence, and management op- physiologists, nurses and specialists who deal with exercise for CVD patients. Additional versions: tions; consideration of benefits and harms; and specific recommendations.1 Some weaknesses Eur J Prev Cardiol. 2013;20:442–467. Funded by: NHMRC Investigator Grant (APP1194105). Expert include limited information on how: studies were selected; consensus was reached; the working group: Individuals from relevant cardiac rehabilitation professional groups (European As- recommendations were formulated; the views of end users/external reviewers informed the sociation of Preventive Cardiology, American Association of Cardiovascular and Pulmonary Rehabil- recommendations; and funding or conflicts of interest potentially influenced guideline itation, Canadian Association of Cardiac Rehabilitation). Consultation with: Not stated. Approved by: development. There was also no systematic review and meta-analysis of the available evidence Expert opinion. Location: https://doi.org/10.1093/eurjpc/zwab007 and no information about dissemination of the recommendations (eg, resource implications, barriers, facilitators). Future aerobic exercise training research should investigate the impact of different ex- Description: The following recommendations on aerobic and resistance exercise prescription align ercise intensities on CVD risk factors and other health indicators under isocaloric conditions, and the with current cardiac rehabilitation position papers or guidelines and exercise prescription for pa- long-term safety of unsupervised high-intensity exercise. Future resistance exercise research should tients with CVD. For aerobic exercise: aim to optimise total energy expenditure rather than one investigate the benefits of different types of resistance exercise as an adjunct to aerobic exercise, the specific training characteristic (eg, intensity); use intensity to design programs that are time-efficient haemodynamic impact of high-intensity versus low-intensity exercise, a standardised approach for and induce short-term clinical benefits; higher intensities increase the relative risk of cardiovascular exercise progression, the impact of functional resistance training in prolonged rehabilitation, and the events in patients with coronary artery disease and heart failure, but the absolute risk remains low; acceptability of self-selected versus imposed exercise intensities. intensity should be decided upon using a shared decision-making approach; commonly used peak indices (eg, VO2peak, HRpeak) should be applied with great caution; assessment of the first venti- Provenance: Invited. Not peer reviewed. latory threshold, VT1, and VT2 during a cardiopulmonary exercise test is a reliable measure of ex- ercise intensity in most CVD patients; the talk test and Borg Rated Percieved Exertion scale should Joshua R Zadro only be considered as adjuncts to objective intensity determination; progress duration of the exercise The University of Sydney, Australia session before progressing intensity; and re-assess VT1 and VT2 as fitness increases to inform in- tensity. For resistance exercise: high-intensity dynamic resistance training (versus low-intensity) Reference leads to greater improvements in muscle strength in patients with elevated CVD risk and is equally safe when properly executed; intensity should be decided upon using a shared decision- 1. Hansen D, et al. Eur J Prev Cardiol. 2022;29:230–245. making approach; to determine intensity, use a , 10 RM test to set the initial weight and the Omnibus Resistance Exercise Scale to track perceived intensity during sessions; and ongoing strength https://doi.org/10.1016/j.jphys.2023.02.011 improvements should be facilitated by progressing repetitions, intensity and rest periods (in that order). 1836-9553/© 2023 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/4.0/).
Journal of Physiotherapy 69 (2023) 65–67 j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j p hy s Editorial Are we giving stratified care a fair trial? Nadine E Foster a,b, Jonathan C Hill b, Jesper Knoop c aSTARS Education and Research Alliance, Surgical Treatment and Rehabilitation Service (STARS), The University of Queensland and Metro North Health, Brisbane, Australia; bSchool of Medicine, Keele University, Keele, UK; cDepartment of Health Sciences, Amsterdam Movement Sciences, Vrije Universiteit Amsterdam, Amsterdam, The Netherlands The recent OCTOPuS study in the Journal of Physiotherapy is unique, treatment matched to their subgroup?’ Imagine a cancer trial, for example, that wanted to test whether people with a certain clinical being the first randomised controlled trial (RCT) to test stratified ex- profile (subgroup) have better outcomes if they receive a specific ercise therapy for patients with knee osteoarthritis.1 Stratified care matched medication (targeted treatment). If trial participants were involves targeting treatment to subgroups of patients based on key not subgrouped in the intended way or did not receive the matched medication, the trial would not be a fair test of stratified care. Simi- characteristics such as prognostic factors, responsiveness to treatment larly, in physiotherapy, where upskilling or training of healthcare and/or underlying mechanisms.2 Most stratified care trials to date have providers is needed for both components of stratified care (sub- focused on people with low back pain – such as the UK STarT Back trial,3 grouping and targeting treatment), trials can fail to be a ‘fair test’ of the MATCH4 and TARGET5 trials in the USA, and trials from Denmark6 stratified care. Below we consider these issues using previous trials of and the Netherlands7 – but OCTOPuS is the first stratified care trial to stratified care. have tested a model of stratified care in people with knee osteoarthritis. Almost two decades ago stratified care was called the ‘Holy Grail’ of The STarT Back risk stratification trial for low back pain from the back pain research,8 and stratified care or precision medicine has UK demonstrated positive results in favour of stratified care.3 All become the zeitgeist in many clinical fields – such as mental health,9 participants were successfully subgrouped (using the STarT Back tool) diabetes10 and cancer11 – with the goal of personalised healthcare. and only 2.6% of those randomised to stratified care did not receive The allure is the potential to maximise treatment benefit, reduce harm the matched treatment for their subgroup. In addition, given that and increase healthcare efficiency by offering the right treatment to the different therapists delivered the control and intervention treat- right patient at the right time.12 However, the findings of the OCTOPuS ments, contamination between arms of the trial was avoided. To trial are not unique, as they concluded that stratified care did not lead to achieve high intervention fidelity in the STarT Back trial, dedicated better clinical1 or cost outcomes.13 new clinics using trained research staff were set up, the stratified care treatment (three subgroups each matched to one treatment protocol) This Editorial considers one particularly compelling explanation was provided by a small team of 13 trained physiotherapists, who were given time for self-reflection and skills practice as part of their for the lack of differences observed in stratified care trials: inter- training, and were supported by ongoing mentoring, feedback, and vention fidelity. It highlights the importance of the two key compo- jointly reviewing the management plans of example patients.15,16 nents of stratified care (subgrouping and targeted treatment) and These processes had all been informed by a pilot study before the makes recommendations for future research to determine feasibility main STarT Back trial that demonstrated the feasibility and accept- ability of the approach.17 and optimise intervention fidelity before conducting full-scale RCTs. The stratified care model in OCTOPuS was based on empirically Intervention fidelity has been a challenge in all subsequent stratified care trials in the field. A subsequent quality improvement observed phenotypes (distinct clinical profiles), a valid stratification study in the UK (the IMPaCT Back study) showed similar results as the algorithm (robust categorisation of participants into subgroups) and STarT Back trial but in real-life general practice,18 although the pos- itive results were attenuated by the lower intervention fidelity previously effective exercise interventions (evidence-based targeted observed (general practitioners followed the risk stratification tool’s treatment options).1 Participants were phenotyped into three sub- recommendation for matched treatment in 71% of participants). The MATCH trial in the USA tested an adapted STarT Back approach but, groups for targeted treatment: the ‘high muscle strength’ subgroup despite education and training of clinicians and embedding the was matched to a targeted program of three to five home-based ex- subgrouping tool into the electronic health record, only about half of ercise sessions over 12 weeks and one booster session; the ‘low participants were successfully subgrouped and there was no evidence muscle strength’ subgroup was matched to eight to 12 supervised that clinical management changed based on the subgroup.4 The sessions plus home exercise sessions and one to two booster sessions; TARGET trial in the USA focused on acute, high-risk patients with low back pain (ie, one patient subgroup) for whom the intended matched and the ‘obesity subgroup’ was matched to 12 to 18 individual exer- intervention was psychologically informed physiotherapy.5 Despite cise sessions with two to three booster sessions plus five to eight creating an automated process to identify the subgroup and generate dietitian sessions and at least one joint consultation with both a psychologically informed physiotherapy referrals, only 36% of par- ticipants in the stratified care arm were referred for psychologically physiotherapist and dietitian. The trial compared a stratified informed physiotherapy and 30% of controls were also referred for approach versus usual exercise therapy involving an average of 10 it. Intervention fidelity was well below expectations, leading the TARGET trial authors to reflect that they had ‘balanced the physiotherapy sessions of education and exercises designed around strength, fitness, function and balance/stabilisation training. The OCTOPuS results showed no added value of stratified care for clinical1 or cost-effectiveness outcomes.13 Frequently reported protocol vio- lations in the stratified care arm were the delivery of too few phys- iotherapy and dietary sessions, no booster sessions and barriers to interprofessional collaboration.1,14 Key questions to ask of trials of stratified care are ‘Were the participants successfully subgrouped?’ and ‘Did they receive the https://doi.org/10.1016/j.jphys.2023.02.014 1836-9553/© 2023 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/4.0/).
66 Editorial pragmatic-explanatory continuum of their RCT more towards prag- based on the results of those RCTs that were, for the most part, matic with regard to flexibility in delivery of the intervention’.5 explanatory trials. Yet most studies that evaluate the effect of implementation strategies of guideline recommendations in RCTs A Danish trial of stratified care amended the STarT Back approach show no benefit on patient clinical outcomes.21–23 We don’t tend to for low back pain for their setting, with training for healthcare pro- then conclude that guidelines are unhelpful, but that we have yet to viders supported by the same training lead from the original UK STarT learn much about the art and science of successful implementation Back trial.6 That trial found no positive effects from stratified care but into complex healthcare systems.24 Another learning might be that noted that treatment was similar in the two trial arms and that, due for trials of stratified care (or indeed any complex intervention), the sample size should be inflated for the likely lower rates of interven- to slow recruitment, clinicians reported insufficient opportunities to tion fidelity that are typically seen in RCTs conducted in real-world practice and develop their confidence and competence in delivering settings. However, this may make conducting such trials neither the matched treatments.6 feasible nor fundable. The UK STarT MSK trial compared an adapted model of stratified It is therefore proposed that future attempts to test the effec- versus usual primary care for adults with the five most common tiveness of stratified care need to give stratified care a ‘fair trial’. It is musculoskeletal pain presentations.19 General practitioners were suggested that more careful staging of the research is needed (rather introduced to stratified care through training sessions, facilitated by like in the cancer field), whereby efficacy studies that are more highly embedding the STarT MSK subgrouping tool in electronic pop-up controlled are needed first (to ensure that participants with the computer templates that recommended matched treatment options phenotype (subgroup) are clearly identified and then receive the treatment they are supposed to), to determine whether there is merit for each participant (there were 15 matched treatment options), and in the intervention and, only if so, then subsequent studies focus on general practitioners were invited to share and discuss feedback on the feasibility of delivering the stratified care intervention. Only when feasibility, at least initially, is shown or improved such that there is their intervention fidelity. Whilst some aspects of clinical decision- adequate intervention fidelity, should larger and more expensive making about treatment changed in a positive direction, there were RCTs be conducted that test the clinical effectiveness of the stratified care intervention on clinical outcomes with attention to careful no overall benefits for participants’ clinical outcomes. The general intervention fidelity, followed by studies that then focus on imple- practitioners used the subgrouping tool for only 30% of all eligible mentation in real-world clinical settings as the primary aim (such as hybrid type 3 implementation trials).24 Such a staged approach is participants, highlighting the challenge of incorporating risk stratifi- similar to the phases recommended in the framework for the cation into their workflows. Only those who completed question- development and evaluation of complex interventions.25 New trial naires, in which the STarT MSK tool was embedded, and provided designs, such as adaptive platform trials, might mean that it becomes consent to participate were included in the trial, and in the stratified more possible to add in new subgroups with matched treatment arms care arm general practitioners reported selecting a matched treat- without stopping and starting a new trial (with all the expense and ment for 77% of those participants.19 Whilst treatment fidelity was effort)26 but intervention fidelity will continue to be a pervasive moderate, it is unlikely that this alone was the reason for the lack of challenge. It is recommended that researchers, clinicians and funders positive findings. Other explanations could be the lack of the effec- better understand and mitigate the challenge of intervention fidelity tiveness of the matched treatments, and the way in which the in future RCTs of stratified care. Tools such as the PRagmatic stratification was restricted to the initial primary care decision alone, Explanatory Continuum Indicator Summary-2 (PRECIS-2) Provider without any further system-level changes in the services that par- Strategies could be helpful aids in the design of future stratified care ticipants were referred to, given their subgroup. trials.27 Lastly, a Dutch stratified care trial found conflicting results with Footnotes: Nil. positive reductions for pain intensity only.7 The trial compared a new Ethics approval: Not applicable. Competing interests: The authors are active researchers in the stratified blended intervention for low back pain (stratified using the field of stratified care. STarT Back tool) comprising a smartphone app integrated with face- Source(s) of support: Nil. Acknowledgements: Nil. to-face physiotherapy versus face-to-face physiotherapy alone. The Provenance: Not invited. Peer reviewed. authors noted that there were few differences in the treatment across Correspondence: Nadine E Foster, STARS Education and Research Alliance, Surgical Treatment and Rehabilitation Service (STARS), The the two trial arms, again suggesting that treatment targeting (where University of Queensland and Metro North Health, Brisbane, treatment is different in stratified care than control and is different Australia. Email: [email protected] for different subgroups of patients) was not successfully achieved.7 References With the addition of the OCTOPuS stratified care trial,1 there is a pattern emerging of similar conclusions that stratified care is not 1. Knoop J, et al. J Physiother. 2022;68:182–190. clinically effective. Is it time to give up on stratified care or are we 2. Foster NE, et al. Best Prac Res Clin Rheum. 2013;27:649–661. failing to give it a fair trial? Perhaps stratified care is truly not su- 3. Hill JC, et al. Lancet. 2011;378:1560–1571. perior to usual care. As proposed by Knoop et al,1 usual physiotherapy 4. Cherkin D, et al. J Gen Intern Med. 2018;33:1324–1336. may not be a ‘one-size-fits-all’ approach but already personalised, for 5. Delitto A, et al. E Clin Med. 2021;34:100795. example in ways that better harness the understanding of the 6. Morsø L, et al. Eur J Pain. 2021;25:2020–2038. importance of an individual patient’s beliefs, so that every patient 7. Koppenaal T, et al. J Med Internet Res. 2022;24:e31675. receives a treatment targeting their personal needs, obstacles to re- 8. Waddell G. J Rheum. 2005;32:395–396. covery and preferences. Thus, no further models of stratified care are 9. Fletcher S, et al. Lancet Psychiatry. 2021;8:202–214. needed. However, if future research aims to develop and test new 10. Brandon R, et al. Front Endocrinol. 2023;13:1091421. models of stratified care that better target treatment to subgroups of 11. Keller RB, et al. JCO Precis Oncol. 2023;7:e2200342. participants, the field should learn from this suite of trials. Stratified 12. Hingorani AD, et al. BMJ. 2013;346:e5793. care requires changes to clinical workflows, training of healthcare 13. Knoop J, et al. Braz J Phys Ther. 2022;27:100469. providers and system-level changes such as different referral pro- 14. Knoop J, et al. BMC Musculoskelet Disord. 2022;23:559. cesses. Yet, to date, stratified care trials have: lacked monitoring or 15. Main CJ, et al. Physiotherapy. 2012;98:110–116. feedback to encourage clinicians to deliver matched treatments with 16. Sowden G, et al. Fam Pract. 2012;29:50–62. high fidelity;1,20 included numerous treatment options, some of 17. Hay EM, et al. BMC Musculoskelet Disord. 2008;9:58. which are difficult to access;4,5,6,20 experienced contamination be- 18. Foster NE, et al. Ann Fam Med. 2014;12:102–111. tween intervention arms;1,5,6 and lacked contrast in the matched 19. Hill JC, et al. Lancet Rheum. 2022;4:e591–e602. treatments for different subgroups.6,19 It could be argued that if stratified care interventions are so complex to deliver in the real world, even if a carefully controlled explanatory RCT shows benefit, then the intervention is likely not worthwhile anyway. However, this is rather like the conundrum of clinical guidelines. Guidelines review individual RCTs (or reviews of individual RCTs), offer recommendations for best clinical practice
Editorial 67 20. Hsu C, et al. Pain Med. 2019;20:1105–1119. 24. Hassett L, et al. J Physiother. 2022;3:210–212. 21. Fillipo R, et al. Implement Sci Commun. 2022;3:57. 25. Skivington K, et al. BMJ. 2021;374:n2061. 22. Suman A, et al. Implement Sci. 2016;11:126. 26. Pallmann P, et al. BMC Med. 2018;16:29. 23. Dziedzic KS, et al. Osteo Cartil. 2018;26:43–53. 27. Norton WE, et al. Implement Sci. 2021;16:7.
Journal of Physiotherapy 69 (2023) 127 j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j p hy s Appraisal Clinimetrics: The Breathing Pattern Assessment Tool Summary the patient are required.1 The BPAT is free to use with the developer’s permission and the tool is published in the supplementary material in Description: The Breathing Pattern Assessment Tool (BPAT) is the original publication.1 designed to identify and objectively measure abnormal breathing patterns.1 It is administered by a clinician who observes and scores a Clinimetric properties: The BPAT has been validated against patient’s breathing pattern whilst they are at rest. The BPAT was devised in 2018 by physiotherapists at The Royal Brompton Hospital, expert physiotherapist-diagnosed BPD in patients with asthma UK, as a practical clinical screening tool for breathing pattern disorder and/or unexplained dyspnoea.1 BPAT scores of 4 have a sensitivity (BPD) in adults with asthma and/or unexplained dyspnoea.1 Since of 0.92 and a specificity of 0.75 for diagnosing BPD in this population, then, the BPAT has primarily been used in this patient population;2,3 and these test characteristics have been reproduced in patients post however, recent interest in BPD has seen the tool used in patients COVID-19 infection.4 Higher BPAT scores have been associated with with other respiratory conditions including post COVID-19 infection.4 No special training, equipment or manual assessment is required to increased breathlessness and hyperventilation symptoms, reduced administer the BPAT, and observations take approximately 1 minute quality of life and exercise capacity, and worse asthma control.1,2 In to complete. patients with asthma, the tool has excellent inter-rater reliability for Instructions and scoring: The BPAT measures seven core biomechanical breathing characteristics that contribute to BPD: chest in-person assessment between physiotherapists (ICC = 0.95, 95% CI wall movement (apical versus abdominal), channels of respiration (oral versus nasal), noise of inspiratory flow, noise of expiratory flow, 0.91 to 0.98), with a second assessor scoring within 1.4 points of the signs of air hunger, respiratory rate and rhythm of breathing. Items first on 95% of occasions.3 The BPAT has good reliability when are scored from 0 to 2 (0 = normal; 2 = present in severe BPD), then conducted by video assessment (ICC = 0.76, 95% CI 0.56 to 0.88), summed to give an overall score out of 14; higher scores indicate an abnormal breathing pattern. The BPAT is evaluated at rest during although less than in-person assessment due primarily, but not quiet breathing with the patient sitting comfortably in a supported seat for 5 minutes prior to assessment, and no special instructions to exclusively, to reduced clarity of sound recordings. On 95% of occasions the video score will be within 2.9 points of the in-person score. Exclusion of respiratory flow sounds improves reproducibility of video assessment.3 Responsiveness to treatment and the minimum clinically important difference in scores to determine meaningful change are unknown. Commentary There is increasing awareness of the need to objectively assess BPD, limitations, it is a relevant tool for physiotherapists working in a and the BPAT is a practical and reliable tool for both clinical and variety of settings to easily measure biomechanical breathing patterns research settings. It does not require specialist training or equipment to in an objective manner and evaluate the effect of interventions. administer, there are no risks or special instructions, and it takes approximately 1 minute to complete while the patient is seated and Provenance: Invited. Not peer reviewed. resting. Telehealth consultations are increasingly being utilised for patient consultations and the BPAT can be reliably administered using Janet Bondarenkoa,b and Anne E Hollanda,b,c videoconferencing technology, provided issues around the sound aPhysiotherapy Department, Alfred Health, Melbourne, Australia recording are taken into consideration, enabling remote objective assessments. Further research is required to determine the validity of bRespiratory Research@Alfred, Central Clinical School, Monash the BPAT in diagnosing BPD in health conditions other than asthma, University, Melbourne, Australia paediatric populations and against other measurements of breathing pattern such as optoelectronic plethysmography. The subscales of each cInstitute for Breathing and Sleep, Melbourne, Australia breathing characteristic in the tool have not been explored; determining the contribution of individual measurements to overall References breathing pattern abnormalities and patient symptoms may provide information to help guide treatment interventions or evaluate the 1. Todd S, et al. Respirology. 2018;23:284–290. effect of treatments in the future. Finally, the BPAT is a relatively new 2. Sedeh FB, et al. Respir Med. 2020;163:105894. instrument, with little known about its responsiveness to treatment 3. Bondarenko J, et al. Clin Exp Allergy. 2021;51:1218–1220. including the minimum clinically important difference. Despite these 4. Hylton H, et al. Clin Med. 2022. https://doi.org/10.7861/clinmed.2021-0759 https://doi.org/10.1016/j.jphys.2022.11.006 1836-9553/© 2022 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/4.0/).
Journal of Physiotherapy 69 (2023) 126 j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j p hy s Appraisal Clinimetrics: The Patient-Specific Functional Scale Summary Disability and Health. A recent systematic review5 found, with low to The Patient-Specific Functional Scale (PSFS) is an individualised moderate certainty, that the PSFS has sufficient test-retest reliability measure of functional limitation widely used in physiotherapy and (ICC = 0.71 to 0.97) and sufficient responsiveness in musculoskeletal rehabilitation.1 Respondents are asked to self-identify three to five ac- conditions such as shoulder pain, neck pain, cervical radiculopathy, tivities they have difficulty performing due to their condition, and then rate each activity on an ordinal scale of 0 (‘unable to perform activity’) to knee amputations and low back pain. It has similar or sometimes 10 (‘able to perform activity at the same level as before injury or disease’). If patients struggle to nominate activities, an activity list from relevant better responsiveness than condition-specific scales in low back pain condition-specific scales (eg, Oswestry Disability Index) can be read to and neck pain.6,7 The minimum important change for musculoskel- them. Scores are averaged and higher scores indicate lesser disability. During repeat assessments, patients are generally not informed of their etal conditions is a 2-point difference in the average score or a 3- initial scores. The scale is freely available, takes less than 5 minutes to complete, does not require any special training, has been translated into point difference for a single activity; however, the evidence for its at least 12 languages,2–5 and can either be self-administered or construct validity is uncertain.5 completed via interviews. It has been used in 87 unique health condi- tions but without evidence of validity and reliability in many conditions.5 Although used in paediatric, neurological and cardiovascular The PSFS predominantly measures the ‘activity limitation’ conditions, the measurement properties of the PSFS in these pop- component of the International Classification of Functioning, ulations are not well studied.5 Evidence with very low certainty suggests that it has sufficient reliability in Parkinson’s disease and coronary artery disease, but insufficient reliability in chronic obstructive pulmonary disease. Responsiveness in Parkinson’s disease was found to be insufficient in a small study.8 Commentary The PSFS offers distinct advantages as an intuitive, patient-centred Overall, the PSFS is a useful clinical tool with which to assess ac- tivity limitations unique to an individual with musculoskeletal con- and personalised scale with minimal patient and clinician burden. The ditions. As it is not intended to provide a comprehensive assessment, scale encourages clinicians and patients to focus on functional activities it should be used alongside other appropriate condition-specific scales that assess impairments and general health. Its utility in pae- important to the patient when planning treatment. Given its simplicity, diatric, neurological and cardiopulmonary conditions is uncertain and it may be easier to complete than longer and nuanced condition- needs further exploration. specific scales, especially for patients from diverse cultural Anupa Pathaka and Saurab Sharmab,c backgrounds and those with low literacy. Overall, it has the potential to aSchool of Public Health, Faculty of Medicine and Health, University of be used in a wide variety of clinical populations, health conditions Sydney, Sydney, Australia and health settings; it can also be used during telerehabilitation, bDepartment of Exercise Physiology, School of Health Sciences, Faculty of Medicine and Health, University of New South Wales, Sydney, Australia as telephone administration has also been shown to be reliable and valid.5 cCentre for Pain IMPACT, Neuroscience Research Australia, Sydney, Australia Despite the potential, there are limitations to its clinical utility. The References first is the uncertainty around its construct validity. In a systematic re- view, the a priori hypotheses that the PSFS would strongly correlate 1. Stratford P, et al. Physiother Can. 1995;47:258–263. 2. Sharma S, et al. J Orthop Sports Phys Ther. 2018;48:659–664. (coefficient . 0.50) with other measures of physical function/disability 3. Alnahdi AH, et al. Disabil Rehabil. 2021;1–8. were not met.5 One plausible reason for this is that in the PSFS, patients 4. Lehtola V, et al. Eur J Physiother. 2013;15:134–138. 5. Pathak A, et al. J Orthop Sports Phys Ther. 2022;52:262–275. select activities that are most affected and highly specific to them. This 6. Cleland JA, et al. Spine. 2006;31:598–602. results in lower baseline scores (ie, greater disability) compared with 7. Hall AM, et al. Eur Spine J. 2011;20:79–86. 8. Burgos-Martinez G. McMaster University (MSc Thesis). 2011. fixed-item scales. Therefore, the correlations with fixed-item scales are 9. Abbott JH, et al. J Clin Epidemiol. 2014;67:681–688. weaker at baseline than at follow-up.5,9 Second, a criticism of the PSFS is 10. Pathak A, et al. Qual Life Res. 2021;30:613–628. that multiple patient data cannot be combined; however, one study from New Zealand reported that PSFS scores can be aggregated for valid group-level comparisons.9 Third, the measurement properties of the scale have not yet been studied in several musculoskeletal and non- musculoskeletal conditions,5 especially its utility in progressive dis- eases. Fourth, the 0 to 10 numeric response format may be problematic for many cultures, especially for uneducated and older adults.10 Finally, the reference to ‘before illness or injury’ might be confusing or inappli- cable for patients with congenital and chronic conditions.10 https://doi.org/10.1016/j.jphys.2022.07.001 1836-9553/© 2022 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/4.0/).
Journal of Physiotherapy 69 (2023) 124 j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j p hy s Appraisal Critically appraised paper: In adults on antidepressants for mild-to-moderate depression, yoga improves several aspects of quality of life Synopsis Summary of: Kumari N, Yadav A, Garg N, Yadav K, Sankhla M, Goothy (physical health, psychological health, social relationships and SSK, et al. Effectiveness of yoga on quality of life among patients with environment), with the scores in each domain ranging from 0 (worst) depression: a randomised controlled trial. J Clin Diagnostic Res. to 100 (best). Participants also gave evaluations of overall quality of life 2022;16:CC22–CC27. and general health, each on a 5-point Likert scale. Results: Sixty participants completed the study. At 3 months, the yoga group had Question: Does a 12-week program of yoga improve quality of life in better improvement on the physical health domain (MD 19, 95% CI 15 adults taking antidepressants for depression? Design: Randomised to 23) and the psychological health domain (MD 13, 95% CI 10 to 17) controlled trial with concealed allocation. Setting: Participants than the control group. Those in the yoga group also improved were recruited from the psychiatry outpatient department of a their general health score more than the control group (MD 1.0, 95% CI medical college in India. Participants: The inclusion criteria were 0.8 to 1.3). The between-group difference was unclear on the mild-to-moderate depression diagnosed using the Hamilton remaining outcomes. Conclusion: A 12-week course of five sessions Depression Rating Scale, aged 18 to 45 years, and receiving per week of supervised yoga improves physical, psychological and stable treatment with conventional antidepressant medication for general health in people who are taking antidepressant medication for 1 month. Severe mental or medical comorbidities and receipt of mild-to-moderate depression. [Mean difference in change scores and cognitive behavioural therapy or psychotherapy were exclusion 95% CIs calculated by the Editor.] criteria. Randomisation of 60 participants allocated 30 to a yoga group and 30 to a control group. Interventions: Both groups received Provenance: Invited. Not peer reviewed. conventional antidepressant medication. Those in the yoga group were asked to also practise yoga postures for 15 minutes, breathing Mark Elkins exercises for 20 minutes and meditation for 20 minutes per session, at Editor, Journal of Physiotherapy least 5 days per week, under the supervision of a trained yoga instructor. Outcome measures: The primary outcome was quality of https://doi.org/10.1016/j.jphys.2023.02.005 life measured at 1 and 3 months with the World Health Organisation Quality of Life-BREF (WHOQOL-BREF), which has four domains 1836-9553/© 2023 Published by Elsevier B.V. on behalf of Australian Physiotherapy Association. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/). Commentary psychotherapies are typically considered first-line treatment ahead of pharmacotherapy. Future studies should consider the use of The intervention in this trial was quite intensive, but it may still be psychotherapy with yoga. Other considerations include supervised cost-effective in the public health system. A qualified yoga instructor compared with independently performed yoga, and follow-up to would be an inexpensive addition in comparison to existing medical assess continuation of yoga. staff and also cater for large groups. Unlike a structured psychotherapy group program, yoga would have flexible entry points Provenance: Invited. Not peer reviewed. (useful in outpatient hospital settings) and few or no side effects. Leon Gantert Physiological mechanisms may be responsible for improving National Centre for Veterans’ Healthcare, Concord Repatriation General physical and mental health in depression, such as by enhancing synaptic transmission in the brain (like the mechanism of Hospital, Sydney, Australia anti-depressant medications) or by increasing endorphin excretion after exercise.1 Improvements in body weight and fitness may also https://doi.org/10.1016/j.jphys.2023.02.006 contribute. As participants were yoga naïve, skill mastery and frequent distraction may have contributed to improvements in the Reference psychological domain. 1. Craft L, Perna F. Prim Care Companion J Clin Psychiatry. 2004;6:104–111. The authors did not specify which antidepressant medication(s) were used or the yoga intensity, which hampered external validity. Also, patients being managed with psychotherapy or cognitive behavioural therapy were excluded. In mild depression, 1836-9553/Crown Copyright © 2023 Published by Elsevier B.V. on behalf of Australian Physiotherapy Association. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Journal of Physiotherapy 69 (2023) 124 j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j p hy s Appraisal Critically appraised paper: In adults on antidepressants for mild-to-moderate depression, yoga improves several aspects of quality of life Synopsis Summary of: Kumari N, Yadav A, Garg N, Yadav K, Sankhla M, Goothy (physical health, psychological health, social relationships and SSK, et al. Effectiveness of yoga on quality of life among patients with environment), with the scores in each domain ranging from 0 (worst) depression: a randomised controlled trial. J Clin Diagnostic Res. to 100 (best). Participants also gave evaluations of overall quality of life 2022;16:CC22–CC27. and general health, each on a 5-point Likert scale. Results: Sixty participants completed the study. At 3 months, the yoga group had Question: Does a 12-week program of yoga improve quality of life in better improvement on the physical health domain (MD 19, 95% CI 15 adults taking antidepressants for depression? Design: Randomised to 23) and the psychological health domain (MD 13, 95% CI 10 to 17) controlled trial with concealed allocation. Setting: Participants than the control group. Those in the yoga group also improved were recruited from the psychiatry outpatient department of a their general health score more than the control group (MD 1.0, 95% CI medical college in India. Participants: The inclusion criteria were 0.8 to 1.3). The between-group difference was unclear on the mild-to-moderate depression diagnosed using the Hamilton remaining outcomes. Conclusion: A 12-week course of five sessions Depression Rating Scale, aged 18 to 45 years, and receiving per week of supervised yoga improves physical, psychological and stable treatment with conventional antidepressant medication for general health in people who are taking antidepressant medication for 1 month. Severe mental or medical comorbidities and receipt of mild-to-moderate depression. [Mean difference in change scores and cognitive behavioural therapy or psychotherapy were exclusion 95% CIs calculated by the Editor.] criteria. Randomisation of 60 participants allocated 30 to a yoga group and 30 to a control group. Interventions: Both groups received Provenance: Invited. Not peer reviewed. conventional antidepressant medication. Those in the yoga group were asked to also practise yoga postures for 15 minutes, breathing Mark Elkins exercises for 20 minutes and meditation for 20 minutes per session, at Editor, Journal of Physiotherapy least 5 days per week, under the supervision of a trained yoga instructor. Outcome measures: The primary outcome was quality of https://doi.org/10.1016/j.jphys.2023.02.005 life measured at 1 and 3 months with the World Health Organisation Quality of Life-BREF (WHOQOL-BREF), which has four domains 1836-9553/© 2023 Published by Elsevier B.V. on behalf of Australian Physiotherapy Association. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/). Commentary psychotherapies are typically considered first-line treatment ahead of pharmacotherapy. Future studies should consider the use of The intervention in this trial was quite intensive, but it may still be psychotherapy with yoga. Other considerations include supervised cost-effective in the public health system. A qualified yoga instructor compared with independently performed yoga, and follow-up to would be an inexpensive addition in comparison to existing medical assess continuation of yoga. staff and also cater for large groups. Unlike a structured psychotherapy group program, yoga would have flexible entry points Provenance: Invited. Not peer reviewed. (useful in outpatient hospital settings) and few or no side effects. Leon Gantert Physiological mechanisms may be responsible for improving National Centre for Veterans’ Healthcare, Concord Repatriation General physical and mental health in depression, such as by enhancing synaptic transmission in the brain (like the mechanism of Hospital, Sydney, Australia anti-depressant medications) or by increasing endorphin excretion after exercise.1 Improvements in body weight and fitness may also https://doi.org/10.1016/j.jphys.2023.02.006 contribute. As participants were yoga naïve, skill mastery and frequent distraction may have contributed to improvements in the Reference psychological domain. 1. Craft L, Perna F. Prim Care Companion J Clin Psychiatry. 2004;6:104–111. The authors did not specify which antidepressant medication(s) were used or the yoga intensity, which hampered external validity. Also, patients being managed with psychotherapy or cognitive behavioural therapy were excluded. In mild depression, 1836-9553/Crown Copyright © 2023 Published by Elsevier B.V. on behalf of Australian Physiotherapy Association. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Journal of Physiotherapy 69 (2023) 123 j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j p hy s Appraisal Critically appraised paper: In patients with chronic obstructive pulmonary disease, the addition of oscillatory positive expiratory pressure to usual care improves cough-specific and general quality of life Synopsis Summary of: Alghamdi SM, Alsulayyim AS, Alasmari AM, Philip KEJ, online video. The intervention window was 3 months from Buttery SC, Banya WAS, et al. Oscillatory positive expiratory pressure randomisation in both groups. Outcome measures: The primary therapy in COPD (O-COPD): a randomised controlled trial. Thorax. outcome was cough-related quality of life, measured using the 2023;78:136–143. https://doi.org/10.1136/thorax-2022-219077. Leicester Cough Questionnaire. Secondary outcome measures included a general quality of life measure with the EuroQol 5 Dimensions 5 Question: In patients with stable chronic obstructive pulmonary Levels. Results: A total of 103 participants completed the study (55 in disease who regularly produce sputum, does the addition of oscillatory the experimental group and 48 in the control group). At 3 months, positive expiratory pressure to usual care improve cough-specific and compared with the control group, the experimental group general quality of life? Design: Randomised controlled trial with demonstrated better scores for cough-related quality of life (MD 1.03, concealed allocation and blinded assessor. Setting: A single tertiary 95% CI 0.71 to 2.10) and general quality of life (MD 4.0, 95% CI 0.5 to hospital in the UK. Participants: Patients with stable chronic 19.8). Conclusion: In patients with stable chronic obstructive obstructive pulmonary disease who produce sputum daily or most pulmonary disease who produce sputum regularly, the addition of days. Exclusion criteria were exacerbation or medication change in the oscillatory positive expiratory pressure to usual care improved preceding month, referral for chest physiotherapy in the preceding cough-specific and general quality of life. year, or already using an oscillatory positive expiratory pressure device. Randomisation of 122 participants allocated 61 to the experimental Provenance: Invited. Not peer reviewed. group and 61 to the control group. Interventions: All participants received leaflets on correct inhaler use, smoking cessation and active Vinicius Cavalheri cycle of breathing technique, as well as a link to online instructional Curtin School of Allied Health, Curtin University, Australia videos. In addition, participants in the experimental group received an oscillatory positive expiratory pressure device (Acapella), were taught https://doi.org/10.1016/j.jphys.2023.02.009 how to use the device, and given written information and a link to an Commentary Chronic obstructive pulmonary disease is a heterogenous disease. less frequently than mucolytic therapies,2 despite questionable Despite the common presence of ‘cardinal’ symptoms (dyspnoea, evidence of their benefit.3 Clinicians should be reassured by these cough, sputum), personalised management is essential due to findings: the intervention is simple to replicate (even amidst variability in how the disease affects individuals. Chronic sputum COVID-affected practice changes), the patient subgroup is easily production can be problematic; however, defining the precise identifiable and adverse outcomes were not reported. Additional evidence-informed role of airway clearance techniques to manage confirmatory long-term studies adopting rigorous methodologies this ‘treatable trait’ has been challenging.1 remain indicated in this space. Additional insights are available in its accompanying editorial.4 The randomised controlled trial by Alghamdi and colleagues offers valuable information for those who self-identify as having this chronic Provenance: Invited. Not peer reviewed. sputum production phenotype. It was performed in a tertiary hospital setting in the United Kingdom and suggested that the addition of 3 Christian Osadnik months of Acapella use (an oscillating positive expiratory pressure Physiotherapy, Monash University, Melbourne, Australia device) on top of the active cycle of breathing technique improved clinically relevant outcomes such as cough-related quality of life, https://doi.org/10.1016/j.jphys.2023.02.008 fatigue and self-reported exacerbations during this time. Subjective benefits were triangulated with quantified cough frequency counts in a References subset of participants who underwent objective actigraphy monitoring. However, the magnitude of the observed benefits for 1. Osadnik CR, et al. Cochrane Database Syst Rev. 2012;3:CD008328. several outcomes did not exceed commonly used minimum important 2. OpenPrescribing. Available at: https://openprescribing.net difference thresholds, which warrants consideration. 3. Poole P, et al. Cochrane Database Syst Rev. 2019;5:CD001287. 4. Lewis A, Osadnik CR. Thorax. 2022;78:113–115. Clinical trials do not always replicate usual clinical practice and rarely change practice in isolation. Airway clearance techniques are still under-represented in respiratory guidelines and are prescribed 1836-9553/© 2023 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/4.0/).
Journal of Physiotherapy 69 (2023) 123 j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j p hy s Appraisal Critically appraised paper: In patients with chronic obstructive pulmonary disease, the addition of oscillatory positive expiratory pressure to usual care improves cough-specific and general quality of life Synopsis Summary of: Alghamdi SM, Alsulayyim AS, Alasmari AM, Philip KEJ, online video. The intervention window was 3 months from Buttery SC, Banya WAS, et al. Oscillatory positive expiratory pressure randomisation in both groups. Outcome measures: The primary therapy in COPD (O-COPD): a randomised controlled trial. Thorax. outcome was cough-related quality of life, measured using the 2023;78:136–143. https://doi.org/10.1136/thorax-2022-219077. Leicester Cough Questionnaire. Secondary outcome measures included a general quality of life measure with the EuroQol 5 Dimensions 5 Question: In patients with stable chronic obstructive pulmonary Levels. Results: A total of 103 participants completed the study (55 in disease who regularly produce sputum, does the addition of oscillatory the experimental group and 48 in the control group). At 3 months, positive expiratory pressure to usual care improve cough-specific and compared with the control group, the experimental group general quality of life? Design: Randomised controlled trial with demonstrated better scores for cough-related quality of life (MD 1.03, concealed allocation and blinded assessor. Setting: A single tertiary 95% CI 0.71 to 2.10) and general quality of life (MD 4.0, 95% CI 0.5 to hospital in the UK. Participants: Patients with stable chronic 19.8). Conclusion: In patients with stable chronic obstructive obstructive pulmonary disease who produce sputum daily or most pulmonary disease who produce sputum regularly, the addition of days. Exclusion criteria were exacerbation or medication change in the oscillatory positive expiratory pressure to usual care improved preceding month, referral for chest physiotherapy in the preceding cough-specific and general quality of life. year, or already using an oscillatory positive expiratory pressure device. Randomisation of 122 participants allocated 61 to the experimental Provenance: Invited. Not peer reviewed. group and 61 to the control group. Interventions: All participants received leaflets on correct inhaler use, smoking cessation and active Vinicius Cavalheri cycle of breathing technique, as well as a link to online instructional Curtin School of Allied Health, Curtin University, Australia videos. In addition, participants in the experimental group received an oscillatory positive expiratory pressure device (Acapella), were taught https://doi.org/10.1016/j.jphys.2023.02.009 how to use the device, and given written information and a link to an Commentary Chronic obstructive pulmonary disease is a heterogenous disease. less frequently than mucolytic therapies,2 despite questionable Despite the common presence of ‘cardinal’ symptoms (dyspnoea, evidence of their benefit.3 Clinicians should be reassured by these cough, sputum), personalised management is essential due to findings: the intervention is simple to replicate (even amidst variability in how the disease affects individuals. Chronic sputum COVID-affected practice changes), the patient subgroup is easily production can be problematic; however, defining the precise identifiable and adverse outcomes were not reported. Additional evidence-informed role of airway clearance techniques to manage confirmatory long-term studies adopting rigorous methodologies this ‘treatable trait’ has been challenging.1 remain indicated in this space. Additional insights are available in its accompanying editorial.4 The randomised controlled trial by Alghamdi and colleagues offers valuable information for those who self-identify as having this chronic Provenance: Invited. Not peer reviewed. sputum production phenotype. It was performed in a tertiary hospital setting in the United Kingdom and suggested that the addition of 3 Christian Osadnik months of Acapella use (an oscillating positive expiratory pressure Physiotherapy, Monash University, Melbourne, Australia device) on top of the active cycle of breathing technique improved clinically relevant outcomes such as cough-related quality of life, https://doi.org/10.1016/j.jphys.2023.02.008 fatigue and self-reported exacerbations during this time. Subjective benefits were triangulated with quantified cough frequency counts in a References subset of participants who underwent objective actigraphy monitoring. However, the magnitude of the observed benefits for 1. Osadnik CR, et al. Cochrane Database Syst Rev. 2012;3:CD008328. several outcomes did not exceed commonly used minimum important 2. OpenPrescribing. Available at: https://openprescribing.net difference thresholds, which warrants consideration. 3. Poole P, et al. Cochrane Database Syst Rev. 2019;5:CD001287. 4. Lewis A, Osadnik CR. Thorax. 2022;78:113–115. Clinical trials do not always replicate usual clinical practice and rarely change practice in isolation. Airway clearance techniques are still under-represented in respiratory guidelines and are prescribed 1836-9553/© 2023 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/4.0/).
Journal of Physiotherapy 69 (2023) 125 j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j p hy s Appraisal Critically appraised paper: Timing and coordination training does not enhance standard treatment benefit on mobility in community-dwelling older adults Synopsis Summary of: Brach JS, Perera S, Shuman V, Gill AB, Kriska A, Nadkarni NK, measures were leg strength, 6-minute walk test, hamstring flexibility and et al. Effect of timing and coordination training mobility and physical the figure 8 walk test. Results: A total of 244 participants (122 in each activity among community-dwelling older adults: A randomized clinical group) completed the study. There was negligible between-group differ- trial. JAMA Netw Open. 2022;5:e2212921. ence in gait speed improvement from baseline, 12 weeks (adjusted mean difference [MD] 0.001 m/s, 95% CI –0.03 to 0.03), 24 weeks (adjusted MD Question: Does a strength and endurance program incorporating timing 0.001, 95% CI –0.04 to 0.04) or 36 weeks (adjusted MD –0.003, 95% CI and coordination training improve walking speed in community-dwelling –0.04 to 0.03). There were no significant between-group differences in older adults? Design: Randomised controlled trial with blinded outcome any of the secondary outcomes at any time point. Conclusion: Among assessment. Setting: Outpatient clinical research setting in United States. independently ambulatory community-dwelling older adults, incorpora- Participants: Community-dwelling adults aged 65 years, walking tion of timing and coordination exercise with strength and endurance without physical assistance or use of an assistive device, and usual gait training does not produce superior improvements in walking speed speed between 0.60 and 1.2 m/s. Randomisation of 249 participants compared with standard strength and endurance training alone. allocated 124 to the intervention group and 125 to the control group. Interventions: Both groups received physiotherapist-supervised exercise [95% CIs calculated by the CAP Editor.] sessions, 50 to 60 minutes, twice weekly for 12 weeks, including lower limb strengthening and moderate-intensity treadmill walking. In Provenance: Invited. Not peer reviewed. addition, the intervention group completed task-specific timing and coordination exercise, comprising challenging stepping patterns that Prudence Plummer were progressed in speed, amplitude and complexity as part of the total Department of Physical Therapy, MGH Institute of Health Professions, walking time. The total standing and walking time was matched between groups. Both groups also received 16 education sessions to improve eating USA and physical activity. Outcome measures: The primary outcome was gait speed, measured at baseline, 12, 24 and 36 weeks. Secondary outcome https://doi.org/10.1016/j.jphys.2023.02.004 1836-9553/© 2023 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/4.0/). Commentary Slow gait speed in older adults is predictive of adverse outcomes1 and Although exploratory subgroup analyses are not encouraged by rigorous limits community ambulation. A search of the Physiotherapy Evidence trialists,3 there is also a call to reduce research waste.4 Use of trial data for Database (PEDro) confirms that gait speed can be improved by exercise exploratory analyses (and in meta-analysis) makes more of the effort of interventions,2 but reveals surprisingly few high-quality trials large investigators, participants and funders. enough to estimate differences between different approaches with adequate precision. Further rigorous large trials comparing approaches are needed for greater understanding of the important question about how best to It would make sense that adding task-specific ‘timing and coordina- improve gait and the impact of this for older adults. tion training’ to standard care would be more effective. The absence of between-group differences does not undermine the importance of un- Provenance: Invited. Not peer reviewed. dertaking large trials such as this one; rather, it reinforces that we should not rely on theory alone to make assumptions about intervention effects. Catherine Sherrington Faculty of Medicine and Health, University of Sydney, Sydney, Australia This trial does not mean that it is time to give up on task-specific training. Both groups undertook quite extensive practice relevant to the https://doi.org/10.1016/j.jphys.2023.02.003 task of walking (ie, up to 40 minutes of supervised treadmill walking at a ‘somewhat difficult’ pace twice a week) with the overall volume of References walking practice matched between groups. The ‘timing and coordination’ training may have had a greater impact on more complex gait tasks, 1. Studenski S, et al. JAMA. 2011;305:50–58. which is arguably more relevant to community ambulation, but harder to 2. Rosendahl E, et al. Aust J Physiother. 2006;52:105–113. assess. 3. Sun X, et al. BMJ. 2010;340:c117. 4. Glasziou P, et al. Lancet. 2014;383:267–276. Gait and its rehabilitation are complex. Trial and clinical populations are often heterogeneous. Further exploration may help understand whether there were different effects of the intervention in different participants. 1836-9553/© 2023 Published by Elsevier B.V. on behalf of Australian Physiotherapy Association. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).
Journal of Physiotherapy 69 (2023) 125 j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j p hy s Appraisal Critically appraised paper: Timing and coordination training does not enhance standard treatment benefit on mobility in community-dwelling older adults Synopsis Summary of: Brach JS, Perera S, Shuman V, Gill AB, Kriska A, Nadkarni NK, measures were leg strength, 6-minute walk test, hamstring flexibility and et al. Effect of timing and coordination training mobility and physical the figure 8 walk test. Results: A total of 244 participants (122 in each activity among community-dwelling older adults: A randomized clinical group) completed the study. There was negligible between-group differ- trial. JAMA Netw Open. 2022;5:e2212921. ence in gait speed improvement from baseline, 12 weeks (adjusted mean difference [MD] 0.001 m/s, 95% CI –0.03 to 0.03), 24 weeks (adjusted MD Question: Does a strength and endurance program incorporating timing 0.001, 95% CI –0.04 to 0.04) or 36 weeks (adjusted MD –0.003, 95% CI and coordination training improve walking speed in community-dwelling –0.04 to 0.03). There were no significant between-group differences in older adults? Design: Randomised controlled trial with blinded outcome any of the secondary outcomes at any time point. Conclusion: Among assessment. Setting: Outpatient clinical research setting in United States. independently ambulatory community-dwelling older adults, incorpora- Participants: Community-dwelling adults aged 65 years, walking tion of timing and coordination exercise with strength and endurance without physical assistance or use of an assistive device, and usual gait training does not produce superior improvements in walking speed speed between 0.60 and 1.2 m/s. Randomisation of 249 participants compared with standard strength and endurance training alone. allocated 124 to the intervention group and 125 to the control group. Interventions: Both groups received physiotherapist-supervised exercise [95% CIs calculated by the CAP Editor.] sessions, 50 to 60 minutes, twice weekly for 12 weeks, including lower limb strengthening and moderate-intensity treadmill walking. In Provenance: Invited. Not peer reviewed. addition, the intervention group completed task-specific timing and coordination exercise, comprising challenging stepping patterns that Prudence Plummer were progressed in speed, amplitude and complexity as part of the total Department of Physical Therapy, MGH Institute of Health Professions, walking time. The total standing and walking time was matched between groups. Both groups also received 16 education sessions to improve eating USA and physical activity. Outcome measures: The primary outcome was gait speed, measured at baseline, 12, 24 and 36 weeks. Secondary outcome https://doi.org/10.1016/j.jphys.2023.02.004 1836-9553/© 2023 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/4.0/). Commentary Slow gait speed in older adults is predictive of adverse outcomes1 and Although exploratory subgroup analyses are not encouraged by rigorous limits community ambulation. A search of the Physiotherapy Evidence trialists,3 there is also a call to reduce research waste.4 Use of trial data for Database (PEDro) confirms that gait speed can be improved by exercise exploratory analyses (and in meta-analysis) makes more of the effort of interventions,2 but reveals surprisingly few high-quality trials large investigators, participants and funders. enough to estimate differences between different approaches with adequate precision. Further rigorous large trials comparing approaches are needed for greater understanding of the important question about how best to It would make sense that adding task-specific ‘timing and coordina- improve gait and the impact of this for older adults. tion training’ to standard care would be more effective. The absence of between-group differences does not undermine the importance of un- Provenance: Invited. Not peer reviewed. dertaking large trials such as this one; rather, it reinforces that we should not rely on theory alone to make assumptions about intervention effects. Catherine Sherrington Faculty of Medicine and Health, University of Sydney, Sydney, Australia This trial does not mean that it is time to give up on task-specific training. Both groups undertook quite extensive practice relevant to the https://doi.org/10.1016/j.jphys.2023.02.003 task of walking (ie, up to 40 minutes of supervised treadmill walking at a ‘somewhat difficult’ pace twice a week) with the overall volume of References walking practice matched between groups. The ‘timing and coordination’ training may have had a greater impact on more complex gait tasks, 1. Studenski S, et al. JAMA. 2011;305:50–58. which is arguably more relevant to community ambulation, but harder to 2. Rosendahl E, et al. Aust J Physiother. 2006;52:105–113. assess. 3. Sun X, et al. BMJ. 2010;340:c117. 4. Glasziou P, et al. Lancet. 2014;383:267–276. Gait and its rehabilitation are complex. Trial and clinical populations are often heterogeneous. Further exploration may help understand whether there were different effects of the intervention in different participants. 1836-9553/© 2023 Published by Elsevier B.V. on behalf of Australian Physiotherapy Association. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).
Journal of Physiotherapy 69 (2023) 100–107 j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j p hy s Research External validation and updating of prognostic models for predicting recovery of disability in people with (sub)acute neck pain was successful: broad external validation in a new prospective cohort Roel W Wingbermühle a,b, Alessandro Chiarotto b, Emiel van Trijffel c, Martijn S Stenneberg a,d, Ronald Kan a, Bart W Koes b,e, Martijn W Heymans f a SOMT University of Physiotherapy, Amersfoort, The Netherlands; b Department of General Practice, Erasmus MC, University Medical Center, Rotterdam, The Netherlands; c Ziekenhuisgroep Twente, ZGT Academy, The Netherlands; d Department of Physiotherapy, Human Physiology and Anatomy, Experimental Anatomy Research Department, Vrije Universiteit Brussel, Belgium; e Department of Sports Science and Clinical Biomechanics, Faculty of Health Sciences, University of Southern Denmark, Odense, Denmark; f Department of Epidemiology and Data Science, Amsterdam University Medical Center, VU University Medical Center, Amsterdam, The Netherlands KEY WORDS ABSTRACT External validation Question: Can existing post-treatment prognostic models for predicting neck pain recovery (primarily in Neck pain terms of disability and secondarily in terms of pain intensity and perceived improvement) be externally Clinical prediction model validated and updated at the end of the treatment period and at 6 and 12 weeks of follow-up in a new Prognosis Dutch cohort of people with neck pain treated with guideline-based usual care physiotherapy? Design: Prognostic model External validation and model updating in a new prospective cohort of three previously developed prog- Recovery nostic models. Participants: People with (sub)acute neck pain and registered for primary care physiotherapy treatment. Outcome measures: Recovery of disability, pain intensity, and perceived recovery at 6 and 12 weeks and at the end of the treatment period. Results: Discriminative performance (c-statistic) of the disability model at 6 weeks was 0.73 (95% CI 0.69 to 0.77) and reasonably well calibrated after intercept recalibration. The disability model at 12 weeks and at the end of the treatment period showed discriminative c-statistic performance values of 0.69 (95% CI 0.64 to 0.73) and 0.68 (95% CI 0.63 to 0.72), respectively, and was well calibrated. Pain models and perceived recovery models did not reach acceptable performance. Cervical mobility added value to the disability models and pain catastrophising to the disability and pain models at 6 weeks. Discussion: Broad external validation of the disability model was successful in people with (sub)acute neck pain and clinicians may use this model in clinical practice with reasonable accuracy. Further research is required to assess the disability model’s clinical impact and generalisability, and to identify additional valuable model predictors. Registration: https://osf.io/a6r3k/ [Wingbermühle RW, Chiarotto A, van Trijffel E, Stenneberg MS, Kan R, Koes BW, Heymans MW (2023) External validation and updating of prognostic models for predicting recovery of disability in people with (sub)acute neck pain was successful: broad external validation in a new prospective cohort. Journal of Physiotherapy 69:100–107] © 2023 Published by Elsevier B.V. on behalf of Australian Physiotherapy Association. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Introduction Prognostic models provide a personalised evidence-based approach by combining multiple predictors simultaneously to esti- Neck pain is common and remains one of the leading causes of mate a patient’s future individual outcomes (eg, neck pain intensity disability in most countries.1,2 Its burden is likely to increase even or neck pain-related disability).9,10 Several prognostic models for neck further, warranting greater need for rehabilitation services in primary pain have been developed; however, methodological shortcomings care.3,4 Identification at intake of patients with neck pain who are are common (eg, small sample size, no correction of overfitting, lack unlikely to recover enables personalised care and supports the of reporting of key performance measures, and limitations in mea- improvement of health outcomes with potential to reduce its burden. surement of predictors and outcomes) and very few models have Recovery of acute neck pain mainly takes place in the first few weeks, been externally validated.11,12 Recently, a model for people with neck otherwise prognosis becomes worse potentially leading to persistent pain was developed and internally validated in a Dutch cohort of pain and disability.5,6 Prognostic factors for predicting neck pain re- patients treated with manual therapy, predicting post-treatment re- covery have largely been established;7,8 however, individual factors covery of disability with good discriminative performance.13 This cannot provide sufficient information to be used for accurate indi- disability model may have good potential to inform primary care vidualised outcome predictions. clinicians about individual prognoses of people with neck pain after https://doi.org/10.1016/j.jphys.2023.02.002 1836-9553/© 2023 Published by Elsevier B.V. on behalf of Australian Physiotherapy Association. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).
Research 101 Table 1 The three models that underwent external validation. Model name Model calculation Area under the curve (95% CI) R2 (95% CI) 0.74 (0.72 to 0.75) 0.21 (0.19 to 0.23) Disability model (DModel) –2.64 1 0.28*Subacute pain 1 0.88*Chronic pain 1 0.11*Baseline disability 1 0.02*Age 1 0.28*Sleeping 0.67 (0.66 to 0.69) 0.09 (0.08 to 0.11) Pain model (PModel) problem 1 0.02*FABQ-PA 0.67 (0.65 to 0.69) 0.09 (0.07 to 0.11) Perceived improvement model (PIModel) –5.94 1 0.18*Subacute pain 1 0.83*Chronic pain 1 0.16*Baseline pain intensity 1 0.34*BNQ-AD 1 0.01*Age 4.54 1 0.14*Subacute pain 1 0.82*Chronic pain 1 0.35*Low back pain 1 0.03*FABQ-PA 1 0.01*Age – 0.03*Baseline disability – 0.4*Previous episode 1 0.33*Sporting activities treatment. However, a model’s broad external validation is a crucial manual therapy at SOMT University of Physiotherapy, Amersfoort (the step before it can be advocated for clinical use14 and there should be Netherlands). an ongoing process of model validation and updating.15,16 Models for recovery of pain and perceived improvement have also been devel- Inclusion criteria were: primary complaints of neck pain grade 1, 2 oped but they do not meet commonly used thresholds for discrimi- or 3 according to the Neck Pain Task Force;19 age 18 years; and native performance criteria; however, they still exhibit reasonable minimum 3 days to maximum 12 weeks of neck pain. Exclusion performance and may benefit from model updating.13 criteria were: past or current cervical fractures; congenital disorders affecting cervical functioning; systemic diseases or neurological dis- Therefore, the research question for this study was: orders affecting cervical functioning; past or actual malignancy; and previous cervical surgery. Patients received interventions in accor- Can existing post-treatment prognostic models for predicting dance with the ‘Clinical Practice Guideline for Physical Therapy neck pain recovery (primarily in terms of disability and second- Assessment and Treatment in Patients with Nonspecific Neck Pain’.20 arily in terms of pain intensity and perceived improvement) be This included information on the benign nature of the condition, externally validated and updated at the end of the treatment advice to stay active, neck muscle strengthening exercises and period and at 6 and 12 weeks of follow-up in a new Dutch cohort cervical spine mobilisations. No high-velocity low-amplitude in- of people with neck pain treated with guideline-based usual care terventions were applied. Participant characteristics and the models’ physiotherapy? predictors were collected by the physiotherapist at baseline and 6 weeks, and models’ outcomes (Dutch version of the Neck Disability Method Index, Numeric Pain Rating Scale and Global Perceived Effect) at 3, 6 and 12 weeks, 6 months, and at the end of the treatment period (only An external validation study of three internally validated models for Neck Disability Index and Numeric Pain Rating Scale), which was for recovery of neck pain was performed. This study was prospec- defined by date of discharge. tively registered on the Open Science Framework and is reported according to the Transparent Reporting of a multivariable prediction Validation procedure model for Individual Prognosis Or Diagnosis (TRIPOD) recommen- dations.17 The models that were externally validated were a disability We described and compared case-mix differences (ie, participant model (DModel), a pain model (PModel) and a perceived improve- characteristics data and outcome occurrence) and study characteris- ment model (PIModel), as shown in Table 1. tics (ie, recruitment period, setting, inclusion/exclusion criteria and treatment) between the development and validation cohorts, and we Development cohort tested the models’ performance in the validation cohort by examining discrimination, calibration and overall performance measures. The The models were previously developed in ANIMO, a prospective number of events in the recovered and non-recovered disability, pain cohort study with 12-month follow-up that ran during the period and perceived improvement outcome groups for a minimum of 100 2007 to 2009, where 345 manual therapists in the Netherlands each to 200 events was checked a priori at each follow-up of the validation recruited five consecutive patients aged between 18 and 80 years cohort, as advised for validation studies that predict binary out- (total n = 1,311) who presented to physiotherapy clinics with non- comes.21 External validation at 6 weeks, 12 weeks and at the end of specific neck pain of any duration, with or without arm pain. Par- the treatment period (if , 12 weeks) was performed. We also eval- ticipants were required to provide baseline data and sign informed uated whether the models could be updated by adding additional consent to be deemed eligible (n = 1,193). Patients with red flags were potential predicting variables. excluded and no serious pathology was assumed. Participants received usual care manual therapy, such as specific joint mobi- Outcomes lisations, high-velocity thrust techniques, myofascial techniques, advice or exercise. Follow-up timepoints were at the end of the Comparable with the derivation study, ‘recovery’ was used as an treatment period and at 12-months follow-up. More detail on this umbrella term for three different constructs: pain intensity, disability cohort is provided elsewhere.18 and perceived improvement. The outcome measure defining recovery with respect to pain was the Numeric Pain Rating Scale (11-point Validation cohort Likert scale), dichotomised into 2 (recovered) or . 2 (not recov- ered). The outcome measure defining recovery with respect to For external validation, data from the PRONEPA cohort, which ran disability was the Neck Disability Index (0 to 50 scale, transformed to from November 2020 to April 2021, with a 12-week follow-up was % by multiplying by 2) dichotomised into , 8% (recovered) or 8% used. PRONEPA was a prospective cohort study (registered at https:// (not recovered). The outcome measure defining recovery with respect osf.io/u8rnw/, ethics committee permission METCZ20200178) that to perceived improvement was Global Perceived Effect (7-point Likert primarily aimed to evaluate prognostic factors that predict develop- scale), where the response options ‘very much better’ or ‘much better’ ment of chronic neck pain in people with (sub)acute neck pain (, 12 defined recovery. weeks), with or without radicular symptoms, registered for physio- therapy treatment. PRONEPA 2020 to 2021 included a convenience Comparison of study characteristics sample of 586 participants with neck pain, recruited by 102 physio- therapists who were graduating from a Master of Science program in Both cohorts were recruited by students graduating from a Master of Science program in manual therapy at the same institution. Usual
102 Wingbermühle et al: External validation of prognostic neck pain models care physiotherapy treatment was provided in both studies. The Table 2 manual therapists in the development study had more work expe- Predictor variables and characteristics of participants in the validation cohort and the rience (mean work experience 19.3 years, SD 7.1) and qualified development study. manual skills compared with the physiotherapists in the validation study (mean work experience 5.4 years, SD 4.7), who were final year Predictor characteristics Validation Development Master of Science in manual therapy students. Therefore, the manual therapists may have added high-velocity thrust techniques and spe- cohort (n = 586) study (n = 1,193) cific joint mobilisation techniques. It was expected that the manual therapy students were providing care according to current Dutch Age (y), mean (SD) 44.0 (15.7) 44.7 (13.7) guidelines.20 Both studies excluded red flags and included people Sex, n (%) female 393 (67) 823 (69) aged . 18 years with non-specific neck pain with or without arm Previous neck pain episode, n (%) 490 (84) 755 (67) pain, and with or without trauma. The validation cohort contained 10 Neck pain duration, n (%) (1.7%) participants aged . 80 years and the development study 464 (79) 420 (39) excluded those aged . 80 years. The development study included acute (0 to 6 wk) 122 (21) 138 (13) people with neck pain of any duration, the validation study included subacute (6 to 12 wk) 513 (48) those who had had neck pain for a minimum of 3 days to a maximum chronic (. 12 wk) 6 (4 to 7) 4.8 (2.1) of 12 weeks. For model updating, additional history, physical exam- Pain intensity (Numerical pain rating scale, ination and psychosocial variables were available in the validation 0 to 10a and 1 to 10b), median (IQR)a and 11 (8 to 16) 12 (8 to 17) cohort. mean (SD)b Disability (Neck Disability Index, 0 to 50), 167 (29) 538 (45) Data analysis median (IQR) 280 (48) 337 (28) Accompanying low back pain, n (%) Missing data Accompanying general sleeping problems, 8 (4 to 13) 11 (6 to 15) We described missing predictor and outcome values and analysed n (%) Fear-Avoidance Beliefs Questionnaire, 5 (2 to 9) 7 (3 to 10) the data to assume the missing data mechanism (Little’s test, t-test, Physical Activity subscale (0 to 24), median chi-square test and logistic regression analysis) to decide whether (IQR) 404 (69) 783 (66) multiple imputation was needed. Neck Bournemouth Questionnaire, Anxiety Validation and Depression subscale (0 to 20)c, median cohort (n = 586) Statistical validation of the models’ performance (IQR) Partaking in sporting activities, n (%) 62 (11) We tested linearity and model assumptions and compared Potential predictors characteristics observed outcomes with those predicted by the models in the vali- 30.5 (21 to 46) dation cohort in terms of discrimination and calibration measures.9 Range of motion (deg, sum score), mean We calculated the model’s linear predictors (lp) and individual (SD) 6 (2 to 12) probability of recovery for disability, pain and perceived improve- Neck flexor muscle endurance (s), median ment as p (y = 1) = 1/(1 1 e–lp) for all participants at 6-weeks and 12- (IQR) weeks follow-up, and at the end of the treatment period.22 Each Pain Catastrophising Scale (0 to 52), model’s overall performance was estimated by Nagelkerke’s R2 and median (IQR) Brier scores. Percentages are rounded to the closest integer. Discriminative performance: Discriminative performance indicates a Pertains to validation cohort. whether a model can distinguish between people with neck pain with b Pertains to derivation cohort. and without recovery. It was calculated as the concordance (c) sta- c Sum score of 11-point numeric subscale of items 4 and 5. tistic, which is comparable with the area under the curve (AUC) of the receiver operating characteristic (ROC) curve for binary data.9 are reliable and their clinical use is supported.31 Measurement error Discriminative performance was a priori considered acceptable if the AUC was 0.70.23 of the Goniometro application using the CROM-device as reference appeared small.32 Endurance of the anterior neck muscles was Calibration performance: Calibration performance refers to the measured by the neck flexor endurance test;33 however, it revealed agreement between a model’s predicted risks and observed out- comes.24 We performed calibration-in-the-large and present the substantial intra-rater and inter-rater reliability and a large standard models’ calibration slopes and calibration plots.24 The models were re-estimated in the validation cohort using the linear predictor (lp) error of measurement of 14.57 seconds and a minimum detectable and model: logit (y) = a 1 b x lp.9,24,25 Calibration was tested as de- change of 40 seconds.34 Pain catastrophising was also considered as a viation from the ideal calibration slope of 1 and the intercept of potential additional predictor.35 Catastrophising is considered a pre- 0 using the model with an offset procedure. Calibration plots’ prob- abilities were calculated to allow observation of whether all decile dictor for persistent pain and disability in people with chronic groups closely fit the perfect 45 deg line of identity.9,24 Statistical musculoskeletal pain and in people with whiplash-related pain.36,37 analyses were performed using commerciala and open sourceb software. Pain catastrophising was measured with the pain catastrophising Updating of models: We evaluated whether updating each model scale, which is a reliable and valid instrument for measuring cata- enhanced its performance through adjustment of the model’s inter- strophic thinking related to pain.38,39 We evaluated whether the cept using the calibration intercept, and the model’s regression co- efficients using the calibration slope.26–28 models improved significantly after including these potential candi- date predictors (p , 0.157) and enhanced model performance.15,27,28 Additional variables were available in the validation cohort, and we tested whether a limited number of potential predictors improved Results the models. First, from physical examination, cervical mobility and endurance of the anterior neck muscles were used for updating the The predictor and outcome characteristics between the validation models as interventions aimed at improving these functions are and derivation cohort are presented in Tables 2 and 3, respectively. effective.29,30 Cervical mobility was measured in degrees by a total Due to the difference in neck pain duration inclusion criteria, the sum score of flexion, extension and both rotations using the mobile validation cohort displayed no participants with chronic neck pain phone applicationc. Smartphone applications measuring spinal ROM and 40.2% more participants with acute neck pain. There were no clinically meaningful differences between the other predictors (Table 2). The amount and percentage of non-recovered participants with neck pain in the validation cohort decreased from 6 to 12 weeks for all outcomes. The percentage of post-treatment non-recovered participants with neck pain between the validation and derivation study was comparable for pain intensity and differed for disability (post-treatment perceived recovery was not registered in the vali- dation study). The number of recovered and non-recovered events in the validation cohort turned out to be between the required mini- mum of 100 to 200 events for all follow-up periods; for disability, the number of events exceeded 200 for all follow-up periods.
Research 103 Table 3 Outcome variables in the validation cohort and development study. Outcome variables Validation cohort Development study 6 weeks (n = 586) Not recovered n (%) (n = 1,193) Not recovered n (%) Pain intensity (Numerical pain rating scale, 0 to 10a and 1 to 10b), median (IQR) Disability (Neck Disability Index, 0 to 50), median (IQR) 2 (1 to 4) 220 (38) 2 (1 to 2) 112 (21) Perceived improvement (Global perceived effect, 1 to 7), n (%) 5 (2 to 9) 349 (60)c 5 (1 to 9) 290 (58)c 174 (30) very much better 127 (22) 127 (24) 107 (21) much better 277 (48) 167 (29) 287 (55) slightly better 132 (23) 272 (47)c 83 (16) no change 150 (26) slightly worse 31 (5) 24 (5) much worse 8 (1) 134 (23) 0 (0) very much worse 3 (1) 274 (48)c 0 (0) 12 weeks 0 (0) 0 (0) Pain intensity (Numerical pain rating scale, 0 to 10a and 1 to 10b), median (IQR) Disability (Neck Disability Index, 0 to 50), median (IQR) 1 (0 to 3) Perceived improvement (Global perceived effect, 1 to 7), n (%) 3 (25 to 75) very much better much better 201 (35) slightly better 232 (40) no change 101 (17) slightly worse much worse 31 (6) very much worse 13 (2) End of treatment period 2 (0) Pain intensity (Numerical pain rating scale, 0 to 10a and 1 to 10b), median (IQR) 2 (0) Disability (Neck Disability Index, 0 to 50), median (IQR) Perceived improvement (Global perceived effect, 1 to 7), n (%) 1 (0 to 2) very much better 3 (1 to 7) much better slightly better no change slightly worse much worse very much worse Percentages are rounded to the closest integer. a Pertains to validation cohort. b Pertains to derivation cohort. c 0 to 50 value transformed to % by multiplying by 2 and dichotomised into , 8% (recovered) or 8% (not recovered). Model performance there was no need for multiple imputation and complete case anal- ysis was acceptable. Missing data The number of variables with missing data and the amount of Model performance measures before updating We tested and evaluated linearity and concluded that non-linear missing data was very low, with the vast majority between 0 and 1%. Six variables had marginally more than 1% missing values: the cer- transformation would not be advantageous.41 The models’ valida- vical mobility predictor 1.2%, the three outcome variables at 6 weeks tion performance is described in Table 4. The disability model at 6 1.4%, post-treatment pain intensity 1.2%, and post-treatment weeks showed acceptable discriminative performance with a c-sta- disability 1.5%. There was little gain from multiple imputation for tistic equal to 0.73 (95% CI 0.69 to 0.77). The disability models at 12 these low proportions of missing data.40 In addition, after analysing weeks and at the end of treatment showed discriminative perfor- the missing data and checking the researcher’s logbooks for reasons mance values of 0.69 (95% CI 0.64 to 0.73) and 0.68 (95% CI 0.63 to for missing outcomes, it was assumed that the missing completely at 0.72), respectively. The pain models and perceived improvement random missingness was plausible. Consequently, it was decided that models did not reach acceptable levels of the performance measures Table 4 Performance of the three models before updating. Model name Discrimination (c-statistic) Nagelkerke’s R2 Brier Score Testing calibration logit (95% CI) in-the-large (intercept) 0.20 (slope) 0.22 Disability model (DModel) 0.73 (0.69 to 0.77) 0.20 0.23 0.60a 1.10 6 weeks 0.69 (0.64 to 0.73) 0.14 –0.06 0.83a,b 12 weeks 0.68 (0.63 to 0.72) 0.12 0.22 –0.05 0.76a,b post-treatmentc 0.19 0.66 (0.62 to 0.71) 0.10 0.17 0.29 0.32a Pain model (PModel) 0.66 (0.61 to 0.71) 0.09 –0.16 0.31a 6 weeks 0.61 (0.56 to 0.67) 0.04 0.21 –0.69a 0.21a 12 weeks 0.19 post-treatmentc 0.53 (0.48 to 0.58) 0.00 –1.91 0.21 0.54 (0.48 to 0.59) 0.00 –2.59 0.31 Perceived improvement model (PIModel) 6 weeks 12 weeks a Significant deviation (intercept from 0, slope from 1) for test lp fit. b Not significant deviation for test intercept and slope separate with offset procedure. c If , 12 weeks.
104 Wingbermühle et al: External validation of prognostic neck pain models Figure 1. Calibration curves of the three models at validation. Disability model calibration curve at 6 weeks (A), 12 weeks (B) and the end of treatment (C). Pain model calibration curve at 6 weeks (D), 12 weeks (E) and the end of treatment (F). Perceived improvement model calibration curve at 6 weeks (G) and 12 weeks (H). (Table 4). The pain models showed discriminative performance values Model updating of 0.66 (95% CI 0.62 to 0.71), 0.66 (95% CI 0.61 to 0.71) and 0.61 (95% CI 0.56 to 0.67) at 6 weeks, 12 weeks and at the end of treatment, We assessed model updating for the disability model and pain respectively. The perceived improvement models showed discrimi- model. Based on the model performance, it was decided that further native performance values of 0.53 (95% CI 0.48 to 0.58) and 0.54 (95% testing of the perceived improvement model was not useful. If CI 0.48 to 0.59) at 6 weeks and 12 weeks, respectively. Calibration intercept and/or slope values differed significantly after testing with curves are displayed in Figure 1. the logit (y) = a 1 b x lp offset procedure, the models were updated
Research 105 Figure 1. Continued. with the values found, and the models’ performance was subse- acceptable performance of c-statistic of 0.73 (95% CI 0.69 to 0.77). quently re-evaluated. The calibration performance of the disability The other models’ calibration performance did not improve, and model at 6 weeks clearly improved from intercept correction using discrimination remained identical. For further testing, the recali- the found 0.6 value (Figure 2A); the discriminative performance did brated 6-week disability model was used and the other models not change after this correction and remained at the same were not adjusted. Figure 2. Calibration curves of the disability model after updating. Adjusted disability model calibration curve at 6 weeks, after recalibration with the 0.6 calibration intercept (A), after adding the cervical mobility and pain catastrophising pre- dictors and recalibration with the intercept (B). Disability model calibration curve at 12 weeks, after adding the cervical mobility predictor and recalibration with the intercept (C). Disability model calibration curve at the end of treatment, after adding the cervical mobility predictor and recalibration with the intercept (D).
106 Wingbermühle et al: External validation of prognostic neck pain models Testing the additional variables revealed that the cervical mobility model’s generalisability. Furthermore, the model’s relatively low and pain catastrophising variables added significantly (p , 0.157) to explained variance indicates that predictors for non-recovery are still the 6-week recalibrated disability model. The cervical mobility vari- missing and the quest for additional valuable predictors continues. able added significantly to the 12-week and post-treatment disability Additionally, it may be of interest to further evaluate the cervical models. The pain catastrophising variable added significantly (p , mobility and pain catastrophising predictors. Although they showed 0.157) to the 6-week pain model. The neck flexor endurance test minimal impact on prognostic performance in this study, being variable showed no additional significant value for the models. modifiable factors, they may have predictive capacity depending on specific treatments. For instance, the cervical mobility predictor may The significantly adding variables and their weights were included in show predictive capacity depending on mobilisation treatment, and the disability models at the three follow-ups and to the pain model at 6 the prognostic effect of the pain catastrophising predictor may weeks, and each model’s performance was then re-evaluated. Adding depend on cognitive-behavioural therapy. cervical mobility and pain catastrophising variables to the 6-week recalibrated disability model slightly improved discrimination to 0.74 Broad external validation of the disability model was successful, (95% CI 0.70 to 0.78). Adding cervical mobility to the 12-week and post- and this model is generalisable to current physiotherapy settings and treatment disability models showed c-statistics of 0.69 (95% CI 0.65 to can be used in clinical practice with reasonable confidence. We 0.73) and 0.69 (95% CI 0.65 to 0.73), respectively. Calibration perfor- advocate that physiotherapists use the disability model at intake for mance of all the disability models was initially overfitted and recovered the prognosis of people with neck pain to assist in clinical decisions after intercept recalibration (Figure 2B, C, D). The discrimination and concerning recovery of neck pain disability at 6 weeks. Further calibration performance of the 6-week pain model did not improve. research is required to assess the disability model’s clinical impact and generalisability. Discussion What is already known on this topic: Clinical use of currently published models for predicting recovery of non- The disability model for prediction of neck pain recovery remained specific neck pain cannot be advised. discriminatory at 6 weeks in a different, external cohort of people with What this study adds: A model for predicting recovery of neck pain, coming from an independent physiotherapy setting with a disability at 6 weeks in people with neck pain was broadly different case-mix. At 12 weeks and at the end of treatment, it showed externally validated and is advised for use in clinical practice. For nearly acceptable performance: c-statistics of 0.69 (95% CI 0.64 to 0.73) clinical use, the disability model’s regression formula was and 0.68 (95% CI 0.63 to 0.72), respectively. The pain model and transformed into a web-based risk calculator, which can be perceived improvement model could not be externally validated, accessed at www.somt.nl/research which was expected since internal validation was also not acceptable.13 Cervical mobility added value to the disability model at all follow-up Footnotes: a BM-SPSS Statistics version 27.0, SPSS Inc., Chicago, periods and pain catastrophising also to the 6-week pain model. USA. Model updating hardly affected discriminative and overall perfor- mance, whereas the different levels of updating were reflected in the b R software version 2021.09.01, R Core Team, Vienna, Austria. shape of the calibration curves. The additional predictors improved c Goniometro, Human Computer Technology, Madrid, Spain. model performance minimally and may have insufficient gain to be eAddenda: Nil. used clinically for purely prognostic purpose. Ethics approval: The METCZ20200178 Ethics Committee(s) approved this study. All participants gave written informed consent Few prognostic models for recovery of non-specific neck pain have before data collection began. been exposed to external validation.11,42 Until now, no non-specific Competing interests: The authors declare that there are no neck pain model has been successfully externally validated with competing interests. reporting of both discrimination and calibration performance mea- Source(s) of support: The cohort study was supported by SOMT sures as recommended by TRIPOD.43 One model stood out as it was University of Physiotherapy. evaluated in several external validation studies, whereby all these Acknowledgements: Nil. studies reported AUC values , 0.70.44–46 Provenance: Not invited. Peer reviewed. Correspondence: Roel W Wingbermühle, SOMT University of The strength of this broad external validation study is that it was Physiotherapy, The Netherlands. Email: [email protected] conducted in a cohort with sufficient power with very few missing values. A model is more challenged in broad than narrow external References validation, indicating a better test for its generalisability.47 This disability model keeps performing well at different follow-up periods 1. Vos T, Allen C, Arora M, Barber RM, Bhutta ZA, Brown A, et al. Global, regional, and in a cohort of people with neck pain with a different case-mix, national incidence, prevalence, and years lived with disability for 310 diseases and especially regarding the duration of pre-existing neck pain com- injuries, 1990–2015: a systematic analysis for the Global Burden of Disease Study plaints. In addition, participants were treated recently, reflecting 2015. Lancet. 2016;388(10053):1545–1602. https://doi.org/10.1016/S0140-6736(16) current physiotherapy guidelines. 31678-6 The 6-week disability model needed to be recalibrated, which is 2. Hay SI, Abajobir AA, Abate KH, Abbafati C, Abbas KM, Abd-Allah F, et al. Global, often needed in validation studies and indicates a difference in baseline regional, and national disability-adjusted life-years (DALYs) for 333 diseases risk between development and validation study that was not reflected and injuries and healthy life expectancy (HALE) for 195 countries and territories, by the model predictors. This could be explained by the difference in 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. non-recovery percentage for disability.26,28,48 Looking at the disability Lancet. 2017;390(10100):1260–1344. https://doi.org/10.1016/S0140-6736(17) models’ calibration slopes revealed that some group mean values were 32130-X still somewhat scattered around the perfect line of identity. This scat- tering may have been less if we had decided that non-linear trans- 3. Safiri S, Kolahi AA, Hoy D, Smith E, Bettampadi D, Mansournia MA, et al. Global, formation was advantageous, at the expense of clinical manageability. regional, and national burden of neck pain in the general population, 1990–2017: Furthermore, use of predictor weights gained by fitting the models Systematic analysis of the Global Burden of Disease Study 2017. BMJ. 2020:368. anew may have improved the model’s predictive performance. How- https://doi.org/10.1136/bmj.m791 ever, this implies model revision and subsequent external validation and is not preferred over simple recalibration.28 4. Cieza A, Causey K, Kamenov K, Hanson SW, Chatterji S, Vos T. Global estimates of the need for rehabilitation based on the Global Burden of Disease study 2019: a Further research is recommended to assess the disability model’s systematic analysis for the Global Burden of Disease Study 2019. Lancet. clinical utility and clinical impact. Additional external validation 2020;396(10267):2006–2017. https://doi.org/10.1016/S0140-6736(20)32340-0 studies in another clinical context (eg, other countries, other healthcare providers and other settings) may add knowledge to the 5. Hush JM, Lin CC, Michaleff Z, Verhagen A, Refshauge KM. Prognosis of acute idiopathic neck pain is poor: a systematic review and meta-analysis. Arch Phys Med Rehabil. 2011;92:824–829. https://doi.org/10.1016/j.apmr.2010.12.025 6. Vasseljen O, Woodhouse A, Bjørngaard JH, Leivseth L. Natural course of acute neck and low back pain in the general population: The HUNT study. Pain. 2013;154:1237–1244. https://doi.org/10.1016/j.pain.2013.03.032
Research 107 7. Verwoerd M, Wittink H, Maissan F, de Raaij E, Smeets RJEM. Prognostic factors for 29. Coulter ID, Crawford C, Vernon H, Hurwitz EL, Khorsan R, Booth MS, et al. persistent pain after a first episode of nonspecific idiopathic, non-traumatic neck Manipulation and mobilization for treating chronic nonspecific neck pain: a sys- pain: A systematic review. Musculoskelet Sci Pract. 2019;42:13–37. https://doi.org/ tematic review and meta-analysis for an appropriateness panel. Pain Physician. 10.1016/j.msksp.2019.03.009 2019;22:E55–E70. 8. Artus M, Campbell P, Mallen CD, Dunn KM, van der Windt DAW. Generic prognostic 30. Martin-Gomez C, Sestelo-Diaz R, Carrillo-Sanjuan V, Navarro-Santana MJ, Bardon- factors for musculoskeletal pain in primary care: a systematic review. BMJ Open. Romero J, Plaza-Manzano G. Motor control using cranio-cervical flexion exercises 2017;7, e012901. https://doi.org/10.1136/bmjopen-2016-012901 versus other treatments for non-specific chronic neck pain: a systematic review and meta-analysis. Musculoskelet Sci Pract. 2019;42:52–59. https://doi.org/10.1016/ 9. Steyerberg EW. Clinical Prediction Models: A Practical Approach to Development, j.msksp.2019.04.010 Validation, and Updating. 2nd ed. Cham: Springer Science and Business Media; 2019. 31. Nuhmani S, Khan MH, Kachanathu SJ, Bari MA, Abualait TS, Muaidi QI. Reliability and validity of smartphone applications to measure the spinal range of motion: 10. Riley RD, Van Der Windt DA, Croft P, Moons KGM. Prognosis Research in Healthcare. asystematic review. Expert Rev Med Devices. 2021;18:893–901. https://doi.org/10. 1st ed. Oxford: Oxford University Press; 2019. 1080/17434440.2021.1962290 11. Wingbermühle RW, van Trijffel E, Nelissen PM, Koes B, Verhagen AP. Few prom- 32. Verstelle K. Reliability of the Cervical Flexion Rotation Test and the C0-C2 Axial ising multivariable prognostic models exist for recovery of people with non- Rotation Test in an Elderly Patient Population. Master Thesis. Free University Brussels; specific neck pain in musculoskeletal primary care: a systematic review. 2021. J Physiother. 2018;64:16–23. https://doi.org/10.1016/j.jphys.2017.11.013 33. Harris KD, Heer DM, Roy TC, Santos DM, Whitman JM, Wainner RS. Reliability of a 12. Wingbermühle RW, Chiarotto A, Koes B, Heymans MW, van Trijffel E. Challenges measurement of neck flexor muscle endurance. Phys Ther. 2005;85:1349–1355. and solutions in prognostic prediction models in spinal disorders. J Clin Epidemiol. https://doi.org/10.1093/ptj/85.12.1349 2021;132:125–130. https://doi.org/10.1016/j.jclinepi.2020.12.017 34. Juul T, Langberg H, Enoch F, Søgaard K. The intra- and inter-rater reliability of 13. Wingbermühle RW, Chiarotto A, van Trijffel E, Koes B, Verhagen AP, Heymans MW. five clinical muscle performance tests in patients with and without neck pain. Development and internal validation of prognostic models for recovery in patients BMC Musculoskelet Disord. 2013;14:339. https://doi.org/10.1186/1471-2474-14- with non-specific neck pain presenting in primary care. Physiotherapy. 339 2021;113:61–72. https://doi.org/10.1016/j.physio.2021.05.011 35. Thompson DP, Urmston M, Oldham JA, Woby SR. The association between 14. Steyerberg EW, Harrell FE. Prediction models need appropriate internal, internal- cognitive factors, pain and disability in patients with idiopathic chronic neck external, and external validation. J Clin Epidemiol. 2016;69:245–247. https://doi. pain. Disabil Rehabil. 2010;32:1758–1767. https://doi.org/10.3109/096382810 org/10.1016/j.jclinepi.2015.04.005 03734342 15. Steyerberg EW, Moons KGM, van der Windt DA, Hayden JA, Perel P, Schroter S, et al. 36. Walton DM, MacDermid JC, Giorgianni AA, Mascarenhas JC, West SC, Zammit CA. Prognosis Research Strategy (PROGRESS) 3: Prognostic Model Research. PLoS Med. Risk factors for persistent problems following acute whiplash injury: update of a 2013;10, e1001381. https://doi.org/10.1371/journal.pmed.1001381 systematic review and meta-analysis. J Orthop Sports Phys Ther. 2013;43:31–43. https://doi.org/10.2519/jospt.2013.4507 16. Debray TPA, Vergouwe Y, Koffijberg H, Nieboer D, Steyerberg EW, Moons KGM. A new framework to enhance the interpretation of external validation studies of 37. Martinez-Calderon J, Jensen MP, Morales-Asencio JM, Luque-Suarez A. Pain cata- clinical prediction models. J Clin Epidemiol. 2015;68:279–289. https://doi.org/10. strophizing and function in individuals with chronic musculoskeletal pain. Clin J 1016/j.jclinepi.2014.06.018 Pain. 2019;35:279–293. https://doi.org/10.1097/AJP.0000000000000676 17. Collins GS, Reitsma JB, Altman DG, Moons KGM. Transparent Reporting of a 38. Sullivan MJL, Bishop SR, Pivik J. The Pain Catastrophizing Scale: Development and Multivariable Prediction Model for Individual Prognosis or Diagnosis (TRIPOD): validation. Psychol Assess. 1995;7:524–532. https://doi.org/10.1037/1040-3590.7.4. The TRIPOD Statement. Eur Urol. 2015;67:1142–1151. https://doi.org/10.1016/j. 524 eururo.2014.11.025 39. Osman A, Barrios FX, Gutierrez PM, Kopper BA, Merrifield T, Grittmann L. The pain 18. Peters R, Mutsaers B, Verhagen AP, Koes BW, Pool-Goudzwaard AL. Prospective catastrophizing scale: Further psychometric evaluation with adult samples. J Behav cohort study of patients with neck pain in a manual therapy setting: design and Med. 2000;23:351–365. https://doi.org/10.1023/A:1005548801037 baseline measures. J Manipulative Physiol Ther. 2019;42:471–479. https://doi.org/ 10.1016/j.jmpt.2019.07.001 40. Madley-Dowd P, Hughes R, Tilling K, Heron J. The proportion of missing data should not be used to guide decisions on multiple imputation. J Clin Epidemiol. 19. Guzman J, Hurwitz EL, Carroll LJ, Haldeman S, Côté P, Carragee EJ, et al. A new con- 2019;110:63–73. https://doi.org/10.1016/j.jclinepi.2019.02.016 ceptual model of neck pain. linking onset, course, and care: The Bone and Joint Decade 2000-2010 Task Force on Neck Pain and Its Associated Disorders. J Manipulative Physiol 41. Sauerbrei W, Perperoglou A, Schmid M, Abrahamowicz M, Becher H, Binder H, Ther. 2009;32:17–28. https://doi.org/10.1016/j.jmpt.2008.11.007 et al. State of the art in selection of variables and functional forms in multivariable analysis—outstanding issues. Diagn Progn Res. 2020;4:3. https://doi.org/10.1186/ 20. Bier JD, Scholten-Peeters WGM, Staal JB, Pool J, van Tulder MW, Beekman E, et al. s41512-020-00074-3 Clinical practice guideline for physical therapy assessment and treatment in pa- tients with nonspecific neck pain. Phys Ther. 2018;98:162–171. https://doi.org/10. 42. Kelly J, Ritchie C, Sterling M. Clinical prediction rules for prognosis and treatment 1093/ptj/pzx118 prescription in neck pain: A systematic review. Musculoskelet Sci Pract. 2017;27:155–164. https://doi.org/10.1016/j.math.2016.10.066 21. Collins GS, Ogundimu EO, Altman DG. Sample size considerations for the external validation of a multivariable prognostic model: A resampling study. Stat Med. 43. Collins GS, Reitsma JB, Altman DG, Moons KGM. Transparent Reporting of a 2016;35:214–226. https://doi.org/10.1002/sim.6787 multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD): The TRIPOD Statement. Ann Intern Med. 2015;162:55. https://doi.org/10.7326/M14- 22. Vergouwe Y, Moons KGM, Steyerberg EW. External validity of risk models: Use of 0697 benchmark values to disentangle a case-mix effect from incorrect coefficients. Am J Epidemiol. 2010;172:971–980. https://doi.org/10.1093/aje/kwq223 44. Schellingerhout JM, Heymans MW, Verhagen AP, Lewis M, de Vet HCW, Koes BW. Prognosis of patients with nonspecific neck pain: development and external 23. Hosmer DW, Lemeshow S, Sturdivant RX. Applied Logistic Regression. 3rd ed. Wiley; validation of a prediction rule for persistence of complaints. Spine. 2010;35:E827– 2013. E835. https://doi.org/10.1097/BRS.0b013e3181d85ad5 24. Steyerberg EW, Vergouwe Y. Towards better clinical prediction models: seven steps 45. Wingbermühle RW, Heymans MW, van Trijffel E, Chiarotto A, Koes B, for development and an ABCD for validation. Eur Heart J. 2014;35(29):1925–1931. Verhagen AP. External validation of prognostic models for recovery in patients https://doi.org/10.1093/eurheartj/ehu207 with neck pain. Braz J Phys Ther. 2021;25:775–784. https://doi.org/10.1016/j.bjpt. 2021.06.001 25. Vergouwe Y, Royston P, Moons KGM, Altman DG. Development and validation of a prediction model with missing predictor data: a practical approach. J Clin Epi- 46. Myhrvold BL, Kongsted A, Irgens P, Robinson HS, Thoresen M, Vøllestad NK. Broad demiol. 2010;63:205–214. https://doi.org/10.1016/j.jclinepi.2009.03.017 external validation and update of a prediction model for persistent neck pain after 12 weeks. Spine. 2019;44:E1298–E1310. https://doi.org/10.1097/BRS.000000000 26. Vergouwe Y, Nieboer D, Oostenbrink R, Debray TP, Murray GD, Kattan MW, et al. 0003144 A closed testing procedure to select an appropriate method for updating prediction models. Stat Med. 2017;36(28):4529–4539. https://doi.org/10.1002/sim.7179 47. Steyerberg EW, Harrell FE. Prediction models need appropriate internal, internal- external, and external validation. J Clin Epidemiol. 2016;69:245–247. https://doi. 27. Moons KGM, Kengne AP, Grobbee DE, Royston P, Vergouwe Y, Altman DG, et al. Risk org/10.1016/j.jclinepi.2015.04.005 prediction models: II. External validation, model updating, and impact assessment. Heart. 2012;98:691–698. https://doi.org/10.1136/heartjnl-2011-301247 48. Van Calster B, McLernon DJ, Van Smeden M, Wynants L, Steyerberg EW. Calibra- tion: The Achilles heel of predictive analytics. BMC Med. 2019;17:1–7. https://doi. 28. Janssen KJM, Moons KGM, Kalkman CJ, Grobbee DE, Vergouwe Y. Updating org/10.1186/s12916-019-1466-7 methods improved the performance of a clinical prediction model in new patients. J Clin Epidemiol. 2008;61:76–86. https://doi.org/10.1016/j.jclinepi.2007.04.018
Journal of Physiotherapy 69 (2023) 108–113 j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j p hy s Research In adults with advanced lung disease, the 1-minute sit-to-stand test underestimates exertional desaturation compared with the 6-minute walk test: an observational study Kathryn Watson a, Peta Winship a,b, Vinicius Cavalheri a,c,d, Caitlin Vicary a, Stephanie Stray a, Natasha Bear e, Kylie Hill d a Physiotherapy Department, Fiona Stanley Hospital, Perth, Australia; b National School of Health Sciences and Physiotherapy, Notre Dame University, Perth, Australia; c Allied Health, Metropolitan Health Service, Perth, Australia; d Curtin School of Allied Health and enAble Institute, Curtin University, Perth, Australia; e Institute for Health Research, Notre Dame University, Perth, Australia KEY WORDS ABSTRACT Exercise testing 6MWT Question: In adults with advanced lung disease, do the 6-minute walk test (6MWT) and 1-minute sit-to- 1-min STS Lung disease stand test (1minSTS) elicit similar cardiorespiratory responses? Can the 6-minute walk distance (6MWD) Physical therapy be estimated from the 1minSTS result? Design: Prospective observational study using data collected dur- ing routine clinical practice. Participants: Eighty adults (43 males) with advanced lung disease, a mean age of 64 years (SD 10) and a mean forced expiratory volume in 1 second of 1.65 L (SD 0.77). Outcome measures: Participants completed a 6MWT and a 1minSTS. During both tests, oxygen saturation (SpO2), pulse rate, dyspnoea and leg fatigue (Borg 0 to 10) were recorded. Results: Compared with the 6MWT, the 1minSTS resulted in higher nadir SpO2 (MD 4%, 95% CI 3 to 5), lower end-test pulse rate (MD –4 beats/minute, 95% CI –6 to –1), similar dyspnoea (MD –0.3, 95% CI –0.6 to 0.1) and greater leg fatigue (MD 1.1, 95% CI 0.6 to 1.6). Among the participants who demonstrated severe desaturation (SpO2 nadir , 85%) on the 6MWT (n = 18), five and ten participants were classified as moderate (nadir 85 to 89%) or mild desaturators (nadir 90%), respectively, on the 1minSTS. The relationship between the 6MWD and 1minSTS was: 6MWD (m) = 247 1 (7 3 number of transitions achieved during the 1minSTS) with poor predictive ability (r2 = 0.44). Conclusion: The 1minSTS elicited less desaturation than the 6MWT and classified a smaller proportion of people as ‘severe desaturators’ on exertion. It is therefore inappropriate to use the nadir SpO2 recorded during a 1minSTS to make decisions about whether strategies are needed to prevent severe transient exertional desaturation during walking-based exercise. Further, the extent to which performance on the 1minSTS can estimate a person’s 6MWD is poor. For these reasons, the 1minSTS is unlikely to be helpful when prescribing walking-based exercise. [Watson K, Winship P, Cavalheri V, Vicary C, Stray S, Bear N, Hill K (2023) In adults with advanced lung disease, the 1-minute sit-to-stand test underestimates exertional desatu- ration compared with the 6-minute walk test: an observational study. Journal of Physiotherapy 69:108– 113] © 2023 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Introduction the prescription of exercise. Specifically, 6-minute walk distance (6MWD) is often used to develop a 20-minute walking goal.4 Adults diagnosed with advanced lung disease who attend respi- Marked desaturation during the 6MWT may trigger an evaluation ratory medicine outpatient clinics are often referred to participate in for ambulatory oxygen prescription5,6 and SpO2, pulse rate and pulmonary rehabilitation programs (PRPs), which include supervised symptom responses during the 6MWT are also useful in monitoring exercise training. These programs reduce dyspnoea and improve disease progression. health-related quality of life and functional capacity in adults with a range of chronic respiratory conditions.1 Prior to commencing a PRP, In recent times, due to the COVID-19 pandemic, physiotherapists the response to exercise is often evaluated using a 6-minute walk have needed to balance the requirement to evaluate the response to test (6MWT).2 During this test, several measures are collected, exercise by supervising the person whilst they perform a 6MWT including peripheral capillary oxygen saturation (SpO2) and pulse with the restrictions related to the pandemic such as lock-downs, rate, using a pulse oximeter, and dyspnoea and leg fatigue, using the personal protective equipment requirements and physical Modified Borg category ratio scale.2,3 These measures are used in distancing. The transition of many services to telehealth due to such restrictions has led to an increased reliance on assessments such as https://doi.org/10.1016/j.jphys.2023.02.001 1836-9553/© 2023 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/4.0/).
Research 109 the 1-minute sit-to-stand test (1minSTS).7 Over the last decade, few within the preceding 3-month period. Given the absence of data studies have compared cardiorespiratory and symptom responses demonstrating the safety profile of desaturation to , 80% at the fa- measured during the 6MWT with the 1minSTS in people with a cility during exercise testing, participants who desaturated to , 80% chronic lung condition. Some8–10 but not all of these studies11 have were instructed to stop and rest and were permitted to recommence demonstrated greater desaturation (w3%) during the 6MWT once their SpO2 was 85%.12,13 compared with the 1minSTS. However, most of the earlier studies excluded people who required long-term oxygen therapy (LTOT) 1-minute sit-to-stand test and/or supplemental oxygen on exertion.8,9,11 It is therefore difficult to generalise these results to PRP in Australia, where . 80% of The testing procedures were informed by recommendations from programs have access to supplemental oxygen during exercise a systematic review of the literature.14 Participants were instructed to testing and training.5 Further, no study has collected responses after transition between sitting and standing as many times as possible in 1 minute of the 6MWT and compared these with responses elicited 1 minute. The height of the chair was kept between 45 and 48 cm and during the 1minSTS. Such a comparison will help to elucidate participants were encouraged not to use their arms to facilitate the whether any difference in SpO2 or end-test pulse rate relates mainly transitions. to differences in test duration. Data analyses The aim of this study was, in a large sample of adults known to the advanced lung disease service at a tertiary hospital (Fiona Data were analysed using commercial softwarea. The distribution Stanley Hospital, Perth, Western Australia) to: explore the agree- of data was assessed using frequency histograms. Data are ment between measures of SpO2, pulse rate, dyspnoea and leg fa- expressed as mean 6 standard deviation. Agreement in measures tigue elicited during the 6MWT and 1minSTS; and explore the was explored using the methods described by Bland-Altman, extent to which the 6MWD could be estimated using the results of including Bland-Altman plots and limits of agreement (LOA).15 the 1minSTS. Therefore, the research questions for this prospective Mean differences between tests are reported along with the corre- observational study using data collected during routine clinical sponding 95% confidence interval. The study also explored the practice were: extent to which the two tests produced a response that would classify each participant as either a mild desaturator (nadir 90%; 1. In adults with advanced lung disease, do the 6-minute walk test included non-desaturators), moderate desaturator (nadir 85 to 89%) (6MWT) and 1-minute sit-to-stand test (1minSTS) elicit similar or severe desaturator (, 85%). These desturation thresholds, albeit cardiorespiratory responses? arbitrary, are consistent with earlier work that describes desatura- tion responses during exercise.16,17 Cohen’s kappa was used to assess 2. Can the 6-minute walk distance (6MWD) be estimated from the agreement (interpretation: , 0.2 poor, 0.2 to , 0.4 fair, 0.4 to , 0.6 1minSTS result? moderate, 0.6 to , 0.8 substantial, 0.8 to , 1 almost perfect).18 To explore differences in sub-groups, participants were grouped ac- Methods cording to their primary respiratory diagnosis as having either obstructive disease (alpha-1 antitrypsin deficiency, asthma, bron- This study presents an analysis of data collected prospectively chiectasis, chronic obstructive pulmonary disease (COPD), cystic during routine clinical practice on all adults known to the advanced fibrosis; n = 31), restrictive disease (asbestosis, interstitial lung lung disease service at Fiona Stanley Hospital who completed a diseases; n = 28) or other disease (bronchiolitis obliterans, pulmo- 6MWT and 1minSTS test between September 2021 and January 2022. nary arterial hypertension, post lung transplant; n = 21). Participants Adults referred to the advanced lung disease service had a chronic were also grouped according to whether they used supplemental lung disease and were being considered for specialist interventions oxygen during the tests. Differences between groups were explored (eg, lung transplant) or enrolment in pharmaceutical trials. All adults using one-way analysis of variance (ANOVA). Linear regression was were exacerbation-free at the time of assessment, and the 6MWT was used to determine the extent to which 6MWD could be estimated performed prior to the 1minSTS with at least 20 minutes of rest be- using the results of the 1minSTS. tween tests. Variables of interest that were recorded were: partici- pant characteristics (eg, age, gender, height, weight and lung Results function); nadir SpO2, pulse rate, dyspnoea and leg fatigue collected on completion of the 6MWT and 1minSTS; and SpO2 and pulse rate, Data were available on 80 participants. The 6MWD was (mean 6 measured at the end of the first minute of the 6MWT (ie, iso-time SD) 419 6 104 m and the number of transitions achieved during the responses). For both tests, SpO2 and pulse rate were monitored 1minSTS was 24 6 10. Participant characteristics are presented in continuously using the Massimo Rad-V pulse oximeter (Massimo Table 1. Australia Pty Ltd, NSW, Australia) and the sensor placement was standardised between tests. For those who had been prescribed Agreement supplemental oxygen, interface and dose were kept identical be- tween tests. Those who used supplemental oxygen during the 6MWT The SpO2 measured during the tests is presented in Figure 1a. transported the cylinder on their own using a trolley or wheeled Regarding SpO2, the nadir was higher during the 1minSTS compared walker. Dyspnoea and leg fatigue were measured on test completion with the 6MWT (mean difference [MD] 4%, 95% CI 3 to 5). This dif- using the modified Borg category ratio scale (0 to 10).3 Approval to ference in nadir SpO2 between the tests was slightly greater in those undertake and submit these analyses for publication was granted who used supplemental oxygen compared with room air by 2% (95% through the Governance Evidence Knowledge Outcomes (GEKO) CI 1 to 4). The difference in nadir SpO2 measured during the tests in system (approval number 42614). participants with ‘obstructive disease’ was 3% (95% CI 2 to 5), ‘restrictive disease’ was 6% (95% CI 4 to 7) and ‘other disease’ was 3% 6-minute walk test (95% CI 1 to 4). The Bland-Altman plot is presented in the Figure 2a; LOA ranged from –4 to 12%. This test was undertaken in accordance with the current technical standard for field-based walking tests, with the exception of a track The SpO2 measured 1 minute into the 6MWT was similar to the length equal to 25 m due to space limitations.12 Only one 6MWT was nadir measured during the 1minSTS (MD 1%, 95% CI 0 to 2). Agree- performed, as 89% of the participants included in this study had ment between the two tests in classifying participants as mild, previously performed a 6MWT, with 83% having performed a 6MWT moderate or severe desaturators, using nadir SpO2, is presented in
110 Watson et al: Assessing exertional desaturation in lung disease Table 1 A Participant characteristics. 100 Characteristic (n = 80) Age (yr), mean (SD) 64 (10) 95 Body mass index (kg/m2), mean (SD) 28 (6) Sex, n (%) male 43 (54) SpO2 (%) 90 Ethnicity, n (%) 72 (90) 85 Caucasian 6 (8) Asian 2 (3) 80 12345 6 ATSI 0 exercise time (minutes) Primary respiratory diagnosis, n (%) 27 (34) Alpha-1 antitrypsin deficiency/asthma/COPD 4 (5) B 140 pulse rate (beats/minute) cystic fibrosis or non-cystic fibrosis bronchiectasis interstitial lung disease 13 (16) 120 interstitial pulmonary fibrosis 14 (18) 100 pulmonary arterial hypertension post lung transplant 7 (9) 80 other 9 (11) FEV1 (L), mean (SD) 6 (7) FVC (L), mean (SD) 1.65 (0.77) On LTOT, n (%) 2.58 (0.84) O2 flow rate used during both tests, n (%) 42 (52) room air 1 to 4 L/min 38 (48) 5 to 8 L/min 26 (33) 9 to 12 L/min 11 (14) 13 to 15 L/min Required at least one rest during the 6MWT, n (%) 2 (3) Used a four-wheeled walker during the 6MWT, n (%) 1 (1) 7 (9) 15 (19) ATSI = Aboriginal/Torres Strait Islander; COPD = chronic obstructive pulmonary disease; FEV1 = forced expiratory volume in 1 second; FVC = forced vital capacity; LTOT = long-term oxygen therapy; 6MWT = 6-minute walk test. Some percentages do not sum to 100 because of rounding. Table 2. Individual participant data are presented in Table 3 on the 60 eAddenda. Agreement between the two tests was fair (k = 0.26, 95% CI 0.09 to 0.36). Of those who demonstrated severe desaturation on the 0123456 6MWT (n = 18), 5 and 10 were classified as moderate or mild desa- turators on the 1minSTS, respectively. Of those who demonstrated exercise time (minutes) moderate desaturation on the 6MWT (n = 25), 20 were classified as a mild desaturators on the 1minSTS test. Figure 1. Measures of peripheral capillary oxygen saturation (SpO2) (1A) and pulse rate (1B) measured at rest and after 1 minute of exercise in both tests, and after 6 minutes The pulse rate measured on completion of both tests is presented of exercise during the 6-minute walk test. Data are mean and standard deviation. Open in Figure 1b. The end-test pulse rate was slightly lower during the circles and dashed lines indicate 6-minute walk test; filled circles and solid lines 1minSTS compared with the 6MWT (MD –4 bpm, 95% CI –6 to –1). indicate 1-minute sit-to-stand test. There was no difference in end-test pulse rate measured between the tests in those who used supplemental oxygen compared with those Discussion who breathed room air (0 bpm, 95% CI –5 to 4). The difference in end- test pulse rate measured during the tests in participants with This study explored cardiorespiratory responses during the 6MWT ‘obstructive disease’ was –3 bpm (95% CI –7 to 1), ‘restrictive disease’ and 1minSTS in a large sample characterised by moderate to severe was –5 bpm (95% CI –10 to –1) and ‘other disease’ was –6 bpm (95% CI lung disease who were known to the advanced lung disease service at –11 to –1). The Bland-Altman plot is presented in Figure 2b; LOA Fiona Stanley Hospital. The data demonstrated that in this clinical ranged from –25 to 16 bpm. The pulse rate measured 1 minute into population, when compared with the 6MWT, the 1minSTS elicited the 6MWT was lower than end-test pulse rate measured during the less desaturation, a lower end-test pulse rate, similar dyspnoea and 1minSTS (MD –8 bpm, 95% CI –11 to –5). greater leg fatigue. The true mean difference in nadir SpO2 between the tests is likely to be between 3 and 5%. Although this difference Regarding dyspnoea, the scores reported on test completion were might seem small, when the nadir SpO2 was used to develop similar for the 6MWT and the 1minSTS (MD –0.3, 95% CI –0.6 to 0.1). thresholds to classify people as mild, moderate or severe desatu- The Bland-Altman plot is presented in Figure 2c; LOA ranged from –3 rators, 83% of those who were classified as severe desaturators on the to 3. Regarding leg fatigue, the score reported on test completion was 6MWT were classified as only mild or moderate on the 1minSTS. greater during the 1minSTS compared with the 6MWT (MD 1.1, 95% Similarly, 80% of those who were classified as moderate desaturators CI 0.6 to 1.6). The Bland-Altman plot is presented in Figure 2d; LOA on the 6MWT were classified as mild desaturators on the 1minSTS. In ranged from –3 to 5. addition, the LOA were wide, indicating that nadir seen during the 1minSTS may be as much as 12% greater than the nadir seen during Linear regression the 6MWT. The association between 6MWD and performance on the 1minSTS The mechanisms underpinning the differences in desaturation was r = 0.66 (Figure 3). The equation that represented this relation- between these tests have not been elucidated. The most likely reason ship was: is simply that the 6MWT is of longer duration than the 1minSTS and therefore people with chronic lung disease are exposed to a longer 6MWD (m) = 247 1 (7 3 number of transitions achieved during period of increased ventilatory demand, dynamic hyperinflation and the 1minSTS). greater V/Q mismatching.19 Our findings that the SpO2 measured 1 minute into the 6MWT was similar to the SpO2 measured on This equation has poor predictive ability (r2 = 0.44). completion of the 1minSTS and that SpO2 continued to decrease
Research 111 A B 40 20 1minSTS – 6MWT end-test pulse rate (beats/minute) 1minSTS – 6MWT 10 20 nadir SpO2 (%) 00 –10 –20 –20 80 85 90 95 100 –40 80 100 120 140 160 75 60 mean nadir SpO2 (%) mean end-test pulse rate (beats/minute) 1minSTS – 6MWT end-testC D1minSTS – 6MWT end-test dyspnoea (0 to 10) leg fatigue (0 to 10) 8 8 6 6 4 4 2 2 0 0 –2 –2 –4 –4 –6 –6 –8 –8 0 2 4 6 8 10 0 2 4 6 8 10 mean end-test dyspnoea (0 to 10) mean end-test leg fatigue (0 to 10) Figure 2. Bland-Altman plots for demonstrating agreement between the measures collected during the 6-minute walk test (6MWT) and 1-minute sit-to-stand test (1minSTS). These measures are: A) nadir peripheral capillary oxygen saturation measured via pulse oximeter (SpO2); B) end-test pulse rate; C) end-test dyspnoea score; and D) end-test leg fatigue score. Solid lines represent bias and dashed lines represent limits of agreement. during the 6MWT suggest that exercise time was an important almost exclusively by muscles of the lower limb, namely the quad- determinant of the difference between tests. riceps, when transitioning between sitting and standing. Of note, following the 1minSTS, earlier work has shown that . 70% of people This study did not enable comment on whether adults with with COPD demonstrate contractile fatigue of the quadriceps, which chronic lung disease demonstrate different nadirs in SpO2 with aligns with our finding that the 1minSTS elicited greater leg fatigue walking and sit-to-stand exercise. This is because the work done than walking.24 This higher lactate associated with overloading the during the two tests was not assessed and is unlikely to be equivalent. quadriceps is a potent ventilatory stimulant.22,23 Compared with Despite this consideration, it is possible that, at equivalent rates of walking-based exercise, people with COPD achieve greater peak oxygen uptake, adults will demonstrate less desaturation with sit-to- ventilation during cycle-based exercise, as they are able to fix their stand exercise than walking exercise. This is because earlier work has arms on the handlebars and use accessory muscles of respiration and shown that when people with COPD exercise at equivalent rates of adopt a forward lean position to optimise the mechanical advantage oxygen uptake, cycle-based exercise elicits less desaturation than and pressure-generating capacity of the diaphragm.25 The greater walking-based exercise20,21 and mechanisms underpinning this levels of ventilation achieved during cycle-based exercise minimise finding may also be applicable to exercise than involves transitioning desaturation.22 Unlike walking, which necessitates arm swing, people between sitting and standing; that is, compared with walking exer- can fix their arms against their body during the 1minSTS (eg, place cise, for a similar amount of work, transitioning between sitting and their hands in their pockets or press them against their trunk). standing is likely to elicit greater lactate.22,23 This is because unlike Although this effect may seem less potent that the advantages offered walking, which is a whole-body exercise, the exercise load is borne during cycle-based exercise (ie, using the handlebars and adopting a forward lean position), an earlier study reported that people with Table 2 COPD reached a similar peak rate of minute ventilation during the Classification of each participant as a mild, moderate or severe desaturator using the 1minSTS when compared with a ramp cycle ergometry test.24 This nadir SpO2 measured during the two tests. mechanism may minimise desaturation during sit-to-stand exer- cise.21 To further explore this idea, ventilation and lactate responses 1-minute sit-to-stand test and at iso-work (or similar peak rates of oxygen uptake) between the exercise modalities need to be evaluated.22 Mild Moderate Severe Total sample The role of supplemental oxygen therapy in people with chronic desaturator desaturator desaturator lung disease who do not qualify for LTOT is contentious and there is very little evidence that transient exertional desaturation is 6-minute Mild 37 0 0 37 (46%) dangerous.26 Nevertheless, people who do not qualify for LTOT but walk test desaturator 20 5 0 25 (31%) who demonstrate transient exertional desaturation are often Moderate 10 5 3 18 (23%) considered for supplemental oxygen on exertion on the basis that desaturator 67 (84%) 10 (12%) 3 (4%) laboratory studies have shown that supplemental oxygen during a Severe 80 single bout of exercise reduces minute ventilation at equivalent work desaturator rates. This in turn delays the onset of dynamic hyperinflation and the Total sample ‘Mild desaturator’ = nadir SpO2 90%; ‘Moderate desaturator’ = nadir SpO2 85 to 89%; ‘Severe desaturator’ = nadir SpO2 , 85%; SpO2 = peripheral capillary oxygen saturation measured via pulse oximeter.
112 Watson et al: Assessing exertional desaturation in lung disease 800 6-minute walk distance (m) 600 400 200 0 0 10 20 30 40 50 60 Transitions achieved during 1minSTS (n) Figure 3. Scatter plot showing the 95% prediction interval for the relationship between the 6-minute walk distance and number of transitions achieved during the 1-minute sit-to- stand test (1minSTS). associated dyspnoea in people with COPD who have severe disease.27 based exercise is inappropriate in this population. Further, the However, it is notable that within the context of PRPs, providing extent to which performance on the 1minSTS can estimate a person’s supplemental oxygen during exercise training to those who do not 6MWD is limited and so the 1minSTS is unlikely to be helpful when qualify for LTOT but demonstrate transient exertional desaturation prescribing walking-based exercise. does not produce any advantage in walking-based outcomes of ex- ercise capacity on program completion.16 Similarly, in this group, What was already known on this topic: In people with providing supplemental oxygen for home use does not produce chronic lung disease, a 6-minute walk test (6MWT) is typically convincing gains in health-related quality of life or participation in used to evaluate the response to exercise prior to commencing a physical activity.28–30 Despite this controversy, clinicians who run pulmonary rehabilitation program. The transition of many pul- PRPs strive to minimise severe transient exertional desaturation5 and monary rehabilitation services to telehealth due to such re- our data demonstrate that relying on the 1minSTS to screen for strictions imposed during the COVID-19 pandemic has led to moderate or severe transient desaturation is problematic. increased use of the 1-minute sit-to-stand test (1minSTS) instead of the 6WMT. Our finding of a relationship (r = 0.66) between the performance What this study adds: The 1minSTS elicited less desaturation on the 1minSTS and 6MWD is consistent with earlier work.8,9,11,31 We than the 6MWTand classified a smaller proportion of patients as present an equation that characterised this relationship. Although ‘severe desaturators’ on exertion. It is therefore inappropriate to clinicians may be tempted to use this equation to estimate the 6MWD use the nadir SpO2 recorded during a 1minSTS to make decisions using the number of transitions achieved during the 1minSTS, it about whether strategies are needed to prevent severe transient explained , 50% of the variance in the measure of 6MWD and so the exertional desaturation during walking-based exercise. Further, estimate derived using this method is poorly predictive for use in the extent to which performance on the 1minSTS can estimate a clinical practice. person’s 6MWT distance is poor. For these reasons, the 1minSTS is not helpful when prescribing walking-based exercise. A strength of this study was its pragmatic approach to include data from all people known to the advanced lung disease service at Footnotes: a Stata version 16, StataCorp, College Station, USA. Fiona Stanley Hospital. In contrast with earlier work in this area, eAddenda: Table 3 can be found online at https://doi.org/10.1016/j. which recruited people who met pre-specified inclusion and exclu- jphys.2023.02.001 sion criteria,8–11 our data are likely to have greater generalisability to Ethics approval: The South Metropolitan Health Service Human the population of adults who have been referred to a tertiary centre Research ethics committee waived the need for ethics approval for management of their lung disease. The use of the facility protocol because the data were collected as part of normal clinical practice. for the 6MWT in which participants who demonstrated a SpO2 , 80% Competing interests: Nil. were instructed to stop and rest until their SpO2 was 85% may be Source(s) of support: This work received local funding at Fiona seen as a limitation to this study. This is because imposing rests to Stanley Hospital through the WA Chief Allied Health Office Building minimise severe desaturation is likely to have reduced the disparity Research Capacity initiative. in nadir SpO2. A small study with 38 participants being assessed for Acknowledgements: Nil. lung transplant, which did not appear to impose rests during the Provenance: Not invited. Peer reviewed. 6MWT, demonstrated very large differences in test nadir (end-test Correspondence: Kathryn Watson, Physiotherapy Department, SpO2 73% [75 to 89] 6MWT vs 88% [81 to 92] 1minSTS).32 It is Fiona Stanley Hospital, Perth, Australia. Email: therefore likely our estimate of the difference was conservative. A [email protected] small number (11%) of participants were naïve to the 6MWT at the time of testing and only performed one test; performing a second test References may have improved their performance. It is also acknowledged that the gold standard method for quantifying arterial oxygen saturation 1. Granger CL, Morris NR, Holland AE. Practical approach to establishing pulmonary requires direct sampling of arterial blood, but this is beyond the scope rehabilitation for people with non-COPD diagnoses. Respirology. 2019;24:879–888. of routine clinical practice. 2. Holland AE, Spruit MA, Singh SJ. How to carry out a field walking test in chronic In conclusion, this study demonstrated differences in cardiore- respiratory disease. Breathe. 2015;11:128–139. spiratory and symptom responses between the 6MWT and 1minSTS in adults referred to an advanced lung disease service. Of importance, 3. Borg GA. Psychophysical bases of perceived exertion. Med Sci Sports Exerc. the data revealed that using the nadir SpO2 recorded during a 1982;14:377–381. 1minSTS to make decisions about whether strategies are needed to prevent severe transient exertional desaturation during walking-
Research 113 4. Lung Foundation Australia. The pulmonary rehabilitation toolkit. https:// 18. McHugh ML. Interrater reliability: the kappa statistic. Biochem Med. 2012;22:276– pulmonaryrehab.com.au/. [Accessed 5th January 2023]. 282. 5. Leung RWM, Alison JA, Jenkins SC, Holland AE, Hill K, Morris NR, et al. Use of 19. Marin JM, Carrizo SJ, Gascon M, Sanchez A, Gallego B, Celli BR. Inspiratory capacity, supplemental oxygen during exercise testing and training for people with chronic dynamic hyperinflation, breathlessness, and exercise performance during the 6- obstructive pulmonary disease: a survey of Australian pulmonary rehabilitation minute-walk test in chronic obstructive pulmonary disease. Am J Respir Crit Care programs. Braz J Phys Ther. 2021;25:97–102. Med. 2001;163:1395–1399. 6. Yang IA, George J, McDonald CF, McDonald V, O’Brien M, Craig S, et al. The COPD-X 20. Turner SE, Eastwood PR, Cecins NM, Hillman DR, Jenkins SC. Physiologic responses Plan: Australian and New Zealand Guidelines for the management of chronic to incremental and self-paced exercise in COPD: a comparison of three tests. Chest. obstructive pulmonary disease; 2021. https://copdx.org.au/copd-x-plan/p- 2004;126:766–773. prevent-deterioration/p10-oxygen-therapy/. [Accessed 11th November 2022]. 21. Hill K, Dolmage TE, Woon L, Coutts D, Goldstein R, Brooks D. Comparing peak and 7. Grove K, Harrold M, Mohd S, Natarajan V, Hurn E, Pearce J, et al. Research lessons submaximal cardiorespiratory responses during field walking tests with incre- during the COVID-19 pandemic: collecting longitudinal physical and mental health mental cycle ergometry in COPD. Respirology. 2012;17:278–284. outcomes. Arch Public Health. 2022;80:14. 22. Hsia D, Casaburi R, Pradhan A, Torres E, Porszasz J. Physiological responses to 8. Fernandes AL, Neves I, Luis G, Camilo Z, Cabrita B, Dias S, et al. Is the 1-minute sit- linear treadmill and cycle ergometer exercise in COPD. Eur Respir J. 2009;34:605–615. to-stand test a good tool to evaluate exertional oxygen desaturation in chronic obstructive pulmonary disease? Diagnostics. 2021;11:159. 23. Mahler DA, Gifford AH, Waterman LA, Ward J, Machala S, Baird JC. Mechanism of greater oxygen desaturation during walking compared with cycling in patients 9. Crook S, Büsching G, Schultz K, Lehbert N, Jelusic D, Keusch S, et al. A multicentre with COPD. Chest. 2011;140:351–358. validation of the 1-min sit-to-stand test in patients with COPD. Eur Respir J. 2017;49. 24. Gephine S, Bergeron S, Tremblay Labrecque PF, Mucci P, Saey D, Maltais F. Cardiorespiratory response during the 1-min sit-to-stand test in chronic obstruc- 10. Reychler G, Boucard E, Peran L, Pichon R, Le Ber–Moy C, Ouksel H, et al. One minute tive pulmonary disease. Med Sci Sports Exerc. 2020;52:1441–1448. sit-to-stand test is an alternative to 6MWT to measure functional exercise per- formance in COPD patients. Clin Respir J. 2018;12:1247–1256. 25. Sharp JT, Drutz WS, Moisan T, Foster J, Machnach W. Postural relief of dyspnea in severe chronic obstructive pulmonary disease. Am Rev Respir Dis. 1980;122: 11. Meriem M, Cherif J, Toujani S, Ouahchi Y, Hmida AB, Beji M. Sit-to-stand test and 6- 201–211. min walking test correlation in patients with chronic obstructive pulmonary dis- ease. Ann Thorac Med. 2015;10:269–273. 26. Jenkins S, Cecins N. Six-minute walk test: observed adverse events and oxygen desaturation in a large cohort of patients with chronic lung disease. Intern Med J. 12. Holland AE, Spruit MA, Troosters T, Puhan MA, Pepin V, Saey D, et al. An official 2011;41:416–422. European Respiratory Society/American Thoracic Society technical standard: field walking tests in chronic respiratory disease. Eur Respir J. 2014;44:1428–1446. 27. O’Donnell DE, D’Arsigny C, Webb KA. Effects of hyperoxia on ventilatory limitation during exercise in advanced chronic obstructive pulmonary disease. Am J Respir 13. Morris NR, Hill K, Walsh J, Sabapathy S. Exercise & Sports Science Australia (ESSA) Crit Care Med. 2001;163:892–898. position statement on exercise and chronic obstructive pulmonary disease. J Sci Med Sport. 2021;24:52–59. 28. Nonoyama ML, Brooks D, Guyatt GH, Goldstein RS. Effect of oxygen on health quality of life in patients with chronic obstructive pulmonary disease with tran- 14. Bohannon RW, Crouch R. 1-minute sit-to-stand test: systematic review of pro- sient exertional hypoxemia. Am J Respir Crit Care Med. 2007;176:343–349. cedures, performance, and clinimetric properties. J Cardiopulm Rehabil Prev. 2019;39:2–8. 29. Sandland CJ, Morgan MDL, Singh SJ. Patterns of domestic activity and ambulatory oxygen usage in COPD. Chest. 2008;134:753–760. 15. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986;1(8476):307–310. 30. Moore RP, Berlowitz DJ, Denehy L, Pretto JJ, Brazzale DJ, Sharpe K, et al. A randomised trial of domiciliary, ambulatory oxygen in patients with COPD and 16. Alison JA, McKeough ZJ, Leung RWM, Holland AE, Hill K, Morris NR, et al. Oxygen dyspnoea but without resting hypoxaemia. Thorax. 2011;66:32–37. compared to air during exercise training in COPD with exercise-induced desatu- ration. Eur Respir J. 2019;53. 31. Ozalevli S, Ozden A, Itil O, Akkoclu A. Comparison of the Sit-to-Stand Test with 6 min walk test in patients with chronic obstructive pulmonary disease. Respir Med. 17. Pearce L. Primary Care Respiratory Society UK - Opinion No.44; Routine use of 2007;101:286–293. oxygen in primary care; 2011. https://www.pcrs-uk.org/sites/default/files/os44_ routine_o2.pdf. [Accessed 11th November 2022]. 32. Kohlbrenner D, Benden C, Radtke T. The 1-minute sit-to-stand test in lung transplant candidates: an alternative to the 6-minute walk test. Respir Care. 2020;65:437–443.
Journal of Physiotherapy 69 (2023) 114–122 j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j p hy s Research Physiotherapists vary in their knowledge of and approach to working with patients who are LGBTQIA1: a qualitative study Megan H Ross a, Jenny Setchell b a RECOVER Injury Research Centre, The University of Queensland, Brisbane, Australia; b School of Health and Rehabilitation Sciences, The University of Queensland, Brisbane, Australia KEYWORDS ABSTRACT LGBT Question: What are the attitudes, beliefs and knowledge gaps of physiotherapists in Australia regarding Health working with patients who identify as LGBTQIA1? Design: Qualitative design using a custom online survey. Gender Participants: Physiotherapists currently practising in Australia. Methods: Data were analysed using reflexive Sexuality thematic analysis. Results: A total of 273 participants met the eligibility criteria. Participating physiothera- Physical therapy pists were predominantly female (73%), aged between 22 and 67 years, residing in a large Australian city (77%) and working in musculoskeletal physiotherapy (57%), in private practice (50%) and hospital (33%) settings. Almost 6% self-identified as part of the LGBTQIA1 community. Only 4% of the participants had received training related to healthcare interactions or cultural safety for working with patients who identify as LGBTQIA1 in the context of physiotherapy. Three main themes regarding approaches to physiotherapy management were identified: treating the whole person in context, treating everyone the same, and treating the body part. Gaps in knowledge were primarily related to understandings of the relevance of sexual orientation and gender identity to physiotherapy and health issues specific to LGBTQIA1 patients. Conclusion: Physiotherapists can be considered to approach gender identity and sexual orientation in three distinct ways and suggest a range of knowledge and attitudes about working with patients who identify as LGBTQIA1. Physiotherapists who consider gender identity and sexual orientation to be relevant to physio- therapy consultations appear to have higher levels of knowledge and understanding of this topic and may also understand physiotherapy as multifactorial and not only biomedical. [Ross MH, Setchell J (2023) Physiotherapists vary in their knowledge of and approach to working with patients who are LGBTQIA1: a qualitative study. Journal of Physiotherapy 69:114–122] © 2023 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Introduction sexuality and/or gender identity, discomfort with touch and Health and access to healthcare continue to be areas of inequity undressing, fear of discrimination and actual discrimination, and a for individuals who identify as lesbian, gay, bisexual, transgender, lack of knowledge of LGBTQIA1-specific health needs.20 At times, queer, intersex, asexual and related identities (LGBTQIA1).1 Health participants reported not disclosing their LGBTQIA1 status, even disparities for sexual and gender minority groups manifest as poorer when it was relevant to their care. Participants suggested or sup- physical2–7 and mental health,8–12 as well as greater barriers to accessing and receiving healthcare, than the general population. ported several ways in which their physiotherapy experience could be Broader sociocultural issues can also have a negative impact on LGBTQIA1 individuals; discrimination, heteronormative assumptions improved, including diversity training, education specific to the and stigma can have profound effects on wellbeing13 and contribute LGBTQIA1 population and provision of open options on intake to depression, anxiety, substance misuse and suicidal thoughts.14–17 forms.20 As suggested by these participants and other related This social context can also have a significant impact on access to research,21,22 an important approach to improving physiotherapy for and utilisation of healthcare.16,18,19 the LGBTQIA1 community is likely to be enhancing professional Individuals who identify as LGBTQIA1 can experience discrimi- reflexivity about cultural norms, and enhancing culturally safe, nation in healthcare settings, which may contribute to decisions to responsive care23 by improving knowledge, attitudes and skills of avoid accessing healthcare. physiotherapists to engage with diverse patient populations.22,24 Recent research in physiotherapy suggests that individuals who There is a clearly identified need to improve culturally responsive identify as LGBTQIA1 may experience assumptions about their and inclusive healthcare services, delivered by practitioners who are skilled and knowledgeable.16,25–27 Significant gaps have been identi- fied in healthcare professionals’ knowledge and comfort with pa- tients who identify as LGBTQIA1, and their ability to provide https://doi.org/10.1016/j.jphys.2023.02.012 1836-9553/© 2023 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/4.0/).
Research 115 culturally responsive care.28 In Australia, 30% of undergraduate health LGBTQIA1 patients. Topics such as this require methodologies that profession students (including physiotherapy students) indicated that can provide consideration of complexity and nuance.33 Qualitative they would feel uncomfortable working with a patient who identified approaches enable exploration of data in sufficient depth and as a lesbian, and 27% with a gay male patient.29 Furthermore, stu- breadth, and can be inductive rather than pre-imposing a constrained dents overwhelmingly (. 75%) reported that their undergraduate hypothesis.34 Relativism assumes that there is no independent reality, program did not adequately prepare them to manage issues relating but that the social world is comprised of multiple realities and per- to sexuality and gender.29 While this research provides some infor- spectives.35 This paper takes into account the varying perspectives of mation on student attitudes and potential knowledge gaps, it is participants and includes these in the results without resolving dif- believed that the attitudes and practices of physiotherapists working ferences in opinion or experiences. with patients who identify as LGBTQIA1 have not yet been explored. Understanding physiotherapists’ current knowledge, attitudes and Data analysis beliefs will provide opportunities to improve provision of culturally responsive physiotherapy for individuals of the LGBTQIA1 Reflexive thematic analysis was used to provide an overview of community. beliefs and attitudes of the participating physiotherapists following processes outlined by Braun and Clarke.36 We used an inductive Therefore, the research question for this qualitative study was: approach to develop themes and subthemes, that is, themes were derived from the data rather than forcing predetermined classifica- What are the attitudes, practices and gaps in knowledge of tions.37 Although no analysis is free from the influence of the re- physiotherapists in Australia regarding providing care for patients searcher’s preconceptions, the researchers attempted to largely put who identify as LGBTQIA1? these aside (by reflexively questioning assumptions) to support a predominantly inductive and descriptive analysis.38 Analyses were Method conducted in three iterative and overlapping steps39 by both authors, who are trained and experienced qualitative researchers, physio- Design therapists, and members of the LGBTQIA1 community. First, the whole data set was read by the lead author and preliminary codes A custom-made mixed-methods survey was developed based on were generated. Second, after discussion with the second author, pre-existing attitude scales and our previous research that identified preliminary codes were grouped and categorised into subthemes, and client perspectives of key gaps in Australian physiotherapists’ finally three higher level themes were generated. A third party (TLD) knowledge, attitudes and beliefs about working with LGBTQIA1 pa- read the entire dataset and preliminary codes to ensure that codes, tients.20 The survey collected qualitative and quantitative data online; subthemes and themes were grounded in the data. Any discrepancies however, this paper only analyses and reports on the qualitative were included in the findings. The study followed the guidelines for components (with quantitative data to be reported elsewhere). quality analysis in reflexive thematic analysis.40 To address the aim of seeking a broad understanding of the Results knowledge, experiences, attitudes and beliefs regarding the provision of physiotherapy care for patients who identify as LGBTQIA1 in Participants Australia, a survey method was chosen in order to garner the views of a large number of participants, in an anonymous environment where A total of 305 participants responded to the online survey, with they were likely to express their opinions relatively freely with little 273 meeting all eligibility criteria (17 were not currently practising in fear of judgement or counter-response.30 Open questions encouraged Australia) and providing responses to the survey. Participants were participants to provide fairly detailed and descriptive accounts of aged between 22 and 67 years (median 34) with 0 to 44 years (me- their experiences.31 dian 10) of clinical experience as a physiotherapist. Participants predominantly identified as female (72%), married (54%), and work- Physiotherapists currently practising in Australia were invited to ing in a large city (76%) in musculoskeletal (57%), private practice complete the online survey. Participants were recruited online via (50%) or hospital (33%) settings. Sixteen participants (6%) self- social media, word of mouth and the national physiotherapy orga- identified in open responses as being part of the LGBTQIA1 com- nisation and its subgroups, which represent approximately 31,000 munity. Two participants identified as neither male nor female (wrote physiotherapists in Australia. Snowballing was encouraged. Partici- ‘non-binary’ and ‘confused’ in the free text box for gender identity). pants provided informed consent online prior to completing the Detailed participant characteristics are provided in Tables 1 to 3. survey and remained anonymous throughout. Almost 80% of participants indicated that they had knowingly pro- vided care for patients who identified as LGBTQIA1, yet only 13% Data were collected using Qualtrics (www.qualtrics.com). The indicated that they had participated in any training related to purpose-built survey comprised three sections: demographic infor- healthcare interactions or cultural safety for working with patients mation (eg, location, gender, area/subdiscipline of physiotherapy); who identify as LGBTQIA1. Only 4% of the total sample had under- open text box questions exploring the attitudes and beliefs of phys- gone additional training specifically in the context of physiotherapy. iotherapists in relation to knowledge, skills and competencies for providing care for patients identifying as LGBTQIA1; and a quanti- Thematic analysis tative section (reported elsewhere) asking how physiotherapists feel about providing care for LGBTQIA1 patients and their level of support Three high-level themes were created through inductive analyses: for strategies to improve physiotherapy care for this group. treating the whole person in context, treating everyone the same and treating the body part. Each theme describes a differing approach that Participants physiotherapists appear to have in working with patients who identify as LGBTQIA1 (Table 4). Through these themes, the analysis Qualified physiotherapists, currently practising in Australia, were suggests that participants hold broad, varying attitudes, beliefs and eligible to complete the online survey. Participant numbers were levels of knowledge and understanding about the relevance of determined when analyses during data collection demonstrated LGBTQIA1 identity to physiotherapy practice and how patient’s sufficient repetition and depth of concepts to produce rigorous LGBTQIA1 status might be relevant when considering biological, qualitative findings.32 psychological, social and other aspects of their healthcare. Across all themes, physiotherapists’ personal experiences outside of their Theoretical underpinnings A qualitative approach, underpinned by relativism, was taken to better understand the experiences of physiotherapists working with
116 Ross and Setchell: Physiotherapists’ approaches to LGBTQIA1 patients Table 1 Participants (n = 273) Table 3 Participants (n = 273) Demographic characteristics of participants. Work characteristics of participants. 34 (22 to 67) 10 (0 to 44) Characteristic Characteristic 196 (72) 155 (57) Age (yr), median (range) 75 (28) Years of clinical experience, median (range) 25 (9) Gender, n (%) 1 (, 1) Primary area of physiotherapy, n (%) 23 (8) 1 (, 1) 17 (6) female 1 (, 1) musculoskeletal 17 (6) male cardiorespiratory 13 (5) cis female 146 (54) pelvic health 9 (3) nonbinary 80 (29) generalist 6 (2) confused 41 (15) neurological 2 (1) Relationship status, n (%) orthopaedics 2 (1) married 5 (2) geriatrics 1 (, 1) partnered/de-facto 1 (, 1) sports 1 (, 1) single oncology 1 (, 1) divorced 113 (41) paediatrics 1 (, 1) in a relationship 65 (24) clinical education State of residence, n (%) 56 (21) disability 138 (51) Queensland 16 (6) management 90 (33) New South Wales 10 (4) vestibular 13 (5) Victoria Primary site of practice, n (%) 12 (4) Western Australia 7 (3) private practice South Australia 6 (2) hospital 8 (3) Australian Capital Territory 1 (, 1) outpatient clinic 5 (2) Northern Territory community 3 (1) Tasmania 207 (76) university 2 (1) Place of residence, n (%) 62 (23) sporting organisation 1 (, 1) large city defence 1 (, 1) regional 4 (2) aged care 218 (80) remote government body 34 (13) onsite workplace 11 (4) Percentages do not sum up to 100% because of rounding. Provided care for LGBTIQ1 patients, n (%) Participated in LGBTIQ1 training, n (%) Table 2 in the context of physiotherapy, n (%) Qualifications of participants. Percentages do not sum up to 100% because of rounding. Highest obtained level of education, n (%) Participants (n = 273) adjusting treatments, education and other elements of interactions to Bachelor suit them. This considered approach extended beyond LGBTQIA1 Postgraduate Masters 89 (33) status and included consideration of the intersection of many factors Bachelor (Hons) 70 (26) that make up identity, including: age, culture, life experiences/trauma Graduate Entry Masters 40 (15) and sociodemographics. For example: PhD 35 (13) Graduate Certificate 14 (5) Every person deserves to be treated with equal respect, compassion Doctor of Physiotherapy 11 (4) and given the same access to care as any other human being, Fellowship regardless of race, gender, sexuality, prison record, age or any other Graduate Diploma 8 (3) factor (P114). 2 (1) Percentages do not sum up to 100% because of rounding. 4 (2) Some participants’ emphasis on acknowledging and celebrating diversity (P82) and difference was demonstrated through descriptions professional lives seemed to be linked to levels of understanding of adapting to provide culturally sensitive care, rather than mini- (knowledge, attitudes and beliefs) and ways of interacting with or mising difference. One participant described: treating patients (both positive and negative). Individual participants were not categorised into only one of the themes or approaches, as . the change in treatment strategy [for LGBTQIA1 patients] is based they often expressed varying levels of knowledge, attitudes and be- on the need to recognise the impact psychosocial stressors can have liefs across their responses to different questions, which may have on an individual’s pain experience, rather than anything else (P212). been categorised across several themes. Individual participant iden- tifiers are included in brackets (eg, P48). Participants described a psychosocial approach to management, treating the person more broadly rather than only focusing on bio- Theme 1: Treating the whole person in context logical factors. This included responses suggesting that participants Responses that were classified within the theme treating the whole considered that gender identity and sexual orientation can be an important part of a person’s life and inseparable from (part of) the person in context considered the biological, psychological, social and ‘whole person’. One participant said: other aspects (such as stigma and power differences) of a person, in the context of society. Physiotherapists who responded in a way that . I believe it would be important for physiotherapists to understand seemed to consider multiple aspects of the person and their context the associated identity, social and mental health challenges that are described the importance of providing inclusive, equitable and indi- often faced by the LGBTIQ1 community, and the effect of this on the vidualised care for all patients, and seeing the whole person and un- biopsychosocial model . approaches and interventions will need to derstanding their background and needs (P36). For example, one be ‘culturally’ sensitive and appropriate (P125). participant said: This response demonstrates a complex and nuanced under- . physiotherapists need to be aware, respectful, and supportive of standing of how individuals who identify as LGBTQIA1 may experi- the identity and background of each and every patient that we ence healthcare overall, and specifically physiotherapy. This encounter. Every person has a story and an identity, and I believe that understanding included acknowledgement of unique health issues it is important that physiotherapists respect every person as an in- faced by individuals who identify as LGBTQIA1 and how the effects of dividual (P99). Many participants stated that they valued diversity and inclusion, treated everyone respectfully and were sensitive to their needs,
Research 117 Table 4 Description of themes, underpinning approach and description of how approaches are enacted in physiotherapy practice. Theme Description Underpinning approach What is done in practice Treating the whole patient in Considered the biological, psychological, Equity and systemic considerations: patients Efforts to celebrate difference (eg, display context social and other aspects (such as stigma and who have faced discrimination may need LGBTQIA1 progress flag in clinic, power differences) of a patient, in the context extra support to get the same outcomes as celebrating days of significance for of society. those who have not. Considerations extend to LGBTQIA1 communities). broader systems change required. Responses considered multiple aspects of the Intentions to minimise power differences patient and their context described the Extends the biopsychosocial model to (eg, avoiding cis/hetero/endonormative importance of providing inclusive, equitable cultural and power aspects. assumptions about gender identity, sexual and individualised care. orientation, sex characteristics). Use of inclusive language, cultural safety promoted through training, and organisational strategies promoting acceptance and inclusion of diverse identities (eg, inclusive intake forms, pronoun badges, gender-affirming anatomy terminology). Reflexive: acknowledges own biases and adjusts in response (eg, undertaken additional training). Treating everyone the same The notion that physiotherapists should treat Equality: everyone should be treated the Patient likely to feel respected and celebrated everyone the same and provide equal care for same regardless of needs. for who they are. all patients, with a focus on treating The intention is that everyone is treated biological aspects of a patient and their Focus is mainly on the biological and some respectfully, but it is assumed that this presentation. psychological aspects of the biopsychosocial generally does not need to be done in a model with little consideration of social and responsive way. other factors such as power. Some attempts to minimise stigma – but in a limited way (eg, by treating everyone the same) implicitly erasing or minimising the importance of difference. Reflective: aware of broader context for LGBTQIA1 patients, but unsure how to address. Aware of their lack of knowledge and open to further learning. Patient likely to feel a little uncomfortable, like the physiotherapist is trying but doesn’t really understand how to avoid stigma. Treating the body part LGBTQIA1 not relevant to physiotherapy Equality: everyone should be treated the Everyone is treated as a body. because physiotherapy is only about the physical body. same with no consideration of potentially discriminatory attitudes. Not reflexive or reflective unaware of or does not want to address their lack of Biological focus only. Complete separation of knowledge. ‘bio’ from the biopsychosocial model – traditional biomedical approach. High potential for patient to feel stigmatised and marginalised. stigmatisation, assumptions and past trauma can affect patient ex- I am a cisgendered woman, and I am aware that I am a member of periences (eg, minority stress, abuse, criminalisation). Another the majority in Australian society . I would like to think that I would participant provided a similar response: use my position of power to advocate for the healthcare rights of patients who identify as LQBTIQ1 if I saw them being infringed upon. . taking into account the added pressures and challenges LGBTIQ1 In that way, I hope to be an ally (P92). patients experience in our society and the impact on their biological, psychological and social health. We talk about their health goals and Although rare, this type of response suggested that some physio- work towards them in a positive way. I think that a wholistic physio- therapists can reflexively analyse the impact of their own identity on therapy approach that takes into account the patient’s total wellbeing patients with the intention of providing additional support for those is important. These factors are important to consider for every patient who experience marginalisation. Some participants demonstrated if they have an impact on the patient’s physiotherapy goals. I believe reflexivity beyond individual bias, through consideration of systemic every patient deserves to receive high-level care in an environment bias and spoke to the need for broader systems change. For example, that they feel safe and comfortable in (P16). one participant highlighted that in addition to being aware of their own bias physiotherapists need to recognise the often-systemic bias built into The last sentence of this response speaks to the participants’ the system that works against culturally diverse patients (P195). desire for culturally safe clinic environments for patients who are LGBTQIA1. Through their responses, some physiotherapists demonstrated high levels of knowledge and understanding of the relevance of Suggesting they are considering the context the patient lives in, LGBTQIA1 status to physiotherapy by treating the whole person in some participants expressed awareness of the disparities in healthcare context. For example, in response to a question about whether and health related outcomes for patients identifying as LGBTQIA1 and LGBTQIA1 status is relevant to physiotherapy, one participant wrote: acknowledged the importance of actively closing the gap (P16). Direct advocacy for the community was highlighted through some partici- I think if we are truly delivering a person-centred model of care then pant responses such as: yes, it is [relevant]. Especially for those who have suffered with
118 Ross and Setchell: Physiotherapists’ approaches to LGBTQIA1 patients gender dysphoria or issues around expressing their identity as a focus on the physical body and implications of a patients’ LGBTQIA1 LGBTIQ1 individual. However, I believe it is as relevant to care as the status on a biological level positions (minimises) patients to a ‘body’ way someone might express their identity as being a Christian or and demonstrates a limited understanding of the potential psycho- Muslim. Or further as a Torres Strait Islander or Aboriginal. It is a part logical and social factors that may be important in physiotherapy of getting to know a patient and how to build appropriate rapport care. and act appropriately . (P125). In contexts beyond the focus on biology, physiotherapists This is an example of some physiotherapists’ cultural respon- demonstrated a more limited understanding of the relevance of siveness and is evidenced by knowledge and complex understanding LGBTQIA1 status to physiotherapy and limited ability to act. In of individual factors that may contribute to development of thera- particular, understanding of sociocultural aspects and additional peutic relationships with diverse patients. considerations that may be needed for individuals who identify as LGBTQIA1 was lacking: Theme 2: Treating everyone the same I’m sure there are hurdles and difficulties that the LGBTIQ1 com- Treating everyone the same was the most common theme. This munity face on a daily basis that as a heterosexual male I am completely unaware of and likely grossly underestimate. I’m not sure theme encompasses the notion that physiotherapists should treat how to open that dialogue to start the conversation. It may make a everyone the same and provide equal care for all patients. Gender huge difference (P141). identity and/or sexual orientation were generally positioned as separate factors that do not influence physiotherapy management. Encouragingly, like the participant in the quote above, many Responses coded under this theme acknowledged the existence of physiotherapists acknowledged their lack of understanding and some biological, psychological and social aspects of healthcare, but focused explicitly stated that they were willing, and are interested, to learn on treating biological (ie, anatomical and physiological) aspects of a more. One participant said they felt like they . would certainly patient and their presentation (ie, consideration of biological differ- benefit from more specific training in relation to how to best support ences related to gender identity), as this was most relevant to their patients [and] gain insight into their experiences . (P42) and another physiotherapy practice. Understanding of psychosocial and contex- said they were . not well aware of specific health issues affecting tual factors was limited to basic considerations that can be applied to LGBTIQ1 people, [and] this is something I would like to know more everyone, and many physiotherapists considered that modifying care about (P44). based on LGBTQIA1 status was largely unnecessary, as they treat everyone the same. For example, one participant said they: Many physiotherapists described beliefs about treating everyone with equal respect and dignity, demonstrating some (basic) aware- . assume that musculoskeletal anatomy and physiology doesn’t ness of stigma. Physiotherapists said that this did not depend on change based too much on gender identity and not based on sexuality LGBTQIA1 status, and that psychological factors are equally relevant and thus when considering the bio-portion of the BPS model, there for all patients: isn’t too much to consider? (P142). I don’t think they [gender identity and sexual orientation] are rele- Another said they would: vant . I treat all patients with dignity and respect and allow them all to discuss relationships, etc openly whether gay or straight or . struggle to think of a time it [gender identity and/or sexual whatever. It may be relevant if a physio was working with a surgeon orientation] would be relevant, except perhaps if someone was going who was doing gender reassignment surgeries, but otherwise I don’t through gender transition treatment and wanting advice around the see how gender identity or sexuality are relevant to physiotherapy gain/loss of muscle mass (P194). consultations (P27). Participant responses such as these highlight attitudes and beliefs Although ostensibly egalitarian, this type of sentiment seemed at about LGBTQIA1 status only being relevant to physiotherapy care times to deny and minimise difference, as is typified in the when the issue is related to, or affected by biology (ie, for non-binary following quote: They’re just normal people – let’s not cater to them, or transgender patients undergoing hormone treatment or gender- let’s not ignore them (P56). Many physiotherapists responded in affirming surgeries). ways that similarly expressed that ‘special’ treatment or care for LGBTQIA1-identifying individuals is unnecessary, as everyone is Some physiotherapists demonstrated high levels of knowledge the same and should be treated the same. One participant said: about the relevance of LGBTQIA1 status to physiotherapy, but only in They [LGBTQIA1 individuals] are just treated the same as anyone else the biological context: (P213) and another said they . interact with these members in the same manner as I would with heterosexual members . [and] follow Gender identity is important, especially those going through transi- the identical assessment and treatment procedures (P214). These tion or are intersex and on medication. Transgendered status and responses appeared to come from a place of intending not to intersex status are relevant as these patients may be on a variety of discriminate, make assumptions or stereotype (stigmatise) medications that have medical implications. Or, if biologically one sex members of the LGBTQIA1 community, while implicitly erasing or but living as another we have to treat the underlying biology with minimising the importance of difference. For example, one regard to the current LGBTIQ1 status, eg, bone health in males who participant said: [It is] important to also make them feel like they’re then cease testosterone. Or females who go through menopause being treated the same as anyone else as I don’t discriminate based prematurely which increases their risk of cardiovascular disease on a gender (P173). Overall, responses suggested that most par- (P116). ticipants believed they had good intentions: I treat all of my pa- tients young, old, male, female, etc, the same way — with respect In this way, some physiotherapists’ responses recognised and (P205) and demonstrated underlying beliefs that gender identity acknowledged the relevance to physiotherapy only when there were and/or sexual orientation do not matter in the physiotherapy things that change bodies on a biological level (ie, visually, surgically context. Some participants assumed that providing LGBTQIA1 or chemically) and only perceived gender identity to be an issue for patients with ‘different’ care would mean poorer care (ie, partici- transgender patients, acknowledging that there may be concerns pants said that they wouldn’t work differently with LGBTQIA1 around being body conscious, disrobing and binding. For example: I patients), for example: If someone hurts, I provide care regardless of think it may only be an issue with transgender who are very body their identity or orientation. I’m unsure why I would treat anyone conscious and are uncomfortable with disrobing (P8) and a person that differently from any background (P222). is binding might not be comfortable with lifting their shirt (P18). This
Research 119 A final aspect of this theme treating everyone the same related to is the same regardless (P126) and another said: I have not seen a need the interpersonal aspects of physiotherapy care. Physiotherapists for diversity training or any type of different or special treatment for acknowledged psychological and social factors as being important LGBTQIA1 clients . (P194). but separated them to being outside of physiotherapy care. Par- ticipants described being ‘careful’ and ‘cautious’ during in- Some participants expressed firm beliefs that LGBTQIA1 patients teractions, so as not to offend patients who identify as LGBTQIA1 should not be treated differently, and explicitly stated that treatment (ie, using neutral pronouns and not making assumptions). For or approaches should not be modified. For example: example, when responding to a question asking whether LGBTQIA1 status was relevant to physiotherapy, one participant Just treat everyone/every patient the same. Stop singling out various said: Not necessarily [is LGBTQIA1 status relevant], only as a means separate populations such as LGBTIQ1/Immigrant/Coloured/ATSI, to avoid someone being offended . I always refer to their significant etc – by specifically identifying them as different and worthy of other as ‘partner’ . (P9). However, participants discussed that this special or different treatment. This only exacerbates underlying or was important for all patients, regardless of LGBTQIA1 status, and ingrained prejudices. Treat everyone exactly the same regardless many said that . we as clinicians should be sensitive in how we (P115). discuss all matters with all patients so it shouldn’t matter how they identify (P11). Responses such as the quote above demonstrate how a focus on treating the body part can result in more explicit erasure of difference In this way, when physiotherapists described treating everyone the (ie, to intentionally ignore, question or deny the existence of differ- same, regardless of gender identity and/or sexual orientation, they ences that are marginalised, rather than implicit minimisation, which demonstrated moderate levels of knowledge and understanding, and was seen in Theme 2 above) and the potential to be discriminatory implicitly omitted other factors of LGBTQIA1 status of relevance to and stigmatising (ie, not ‘worthy of special or different treatment’). physiotherapy. For example: Participants who focused on a ‘biomedical’ approach completely Gender identity and sexuality are relevant in consultations specif- separated ‘bio’ from the biopsychosocial approach (and only focused ically around terminology used by the physiotherapist. It is important on the anatomy). Many expressed the belief that LGBTQIA1 status to use correct pronouns, and to use gender neutral terms such as was not relevant to musculoskeletal physiotherapy in particular. One partner when asking about social history, in order to create a participant said it was not at all relevant because they were treating comfortable and accepting environment for the patient (P15). musculoskeletal problems (P215) and another said they didn’t think it’s really relevant, and don’t ask about sexuality when managing muscu- These responses highlight participants’ beliefs about the impor- loskeletal conditions (P216). This was often because gender identity tance of LGBTQIA1 status being only relevant in an interpersonal and/or sexuality does not change anything regarding their anatomy and context (ie, not to offend by making heteronormative assumptions), pathology (P272). At times, these approaches demonstrated both omitting other factors of LGBTQIA1 identity that may be relevant to implicit and explicit discrimination and stigmatisation of the physiotherapy. LGBTQIA1 community. This was particularly evident through reducing the community to only focusing on sexuality or sexual Physiotherapists often self-identified their limitations and indi- orientation, and ‘blurring out’ gender. One participant said LGBTQIA1 cated a willingness and openness to learn more, demonstrating status was not relevant because: we are here to treat the client and reflective practice: I think it is very relevant and something I wish there their physio-related issues, which does not mean their sexual preference was more training available for health professionals (P196). Another (P139) and another said nobody’s sexuality has anything to do with participant said they would feel more confident in conversations and their healthcare (P30). Other times, responses demonstrated a lack of would be able to help or direct them [the client] on a better pathway to knowledge and understanding, with no or little consideration for recovery if they had greater general understanding of gender identi- their potential to be stigmatising and indication of willingness to ties. Through quotes like this, many physiotherapists identified a learn more (ie, practice was not reflexive). For example: desire for further education and noted the difference greater knowledge would have on their confidence to work sensitively with I don’t think LGBTQIA1 are any more prone to any other condition the LGBTQIA1 community. that is relevant for physiotherapy care . I understand the education for Aboriginal Health, who get poorer health outcomes. But LGBTIQ1, Theme 3: Treating the body part does it really produce much worse outcomes for interventions or The final theme, treating the body part, encompasses responses conditions that physiotherapist can treat? As far as I know, not really? (P176). that suggest physiotherapists hold attitudes and beliefs indicating that LGBTQIA1 status is not at all relevant to physiotherapy practice. As well as erasure of people’s experiences of marginalisation and Although not the majority, there were a considerable number of re- discrimination, this response is stigmatising, through its omission of sponses coded within this theme. This theme is related to the second first-person language and omission of consideration for the com- theme treating everyone the same but differs from it because re- plexities of interacting with patients who identify as LGBTQIA1 sponses focus around removing the ‘person’ from the ‘body part’, beyond potential health outcomes of physiotherapy management. rather than treating all persons the same. In this way, physiothera- pists actively exclude or do not acknowledge the ‘psychosocial’ as- Discussion pects of care. Responses within this theme separate gender identity and/or sexual orientation from the ‘rest’ of the person and society. This is the first study to explore the knowledge, attitudes and This approach aligns with the idea that physiotherapy does not need beliefs of Australian physiotherapists regarding working with patients to be tailored or modified, at all, for members of the LGBTQIA1 who identify as LGBTQIA1. This analysis produced three main community. themes, each describing a different approach that physiotherapists take to work with patients who identify as LGBTQIA1. Many phys- In this theme, the responses suggest that LGBTQIA1 status is not iotherapists’ responses indicated an ‘egalitarian’ approach, that is, relevant to physiotherapy practice, because physiotherapists treat they took the approach of treating everyone the same. Fewer partici- bodies. This demonstrates a complete lack of understanding or pants’ responses conveyed inclusive equitable approaches to phys- dismissal of broader factors such as social and cultural aspects that iotherapy that considered treating the whole person in context and a can affect how any condition is understood and managed, and/or a biomedical and potentially explicitly stigmatising approach that lack of acceptance that there might be a need for culturally safe care focused on treating the body part. In this study, inclusive and reflective for patients who identify as LGBTQIA1. One participant said they: practitioners demonstrated the highest levels of knowledge and don’t think it [gender identity and/or sexual orientation] should affect physiotherapy consultations nor the care they receive . an ankle sprain
120 Ross and Setchell: Physiotherapists’ approaches to LGBTQIA1 patients understanding about gender and sexual orientation and how these there was no consideration that they lacked important knowledge may relate to physiotherapy, and expressed confidence in sensitively and understanding to provide sensitive and culturally safe care for engaging with patients who identify as LGBTQIA1. the LGBTQIA1 community. In the first theme, physiotherapists considered the whole person Considering these themes together, these results suggest that within society and described the importance of adapting care to be many physiotherapists lacked knowledge required to understand the inclusive and sensitive to individual contextual factors. This relevance of LGBTIQA1 status to physiotherapy, beyond basic bio- approach is underpinned by equity; physiotherapists understood logical factors. This may partly underpin the negative experiences of that individuals who have faced discrimination and/or margin- physiotherapy reported by individuals who identify as LGBTQIA1.20 alisation may need additional (or different) support to achieve the Improving cultural responsiveness may facilitate physiotherapists to same outcomes as those who have not. These physiotherapists adequately address the specific needs of members of the LGBTQIA1 demonstrated reflexivity; the ability to acknowledge their own community.53,54 Developing cultural responsiveness within the pro- biases and adjust and respond reflexively to individual client cir- fession requires individuals to acquire skills, attitudes and behaviours cumstances. Physiotherapists acknowledged that broader organ- consistent with inclusive and sensitive care, as well as professional isational and systems change are required (Table 4). They policies supporting wider systemic cultural change.55 It requires demonstrated high levels of knowledge beyond the biopsychosocial integration of knowledge and understanding, appreciation of indi- model, not only of individual (eg, intersectionality) and environ- vidual differences, and the ability to identify personal biases and mental contextual factors (eg, creating a safe space by displaying assumptions that may influence effective and appropriate commu- diverse images of people and LGBTQIA1 progress flags in clinics) but nication and engagement with patients of diverse cultures and also cultural (including historical) and power aspects that are backgrounds. Not only is this essential to prevent discriminatory important for improving healthcare for individuals who identify as practices, physiotherapists who are reflexive and culturally sensitive LGBTQIA1.41 Often, these physiotherapists had participated in can use their knowledge and understanding to improve the provision additional LGBTQIA1 and/or diversity training and promoted cul- of care, experience of healthcare and, potentially, patient out- tural safety in their practice (Table 4). This individualised and comes.56,57 Consistent with the lack of education found in this study, person-centred approach is aligned with best practice,42 and likely research consistently highlights a lack of education about providing to facilitate positive client experiences when engaging with phys- care for LGBTQIA1 patients in physiotherapy.58,59 iotherapy.43 Clients are likely to feel accepted and celebrated,22 promoting the development of a strong therapeutic relationship One solution to address challenges experienced by patients who and high-value physiotherapy care. identify as LGBTQIA1 when accessing physiotherapy care20 is to enrich the curriculum with LGBTQIA1 health content in tertiary ed- The second theme, the approach of treating everyone the same, was ucation. Consistent with the findings of this study, research identifies the most common. Physiotherapists described treating everyone the a clear need for evidence-based curricula covering terminology, same, regardless of LGBTQIA1 status. Literature suggests that this stigma and health issues specific to LGBTQIA1 individuals to improve approach is not specific to physiotherapists, with studies in other the understanding of how LGBTQIA1 status is relevant to physio- health professions also reporting the treating everyone the same therapy.20 Studies investigating the effects of health practitioner approach to providing care for LGBTQIA1 patients.44,45 This egali- training suggest positive outcomes for improving knowledge, with tarian or one-size-fits-all approach focuses on equal treatment rather less evidence to support that education alone changes attitudes and than equal outcomes, and may prohibit individualisation and beliefs.60,61 For cultural responsiveness training for populations tailoring of physiotherapy care to sensitively meet specific client beyond the LGBTQIA1 population, there is some evidence of needs and inequitable client outcomes.46 In this approach, physio- improved patient experiences and outcomes.56 Future research could therapists focused predominantly on the biological (and some generic evaluate the implementation and longitudinal impact of education on psychological) aspects of their patients and minimised or inadver- health providers knowledge, attitudes and skills, with a focus on tently missed broader psychological, social and contextual factors changes in clinician practice and patient outcomes and experiences. that may be relevant to LGBTQIA1 identity and important to care (ie, assumptions, erasure, history of trauma).47 Physiotherapists who took Findings from this study suggest that overall, training for phys- this approach were reflective; they were aware that they lacked iotherapists about providing culturally responsive and inclusive knowledge in this space and were unsure how to approach or address healthcare for LGBTQIA1 patients would be well received. It also these topics sensitively with patients. Egalitarian practitioners suggests that many physiotherapists desire training that is specific to demonstrated some attempts to minimise stigma and their state- their profession. The lack of understanding about the relevance of ments suggest that they believe that treating everyone the same, and LGBTQIA1 identity to physiotherapy demonstrated in this study not discussing potentially sensitive topics, is the right thing to do. further highlights the need for tailored, profession-specific cultural Despite good intentions (ie, not wanting to reduce patients to their competency training. Other research has suggested that cultural LGBTQIA1 status by focusing on difference), ignoring difference is competency training is effective but needs to be context specific.62 counterproductive to positive client experiences with health services Likewise, findings from this study suggest that lack of knowledge and may lead to feelings of invisibility, discomfort and stigmatisa- and/or understanding contributes to the one-size-fits-all or treat tion48 or receiving inequitable care.49,50 everyone the same approach, which may manifest in unintentionally hostile and/or stigmatising attitudes towards LGBTQIA1 patients. In the third theme, physiotherapists described treating everyone This lack of knowledge and understanding may be effectively tar- the same, based on a belief that LGBTQIA1 identity does not change geted by education, which has the potential to positively influence aspects of biology that are relevant to physiotherapy practice. In this attitudes and behaviours and ultimately the experience of accessing approach, physiotherapists completely separated biology from the physiotherapy for patients who identify as LGBTQIA1. biopsychosocial model and took a traditional biomedical approach.51 Physiotherapists said that because all bodies are the same, everyone There are some methodological considerations relevant for inter- should be treated equally, with no consideration that this approach pretation of the findings of this study. First, because we wanted to may be potentially threatening, discriminatory or stigmatising for explore physiotherapists’ attitudes and beliefs without the influence individuals whose bodies are different (ie, do not fit traditional of social desirability (ie, feeling the need to respond in a manner that gender stereotypes/cisnormative assumptions) or those who have a will be viewed favourably), we used a confidential and anonymous different relationship with their body. This may have a negative survey design to collect open responses from a large sample of par- impact on patient experience with physiotherapy and health-seeking ticipants.31 It is possible that those with strong views (both positive behaviours in a population that already experiences health in- and negative) may have been more likely to respond. However, with equities.52 This is concerning because when individuals do not feel the large sample size and diverse participant characteristics, the heard, accepted or acknowledged, they may actively disengage or be findings are likely to provide a good insight into the Australian harmed. Physiotherapists who took this approach were not reflective; physiotherapy population. We have attempted to facilitate trans- ferability by providing details on context, participant demographics
Research 121 and data collection and analysis; however, these findings may be 7. Conron KJ, Mimiaga MJ, Landers SJ. A population-based study of sexual orientation most relevant to other similar contexts (eg, other Western cultures) identity and gender differences in adult health. Am J Public Health. 2010;100:1953– and potential differences across cultures and settings should be 1960. considered. Both authors identify as part of the LGBTQIA1 commu- nity, with experience as clinical physiotherapists and qualitative 8. Cochran SD, Mays VM, Alegria M, Ortega AN, Takeuchi D. Mental health and research. These factors are likely to have increased the relevance of substance use disorders among Latino and Asian American lesbian, gay, and the survey design and the insights garnered through analysis. bisexual adults. J Consult Clin Psychol. 2007;75:785–794. In conclusion, physiotherapists appear to approach gender iden- 9. Gilman SE, Cochran SD, Mays VM, Hughes M, Ostrow D, Kessler RC. Risk of psy- tity and sexual orientation in three key ways and demonstrate a range chiatric disorders among individuals reporting same-sex sexual partners in the of knowledge and attitudes about working with patients who identify National Comorbidity Survey. Am J Public Health. 2001;91:933. as LGBTQIA1. Physiotherapists who consider gender identity and sexual orientation to be relevant to physiotherapy consultations 10. Koh AS, Ross LK. Mental health issues: a comparison of lesbian, bisexual and appear to have higher levels of knowledge and understanding of this heterosexual women. J Homosex. 2006;51:33–57. topic and may also understand physiotherapy as multifactorial and not only biomedical. Physiotherapists desire education about the 11. McNeil J, Ellis SJ, Eccles FJR. Suicide in trans populations: A systematic review of LGBTQIA1 community that is physiotherapy specific. Addressing prevalence and correlates. Psychol Sex Orientat Gend Divers. 2017;4:341–353. reflexivity and gaps in knowledge about the relevance of gender identity and sexual orientation in a physiotherapy context may 12. Haas AP, Eliason M, Mays VM, Mathy RM, Cochran SD, D’Augelli AR, et al. Suicide facilitate the provision of socially and culturally competent physio- and suicide risk in lesbian, gay, bisexual, and transgender populations: review and therapy care, including equitable access to safe, high-quality and recommendations. J Homosex. 2011;58:10–51. evidence-based care. 13. Bockting WO, Miner MH, Swinburne Romine RE, Hamilton A, Coleman E. Stigma, What was already known on this topic: Patients who mental health, and resilience in an online sample of the US transgender popula- identify as LGBTQIA1 may experience assumptions about tion. Am J Public Health. 2013;103:943–951. gender identity and/or sexual orientation, discomfort about physical touch, observation and undressing, and report a lack of 14. Bariola E, Lyons A, Leonard W. The mental health benefits of relationship formal- physiotherapist education regarding LGBTQIA1-specific health isation among lesbians and gay men in same-sex relationships. Aust N Z J Public issues. Health. 2015;39:530–535. What this study adds: Physiotherapists’ understanding of gender identity and/or sexual orientation and their relationship to 15. Bariola E, Lyons A, Leonard W, Pitts M, Badcock P, Couch M. Demographic and physiotherapy ranged from highly nuanced to implicitly discrim- psychosocial factors associated with psychological distress and resilience among inatory. There are knowledge gaps about the relevance of gender transgender individuals. Am J Public Health. 2015;105:2108–2116. identity and sexual orientation in a physiotherapy context. Training and education may facilitate the provision of socially and 16. McCann E, Sharek D. Survey of lesbian, gay, bisexual, and transgender people’s culturally competent physiotherapy care, including equitable ac- experiences of mental health services in Ireland. Int J Ment Health Nurs. cess to safe, high-quality and evidence-based care. 2014;23:118–127. Ethics approval: This study received ethics approval from The 17. McCann E, Sharek D. Mental health needs of people who identify as transgender: A University of Queensland Human Research Ethics Committee review of the literature. Arch Psychiatr Nurs. 2016;30:280–285. (2020000770). 18. Bauer GR, Hammond R, Travers R, Kaay M, Hohenadel KM, Boyce M. “I don’t think Competing interests: The authors have no conflicts of interest to this is theoretical; this is our lives”: how erasure impacts health care for trans- report. gender people. J Assoc Nurses AIDS Care. 2009;20:348–361. Source(s) of support: This research is funded by a 2019 Physio- 19. Grant JM, Mottet L, Tanis JE, Harrison J, Herman J, Keisling M. Injustice at every turn: therapy Research Foundation Seeding Grant (MHR). Dr Jenny Setchell A report of the national transgender discrimination survey. National Center for is supported by an NHMRC Fellowship (APP1157119). Transgender Equality; 2011. Acknowledgements: The authors would like to thank the partici- 20. Ross MH, Setchell J. People who identify as LGBTIQ1 can experience assumptions, pants for their time completing the survey, and Tami-Lorrie Dawson discomfort, some discrimination, and a lack of knowledge while attending phys- for her contribution to data analysis. iotherapy: a survey. J Physiother. 2019;65:99–105. Correspondence: Megan H Ross, RECOVER Injury Research Centre, 21. Baiocco R, Pezzella A, Pistella J, Kouta C, Rousou E, Rocamora-Perez P, et al. LGBT1 The University of Queensland, Brisbane, Australia. Email: training needs for health and social care professionals: a cross-cultural comparison [email protected] among seven European countries. Sex Res Social Policy. 2021:1–15. References 22. Ross MH, Hammond J, Bezner J, Brown D, Wright A, Chipchase L, et al. An explo- ration of the experiences of physical therapists who identify as LGBTQIA1: Navi- 1. Bonvicini KA. LGBT healthcare disparities: What progress have we made? Patient gating sexual orientation and gender identity in clinical, academic, and Educ Couns. 2017;100:2357–2361. professional roles. Phys Ther. 2022;102:pzab280. 2. Blosnich J, Foynes MM, Shipherd JC. Health disparities among sexual minority 23. Jensen GM. Time to shine the light. Phys Ther. 2022;102:pzab257. women veterans. J Women’s Health. 2013;22:631–636. 24. Primeau CA, Vader K, Philpott HT, Xiong Y. A need for greater emphasis on 3. Farmer GW, Jabson JM, Bucholz KK, Bowen DJ. A population-based study of car- 2SLGBTQIA1 health among physiotherapists in Canada. Physiother Can. diovascular disease risk in sexual-minority women. Am J Public Health. 2022;74:117–120. 2013;103:1845–1850. 25. Neville S, Henrickson M. Perceptions of lesbian, gay and bisexual people of primary healthcare services. J Adv Nurs. 2006;55:407–415. 4. Fredriksen-Goldsen KI, Kim HJ, Barkan SE, Muraco A, Hoy-Ellis CP. Health dispar- 26. Rondahl G, Bruhner E, Lindhe J. Heteronormative communication with lesbian ities among lesbian, gay, and bisexual older adults: results from a population- families in antenatal care, childbirth and postnatal care. J Adv Nurs. based study. Am J Public Health. 2013;103:1802–1809. 2009;65:2337–2344. 27. Rondahl G, Innala S, Carlsson M. Heterosexual assumptions in verbal and non- 5. Case P, Austin SB, Hunter DJ, Manson JE, Malspeis S, Willett WC, et al. Sexual verbal communication in nursing. J Adv Nurs. 2006;56:373–381. orientation, health risk factors, and physical functioning in the Nurses’ Health 28. Eckstrand KL, Potter J, Bayer CR, Englander R. giving context to the physician Study II. J Women’s Health. 2004;13:1033–1047. competency reference set: adapting to the needs of diverse populations. Acad Med. 2016;91:930–935. 6. Bolderston A, Ralph S. Improving the health care experiences of lesbian, gay, 29. Jones MK, Pynor RA, Sullivan G, Weerakoon P. A study of attitudes toward sexuality bisexual and transgender patients. Radiography. 2016;22:e207–e211. issues among health care students in australia. J Lesbian Stud. 2002;6:73–86. 30. Punch K. Survey research: The basics. London: Sage; 2003. 31. Kvale S, Brinkmann S. Interviews: Learning the craft of qualitative research inter- viewing. London: Sage; 2009. 32. Braun V, Clarke V. To saturate or not to saturate? Questioning data saturation as a useful concept for thematic analysis and sample-size rationales. Qual Res Sport Exerc Health. 2021;13:201–216. 33. Bhattacharya K. Fundamentals of qualitative research: A practical guide. Abingdon: Routledge; 2017. 34. Thomas DR. A general inductive approach for analyzing qualitative evaluation data. Am J Eval. 2006;27:237–246. 35. Dieronitou I. The ontological and epistemological foundations of qualitative and quantitative approaches to research. Int J Economics Commerce Manage. 2014;2:1–17. 36. Braun V, Clarke V. Reflecting on reflexive thematic analysis. Qual Res Sport Exerc Health. 2019;11:589–597. 37. Corbin J, Strauss A. Basics of qualitative research: Techniques and procedures for developing grounded theory. London: Sage; 2014. 38. Hsieh H-F, Shannon SE. Three approaches to qualitative content analysis. Qual Health Res. 2005;15:1277–1288. 39. Elo S, Kyngäs H. The qualitative content analysis process. J Adv Nurs. 2008;62:107–115. 40. Braun V, Clarke V. One size fits all? What counts as quality practice in (reflexive) thematic analysis? Qual Res Psych. 2021;18:328–352. 41. Wahlert L, Fiester A. Repaving the road of good intentions: LGBT health care and the queer bioethical lens. Hastings Cent Rep. 2014;44:S56–S65. 42. Sharma T, Bamford M, Dodman D. Person-centred care: an overview of reviews. Contemp Nurse. 2015;51:107–120.
122 Ross and Setchell: Physiotherapists’ approaches to LGBTQIA1 patients 43. Gibson BE, Terry G, Setchell J, Bright FAS, Cummins C, Kayes NM. The micro-politics practitioners on experiences and needs of older LGBT1 people: Findings from a of caring: tinkering with person-centered rehabilitation. Disabil Rehabil. systematic review. Nurse Educ Pract. 2019;40:102625. 2019;42:1–10. 54. Fredriksen-Goldsen KI, Hoy-Ellis CP, Goldsen J, Emlet CA, Hooyman NR. Creating a vision for the future: Key competencies and strategies for culturally competent 44. Beagan BL, Fredericks E, Goldberg L. Discourse/discours-nurses’ work with practice with lesbian, gay, bisexual, and transgender (LGBT) older adults in the LGBTQ Patients: “They’re just like everybody else, so what’s the difference”. Can J health and human services. J Gerontol Soc Work. 2014;57:80–107. Nurs Res. 2012;44:44–63. 55. McKivigan JM. Cultural diversity in the practice of physical therapy: a review of the literature. Int J Humanit Soc Sci Edu. 2021;8:35. 45. Simpson P, Almack K, Walthery P. ‘We treat them all the same’: the attitudes, 56. Lie DA, Lee-Rey E, Gomez A, Bereknyei S, Braddock CH. Does cultural competency knowledge and practices of staff concerning old/er lesbian, gay, bisexual and trans training of health professionals improve patient outcomes? A systematic review residents in care homes. Ageing Soc. 2018;38:869–899. and proposed algorithm for future research. J Gen Intern Med. 2011;26:317–325. 57. O’Shaughnessy DF, Tilki M. Cultural competency in physiotherapy: a model for 46. Fredman S. Substantive equality revisited. Int J Const Law. 2016;14:712–738. training. Physiotherapy. 2007;93:69–77. 47. Coles E, Wells M, Maxwell M, Harris FM, Anderson J, Gray NM, et al. The influ- 58. Parameshwaran V, Cockbain BC, Hillyard M, Price JR. Is the lack of specific lesbian, gay, bisexual, transgender and queer/questioning (LGBTQ) health care education in ence of contextual factors on healthcare quality improvement initiatives: what medical school a cause for concern? Evidence from a survey of knowledge and works, for whom and in what setting? Protocol for a realist review. Syst Rev. practice among UK medical students. J Homosex. 2017;64:367–381. 2017;6:168. 59. McCann E, Brown M. The inclusion of LGBT1 health issues within undergraduate 48. Croghan CF, Moone RP, Olson AM. Working with LGBT baby boomers and older healthcare education and professional training programmes: A systematic review. adults: Factors that signal a welcoming service environment. J Gerontol Soc Work. Nurse Educ Today. 2018;64:204–214. 2015;58:637–651. 60. Jurcek A, Downes C, Keogh B, Urek M, Sheaf G, Hafford-Letchfield T, et al. Educating 49. Moone RP, Croghan CF, Olson AM. Why and how providers must build culturally health and social care practitioners on the experiences and needs of older LGBT1 competent, welcoming practices to serve LGBT elders. Generations. 2016;40:73–77. adults: Findings from a systematic review. J Nurs Man. 2021;29:43–57. 50. Makadon HJ. Ending LGBT invisibility in health care: the first step in ensuring 61. Sekoni AO, Gale NK, Manga-Atangana B, Bhadhuri A, Jolly K. The effects of equitable care. Cleve Clin J Med. 2011;78:220–224. educational curricula and training on LGBT-specific health issues for healthcare 51. Giannou D, Ioakimidis V. ‘Neither invisible nor abnormal!’ Exploring the invisi- students and professionals: a mixed-method systematic review. J Int AIDS Soc. bility and pathologisation of LGBT people in the Greek National Health System. Crit 2017;20:21624. Soc Policy. 2019;40:370–388. 62. Kruse SD, Rakha S, Calderone S. Developing cultural competency in higher edu- 52. Bretherton I, Thrower E, Zwickl S, Wong A, Chetcuti D, Grossmann M, et al. The cation: an agenda for practice. Teach High Educ. 2018;23:733–750. health and well-being of transgender Australians: a national community survey. LGBT Health. 2020;8:42–49. 53. Higgins A, Downes C, Sheaf G, Bus E, Connell S, Hafford-Letchfield T, et al. Peda- gogical principles and methods underpinning education of health and social care
Journal of Physiotherapy 69 (2023) 70–78 j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j p hy s Invited Topical Review Physiotherapy management of blood cancers Catherine L Granger Department of Physiotherapy, The University of Melbourne, Melbourne, Australia; Department of Physiotherapy, The Royal Melbourne Hospital, Melbourne, Australia KEY WORDS [Granger CL (2023) Physiotherapy management of blood cancers. Journal of Physiotherapy 69:70–78] © 2023 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under Hematologic neoplasms Physical therapy the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Exercise Physical activity Rehabilitation Introduction has improved rapidly due to advances in medical treatment4 (5-year survival in the early 1990s was 52%).5 Physiotherapy is an important part of the management of people with blood cancer. Blood cancer is associated with significant With increasing cases and improved survival rates, the prevalence physical and psychological burden. The goals of physiotherapy of blood cancer is rising.1 There were approximately 112,000 people include facilitating the individual to prepare physically for medical with blood cancer in Australia in 2018 and this is expected to reach treatment, maintain health during treatment, recover after treat- 275,000 people by 2035.1 Survivors of blood cancer commonly ment, and maximise health and wellbeing in the survivorship and experience persistent symptoms and complications from the medical palliative stages of disease. Exercise training and physical activity are treatment, resulting in ongoing physical and psychological impair- critical components of physiotherapy management, and this is sup- ments.6 Therefore, there is a growing number of people in the com- ported by growing evidence of the benefit of exercise for people munity with unmet supportive care needs related to blood cancer, with blood cancer. This review summarises the clinical presentation some of which can be addressed by physiotherapy treatment, as of blood cancer, the disease burden, the management of blood discussed in this paper. However, access to supportive care and cancer with a focus on physiotherapy treatment, and future di- physiotherapy is limited:1 a recent Australian survey of 3,227 people rections for research and clinical practice. Whilst blood cancers with blood cancer found that 25% of them wished they had access to affect people of all ages, this review focuses on the management of physiotherapy or occupational therapy for their blood cancer – but adults with blood cancer. they did not.1 One factor contributing to the practice gap is lack of knowledge of clinicians related to this clinical area;7 therefore, ups- What is blood cancer? killing to be informed about blood cancer, the common impairments that patients experience and the evidence about physiotherapy Blood cancer, also known as haematological cancer or haemato- treatments is critical. logical malignancy, is a term used to describe cancers that originate in blood-forming tissues such as bone marrow and the immune system.1 Medical treatments of blood cancer There are five main classifications of blood cancer (Box 1), although there are numerous subtypes within these classifications, including The medical treatments for blood cancer have improved dramat- over 40 subtypes of leukaemia and 50 subtypes of lymphoma.2,3 The ically over recent times. The treatment is specific to the cancer type prognosis and medical treatment vary between subtypes.2–4 and dependent on factors including age, comorbidities, performance status and cancer stage.8 Treatments are given in combination and Blood cancer is the third most common cancer diagnosed in can include chemotherapy, radiotherapy, oral drug therapy such as Australia after breast cancer and colorectal cancer,1,5 and represents tyrosine kinase inhibitors, immunotherapy, blood transfusions and approximately 10% of all cancers diagnosed worldwide.4 The inci- stem cell transplantation. Stem cell transplantation, also known as dence is rising and this is predominately due to the ageing population haematopoietic cell transplantation or bone marrow transplantation, and improvements in diagnosis.1,5 It is more common in males than is a common form of treatment for blood cancer, and a potentially females (age-standardised incidence rate 75 cases per 100,000 males curative treatment option for some patients.8 The rates of stem cell and 48 cases per 100,000 females in Australia)5 and is diagnosed in transplantation are rising in Australia and it is used more frequently people of all ages, although the highest incidence is in people in their in older individuals who previously would have been considered 70s and 80s.1 Blood cancer is the leading cause of non-preventable ineligible due to their age.9 Due to the aggressive nature of stem cell cancer death in Australia.1 The 5-year survival rate is 68%5 and this transplantation, patients are required to have good functional status https://doi.org/10.1016/j.jphys.2023.02.015 1836-9553/© 2023 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/4.0/).
Invited Topical Review 71 Box 1. Common classifications of blood cancer. In autologous stem cell transplantation, the individual’s own stem cells are used. This type of transplant is commonly used to treat Leukaemia multiple myeloma and non-Hodgkin lymphoma.8 The individual’s cancer that develops in the blood and bone marrow usually stem cells are harvested, they then undergo a period of high-dose affecting white blood cells (lymphocytes or myeloid cells) chemotherapy and then the stem cells are reinfused back into the body.8 Autologous stem cell transplantations are associated with less Lymphoma severe side effects than allogeneic transplantations.8 There is a risk of cancer that develops in the lymphatic system complications following both types of transplants, although autolo- gous stem cell transplantation does not have the added risk of GVHD Myeloma and prolonged immunosuppression.8 cancer that develops in the bone marrow in plasma cells The medical treatments for blood cancer are often intense and Myeloproliferative neoplasm prolonged, and patients can be hospitalised for extended periods cancer that originates from a mutation in stem cells in the (Figure 1). The planned and prolonged period of inpatient medical bone marrow and causes overproduction of blood cells treatment makes this patient population somewhat unique compared with many other patient groups that physiotherapists manage. For Myelodysplastic syndrome some types of blood cancer, treatment protocols can continue in the cancer that originates from an acquired mutation of stem outpatient setting for up to 3 years; such is the case for acute cells in the bone marrow and causes a reduction in healthy lymphoblastic leukaemia, where patients undergo chemotherapy in blood cells three phases: induction, consolidation and then maintenance treat- ments.1 The medical treatments used in blood cancer are associated pre-transplant to be eligible for the procedure.8 Physiological age, with toxicities and significant side effects and complications, some of rather than chronological age, is an important consideration for which can be long-standing. A summary of potential early and late suitability.8 complications from stem cell transplantation is provided in Table 1. In allogeneic stem cell transplantation, donor cells from a ‘healthy’ In Australia, patients are usually hospitalised for 3 to 6 weeks to person who is human leukocyte antigen-compatible are used. This receive treatments such as chemotherapy and a stem cell trans- type of transplantation is commonly used to treat acute myeloid plantation, and then stay close to the hospital to be able to attend leukaemia and acute lymphoblastic leukaemia.8 Prior to allogeneic regular outpatient medical appointments for assessment and ongoing stem cell transplantation, the patient must undergo conditioning, medical treatments until 100 days post-transplant. Stem cell trans- which involves chemotherapy, serotherapy and/or whole body plantations are only conducted at a few hospitals in Australia;12 radiotherapy, with the aim of destroying malignant cells in the bone therefore, there is a significant burden associated with the potential marrow and triggering immunosuppression.8 Stem cell trans- requirement for many patients to temporarily relocate for treatment. plantation then occurs, during which donor stem cells from bone Accommodation and travel support is offered by charities such as the marrow, peripheral blood or cord blood are intravenously infused Leukaemia Foundation.13 However, from a physiotherapy perspective, into the body.9 A major complication of allogeneic stem cell trans- this means that patients are likely to be staying in, or close to, major plantation is graft-versus-host disease (GVHD),8 which results in transplant hospitals for months at a time and, on a practical level, chronic inflammation and fibrosis and can affect multiple organs of should be more easily able to participate in physiotherapy and exer- the body.10 Chronic GVHD requires prolonged systemic immunosup- cise programs during this acute phase of treatment (Figure 1).14,15 pressive treatment11 and is discussed in greater detail in subsequent sections. PRE-TREATMENT DURING TREATMENT POST-TREATMENT SURVIVORSHIP PALLIATIVE (months) (weeks to months) (months to years) CARE Medical Home with frequent Inpatient hospital admission/s Accommodation near hospital with Home with Inpatient outpatient frequent outpatient appointments infrequent admission/s or in Physiotherapy treatment Inpatient medical treatments such as outpatient options appointments r day chemotherapy, radiotherapy, Close monitoring; ongoing outpatient appointments community medical treatments such as admission/s immunotherapy, blood transfusions transfusions or oral therapies Surveillance and Support with end-of- and/or stem cell transplantation monitoring life and palliative care Work-up for treatments, Outpatient physiotherapy Inpatient physiotherapy Outpatient Inpatient or home- extensive investigations r Post-treatment Ax: screening for physiotherapy based Ax: mobilisation; low- to moderate- medical treatment complications; Ax: screening for commencement medical intensity exercise program; range of physiotherapy treatment targeting complications and physiotherapy treatment as outpatient or motion/stretching; balance training; Ax: treatment impairments; late effects; directed by patient day oncology ADL training; discharge moderate- to high-intensity exercise physiotherapy to goals: gentle planning/support target impairments; exercise, provision Outpatient program (outpatient or home); moderate- to high- of gait aids, ADL physiotherapy three to five times per week, once or education on return to physical intensity exercise support, symptom twice a day activity and work/hobbies; education program; education management, Initial Ax: on healthy lifestyle behaviours on physical activity relaxation or moderate- to high- and healthy lifestyle breathing exercises intensity exercise program (outpatient or behaviours home-based); education on physical activity Figure 1. Physiotherapy management across the disease and treatment pathway for blood cancers. The medical treatment pathway varies between the different types of blood cancer and is individualised to the patient. This is a simplistic representation of a common treatment pathway and a summary of physiotherapy treatments that may be appropriate at each timepoint. ADL = activities of daily living, Ax = assessment.
72 Granger: Physiotherapy management of blood cancers Table 1 In the longer term, approximately 50% of survivors report at least one Complications after stem cell transplantation. Adapted from Saad et al.8 moderate-to-high level of unmet need, with physical aspects of daily living and psychological needs being the most commonly reported.19 Early complications Late complications Within 5 years of treatment, about 35% of blood cancer survivors (occurring within 100 days) (occurring after 100 days) report ongoing problems with mobility and activities of daily living.6 Poor HRQoL is an ongoing problem; treatment-related side effects and Acute GVHD* Chronic GVHDa fear of recurrence contribute to this.6 Worse HRQoL is seen in survivors of Infection Infection blood cancer who are younger, women or unemployed.6 Prolonged cytopenia and Organ dysfunction Late radiation-related toxicities including Graft-versus-host disease graft failure Organ toxicities cataracts and hypothyroidism Graft-versus-host disease is a major complication of allogeneic Sinusoidal obstruction Late chemotherapy-related toxicities stem cell transplantation (Table 1). It can be mild, affecting single organs, or it can be more severe, affecting multiple organs. Acute syndrome including heart failure GVHD is seen in 20 to 80% of patients, typically affects the skin, liver Secondary malignancies including and gastrointestinal tract, and clinically presents as a rash, gastroin- testinal complications and hyperbilirubinaemia.8 Chronic GVHD myelodysplastic syndromes occurs in approximately 42% of patients within 3 years of trans- plantation,10 although it usually develops within the first 100 to 200 GVHD = graft-versus-host disease. days.10 Chronic GVHD is associated with significant morbidity and a allogeneic stem cell transplantation only. mortality. It requires prolonged systemic immunosuppressive treat- ment11 and infection is the main cause of death.11 The burden of blood cancer The clinical manifestations of GVHD vary depending on the organs Blood cancer is associated with a high number of symptoms, side involved. This can range from the skin, mouth and oral cavity, eyes, effects (Box 2) and complications (Table 1) caused by the underlying genitalia, gastrointestinal tract, liver, lungs, haematopoietic, neuro- cancer and medical treatments. Symptoms and side effects vary logical, immune and/or musculoskeletal systems.11 In the lungs, across the disease and treatment journey; however, people consis- GVHD can cause bronchiolitis obliterans syndrome, airflow obstruc- tently report the presence of a high number of co-occurring symp- tion and recurrent infections,11 clinically presenting as dyspnoea, toms. This is on average 10 symptoms at the time of diagnosis, eight wheeze and cough.11 Neurologically, polyneuropathy, myositis and symptoms in stable disease, 12 symptoms in refractory disease and myasthenia can develop, clinically presenting as muscle weakness, eight symptoms during relapse.16 Many symptoms are seen in people wasting and pain.11 Musculoskeletal manifestations from chronic undergoing medical treatment, in inpatients and those with advanced GVHD and the ongoing use of systemic corticosteroid treatments disease.16 Cancer-related fatigue and feeling worried are the most include osteoporosis, steroid-induced myopathy, fasciitis and scle- common physical and psychological symptoms (Box 2).16 rotic contractures,11 clinically presenting as joint contractures, restricted range of motion, limb swelling, muscle weakness and Cancer-related fatigue is ‘a distressing, persistent, subjective sense wasting.11 There is significant heterogeneity in the clinical presenta- of physical, emotional, and/or cognitive tiredness or exhaustion tion of patients with chronic GVHD; therefore, physiotherapists related to cancer and treatment that is not proportional to recent working with patients who have developed GVHD must perform a activity and interferes with usual functioning’.17 It can be persistent in thorough whole body assessment to identify impairments. blood cancer survivors years after diagnosis/treatment.18 It is the most common ongoing concern of survivors,19 a key barrier to exer- The National Institutes of Health have published a consensus pa- cise20 and is associated with poorer health-related quality of life per on supportive care recommendations for the prevention and (HRQoL).21 treatment of complications related to chronic GVHD.11 The recom- mendations in their entirety are important for physiotherapists to be Distress in blood cancer is high across the domains of physical, aware of; however, several aspects are worth highlighting due to their psychological and global health.16 Younger adults with blood cancer direct relevance to physiotherapists.11 These are summarised in are more likely to experience distress.22 Psychological distress is most Table 2. Recommended treatments within physiotherapy scope of common in patients with a new diagnosis and those who are hos- practice include pulmonary rehabilitation,24 bronchodilators and pitalised.16 Distress associated with physical symptoms is more supplemental oxygen therapy for lung complications; rehabilitation, common in patients with refractory or progressive disease.16 falls prevention and bracing/splints for neurological complications; Following completion of treatment, patients describe distress asso- and exercise training for musculoskeletal complications (Table 2).11 ciated with their ongoing physical impairments and loss of physical health compared with pre-diagnosis.23 Box 2. Physical and psychological symptoms in people with blood cancer.a Adapted from Manitta et al.16 Cancer-related fatigue 69% Changes in skin 24% Physical impairments Feeling worried 50% Feeling bloated 23% Difficulty sleeping 41% Sweats 23% Patients with blood cancer can experience a multitude of physical Drowsiness 41% Nausea and vomiting 22% and 8% impairments. There is a complex pattern of fatigue, reduced physical Feeling sad 41% Itching 21% activity and deconditioning, combined with the impact of the cancer Drowsiness 41% Constipation 20% and medical treatments, which contribute to the development and Dry mouth 40% Weight loss 19% worsening of physical impairments.21 At diagnosis, in the months Pain 39% Difficulty swallowing 18% before treatment, individuals have been found to have reduced Numbness 38% Changes in food taste 18% functional exercise capacity and muscle strength compared with ex- Breathlessness 36% Problems with urination 18% pected values.21,26 In 29 patients with lymphoma undergoing Irritable 36% Loss of sexual interest 18% chemotherapy, pre-treatment 6-minute walk distance and quadriceps Difficult concentrating 34% Diarrhoea 17% force were 89% predicted (IQR 38 to 111) and 82% predicted (SD 16) of Cough 33% Lymphoedema 17% normative values, respectively.21 Reductions in muscle force before Feeling nervous 33% Don’t look like self 16% treatment may be due to a combination of tumour-related factors Lack of appetite 27% Mouth sores 15% stimulating enhanced protein catabolism and decline in protein Dizziness 24% Hair loss 14% synthesis, and increased levels of tumour necrosis factor resulting in muscle wasting and contractile dysfunction.21,27 There is also the a Listed in order of prevalence from a broad range of patients with blood influence of lack of physical activity.21 In a sample of 71 hospitalised cancer (type, stage, treatment and time since diagnosis) as described by Manitta et al.16
Invited Topical Review 73 Table 2 Recommendations within physiotherapy scope of practice for the prevention and treatment of complications from chronic graft-versus-host disease. Adapted from Carpenter et al11 and Majhail et al.25 Organ affected Prevention strategy Treatments Skin Strict sun protection including avoidance of sun exposure 10 am to 4 Deep muscle massage for treatment of stiffness or contractures pm, use of sunscreen and protective clothes Stretches to improve ROM Lung Education against smoking Pulmonary rehabilitation Supplemental oxygen Neurological system Falls prevention Neurological rehabilitation Resistance training Orthotics Musculoskeletal Monitor for reduced limb ROM every 3 to 12 months whilst on Frequent assessments of muscle strength and function (sit to stand immunosuppressive therapy ability) Daily stretching at home Resistance training (isometric, isotonic, isokinetic) Stretching by a physiotherapist two to three times/week for severe Aerobic training Weight-bearing exercise for osteoporosis cases Follow general physical activity guidelines Cardiac and vascular Education on healthy lifestyle: exercise, healthy weight, no smoking and diet Psychological Education on important return to regular exercise and sexual activities Education on management of sexual functioning, body image, sleep, fatigue and musculoskeletal symptoms Regular assessment of carer psychological adjustment and functioning ROM = range of motion. patients undergoing chemotherapy for a range of blood cancers and progressive functional impairment and is a common cause of cancer death.28,34 compared with healthy controls, there were reductions in: skeletal The optimal treatments for sarcopenia and cachexia for patients muscle mass index (whole body) of 4.1 (SD 0.9) versus 4.6 (SD 0.7) kg/ with cancer are unknown.31,35 Exercise is recommended and may m2; quadriceps strength of 44 (SD 17) versus 52 (SD 17) % body diminish the effects of sarcopenia and cachexia through stimulation of muscle protein synthesis and modulation of inflammation and weight ratio; and hand-grip strength of 39 (SD 12) versus 53 (SD 13) % insulin sensitivity.36,37 However, there are limited data on exercise for body weight ratio.26 During treatment patients experience further cancer cachexia and scant data specifically for blood cancers.36 A recent Cochrane review concluded that the effectiveness, accept- reductions in functional exercise capacity, muscle strength and ability and safety of exercise for cancer cachexia (in any cancer type) mass,21 potentially stimulated by the cytotoxic effects of chemo- is uncertain and further research is required.36 Clinically, at present therapy more so than reduced physical activity.21 for patients with cancer, a multimodal treatment approach is rec- ommended including nutritional therapy, exercise training and or Sarcopenia and cachexia are common disorders in blood cancer. anti-inflammatory interventions.31,35 Given the common occurrence Sarcopenia is the age-related loss of muscle mass.28 It is seen in of sarcopenia and cachexia in blood cancer, it is important that physiotherapists include an assessment of muscle strength as well as approximately 51% of patients with blood cancer before treatment, consideration of the nutritional status of their patient prior to, and confirmed with either bioimpedance analysis29 or computed to- alongside, exercise prescription. In the case of sarcopenia or cachexia, mography.30 Sarcopenia is a major concern in cancer as it contributes establishing adequate nutrition is critical; many centres have an to poor tolerance to chemotherapy.31 In blood cancer, sarcopenia is automatic referral process to dieticians for nutritional assessment and management; however, if this is not the case, referral on from the associated with worse outcomes including lower muscle strength, physiotherapist is likely warranted. worse physical function, higher fatigue, greater pain, poorer HRQoL Physical activity recommendations and barriers to compliance in and worse overall survival.29,32,33 Cancer-associated cachexia is the blood cancer cytokine-mediated degradation of muscle and adipose deposits.28 It is The role of physical activity in the prevention and treatment of a complex metabolic syndrome, multifactorial and driven through an cancer is well established.38,39 There is substantial and strong evi- dence for the prevention of many cancers, predominately solid tu- inflammatory response, resulting in it being a common complication mours.39 There is less evidence in blood cancer, although this is in cancer.34 It affects approximately 25% of patients with blood can- growing and the recent recommendations concluded that there is cer,26 although rates are lower than for many other solid tumours.35 now ‘limited’ evidence for the protective effect of physical activity in blood cancer.39–41 A systematic review with 27 trials demonstrated Cancer-associated cachexia leads to muscle wasting, weight loss that higher physical activity levels were protective against lym- phomas (non-Hodgkin lymphoma and Hodgkin lymphoma com- Table 3 bined) (RR 0.89, 95% CI 0.81 to 0.98).40 Data from seven trials demonstrated a dose response with a 1% reduction in lymphoma American College of Sports Medicine physical activity recommendations for the incidence per 3 metabolic equivalent of task (MET) hours/week of prevention and treatment of cancer in general.38 physical activity increase (RR 0.99, 95% CI 0.98 to 1.00, 6,019 partic- ipants).40 Further research on physical activity and blood cancer Recommendations for prevention Recommendations for cancer survivors prevention is warranted. of cancera The recommendations for physical activity for people with cancer Participate in 150 to 300 minutes/wk Avoid inactivity (Table 3) are predominately drawn from solid tumours or mixed of moderate aerobic exercise, or 75 Increase to three 30-minute sessions to 150 minutes/wk of vigorous aerobic exercise per week of moderate intensity aerobic exercise and 20 to 30 minutes, two sessions per week of resistance training Physical activity is associated with improvements in common cancer-related health outcomes including physical function, fatigue, anxiety and depression a Data for the prevention of bladder, breast, colon, endometrial, oesophageal, kidney and stomach cancers.
74 Granger: Physiotherapy management of blood cancers cancer types.38 Research in blood cancer is less advanced,42,43 partly physiotherapy interventions. Exercise training is safe and generally feasible across these timepoints;50 however, the rationale and pre- due to the field being slower to adopt exercise due to safety concerns scription vary considerably. Prescription can range from assisted of an increased risk of infection and bleeding.42 Concern has dimin- mobilisation or gentle chair-based exercises at times during inpatient treatment when blood counts are low, up to moderate- to high- ished with consistent data from trials in blood cancer showing that intensity aerobic and resistance exercise training before or after exercise is safe when conducted within safe parameters.42,43 The treatment in outpatient settings. At each timepoint a unique exercise plan should be developed for the patient50 and be frequently current public message about physical activity for blood cancer is to reviewed and adjusted based on symptom profile and safety pa- rameters. This section describes the ideal scenario of evidence-based avoid inactivity and return to usual activities when able after physiotherapy management; however, in some hospitals and centres diagnosis.44 this may not yet be reflected in current clinical practice. Survivors of blood cancer are poorly compliant with the physical Assessment activity guidelines. In a large population-based Canadian study with An initial physiotherapy assessment is recommended for patients being worked-up for treatment (Figure 1), ideally as part of a clinical 606 blood cancer survivors, only 22% met the physical activity pathway51 and alongside other disciplines including dieticians. The purpose is to assess physical fitness and identify impairments, to be guidelines and these individuals were more likely to be younger and able to target physiotherapy treatments, and monitor deterioration or university educated.45 Another 22% met the aerobic component only complications with treatment.51 Suggested components of an initial assessment are provided in Table 4.51 Physiotherapists may also be ( 150 minutes/week) and 10% met the resistance component only able to access results of respiratory function tests. Physiotherapy (two or more sessions).45 Yet intentions to exercise were high and assessment prior to treatment is not always part of standard care. Often, the first time that a physiotherapist sees the patient it is as an 77% of participants intended to meet part of or the complete inpatient, once treatment has commenced or following referral from guideline.45 People with blood cancer value the importance of exer- medical or nursing colleagues if the patient has mobility, functional cise.46,47 It is viewed as: a positive contribution that they can make or respiratory issues. Assessment during hospitalisation should occur prior to each physiotherapy encounter50 (see next section), and after toward their physical and mental wellbeing; providing a sense of treatment should occur on initial contact with the therapist and then at regular intervals (recommended to be monthly)51 (Figure 1). After empowerment and control; associated with hope for the future; treatment, physiotherapists should screen for signs of complications such as GVHD and target treatment accordingly (Table 2).11 helping to prepare for treatment, manage symptoms; and reducing the risk of recurrence and progression.46,47 Acute physiotherapy during medical treatment Barriers to physical activity with blood cancer are numerous and include: symptoms and side effects, including fatigue, pain, short- ness of breath and lack of strength; comorbidities and age-related inactivity; injuries; and lack of self-confidence and motiva- tion.20,46,47 Fatigue is the most commonly reported barrier,20 whereas social barriers such as time and cost are less frequent.20 Activity levels are lowest during treatment48,49 and consistently do not return back to pre-diagnosis levels.20,48 The leading predictor of post-diagnosis physical activity is pre-diagnosis physical activity.20 This suggests that physiotherapy assessment of physical activity history is vital, and those individuals who have been less active may need additional support. Reassurance and early education that exercise will improve and not worsen fatigue are important for patients and families.20 Physiotherapy management of blood cancer Acute physiotherapy is typically delivered in the inpatient hospital Physiotherapy management in blood cancer is centred around setting whilst patients are hospitalised for medical treatment. The physical activity and exercise training. Figure 1 depicts the common timepoints at which physiotherapists can intervene during the goals of physiotherapy treatment are maintenance of strength and treatment pathway and Figure 2 summarises the level of evidence for function during this period of time when patients usually experience significant deconditioning.50 Physiotherapy routinely includes mobi- lisation and exercise training;51 however, other treatments including Benefit from the intervention Uncertain effects Exercise training continuing across treatment pathway (before, during and after treatment) Systematic reviews Exercise training continuing across treatment of multiple RCTs pathway (during and after treatment) Exercise training and mobilisation during hospitalisation Exercise training starting after treatment Single high-quality Inspiratory muscle training RCT or several low- Exercise training only pre-treatment quality RCTs Whole body vibration training Uncontrolled trials Physiotherapy in palliative care or expert opinion Deep tissue massage or stretches in cGVHD with contractures Pulmonary rehabilitation for lung cGVHD Figure 2. Interventions for the management of blood cancer with associated levels of evidence. ‘Exercise training only pre-treatment’ refers to data from studies that only included exercise at this time point and did not continue with an exercise program during or after treatment. cGVHD = chronic graft-versus-host disease, RCTs = randomised controlled trials.
Invited Topical Review 75 Table 4 tigue compared with decline in control groups,43,50 although the re- sults are inconsistent, trials are small and the risk of bias is high.43,50 Physiotherapy assessment pre-treatment in blood cancer. Adapted from Mohammed et al.51 Many studies have been unable to attenuate deconditioning experi- enced during treatment.14,43 There may be earlier recovery of blood Outcome/component Example measures counts,52 but other studies have not demonstrated between-group differences.43 Commonly, exercise is tested as part of a broader pre- Subject assessment/history Comprehensive physiotherapy subjective to-post exercise program. The meta-analysis by Abo et al43 of seven assessment including: trials commencing exercise either before or during stem cell trans- medications plantation demonstrated a reduced hospital length of stay in favour existing musculoskeletal and comorbid of exercise training (MD 2.1 days, 95% CI 0.4 to 3.7, 379 participants, low-certainty evidence).43 Cost-effectiveness of such programs is conditions unknown. occupation and return to work or study goals physical activity levelsa – current and past Daily assessment before physiotherapy treatment is important functional impairments symptomsa and should include symptom levels, vital signs and blood counts anxiety, depression or distressa (haemoglobin, platelets and white blood cells).50 Clinical decision- Functional exercise 6-minute walk test or making is required to tailor an appropriate mobilisation or exercise capacity incremental shuttle walk test treatment that takes the patient to their highest functional level Physical function 30-second chair stand test, Short Physical Performance Battery or within safety limits. Discussion with the treating medical team is Timed Up and Go recommended regarding the planned physiotherapy treatment and Self-reported function Lower Extremity Functional Scale or Disabilities of the Arm, Shoulder and Hand specific precautions or contraindications for the patient at that time. Questionnaire Wiskemann et al53 and Abo et al14 utilised models where patients Peripheral muscle and Hand-held and hand-grip dynamometers self-rated their wellbeing prior to each session and this rating was hand-grip strength used by therapists to adjust the exercise plan (Figure 3). Peripheral muscle mass Ultrasound Patients are likely to have anaemia, thrombocytopenia and neu- Frailty Clinical frailty scale tropenia, which impact the appropriateness of physiotherapy treat- Upper and lower limb Range measured with a goniometer range of motion ments. There is a balance between bed rest to prevent bleeding, HRQoL EORTC QLQ-C30 infection or clinical compromise versus the risk of functional decline EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer and deconditioning associated with prolonged sedentary time. Quality of Life questionnaire, HRQoL = health-related quality of life. Several publications have reported haematological safety criteria for a Can also be assessed using self-reported questionnaires. physiotherapy or exercise during blood cancer treatment.15,50,51,54 range of motion, balance training and activities of daily living training Table 5 provides an example of a safety protocol based on the may be required (Figure 1). The exercise programs tested in the consensus-based recommendations for acute physiotherapy for literature to date vary, although they are most commonly individu- patients with multiple myeloma undergoing chemotherapy and alised, supervised, low- to moderate-intensity aerobic exercises,50,51 stem cell transplantation,50 which is endorsed by the Canadian sometimes supplemented with resistance exercises and unsuper- Physiotherapy Association. Physiotherapists should also refer to their vised exercise on non-physiotherapy days.50,51 Physiotherapy is often local hospital policies and work in consultation with the medical team. delivered on a 1:1 basis either in the patient’s single room or on a pressure-filtered ward to minimise infection risk;43 however, Other emerging physiotherapy treatments investigated in studies emerging data suggest that group-based exercise on a pressure- to date include inspiratory muscle training, whole body vibration55 filtered ward is safe14 and potentially adds the psychological benefit and mindful breathing.56 In the trial by Almeida et al,57 57 in- of peer support.48 patients undergoing stem cell transplantation received standard care The evidence to date suggests that exercise during treatment is physiotherapy (early mobilisation, breathing exercises and aerobic safe and feasible, with very few adverse events reported. The evi- training). The intervention group additionally performed inspiratory dence for efficacy of exercise during treatment is inconsistent. Results of recent systematic reviews42,43 show neutral or positive benefits of muscle training, 5 days/week, at 40% baseline maximal inspiratory exercise on patient outcomes of muscle strength, endurance and fa- pressure, through an inspiratory pressure threshold loading device and supervised by a physiotherapist.57 The inspiratory muscle training group had improved maximal inspiratory pressure at hos- pital discharge (117%) compared with a reduction in the control group (–20%) from admission.57 Both groups maintained muscle strength, functional exercise capacity and physical function from admission to hospital discharge (median 21 days) attributed to the standard care physiotherapy.57 Self-rating Score of out 10ᵃ Health status Exercise prescription Green 8 to 10 Good or normal health Moderate-intensity No symptoms aerobic (Borg 3 to 5 out of 10) and No concerns resistance (RPE 12 to 14 out of 20) exercises Orange 5 to 7 Medium health Low- to moderate-intensity Red 0 to 5 Mild symptoms of fatigue aerobic (Borg 1 to 3 out of 10) and resistance (RPE 9 to 11 out of 20) exercises and nausea Very mild pain Low-intensity aerobic exercises usually chair based (Borg 1 to 3 out of 10) and functional Low health Fatigue, nausea or pain exercises (RPE 9 to 11 out of 20) without Borderline blood counts resistance Figure 3. Traffic light system (patient-self rating) used to assess patient wellbeing and inform daily adjustments to inpatient exercise prescription, as utilised by Abo et al.14 Borg = Modified Borg Dyspnoea scale, RPE = Rating of Perceived Exertion scale. a Rating from 0 (feeling very unwell, significant symptoms, difficult to get out of bed) to 10 (feeling well, normal health status, no issues).
76 Granger: Physiotherapy management of blood cancers Table 5 Safety considerations for acute physiotherapy (mobilisation and exercise prescription) for blood cancer. Adapted from Jeevanantham et al.50 Condition Precautions and contraindications50 Anaemia If haemoglobin , 80 g/L Thrombocytopenia Exercise training generally contraindicated Consult with physician before commencement Leukopenia and neutropenia Physiotherapy intervention with caution Bony lesions or skeletal issues Mobilisation with caution and monitor closely for vital signs and adverse events including chest pain, pallor, leg cramps, dizziness, arrhythmias, shortness of breath, respiratory distress, SBP . 200 mmHg, DBP . 110 mmHg, drop in SBP . 10 mmHg from baseline, HR . 120 bpm, SpO2 , 88% on activity, positive orthostatic response During red blood cell transfusion Mobilisation with caution and monitor closely for abnormal vital signs and adverse events including dislodgment of intravenous site, syncope or reaction to blood products Monitor for signs of bleeding Educate patients about signs of bleeding and precautions Caution with equipment and clothing to prevent bleeding Consult with physician about appropriateness of physiotherapy, weighing up the balance for bed rest to prevent bleeding versus functional decline with inactivity If platelet count , 10,000/mL Only essential ambulation such as to bathroom, with assistance or supervision to prevent falls Gentle ROM in sitting or lying If platelet count 10,000 to 20,000 mL Gentle ROM and exercises without resistance or strain Exercise in standing or mobilisation only if patient ambulates unassisted, independently and has no signs of bleeding If platelet count 20,000 to 40,000/mL Low-intensity (gentle) aerobic exercise permitted Light resistance exercises using elastic bands without strain and only if no signs of bleeding If platelet count . 40,000/mL Low-intensity (gentle) aerobic exercise including stationary cycling permitted If platelet count , 50,000/mL Avoid high-intensity exercise Patient to wear a face mask when ambulating outside of room Patient to perform regular hand hygiene, including before and after physiotherapy Equipment to be cleaned before and after use Physiotherapy intervention delivered in patient’s room Exercise under supervision Modify exercise program to patient situation Avoid high-impact activities and end range of motion. Exercises should avoid twisting, bending, overhead reaching, pushing, pulling and lifting weights. Ensure correct technique, including back care, and use gait aids and braces as needed DBP = diastolic blood pressure, HR = heart rate, ROM = range of motion, SBP = systolic blood pressure, mL = microlitre. Exercise training before and after treatment participants, moderate-certainty evidence) and a small effect for Physiotherapy before and after treatment is mainly focused on reduction in the depression subscale of HRQoL (SMD 0.19, 95% CI 0.0 exercise training (Figure 1). Variations in clinical practice exist in terms of the degree of physiotherapy input available; however, pre- to 0.38, 445 participants, low-certainty evidence). No clear between- habilitation and rehabilitation programs are more commonly group differences were observed for mortality (RR 1.10, 95% CI 0.79 to becoming standard care for this patient group in line with growing evidence. For patients after transplantation, if there is not a hospital/ 1.52, 1,172 participants, low-certainty evidence), global HRQoL, outpatient program available, home-based exercise is encouraged physical function HRQoL or anxiety-domain HRQoL. The HRQoL re- and it is recommended to avoid exercising with the public, such as in a gym, until there has been full recovery of the immune system.38 sults did not change in sub-analyses of patients receiving chemo- Before treatment, exercise aims to maximise muscle strength and therapy only or stem cell transplantation. Meta-analyses were unable exercise capacity in order to attenuate the decline with treatment. Higher strength and exercise capacity going into treatment are to be completed for exercise capacity or strength, due to heteroge- associated with better performance in these measures after treat- neity in measurement of these outcomes; individual trials report ment.58 After hospitalisation and treatment, exercise aims to target conflicting results (eight RCTs benefit and five RCTs no difference).42 residual impairments, symptoms and side effects. At this point, The authors suggest that variations in the exercise programs, medi- symptoms and wellbeing fluctuate, and exercise volume and intensity should be reviewed and adapted daily depending on patient cal treatments and standard care may have influenced their presentation.38 outcomes.42 Knips et al recently updated their Cochrane review of aerobic The systematic review by Abo et al43 focused specifically on pa- exercise for patients with blood cancer.42 The update includes 18 tients with blood cancers undergoing stem cell transplantation and randomised controlled trials (RCTs) and 1,892 participants with blood cancer exercising at different points across the treatment pathway. included 27 studies (24 RCTs), with one-third of studies published in The earliest trial started recruitment in 200259 and patients received the last 5 years. Only two studies focused entirely on pre-transplant a stem cell transplant in half of the trials.42 The exercise programs exercise;61,62 both were small feasibility studies. Neither study tested in trials varied considerably: home, inpatient or outpatient demonstrated significant changes in patient outcomes before and based; consisting of either aerobic, resistance or mixed training. The after the program, and both reported low to moderate feasibility.61,62 timing of programs varied across before, during and after treatment. Abo et al43 included 10 studies with exercise entirely during treat- The longest program was 30 weeks60 and shortest program commenced 6 days before transplant and finished the day before ment, six entirely after treatment/hospitalisation, and 10 with exer- hospital discharge after treatment.59 Meta-analyses demonstrated cise across the pre-, during and or post-treatment settings. The mean significant benefit on fatigue (SMD 0.31, 95% CI 0.13 to 0.48, 826 PEDro score of RCTs was 5 (SD 1), which is illustrative of the high risk of bias within the RCTs published to date. Meta-analyses demon- strated positive effects of exercise compared with no exercise on functional exercise capacity (6-minute walk distance MD 29 m, 95% CI 13 to 45, 542 participants, moderate-certainty evidence) and upper and lower body muscle strength (lower body strength SMD 0.39, 95% CI 0.20 to 0.58, 431 participants, low-certainty evidence).43 The dif- ference between these results and Knips et al42 may be due to the narrower inclusion (only stem cell transplantation recipients) and
Invited Topical Review 77 addition of several new trials. Meta-analysis results concurred with related to exercise and safety are now lessened. Given the significant the findings by Knips et al42 in terms of positive effects on global disease burden and residual complications experienced by patients HRQoL and fatigue.43 Effects on functional exercise capacity and well after treatment has concluded, there is a strong rationale for fatigue were larger in patients undergoing allogeneic stem cell physiotherapy and exercise training to address these impairments. transplant than autologous, although certainty for these sub-analyses With prevalence of blood cancer increasing, there is urgency for high- was low.43 quality evidence to inform clinical practice. Questions remain around: the optimal type, timing and setting of exercise training; the cost- The feasibility of pre- and post-treatment exercise programs varies effectiveness of programs; the impact on overall survival; and in the literature. For example, in patients undergoing a transplant, longer-term patient outcomes beyond 1 year post-treatment. Addi- consent varies from 23 to 99% and adherence from 24 to 92%.43 In the tionally, many of the trials conducted to date are limited by high risk pre-treatment setting, home-based and outpatient-based programs of bias and most conclusions from recent meta-analyses are low have been tested, but both appear to have problems related to feasi- certainty. Future trials are still needed and should avoid prior limi- bility. Van Haren et al61 included 29 patients of mixed blood cancer tations, including lack of assessor blinding, incomplete outcome data types undergoing either allogeneic or autologous stem cell transplant. and selective reporting of outcomes without prior trial registration. The intervention was a physiotherapist-supervised, moderate-to high- intensity aerobic and resistance training program, twice/week, for 4 to There is a huge interest in this field and numerous trials in 6 weeks pre-transplant. Adherence to the program was 69% but the progress, which will soon significantly improve understanding (there median number of sessions attended was only six. Wood et al62 are currently 25 open exercise studies registered on ClinicalTrials. included 34 patients who were awaiting allogeneic stem cell trans- gov). Several of these trials include multi-modal interventions offer- plantation. The intervention was an unsupervised, home-based, high- ing exercise combined with nutritional support, relaxation strategies, intensity interval aerobic exercise program supported by weekly mindfulness, cognitive behavioural therapy, psychology, and/or input motivational phone calls and provision of an activity tracking device. from other allied health disciplines as both a multidisciplinary pre- Recruitment was ceased early due to challenges, and intervention habilitation or rehabilitation program. adherence was poor. The authors recommended that future trials maintain high frequency of patient contact (such as through remote Due to the COVID-19 pandemic, clinical practice in the field was monitoring) and focus on exercise close to and during the trans- forced to adapt rapidly and, in many centres, exercise programs for plantation period, as most other trials have done. The largest RCT in patients with blood cancer delivered by telehealth (such as video- exercise and blood cancer was a four-arm RCT (self-directed exercise, conferencing) are now part of standard care. In this situation, clinical self-administered relaxation, both and neither) with 711 patients.63 practice moved forward by implementing telehealth-delivered This trial demonstrated no difference in outcomes, including HRQoL programs before evidence was generated. A major emphasis of the at 100 days, distress, sleep or survival, and suggested that a hands-off, current trials is investigating the efficacy of exercise delivered via self-directed approach may not be effective in this population.63 Early telehealth. Lee et al66 are investigating the benefit of a home-based after discharge from hospital, patients routinely attended outpatient exercise program, supervised by a clinical exercise physiologist, via medical appointments up to three times/week for the first 100 days. telehealth for 60 participants with multiple myeloma 30 to 180 days This provides an opportunity for outpatient hospital-based physio- after stem cell transplantation. The 8-week individualised exercise therapy input including exercise programs (Figure 1). Local experience program includes 24 videoconferencing exercise sessions, run for at is that there is good feasibility of offering an exercise program for least 30 minutes on 3 days/week. The primary outcome is physical patients to join at the hospital when attending their medical function measured by the Short Physical Performance Battery. The appointments during this time.15 trial is currently recruiting and estimated to be completed at the end of 2024. Given the issues of feasibility described by some prior trials, Education forms an important part of physiotherapy treatment for telehealth is an appealing model so this is likely to inform ongoing people with blood cancer. This includes advice to keep active and use of this model in clinical practice if the efficacy data are favourable. safely return to physical activity when able; physical activity guide- lines; barriers to being active; symptom management; healthy eAddenda: Nil. lifestyle behaviours including diet; infection control; and safety Ethics approval: Not applicable. considerations related to exercise. For patients who do not have Competing interests: Nil. access to physiotherapy or an exercise program at their treating Source(s) of support: Nil. hospital, referral to a community oncology rehabilitation or pulmo- Acknowledgements: Dr Shaza Abo. nary rehabilitation program is recommended.64 Provenance: Invited. Peer reviewed. Correspondence: Catherine L Granger, Department of Physio- Physiotherapy in palliative care therapy, The University of Melbourne, Melbourne, Australia. Email: [email protected] Physiotherapy for patients in palliative or end-of-life care is References typically focused on supporting the individual to maintain their in- dependence, comfort and HRQoL, as well as reduce carer burden. 1. Leukaemia Foundation. State of the Nation: Blood Cancer in Australia. 2019. Physiotherapists work closely with the palliative care multidisci- 2. Khoury JD, Solary E, Abla O, Akkari Y, Alaggio R, Apperley JF, et al. The 5th edition of plinary team in this setting. Treatment should relate directly to the goals of the patient; it may focus on alleviating symptoms, main- the World Health Organization Classification of Haematolymphoid Tumours: taining physical function and ability to perform activities of daily Myeloid and Histiocytic/Dendritic Neoplasms. Leukemia. 2022;36:1703–1719. living.65 Treatments are individualised to the patient and can include 3. Alaggio R, Amador C, Anagnostopoulos I, Attygalle AD, Araujo IB, Berti E, et al. The mobilisation, less intensive exercise, provision of gait aids, activities 5th edition of the World Health Organization Classification of Haematolymphoid of daily living training/support, symptom management, and relaxa- Tumours: Lymphoid Neoplasms. Leukemia. 2022;36:1720–1748. tion or breathing exercises (Figure 1). 4. Siegel R, Miller K, Wagle N, Jemal A. Cancer statistics, 2023. CA. 2023;73:17–48. 5. Australian Institute of Health Welfare. Cancer data in Australia. Canberra: AIHW; Future directions for research and practice 2022. 6. Immanuel A, Hunt J, McCarthy H, van Teijlingen E, Sheppard ZA. Quality of life in Physiotherapy plays an important role in the management of survivors of adult haematological malignancy. Eur J Cancer. 2019;28:e13067. blood cancer across the disease and treatment pathway. There has 7. McCourt O, Yong K, Ramdharry G, Fisher A. Physical Activity During and After Hae- been significant growth in the evidence, especially in the last 5 years, matological Cancer Treatment: A Cross-Sectional Survey of Haematology Healthcare mainly focused on the efficacy of exercise training. Early concerns Professionals in the United Kingdom. J Multidiscip Healthc. 2021;14:1659. 8. Saad A, de Lima M, Anand S, Bhatt VR, Bookout R, Chen G, et al. Hematopoietic Cell Transplantation, Version 2.2020, NCCN Clinical Practice Guidelines in Oncology. JNCCN. 2020;18:599–634. 9. Nivison-Smith I, Bardy P, Dodds AJ, Ma DD, Aarons D, Tran S, et al. A Review of Hematopoietic Cell Transplantation in Australia and New Zealand, 2005 to 2013. Biol Blood Marrow Transplant. 2016;22:284–291.
78 Granger: Physiotherapy management of blood cancers 10. Bachier CR, Aggarwal SK, Hennegan K, Milgroom A, Francis K, Dehipawala S, et al. 40. Davies GA, Strader C, Chibbar R, Papatheodorou S, Dmytriw AA. The relationship Epidemiology and Treatment of Chronic Graft-versus-Host Disease Post-Allogeneic between physical activity and lymphoma: a systematic review and meta analysis. Hematopoietic Cell Transplantation: A US Claims Analysis. Transplant Cell Ther. BMC cancer. 2020;20:962. 2021;27:504.e501–504. 41. Jochem C, Leitzmann MF, Keimling M, Schmid D, Behrens G. Physical Activity in 11. Carpenter PA, Kitko CL, Elad S, Flowers ME, Gea-Banacloche JC, Halter JP, et al. Relation to Risk of Hematologic Cancers: A Systematic Review and Meta-analysis. National Institutes of Health Consensus Development Project on Criteria for Clin- Cancer Epidemiol Biomarkers Prev. 2014;23:833–846. ical Trials in Chronic Graft-versus-Host Disease: V. The 2014 Ancillary Therapy and Supportive Care Working Group Report. Biol Blood Marrow Transplant. 42. Knips L, Bergenthal N, Streckmann F, Monsef I, Elter T, Skoetz N. Aerobic physical 2015;21:1167–1187. exercise for adult patients with haematological malignancies. Cochrane Database Syst Rev. 2019;2019:undefined-undefined. 12. Australian Bone Marrow Donor Registry. Information for Patients: Adult transplant centres. https://www.abmdr.org.au/patients/adult-transplant-centres/. Published 43. Abo S, Denehy L, Ritchie D, Lin KY, Edbrooke L, McDonald C, et al. People With 2023. [Accessed 27/01/2023]. Hematological Malignancies Treated With Bone Marrow Transplantation Have Improved Function, Quality of Life, and Fatigue Following Exercise Intervention: A 13. Leukaemia Foundation. About Leukaemia Foundation accommodation services. Systematic Review and Meta-Analysis. Phys Ther. 2021;101. https://www.leukaemia.org.au/how-we-can-help/accommodation-services/about- accommodation-services/. Published 2023. [Accessed 27/01/2023]. 44. Cormie P, Atkinson M, Bucci L, Cust A, Eakin E, Hayes S, et al. Clinical Oncology Society of Australia position statement on exercise in cancer care. Med J Aust. 14. Abo S, Ritchie D, Denehy L, Panek-Hudson Y, Irving L, Granger CL. Feasibility of 2018;209:184–187. early-commencing group-based exercise in allogeneic bone marrow trans- plantation: the BOOST study. Bone Marrow Transplant. 2021. 45. Vallerand JR, Rhodes RE, Walker GJ, Courneya KS. Correlates of meeting the combined and independent aerobic and strength exercise guidelines in hemato- 15. Abo S, Ritchie D, Denehy L, Panek-Hudson Y, Irving L, Granger CL. A hospital and logic cancer survivors. IJBNPA. 2017;14:44. home-based exercise program to address functional decline in people following allogeneic stem cell transplantation. Support Care Cancer. 2018;26:1727–1736. 46. Colton A, Smith M, Broadbent S, Rune K, Wright H. Perceptions of Older Adults with Hematological Cancer on Diet and Exercise Behavior and Its Role in Navi- 16. Manitta V, Zordan R, Cole-Sinclair M, Nandurkar H, Philip J. The symptom burden gating Daily Tasks. In. Int J Environ Res Public Health. Vol 19: MDPI AG; 2022: of patients with hematological malignancy: a cross-sectional observational study. 15044–15044. J Pain Symptom Manage. 2011;42:432–442. 47. Abo S, Parry SM, Ritchie D, Sgro G, Truong D, Denehy L, et al. Exercise in allogeneic 17. Bower JE, Bak K, Berger A, Breitbart W, Escalante CP, Ganz PA, et al. Screening, bone marrow transplantation: a qualitative representation of the patient assessment, and management of fatigue in adult survivors of cancer: an American perspective. Support Care Cancer. 2022;30:5389–5399. Society of Clinical oncology clinical practice guideline adaptation. J Clin Oncol. 2014;32:1840–1850. 48. Vallance JKH, Courneya KS, Jones LW, Reiman T. Differences in quality of life be- tween non-Hodgkin’s lymphoma survivors meeting and not meeting public health 18. Hjermstad MJ, Fosså SD, Oldervoll L, Holte H, Jacobsen AB, Loge JH. Fatigue in exercise guidelines. Psycho-Oncol. 2005;14:979–991. Long-Term Hodgkin’s Disease Survivors: A Follow-Up Study. J Clin Oncol. 2005;23:6587–6595. 49. Danaher EH, Ferrans C, Verlen E, Ravandi F, van Besien K, Gelms J, et al. Fatigue and physical activity in patients undergoing hematopoietic stem cell transplant. Oncol 19. Barata A, Wood WA, Choi SW, Jim HSL. Unmet Needs for Psychosocial Care in Nurs Forum. 2006;33:614–624. Hematologic Malignancies and Hematopoietic Cell Transplant. Curr Hematol Malig Rep. 2016;11:280–287. 50. Jeevanantham D, Rajendran V, McGillis Z, Tremblay L, Larivière C, Knight A. Mobilization and Exercise Intervention for Patients With Multiple Myeloma: 20. Craike M, Hose K, Livingston PM. Physical activity participation and barriers for Clinical Practice Guidelines Endorsed by the Canadian Physiotherapy Association. people with multiple myeloma. Support Care Cancer. 2013;21:927–934. Phys Ther. 2021;101. 21. Vermaete N, Wolter P, Verhoef G, Gosselink R. Physical activity and physical fitness 51. Mohammed J, Aljurf M, Althumayri A, Almansour M, Alghamdi A, Hamidieh AA, in lymphoma patients before, during, and after chemotherapy: a prospective et al. Physical therapy pathway and protocol for patients undergoing hematopoi- longitudinal study. Ann Hematol. 2014;93:411–424. etic stem cell transplantation: Recommendations from The Eastern Mediterranean Blood and Marrow Transplantation (EMBMT) Group. Hematol Oncol Stem Cell Res. 22. Raphael D, Frey R, Gott M. Psychosocial distress in haematological cancer survi- 2019;12:127–132. vors: An integrative review. Eur J Cancer. 2017;26:e12640. 52. Kim SD, Kim HS. A series of bed exercises to improve lymphocyte count in 23. Raphael D, Frey R, Gott M. The nature and timing of distress among post-treatment allogeneic bone marrow transplantation patients. Eur J Cancer Care. 2006;15: haematological cancer survivors. Eur J Cancer Care. 2019;28:e12951. 453–457. 24. Tran J, Norder EE, Diaz PT, Phillips GS, Elder P, Devine SM, et al. Pulmonary reha- 53. Wiskemann J, Dreger P, Schwerdtfeger R, Bondong A, Huber G, Kleindienst N, et al. bilitation for bronchiolitis obliterans syndrome after hematopoietic stem cell Effects of a partly self-administered exercise program before, during, and after transplantation. Biol Blood Marrow Transplant. 2012;18:1250–1254. allogeneic stem cell transplantation. Blood. 2011;117:2604–2613. 25. Majhail NS, Rizzo JD, Lee SJ, Aljurf M, Atsuta Y, Bonfim C, et al. Recommended 54. Mina DS, Langelier D, Adams SC, Alibhai SM, Chasen M, Campbell KL, et al. Exercise screening and preventive practices for long-term survivors after hematopoietic cell as part of routine cancer care. Lancet Oncol. 2018;19:e433–e436. transplantation. Biol Blood Marrow Transplant. 2012;18:348–371. 55. Pahl A, Wehrle A, Kneis S, Gollhofer A, Bertz H. Whole body vibration training 26. Fukushima T, Nakano J, Ishii S, Natsuzako A, Hirase T, Sakamoto J, et al. Charac- during allogeneic hematopoietic cell transplantation-the effects on patients’ teristics of muscle function and the effect of cachexia in patients with haemato- physical capacity. Ann Hematol. 2020;99:635–648. logical malignancy. Eur J Cancer. 2019;28:e12956. 56. Ng DL, Gan GG, Anuar NA, Tung YZ, Lai NZ, Tan YW, et al. The effect of a single 27. Reid MB, Lännergren J, Westerblad H. Respiratory and Limb Muscle Weakness Induced session of 30-min mindful breathing in reducing fatigue among patients with by Tumor Necrosis Factor-a. Am J Respir Crit Care Med. 2002;166:479–484. haematological cancer – a randomised controlled trial. BMC Palliative Care. 2021;20:1–9. 28. Peterson SJ, Mozer M. Differentiating Sarcopenia and Cachexia Among Patients With Cancer. Nutr Clin Pract. 2017;32:30–39. 57. Almeida LB, Laterza MC, Rondon M, de Matos LD, Granger CL, Denehy L, et al. Inspiratory muscle training in addition to conventional physical rehabilitation in 29. Morishita S, Kaida K, Tanaka T, Itani Y, Ikegame K, Okada M, et al. Prevalence of hospitalized patients undergoing hematopoietic stem cell transplantation: a ran- sarcopenia and relevance of body composition, physiological function, fatigue, and domized controlled trial. Support Care Cancer. 2022;30:9393–9402. health-related quality of life in patients before allogeneic hematopoietic stem cell transplantation. Support Care Cancer. 2012;20:3161–3168. 58. Ishikawa A, Otaka Y, Kamisako M, Suzuki T, Miyata C, Tsuji T, et al. Factors affecting lower limb muscle strength and cardiopulmonary fitness after allogeneic he- 30. Xiao DY, Luo S, O’Brian K, Ganti A, Riedell P, Sanfilippo KM, et al. Impact of sar- matopoietic stem cell transplantation. Support Care Cancer. 2019;27:1793–1800. copenia on treatment tolerance in United States veterans with diffuse large B-cell lymphoma treated with CHOP-based chemotherapy. Am J Hematol. 2016;91:1002– 59. Baumann F, Kraut L, Schüle K, Bloch W, Fauser A. A controlled randomized study 1007. examining the effects of exercise therapy on patients undergoing haematopoietic stem cell transplantation. Bone Marrow Transplant. 2010;45:355–362. 31. Bozzetti F. Forcing the vicious circle: sarcopenia increases toxicity, decreases response to chemotherapy and worsens with chemotherapy. Ann Oncol. 60. Coleman EA, Coon SK, Kennedy RL, Lockhart KD, Stewart CB, Anaissie EJ, et al. 2017;28:2107–2118. Effects of exercise in combination with epoetin alfa during high-dose chemo- therapy and autologous peripheral blood stem cell transplantation for multiple 32. Go S, Park M, Song H, Kim HG, Kang MH, Kang JH, et al. A comparison of pectoralis myeloma. Oncol Nurs Forum. 2008;35:E53–E61. versus lumbar skeletal muscle indices for defining sarcopenia in diffuse large B-cell lymphoma - two are better than one. Oncotarget. 2017;8:47007–47019. 61. van Haren I, Staal JB, Potting CM, Blijlevens NM, Staal JB, Nijhuis-van der Sanden MW. Physical exercise prior to hematopoietic stem cell transplantation: A feasibility study. 33. Sun Q, Cui J, Liu W, Li J, Hong M, Qian S. The Prognostic Value of Sarcopenia in Physiother Theory Pract. 2018;34:747–756. Acute Myeloid Leukemia Patients and the Development and Validation of a Novel Nomogram for Predicting Survival. Front Oncol. 2022;12:828939. 62. Wood WA, Weaver M, Smith-Ryan AE, Hanson ED, Shea TC, Battaglini CL. Lessons learned from a pilot randomized clinical trial of home-based exercise prescription 34. Fearon K, Strasser F, Anker SD, Bosaeus I, Bruera E, Fainsinger RL, et al. Definition before allogeneic hematopoietic cell transplantation. Support Care Cancer. 2020;28: and classification of cancer cachexia: an international consensus. Lancet Oncol. 5291–5298. 2011;12:489–495. 63. Jacobsen PB, Le-Rademacher J, Jim H, Syrjala K, Wingard JR, Logan B, et al. Exercise 35. Baracos V, Martin L, Korc M, Guttridge DC, Fearon KC. Cancer-associated cachexia. and stress management training prior to hematopoietic cell transplantation: Blood Nat Rev Dis Primers. 2018;4:17105. and Marrow Transplant Clinical Trials Network (BMT CTN) 0902. Biol Blood Marrow Transplant. 2014;20:1530–1536. 36. Grande AJ, Silva V, Sawaris Neto L, Teixeira Basmage JP, Peccin MS, Maddocks M. Exercise for cancer cachexia in adults. Cochrane Database Syst Rev. 2021;3: 64. Dennett A, Sarkies M, Shields N, Peiris CL, Williams C, Taylor NF. Multidisciplinary, Cd010804. exercise-based oncology rehabilitation programs improve patient outcomes but their effects on healthcare service-level outcomes remain uncertain: a systematic 37. Maddocks M, Murton AJ, Wilcock A. Therapeutic exercise in cancer cachexia. Crit review. J Physiother. 2021;67:12–26. Rev Oncog. 2012;17:285–292. 65. Chowdhury R, Brennan F, Gardiner M. Cancer Rehabilitation and Palliative Care- 38. Campbell KL, Winters-Stone KM, Wiskemann J, May AM, Schwartz AL, Exploring the Synergies. J Pain Symptom Manag. 2020;60:1239–1252. Courneya KS, et al. Exercise Guidelines for Cancer Survivors: Consensus Statement from International Multidisciplinary Roundtable. Med Sci Sports Exerc. 66. Lee K, Nathwani N, Shamunee J, Lindenfeld L, Wong FL, Krishnan A, et al. Telehealth 2019;51:2375–2390. exercise to Improve Physical function and frailty in patients with multiple myeloma treated with autologous hematopoietic Stem cell transplantation (TIPS): 39. Patel AV, Friedenreich CM, Moore SC, Hayes SC, Silver JK, Campbell KL, et al. protocol of a randomized controlled trial. Trials. 2022;23:921. American College of Sports Medicine Roundtable Report on Physical Activity, Sedentary Behavior, and Cancer Prevention and Control. Med Sci Sports Exerc. 2019;51:2391–2402.
Journal of Physiotherapy 69 (2023) 129–132 Appraisal j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j p hy s Research Note: Non-inferiority trials Randomised controlled trials are the gold standard experimental with an established treatment of known efficacy.11 The rationale, design for establishing the efficacy of a new intervention or treatment hypotheses, design, analysis and interpretation of non-inferiority approach in health and medical research.1 The most common aim of trials share similarities with superiority trials, although there are these trials is to demonstrate a difference in outcome between two some differences (Table 2).10,12,13 The typical underlying premise of treatments, where both of these may be active treatments or one may a non-inferiority trial is that the alternative treatment under be a standard of care, a placebo or no treatment at all. Such trials are investigation is better than the established treatment in some referred to as superiority trials: which treatment leads to superior important aspects (eg, costs, side effects/harms, convenience, outcomes? However, the aim of a trial is sometimes not to detect a acceptability, accessibility, ease of delivery), even though it is not difference between treatments but rather to determine whether one believed to be clinically superior or may even be a little less effec- treatment is no worse than another. Consider the scenario where tive. Therefore, the aim of a non-inferiority trial is to demonstrate superiority trials have shown that in-person clinician-delivered that an alternative treatment is clinically not inferior or not exercise-based care is better than no exercise for patients with knee demonstrably worse than the established treatment. The ‘accept- osteoarthritis.2 There can sometimes be geographical barriers to an able’ degree of inferiority of the alternative treatment is referred to individual attending a clinic in-person (eg, those in rural or remote as the non-inferiority margin. This non-inferiority margin is the areas may be unable to access clinics) and some researchers have margin of difference (in the average values of the primary trial proposed telehealth-delivered care as a more accessible and conve- outcome(s)) between the alternative and established treatments nient alternative. Researchers do not expect that telehealth-delivered that represents how much worse the alternative treatment can be exercise care will be superior to in-person care, but if they can show yet can still be considered clinically similar or not unacceptably that it is no worse, people with knee osteoarthritis could instead be worse than the established treatment.12 offered a more accessible and convenient alternative.3 This is the aim of a non-inferiority trial: determining whether an alternative inter- In a standard superiority trial, the null hypothesis is that the vention or treatment approach is not unacceptably worse than the average outcome under treatment A is the same as the average existing accepted treatment. In other words: non-inferiority trials outcome under treatment B (or usual physiotherapy care or no aim to show that the alternative intervention or treatment approach treatment or placebo) for participants with condition X. The alter- does not reduce treatment efficacy too much, where the threshold for native hypothesis is that the average outcomes under the two treat- ‘too much’ must be prespecified by researchers and is known as the ments differ. In non-inferiority trials, the non-inferiority margin itself ‘margin of non-inferiority’. forms part of the null and alternative hypotheses. In non-inferiority trials, the null hypothesis is that the average outcome under treat- Although superiority trials are more common in physiotherapy ment A is worse than that under treatment B by the non-inferiority research, non-inferiority designs are increasingly being used to margin or more. The alternative hypothesis is that the difference determine the effectiveness of innovative models of care and service between the average outcome under treatment A and the average delivery (see Table 1 for examples). These include, but are not limited outcome under treatment B is no worse (ie, non-inferior) than the to: telehealth3 and digital technologies4 to deliver physiotherapy care non-inferiority margin. Determination of non-inferiority is based on remotely (rather than in-person), direct access to subsidised physio- where the confidence intervals of the observed between-group therapy (rather than needing a referral from a general practitioner),5 treatment differences lie with respect to the predetermined non- home-based unsupervised care (rather than supervised in hospital inferiority margin (Figure 1). If the 95% CI does not extend beyond settings),6 health service role substitution (using physiotherapists to the non-inferiority margin, non-inferiority of the alternative treat- deliver care normally delivered by other professionals),7 group-based ment can be concluded. This contrasts with a superiority trial, where exercise approaches (instead of individual exercise-based care)8 and superiority of a new treatment is claimed when the two-sided 95% physiotherapy care as an alternative to surgery.9 Although the aims of CIs lie entirely on the favourable side of zero (or one); although it superiority and non-inferiority trials may differ, they do share many must be emphasised that superiority trials should also be interpreted characteristics, and have been referred to as ‘two sides of the same in light of the effect size considered in the sample size calculation. coin’.10 This Research Note discusses how non-inferiority trials differ from and are similar to superiority trials, as well as the key issues that Equivalence trials, a third type of trial design, are designed to should be considered when designing and interpreting a non- determine whether an alternative treatment approach is clinically inferiority trial. For completeness, equivalence trials are also briefly equivalent (similar to) to an established treatment.14 In an equiva- mentioned. lence trial, the null hypothesis is that the average outcome under treatment A differs to that under treatment B by a pre-defined What is a non-inferiority trial and what is a non-inferiority ‘margin of equivalence’ or more (in either direction), whilst the margin? alternative hypothesis is that any difference between treatments is less than that equivalence margin. That is, to be declared equivalent, Non-inferiority trials help to determine whether an alternative the difference between the alternative and established treatments (new or existing) treatment approach is ‘good enough’ compared and the associated confidence interval must lie between the upper and lower margins of equivalence. https://doi.org/10.1016/j.jphys.2023.01.001 1836-9553/© 2023 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/ 4.0/).
Table 1 Examples of non-inferiority trials in physiotherapy. Study population Alternative treatment/ Established treatment/approach Advantages of alternative approach Adults with knee Telehealth-delivered In-person physiotherapist Increased convenience osteoarthritis3 (videoconferencing) consultations Reduced travel time and cos physiotherapist Increased accessibility to ph Adults with a consultations Face-to-face physiotherapy in musculoskeletal Remotely supported home hospital More timely delivery and in condition4 exercise program accessibility to care Adults with Publicly subsidised Reduced cost of physiothera musculoskeletal pain5 Direct access to publicly physiotherapy with GP referral Easier access to physiothera subsidised physiotherapy Faster recovery Adults with chronic without GP referral Centre-based supervised obstructive pulmonary Home-based unsupervised pulmonary rehabilitation Increased choice for patients disease6 pulmonary rehabilitation Increased access to pulmona Corticosteroid and local rehabilitation Adults with shoulder pain7 Corticosteroid and local anaesthetic injection by an anaesthetic injection by a orthopaedic surgeon Improved access to care physiotherapist Reduced costs Reduced waiting times Women with urinary Group-based pelvic floor Individual pelvic floor muscle Increased access to care incontinence8 muscle training training Reduced treatment costs Increased peer support and Adults with knee Physiotherapist-delivered Early surgery with arthroscopic Reduced inappropriate knee meniscus tears9 exercise program partial meniscectomy Increased choice for patients Reduced treatment costs CRQ-SR = Chronic Respiratory Questionnaire Self-Report, score range 1 to 7; GP = general practitioner; IKDC = International Kn SF-12 = 12-item Short Form Health Survey, score range 0 to 100; SPADI = Shoulder Pain and Disability Index, score range 0 a In this trial, negative values of the difference between the average outcomes of home-based treatment and centre-based difference score was calculated as new treatment minus the established treatment), thus the lower bound of the 95% CI mu
130 Appraisal treatment Primary outcome(s) Non-inferiority Sample Conclusion margin size (n) sts Knee pain on walking 394 Trial ongoing hysiotherapy (NRS) Pain = 0.95 units Trial ongoing ncreased Physical function Function = 5.44 units 210 Trial ongoing apy care (WOMAC) apy Function (patient- 1.5 units 3,422 specific functional scale) 2 units Physical health score (SF-12) s Dyspnoea (CRQ-SR) 0.5 units 287 Indeterminate evidence that home-based ary 15 units 64 pulmonary rehabilitation was non-inferior to Shoulder pain and centre-based pulmonary rehabilitation (mean disability (SPADI) 362 difference –0.24, 95% CI –0.61 to 0.12)a 321 Injection by the physiotherapist declared to be motivation % reduction in number 10% non-inferior to injection by surgeon (mean e surgery of urinary incontinence 8 units difference 3.96, 95% CI –5.41 to 13.34 at 6 episodes (bladder diary) weeks; –4.65, 95% CI –16.46 to 7.17 at 12 s Knee function (IKDC) weeks) Group-based pelvic floor muscle training is not inferior to individual-based training (mean difference 24%, 95% CI –10 to 7%) Physiotherapy was declared non-inferior to early arthroscopic partial meniscectomy (mean difference 3.6, 97.5% CI 2N to 6.5) nee Documentation Committee Subjective Knee Form, score range 0 to 100; NRS = numerical rating scale, score range 0 to 10; to 100; WOMAC = Western Ontario and McMaster Universities (WOMAC) Osteoarthritis Index, score range 0 to 68. d treatment favoured the established treatment/approach (ie, higher values of the primary outcome were favourable and the ust be interpreted. Research Note
Appraisal Research Note 131 Table 2 Similarities and differences between non-inferiority and superiority trials. Non-inferiority Superiority Null hypothesis Alternative treatment has a worse average outcome than the average outcome Average outcome under the new treatment is the same as that under the comparison treatment under the comparison treatment Alternative hypothesis Alternative treatment has an average outcome that is non-inferior to the average New treatment has an average outcome that is not the same as outcome under the comparison treatment that under the comparison treatment Comparator Established treatment of proven efficacy Another treatment, no treatment, usual care or placebo Sample size Based on a pre-determined non-inferiority margin in primary outcome(s), which Based on a pre-determined effect size or minimum clinically will often be the minimum clinically important difference in the primary important difference in primary outcome(s) outcome(s) One-sided or two-sided One-sided or two-sided Typically two-sided test of null hypothesis and confidence intervals Decision-making When lower values of the outcome are favourable, the upper bound of the Upper and lower bounds of the confidence intervals for the confidence interval for the average outcome for the new treatment minus the treatment effect do not span the null value (usually zero for average outcome for the established treatment is smaller than the non-inferiority differences in average outcomes, or one, in the case of relative margin risks) When is a non-inferiority trial appropriate? Sample size and analysis considerations Non-inferiority trials may be appropriate when an alternative Sample size calculations for non-inferiority trials rely on the non- treatment, or method for delivering treatment, is developed and may inferiority margin in the same way that sample size calculations for be preferred because of its inherent advantages over an established superiority trials rely on the effect size that researchers wish to treatment even though it may be slightly less clinically effective. In detect. There is debate about whether confidence intervals should be contrast, an equivalence trial is appropriate to evaluate whether an one-sided or two-sided (which has implications for whether to alternative treatment is ‘equivalently effective’ to an established consider one-sided or two-sided p-values when calculating sample treatment.14 New treatment approaches that potentially offer greater sizes), as summarised by Dunn et al.10 However, it is noted that two- efficacy should be evaluated in a superiority trial. In non-inferiority sided 95% CIs will have the same upper limit as one-sided 97.5% CIs in trials, the efficacy of the established treatment should already have the context of the scenario in Figure 1. Sample size calculations for been convincingly determined in previous robust superiority trials.11 superiority trials that aim to detect a minimum clinically important Non-inferiority trials can also be used to determine treatment supe- difference in the primary outcome will have similar sample size re- riority in either direction.12 quirements to non-inferiority trials where the non-inferiority margin is prespecified as the minimum clinically important difference.10 Choosing the non-inferiority margin Sample sizes for non-inferiority trials are often larger than for su- periority trials because researchers often (optimistically) select larger One of the most important decisions when designing a effect sizes when designing superiority trials.10 non-inferiority trial is predetermining the non-inferiority margin, as this has implications for sample size calculations and drawing There has been discussion about the use of intention-to-treat conclusions from the analysis. There is no accepted method for analyses and per-protocol analyses in non-inferiority trials, also determining an appropriate non-inferiority margin.11 Some authors summarised by Dunn et al.10 Traditionally, non-inferiority trials argue that the non-inferiority margin is conceptually identical to have tended to give greater consideration to ‘per-protocol analyses’ the smallest clinically meaningful difference between the two (which exclude participants who have violated their allocated treatments,10 others suggest that it should be smaller than the intervention protocol and/or not adhered to their intervention) clinically important effect of the established treatment14 and others because of fears that including these participants would reduce the advocate for a margin validated by expert consensus of clinicians observed difference in treatment outcomes and bias findings to- and patients.11 Regulatory bodies may recommend that a wards non-inferiority. It is recommended that, as for superiority non-inferiority margin should be a fraction (often half) of the lower trials, all randomised patients should be included in the analysis of bound of the 95% CI of the effect of the established (comparator) data from non-inferiority trials, in their originally randomised treatment.15 group (ie, an intention-to-treat analysis should be conducted), with options for imputation of missing outcomes considered as for su- periority trials. Per-protocol analyses may not compare randomised groups of participants and thus may be difficult to interpret. In general, unless non-inferiority margins have been pre-specified for secondary outcomes, these should be analysed and interpreted in the same manner as analyses of secondary outcomes in superiority trials. Figure 1. Possible likely conclusions from non-inferiority trials where positive treat- Interpretation of findings ment effects favour the established treatment, adapted from elsewhere.13,19 The x-axis represents the treatment effect (between-group mean difference in this example Interpreting findings from non-inferiority trials is more complex (calculated as alternative treatment minus the established treatment for an outcome than for superiority trials. A range of conclusions are possible from a where lower scores are favourable) with 0 indicating no difference). The horizontal non-inferiority trial (Figure 1). In scenario A, the upper bound of the lines indicate the point estimate (black squares) and 95% CIs of the treatment effect 95% CI does not exceed the non-inferiority margin and does not under five different scenarios. The green shaded area represents the zone of non- cross zero, thus both non-inferiority and superiority of the alter- inferiority of the alternative treatment compared with the established treatment. native treatment can be claimed. In scenario B, the upper bound of the 95% CI crosses zero but does not exceed the non-inferiority margin, thus non-inferiority of the alternative treatment can be claimed but not superiority. In scenario C, the lower bound of the 95% CI crosses zero and the upper bound exceeds the non-inferiority margin, thus the outcome is indeterminate, as the alternative
132 Appraisal Research Note treatment could be either inferior or non-inferior. However Do the extra benefits of the alternative treatment outweigh any non-inferiority cannot be claimed. In scenario D, the 95% CI does not observed loss of treatment efficacy? span zero but it does span the non-inferiority margin, thus non-inferiority has not been shown. In scenario E, the upper and Does the trial provide results regarding these desirable extra benefits? lower bounds of the confidence interval exceed the non-inferiority margin, thus the alternative treatment can be claimed as inferior. It Conclusion is noted that p-values can also be calculated for non-inferiority hypothesis tests but it is recommended that readers focus on Non-inferiority trials are a subset of randomised controlled trials confidence intervals rather than whether a p-value is below a with objectives, design features, methods of analysis, and interpre- certain threshold (as has been more broadly recommended16). tation that may differ from standard superiority trials. This Research Note has discussed the important issues that researchers should Reporting of findings consider when designing, conducting and interpreting a non- inferiority trial, and highlighted factors that clinicians should Quality of reporting of non-inferiority trials is often sub-optimal,17 consider when reading research reports that claim non-inferiority. which can make it challenging for clinicians reading research reports to decide whether claims of ‘non-inferiority’ are warranted. This is Rana S Hinmana and Jessica Kaszab further complicated when a superiority trial that finds evidence of no aDepartment of Physiotherapy, University of Melbourne, Australia difference between treatment groups inappropriately concludes that bSchool of Public Health and Preventive Medicine, Monash University, the interventions are similar or equally effective.18 Non-inferiority or equivalence can only be claimed if a non-inferiority or equivalence Australia margin has been pre-specified. An extension to the CONSORT (Consolidated Standards of Reporting Trials) Statement for non- References inferiority and equivalence randomised trials13 has been developed to help researchers improve the reporting of findings from these 1. Sibbald B, Roland M. BMJ. 1998;316:201. trials. Nonetheless, clinicians should not take conclusions of non- 2. Juhl C, et al. Arthritis Rheumatol. 2014;66:622–636. inferiority for granted and should ask themselves: 3. Hinman RS, et al. BMC Musculoskelet Disord. 2020;21:522. 4. Withers HG, et al. BMJ Open. 2021;11:e041242. Has the trial been designed as a non-inferiority trial? 5. Dahlerup J, et al. Contemp Clin Trials. 2022;113:106648. 6. Horton EJ, et al. Thorax. 2018;73:29–36. What are the hypotheses being tested and is a non-inferiority design 7. Marks D, et al. PLoS One. 2016;11:e0162679. appropriate? 8. Dumoulin C, et al. JAMA Intern Med. 2020;180:1284–1293. 9. van de Graaf VA, et al. JAMA. 2018;320:1328–1337. How was the non-inferiority margin chosen, was it prespecified and 10. Dunn DT, et al. Trials. 2018;19:499. does it seem clinically acceptable/reasonable? 11. Scott IA. Med. J. Aust. 2009;190:326–330. 12. Leung JT, et al. Heart. 2020;106:99–104. Does the sample of recruited participants adequately reflect the 13. Piaggio G, et al. JAMA. 2012;308:2594–2604. population of interest and the patients I typically see in clinical 14. Stefanos R, et al. Intern Emerg Med. 2020;15:1085–1091. practice? 15. Schumi J, Wittes JT. Trials. 2011;12:106. 16. Wasserstein R, et al. Am Stat. 2019;73:1–19. Have the findings been interpreted with respect to 95% CIs and the 17. Rehal S, et al. BMJ Open. 2016;6:e012594. non-inferiority margin? 18. Boutron I, et al. JAMA. 2010;303:2058–2064. 19. Middleton LJ. BJOG. 2021;128:1916.
Search
