Journal of Physiotherapy 68 (2022) 209 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 Physical Therapist Management of People With Rheumatoid Arthritis Date of latest update: February 2021. Date of next update: Not stated. Patient 1 require brief education and exercise/movement instruction to encourage self- group: Adults with rheumatoid arthritis. Intended audience: Physiotherapists. management. Profile 2 require brief supervised exercise to address condition Additional versions: Revision of 2008 Royal Dutch Society for Physical Therapy complexity/severity, or limited self-management skills. Profile 3 require an guideline for physical therapy for patients with rheumatoid arthritis. Funded by: intensive period of supervised exercise due to serious comorbidities or compli- Dutch Society of Physical Therapy. Expert working group: Expert clinical and cations of the disease or its treatment. Exercise should: address functional needs; research physiotherapists. Consultation with: Representatives of consumer align with the patient’s request for help; align with public health recommen- groups, Dutch National Health Care Institute, health insurers, and 15 Dutch dations for physical activity; and adhere to FITT principles (frequency, intensity, professional associations for physiotherapists, rheumatologists, orthopaedic type and time-related characteristics of exercises). Passive interventions such as surgeons, clinical nurse specialists in rheumatology, sports doctors, general massage, electrotherapy, thermotherapy, low-level laser therapy, ultrasound and practitioners, hand therapists, podiatrists, and exercise therapists. Approved by: taping should be limited or avoided. All representatives. Location: https://doi.org/10.1093/ptj/pzab127.1 Description: Assessment and treatment guidelines were developed through Provenance: Invited. Not peer reviewed. author focus groups, guideline panels and review panels. Assessment recommendations (n = 4) include: subjective interview regarding health status, Leanne Johnston disease impact, disease history and medical treatment; physical examination of The University of Queensland, Australia disease activity, structural joint damage, deformities, exercise tolerance and muscle function; use of recommended patient-reported and performance- https://doi.org/10.1016/j.jphys.2022.05.006 based instruments (eg, Patient-Specific Complaint Instrument, Pain and Fatigue rating scales, Health Assessment Questionnaire Disability Index, and Reference 6-minute walk test); and classification of patients into one of three treatment profiles. Treatment recommendations (n = 4) include: education, comprising 1. Peter WF, et al. Clinical Practice Guideline for Physical Therapist Management of tailored advice to support self-management, health and wellbeing; and three People With Rheumatoid Arthritis. Phys Ther. 2021:101. recommendations for exercise according to treatment profile group. Profile 1836-9553/Crown Copyright © 2022 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/). Appraisal of Clinical Practice Guideline: Physiotherapy for epidermolysis bullosa Date of latest update: November 2020. Date of next update: 2023 to 2025. and optimising interaction with the community (eg, environmental modifi- Patient group: People with epidermolysis bullosa (EB). Intended audience: cations when necessary). A systematic literature review was conducted and Physiotherapists, patients with EB and their caregivers, healthcare workers, articles were critically analysed by an international panel. The strength of the educational staff and employers of individuals with EB. Additional versions: recommendations was graded as ‘D’ and based on quality of evidence level 3 None. Expert working group: Thirteen members (physiotherapists, occupa- (case reports) and 4 (expert opinion). Recommendations 1 to 5 were judged tional therapist, nurses, dermatologists, parent/caregiver, person with EB) of as ‘desirable consequences probably outweigh undesirable consequences’, the International Dystrophic EB Research Association (DEBRA). Funded by: while for recommendation 6 ‘desirable and undesirable consequences were DEBRA of America. Consultation with: External reviewer panel (healthcare closely balanced or uncertain’. This guideline lays the foundational work for providers and healthcare consumer groups). Approved by: DEBRA Interna- physiotherapists throughout the world to provide high-quality services, tional CPG Network for EB-CLINET. Location: Journal article: https://ojrd. while improving and maintaining functional mobility and independence biomedcentral.com/articles/10.1186/s13023-021-01997-w. Webinar: https:// within the EB community. The Clinical Practice Guideline outlines limitations www.debra-international.org/physiotherapy-cpg. in the available evidence and possible future research needed to improve physiotherapy practice. Description: Epidermolysis bullosa is a rare genetic disorder characterised by skin fragility with blister formation spontaneously occurring or following Provenance: Invited. Not peer reviewed. minor trauma such as gentle pressure or friction. A survey was conducted within the EB community and six outcomes were identified as a priority to Rik Gosselink a and Joshua Zadro b address in physiotherapy management: attaining developmental motor aUniversity of Leuven, Belgium milestones (eg, prevent contracture, handle carefully to avoid blisters); bUniversity of Sydney, Australia identifying safe and functional mobility in the natural environment (eg, exercise the feet to keep them in good health); encouraging ambulation https://doi.org/10.1016/j.jphys.2022.05.001 endurance (eg, provide mobility aids for safe longer distance use); sup- porting safe ability to bear weight (eg, knee padding and soft special shoes to prevent blistering in infants); improving access to physiotherapy services; 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 68 (2022) 209 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 Physical Therapist Management of People With Rheumatoid Arthritis Date of latest update: February 2021. Date of next update: Not stated. Patient 1 require brief education and exercise/movement instruction to encourage self- group: Adults with rheumatoid arthritis. Intended audience: Physiotherapists. management. Profile 2 require brief supervised exercise to address condition Additional versions: Revision of 2008 Royal Dutch Society for Physical Therapy complexity/severity, or limited self-management skills. Profile 3 require an guideline for physical therapy for patients with rheumatoid arthritis. Funded by: intensive period of supervised exercise due to serious comorbidities or compli- Dutch Society of Physical Therapy. Expert working group: Expert clinical and cations of the disease or its treatment. Exercise should: address functional needs; research physiotherapists. Consultation with: Representatives of consumer align with the patient’s request for help; align with public health recommen- groups, Dutch National Health Care Institute, health insurers, and 15 Dutch dations for physical activity; and adhere to FITT principles (frequency, intensity, professional associations for physiotherapists, rheumatologists, orthopaedic type and time-related characteristics of exercises). Passive interventions such as surgeons, clinical nurse specialists in rheumatology, sports doctors, general massage, electrotherapy, thermotherapy, low-level laser therapy, ultrasound and practitioners, hand therapists, podiatrists, and exercise therapists. Approved by: taping should be limited or avoided. All representatives. Location: https://doi.org/10.1093/ptj/pzab127.1 Description: Assessment and treatment guidelines were developed through Provenance: Invited. Not peer reviewed. author focus groups, guideline panels and review panels. Assessment recommendations (n = 4) include: subjective interview regarding health status, Leanne Johnston disease impact, disease history and medical treatment; physical examination of The University of Queensland, Australia disease activity, structural joint damage, deformities, exercise tolerance and muscle function; use of recommended patient-reported and performance- https://doi.org/10.1016/j.jphys.2022.05.006 based instruments (eg, Patient-Specific Complaint Instrument, Pain and Fatigue rating scales, Health Assessment Questionnaire Disability Index, and Reference 6-minute walk test); and classification of patients into one of three treatment profiles. Treatment recommendations (n = 4) include: education, comprising 1. Peter WF, et al. Clinical Practice Guideline for Physical Therapist Management of tailored advice to support self-management, health and wellbeing; and three People With Rheumatoid Arthritis. Phys Ther. 2021:101. recommendations for exercise according to treatment profile group. Profile 1836-9553/Crown Copyright © 2022 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/). Appraisal of Clinical Practice Guideline: Physiotherapy for epidermolysis bullosa Date of latest update: November 2020. Date of next update: 2023 to 2025. and optimising interaction with the community (eg, environmental modifi- Patient group: People with epidermolysis bullosa (EB). Intended audience: cations when necessary). A systematic literature review was conducted and Physiotherapists, patients with EB and their caregivers, healthcare workers, articles were critically analysed by an international panel. The strength of the educational staff and employers of individuals with EB. Additional versions: recommendations was graded as ‘D’ and based on quality of evidence level 3 None. Expert working group: Thirteen members (physiotherapists, occupa- (case reports) and 4 (expert opinion). Recommendations 1 to 5 were judged tional therapist, nurses, dermatologists, parent/caregiver, person with EB) of as ‘desirable consequences probably outweigh undesirable consequences’, the International Dystrophic EB Research Association (DEBRA). Funded by: while for recommendation 6 ‘desirable and undesirable consequences were DEBRA of America. Consultation with: External reviewer panel (healthcare closely balanced or uncertain’. This guideline lays the foundational work for providers and healthcare consumer groups). Approved by: DEBRA Interna- physiotherapists throughout the world to provide high-quality services, tional CPG Network for EB-CLINET. Location: Journal article: https://ojrd. while improving and maintaining functional mobility and independence biomedcentral.com/articles/10.1186/s13023-021-01997-w. Webinar: https:// within the EB community. The Clinical Practice Guideline outlines limitations www.debra-international.org/physiotherapy-cpg. in the available evidence and possible future research needed to improve physiotherapy practice. Description: Epidermolysis bullosa is a rare genetic disorder characterised by skin fragility with blister formation spontaneously occurring or following Provenance: Invited. Not peer reviewed. minor trauma such as gentle pressure or friction. A survey was conducted within the EB community and six outcomes were identified as a priority to Rik Gosselink a and Joshua Zadro b address in physiotherapy management: attaining developmental motor aUniversity of Leuven, Belgium milestones (eg, prevent contracture, handle carefully to avoid blisters); bUniversity of Sydney, Australia identifying safe and functional mobility in the natural environment (eg, exercise the feet to keep them in good health); encouraging ambulation https://doi.org/10.1016/j.jphys.2022.05.001 endurance (eg, provide mobility aids for safe longer distance use); sup- porting safe ability to bear weight (eg, knee padding and soft special shoes to prevent blistering in infants); improving access to physiotherapy services; 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 68 (2022) 157 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 Call for Applications Call for applications for membership of the Editorial Board The Editorial Board currently consists of 14 members: ten Australian and four international. Applications are invited to fill the following Editorial Board vacancies beginning in 2023: two Australian. All incumbents are entitled to re-apply in the current round. Editorial Board members are given portfolios with substantial responsibilities. This might involve, for example, soliciting submissions and editing contributions for one of the Journal’s ‘Appraisal’ sections. Potential applicants who are not prepared to take on portfolio responsibilities should not apply. The initial term of office commences on 1 January 2023 and expires on 31 December 2025. Editorial Board members are entitled to renominate for a further two successive terms. Selection criteria: 1. a sustained depth and breadth of research experience 2. extensive experience in the review and publication of research 3. prior editorial board experience (highly desirable but not essential) 4. excellent communication skills 5. good working knowledge of the physiotherapy profession and an interest in its future 6. demonstrated international reputation in research relevant to physiotherapy. Eligibility criteria: 7. hold a PhD 8. be a physiotherapist member of the Australian Physiotherapy Association (APA) 9. be a financial member of the APA at the time of application. Responsibilities: contribute to the development of policies that guide the publication of the Journal participate in the activities of the Editorial Board as a voting member manage or co-manage one of the journal portfolios attend regular Editorial Board teleconferences and a two-day face-to-face meeting annually meet and liaise with other members of the Editorial Board and the Scientific Editor as required contribute to the mentoring of Journal of Physiotherapy Editorial Fellows undertake specific tasks from time to time to promote the standing of the Journal. To be considered, physiotherapists applying for positions must submit: 1. a cover letter addressing the numbered criteria, above 2. a brief CV (maximum 3 pages), which includes a clear explanation of the impact of any career interruption(s) over the last 5 years and/or any relative to opportunity considerations. Applicants will be assessed according to the selection criteria listed above, relative to opportunity, and with attention to diversity, equity and inclusion considerations. Applications close Friday, 7 October 2022 and should be directed to Marko Stechiwskyj at [email protected] The Journal of Physiotherapy promotes diversity, equity and inclusion and encourages eligible applicants of all backgrounds to apply. https://doi.org/10.1016/j.jphys.2022.05.013 1836-9553/
Journal of Physiotherapy 68 (2022) 153–155 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 Clinical education of physiotherapy students Alan Reubenson a, Mark R Elkins b,c a Curtin School of Allied Health, Curtin University, Perth, Australia; b Editor, Journal of Physiotherapy; c Faculty of Medicine and Health, University of Sydney, Sydney, Australia This Editorial introduces Journal of Physiotherapy’s article and educator perspectives of the PAL model.8 The respondents re- collection on clinical education of physiotherapy students. The article ported that PAL can help to position students as active learners collection has been curated from papers published in the journal to through reduced dependence on the clinical educator, heightened facilitate access to important findings in this field, highlight trends in roles in observing practice, and making and communicating evalua- the research and summarise avenues for further investigation. The tive judgments about quality of practice. The authors also concluded collected articles show educational models, novel approaches to that a more flexible use of PAL activities might provide students and finding clinical placements, early educational initiatives taken in educators with more agency, and empowering educators to design response to the coronavirus disease 2019 (COVID-19) pandemic, the worthwhile PAL activities might lead to increased adoption and current status of entry-level physiotherapy assessment, and insights acceptance of the PAL clinical placement model. into future directions for physiotherapy education. Simulation Models of clinical education Chipchase et al described how some in the profession have suc- In 2021, the total number of entry-level physiotherapy programs cessfully embedded the use of simulation into entry-level and post- in the member nations of World Physiotherapy had risen to about graduate training and assessment.3 They questioned whether simu- 3,800.1,2 For example, in Australia and New Zealand the number of lation practices could extend further; for example, by using students universities offering such courses has risen to 27 and programs of- as simulated patients, by using simulation to a greater extent in post- fering entry-level physiotherapy programs have increased sharply graduate training and even by using simulation to provide training for over the past two decades (Figure 1).3 Consequently, physiotherapy other workplace settings (eg, private practice). They concluded with student enrolments have also increased, with over 10,000 students some questions and challenges to all stakeholders (clinicians, edu- enrolled in entry-level courses in Australia at 30 June 2020.4 This has cators, researchers and professional bodies) in meeting the future increased competition for clinical placements. Many undergraduate needs of the profession, calling for informal and formal leadership to physiotherapy programs also face financial pressures; for example, drive cultural transformation as well as the need to work together in recent changes to higher education funding in Australia have caused a strong, collaborative partnerships in order to educate the next net decrease in funding support of around 9%.5 These changes generation of safe and effective practitioners. highlight the need for more-efficient models of education and the importance of willingness to embrace organisational change.6 Private practice placements Peer-assisted learning Traditionally, most physiotherapy clinical placements in Australia take place in public hospitals, yet most graduates work in the private Peer-assisted learning (PAL) in clinical placements involves stu- sector. A mixed-methods study quantified and characterised the use dents working in pairs or larger groups, who undertake structured of private practice for clinical placements in Australia.9 Clinical edu- and/or informal learning activities together (eg, provide social sup- cation managers (CEMs) were also interviewed to discern their port and peer feedback on performance). Potential benefits of PAL perceived benefits, risks, barriers and enablers of private practice models include those for students (eg, reduce anxiety by creating a clinical placements.9 With a 95% response rate from eligible Austra- safe learning environment and develop collaborative skills), clinical lian universities, the study provides very robust data. Based on 2017 educators (eg, reduce workload as students support each other and data, 44% of students undertook at least one 5-week placement in can undertake structured learning tasks without direct involvement) private practice and this accounted for 9% of all 5-week placements – and education providers (eg, build clinical placement capacity). A an encouraging increase compared with previous reports. Key find- crossover trial by Sevenhuysen et al gave third-year physiotherapy ings from the CEMs’ perspective were that private practice place- students an opportunity to experience a structured PAL clinical ments were safe and beneficial for students, practices and placement and a traditional clinical placement.7 The PAL model was universities. The main identified risks were related to quality and new to students and educators so training in the use of the PAL model consistency of student experience with no real risk to clients or ser- was provided prior to the 5-week clinical placement. Student per- vice. The main barriers that were highlighted were time costs (to both formance was similar with the two models. Although the PAL model private practitioners and CEMs), lack of space in some private prac- reduced educator workload and increased student feedback, both tices and potential loss of earning as the CEMs recognised that su- educators and students were more satisfied with the traditional pervising students does take time. The CEMs felt that there was no model.7 Sevenhuysen et al then used focus groups to explore student single superior educational model and that with more time and https://doi.org/10.1016/j.jphys.2022.05.012 1836-9553/© 2022 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/).
154Physiotherapy courses (n) Editorial 50 Extended graduate entry Masters degrees Graduate entry Masters degrees 40 Bachelor and Bachelor with Honours degrees 30 20 10 0 1910 1915 1920 1925 1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020 2022 Year Figure 1. Growth in the number of accredited physiotherapy programs in Australia and New Zealand.a,b Modified from Chipchase et al.3 a When a university offers both a Bachelor degree and a Bachelor degree with Honours degree, these have been counted as one program because the Honours program is generally nested within the Bachelor degree. b Some universities operate programs across different locations and in these cases are counted as multiple programs because they require staffing and resources at multiple sites. resources there were opportunities for working with private practi- help to address the unmet healthcare needs of communities in tioners in partnership to enhance learning experiences and build non-metropolitan areas.13 capacity for clinical placements in the private sector. To facilitate this, the paper lists a range of available training and support resources.9 Clinical education in the COVID-19 pandemic Expansion of clinical placements in private practice raises the The reduction in face-to-face interactions and changes in health- question: do students who undertake more private practice place- care delivery during the pandemic have created many challenges14 ments achieve equivalent performance outcomes as their peers who but also provided opportunities for innovation in clinical education. undertake fewer private practice placements? Lawton et al undertook a large-scale, longitudinal, retrospective study comparing final Clinical research placements Assessment of Physiotherapy Practice (APP) results between public and private sector placements of 284 students at one university.10 Student One such example is the advent of the clinical research placement, performance did not substantially differ between the two sectors. where clinical education is delivered in the context of research, with Subsequently, the 284 APP results from the final placement in the students involved in research projects that include delivery of course were compared with final placement results from 517 students evidence-based care.15 Dario and Simic15 described the planning, completing a similar course at other universities, which again found no development and implementation of pilot projects where clinical important differences. This support for expanding private practice placements were embedded in research trials: one in knee osteoar- placements into the private sector should help meet the growing thritis and one in low back pain. The model was co-designed by ac- demand for clinical placements and permit better matching of ademics and researchers. Telehealth trials were identified as the best placement exposure to workforce employment destinations. option in the pandemic and shared care models were used, with experienced physiotherapists leading the interventions and students Non-metropolitan placements supporting and performing some clinical and research-related tasks. The placements were assessed using the APP across the 5-week Another way to respond to increasing demands for clinical placement. Following successful trials, these placements are now placements would be to increase placements in non-metropolitan embedded in the physiotherapy program at the University of Sydney centres. Francis-Cracknell et al11 interviewed first-year students and in partnership with a local health service (clinical activities) and about their perceptions of upcoming non-metropolitan placements, research institute (research activities). Quality assurance processes third-year and fourth-year students about their perceptions of suggest positive outcomes as well as some barriers. Although this completed non-metropolitan placements, and clinical educators from placement addressed an immediate need, it also provides a future non-metropolitan sites. Roughly half of the first-year students (53%) opportunity to expand clinical placement capacity and develop had an unfavourable perception of non-metropolitan placements, research and evidence-based practice skills. especially those from a metropolitan upbringing. The third-year and fourth-year students and the clinical educators suggested strategies Simulation-based assessment of clinical competence to support and prepare students for non-metropolitan placements: tailoring preparation for students, paired rather than individual Pandemic restrictions also challenged assessment of practical placements and near-peer presentations for physiotherapy students skills when determining clinical competence in entry-level physio- prior to undertaking non-metropolitan placements. Other possible therapy education16 and with overseas-qualified physiotherapists.17 facilitators were dedicated clinical coordinator positions, travel sub- Tognon et al17 described the steps taken (eg, environmental scan sidies and affordable accommodation. Students who have positive and analysis of existing mechanisms of remote assessment of clinical experiences in non-metropolitan clinical placements are more likely competence) and considerations explored (eg, whether assessing to seek employment in these settings.12 Another advantage of one’s evaluative judgement of a physical task would be sufficient increasing non-metropolitan placements and their acceptability may
Editorial 155 without testing the actual performance) by the Australian Physio- et al15 recommended integrating clinical education at the inception of therapy Council in determining how assessment of clinical compe- research studies. Future research should target emerging areas of tence could be improved and future-proofed to deal with such physiotherapy practice and workforce development needs such as disruptions. The authors did not identify any model of remote those detailed in the Australian Physiotherapy Association’s recent A assessment that might replace the current simulation-based clinical Strong Physiotherapy Workforce for a Healthy Australia publication,23 assessment but did provide interesting considerations for the future. which covers areas such as preventative health, aged care, mental health24 and Aboriginal and Torres Strait Islander health. Assessment of Physiotherapy Practice tool Footnotes: Nil. The APP tool was originally developed more than a decade ago.18 It eAddenda: Nil. contains 20 items covering seven domains of physiotherapy practice: Ethics approval: Not applicable. professional behaviour, communication, assessment, analysis/plan- Competing interest: Nil. ning, intervention, evidence-based practice, and risk management. Source(s) of support: Nil. Each item is scored on a 5-point scale (0 = infrequently/rarely dem- Acknowledgements: Nil. onstrates performance indicators to 4 = demonstrates most perfor- Provenance: Invited. Peer reviewed. mance indicators to an excellent standard). The tool also has a global Correspondence: Mark Elkins, Centre for Education & Workforce rating scale. This article collection features two of the original Development, Sydney Local Health District, Sydney, Australia. Email: papers19,20 that evaluated the validity and reliability of the APP and a [email protected] more recent paper21 offering an alternative scoring protocol and interpretation. The validity study19 performed Rasch analysis on data References from nine universities to establish construct validity and unidimen- sionality of the tool. These findings supported the summing of the 1. World Physiotherapy. Profile of the global profession (Reference year 2021). 20 items to provide an overall score of clinical competence between https://world.physio/membership/profession-profile. Accessed April 2, 2022. 0 and 80 (ie, a one-factor model). Furthermore, the tool was able to discriminate four levels of competence using the global rating scale. 2. Wikipedia. List of countries and dependencies by population (Reference year 2021/ The APP item scales performed in a consistent way regardless of the 2022). https://en.wikipedia.org/wiki/List_of_countries_and_dependencies_by_ characteristics of the student, clinical educator or placement context. population. Accessed April 2, 2022. The reliability study20 established high inter-rater reliability on 5-week clinical placements. Together, these studies provided confidence for 3. Chipchase L, et al. J Physiother. 2018;64:205–207. physiotherapy educators to use the APP to assess entry-level compe- 4. Australian Health Practitioner Regulation Agency. Annual Report 2019/20. https:// tence and the APP has subsequently been adopted by all entry-level physiotherapy programs in Australia and New Zealand. Some univer- www.ahpra.gov.au/Publications/Annual-reports/Annual-Report-2020.aspx. Accessed sities in at least nine other countries already use the APP or are April 11, 2022. translating it. The third APP paper21 differed from the previous cross- 5. Australian Government Department of Education, Skills and Employment. sectional studies19,20 by using an archival, longitudinal study design to Improving accountability and information for providers - Department of Educa- progress the psychometric evaluation of the APP tool. This paper pri- tion, Skills and Employment, Australian Government (dese.gov.au). https://www. marily set out to establish whether clinical performance scores ob- dese.gov.au/job-ready/improving-accountability-information-providers. Accessed tained via the APP were best represented by one or two factors. Using April 2, 2022. factor analyses, the results demonstrated that clinical performance 6. Pardo del Val MP, et al. Manage Decis. 2003;41:148–155. using the APP are best characterised by two factors, representing 7. Sevenheuysen S, et al. J Physiother. 2014;60:209–216. professional skills (items 1 to 4) and clinical skills (items 5 to 20). 8. Sevenheuysen S, et al. J Physiother. 2015;61:87–92. Furthermore, these interpretations and item scaling were mostly 9. Peiris CL, et al. J Physiother. 2022;68:61–68. consistent over time (across four clinical placements) and placement 10. Lawton V, et al. J Physiother. 2021;67:56–61. context (cardiorespiratory, neurology, musculoskeletal and miscella- 11. Francis-Cracknell A, et al. J Physiother. 2017;63:243–249. neous). A large scale, multi-site (approximately 20 universities across 12. Dalton L, et al. Rural Remote Health. 2008;8:962. Australia and New Zealand) replication study is currently underway to 13. Australian Government Department of Health and Ageing. Report on the Audit of determine the robustness of these findings across different sites, Health Workforce in Rural and Regional Australia. Canberra: 2008. https://apo.org. geographical locations and supervisor demographics. au/node/3628. Accessed April 11, 2022. 14. Haines KJ, et al. J Physiother. 2020;66:67–69. Several additional avenues for future research are recommended in 15. Dario A, et al. J Physiother. 2021;67:235–237. the papers in this article collection. Rivers et al22 suggested exploring 16. Nahon I, et al. Aust J Clin Educ. 2021;9:17–27. career satisfaction and non-financial incentives for choosing and 17. Tognon K, et al. J Physiother. 2021;67:79–81. maintaing a career in physiotherapy. Peiris et al9 recommended 18. Dalton M, et al. Development of the Assessment of Physiotherapy Practice (APP): A examining private practitioners’ views on clinical education. Dario standardised and valid approach to assessment of clinical competence in physio- therapy [Australian Learning and Teaching Council (ALTC) Final report. 2009 6–28]. https://ltr.edu.au/resources/grants_pp_physiotheraphy_instrument_griffith_report_ 2009.pdf. Accessed April 11, 2022. 19. Dalton M, et al. J Physiother. 2011;57:239–246. 20. Dalton M, et al. J Physiother. 2012;58:49–56. 21. Reubenson A, et al. J Physiother. 2020;66:113–119. 22. Rivers G, et al. J Physiother. 2015;61:148–154. 23. Australian Physiotherapy Association. A strong physiotherapy workforce for a healthy Australia. https://australian.physio/sites/default/files/APA_Pre-Budget_ Submission_2021.pdf. Accessed April 2, 2022. 24. Andrew E, et al. J Physiother. 2019;65:222–229.
Journal of Physiotherapy 68 (2022) 208 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: Assessment of generalised joint hypermobility: the Beighton score Summary Description: Joint hypermobility (JH) is defined as the ability of a joint - Fifth finger extension: with the palm of the hand and forearm resting to move beyond normal limits along physiological axes. Joint hypermo- on a flat surface with the elbow flexed at 90 deg, the metacarpal- phalangeal joint of the fifth finger can be hyperextended . 90 deg bility is relatively common, occurring in 7 to 36% of children and ado- with respect to the dorsum of the hand. lescents, and in 2 to 57% of adult populations,1–7 with higher rates - Elbow extension: with the arms outstretched to the side and forearm in amongst children, females, and Asian and African racial groups. supination, the elbow extends . 10 deg. When JH is observed in fewer than five joints, it may be defined as localised - Knee extension: while standing, with knees locked in genu recurvatum, joint hypermobility. Localised joint hypermobility affects a single small or the knee extends . 10 deg. large joint and may be bilateral, and may be an inherited or acquired trait - Forward bending: with knees locked straight and feet together, the patient can bend forward to place the total palm of both hands flat on related to trauma or training. The term generalised joint hypermobility (GJH) the floor just in front of the feet. is preferred in individuals with JH at multiple sites (usually five or more).8 A score is obtained for each bilateral item (0: negative, 1: positive), Joint hypermobility may occur as a result of non-pathological or with a total score ranging from 0 to 9. Recent research recommends variable cut-offs for the age, sex and cohort of interest.11 In children, GJH is pathological causes. In the non-pathological group, subjects do not established at a cut-off value of 6; in adults to 50 years 5/9, whereas develop severe musculoskeletal complaints5 and JH is even an advantage in adults aged . 50 years 4/9 is positive.9,12 for some (eg, dance and music).6 In the pathological group, JH is associ- Psychometric properties of the Beighton score: Concurrent validity ated with musculoskeletal complications like pain and joint instability, of the BS score, according to the COSMIN criteria, is poor to fair; more studies are needed to confirm the validity.12 A recent systematic review which can lead to dislocations and distortions. Generalised joint hyper- included 24 studies in 1,333 patients (age range 4 to 71 years). Inter-rater mobility is frequently observed in hereditary connective tissue disorders, and intra-rater reliability were moderate to excellent using intraclass correlation coefficients. The authors concluded that the BS is a highly which are characterised by pathological connective tissue fragility in reliable clinical tool with substantial to excellent inter-rater and intra- multiple organ systems and can be molecularly confirmed in most cases.8 rater reliability when used by raters of variable backgrounds and expe- In addition, the current label for subjects with JH and musculoskeletal rience levels.13 While individual components of risk of bias among studies complications, who do not fulfil the criteria for hypermobile Ehlers- demonstrated large discrepancy, most of the items were adequate to very Danlos syndromes, is hypermobility spectrum disorders.9 good.13 The Beighton score: The Beighton score (BS) is considered to be the ‘gold standard’ for assessing GJH and is the most widely used.10 The BS consists of five standardised tests, including four bilateral tests: - First finger opposition: with arms outstretched forward but hand pro- nated, the thumb can be passively moved to touch the ipsilateral forearm. Commentary Raoul HH Engelberta and Lies Rombautb aCenter of Expertise Urban Vitality, Faculty of Health, Amsterdam The strengths of the BS are that it is a quick (5 minutes) and reliable University of Applied Sciences, and Department of Rehabilitation tool to screen for GJH, is known worldwide, and requires no equipment, and Pediatrics (Emma Children’s Hospital), University of Amsterdam, with the exception of a goniometer when joint range is equivocal. Limitations of the BS include that the evidence with respect to the Amsterdam, Netherlands validity is weak, and that it assesses a limited number of joints, mainly bCenter for Medical Genetics, Ghent University Hospital/Ghent upper limb, while assessment of other joints that are commonly painful (eg, shoulder, hip, patellofemoral and ankle) remains unassessed. Also, University, Ghent, Belgium the BS only assesses motion in the sagittal plane. Therefore, other comprehensive tools for JH have been developed and provide more References detailed information about more joints in multiple planes of movement (eg, the Upper Limb Hypermobility Assessment Tool14 and Lower Limb 1. Mikkelsson M, et al. J Rheumatol. 1996;23:1963–1967. Assessment Score).15,16 2. El-Metwally A. Pain. 2004;110:550–559. 3. Juul-Kristensen B, et al. Pediatrics. 2009;124:1380–1387. Reliable, accurate and precise measures of JH and for identifying GJH 4. Remvig L, et al. Int Musculoskelet Med. 2011;33:137–145. are essential in children, adolescents and adults with musculoskeletal 5. Castori M, et al. Am J Med Genet C Semin Med Genet. 2017;175:148–157. complaints, and contribute to diagnostics and tailored interventions. 6. Baeza-Velasco C, et al. Curr Sports Med Rep. 2013;12:291–295. Generalised joint hypermobility is a risk factor for musculoskeletal con- 7. Remvig L, et al. J Rheumatol. 2007;34:798–803. ditions such as joint instability, impingement and sprains.13 Furthermore, 8. Castori M, et al. Am J Med Genet C Semin Med Genet. 2017;175:148–157. determining the localisation and generalisability of JH alongside func- 9. Malfait F, et al. Am J Med Genet C Semin Med Genet. 2017;175:8–26. tional assessments of joint stability, muscle strength, proprioception, pain, 10. Beighton P, et al. Ann Rheum Dis. 1973;32:413–418. fatigue, disability and participation should lead to tailored, interdisci- 11. Singh H, et al. Rheumatol. 2017;56:1857–1864. plinary management. 12. Juul-Kristensen B, et al. Am J Med Genet C. 2017;175:116–147. 13. Bockhorn LN, et al. Orthop J Sports Med. 2021;9. Provenance: Invited. Not peer reviewed. 14. Nicholson LL, et al. Musculoskelet Sci Pract. 2018;35:38–45. 15. Ferrari J, et al. Clin Exp Rheumatol. 2005;23:413–420. 16. Meyer KJ, et al. BMC Musculoskelet Disord. 2017;18:514. https://doi.org/10.1016/j.jphys.2022.02.004 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 68 (2022) 207 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 Central Sensitisation Inventory: a useful screening tool for clinicians, but not the gold standard Summary pain,8 and also psychophysiological measures of central sensitisation such as decreased pain thresholds at local and remote body sites,9,10 conditioned pain Description: The Central Sensitisation Inventory (CSI) is a patient-reported modulation9 and the widespread pain index.8 Improvements in CSI scores in outcome measure that has been widely used to study symptoms of central sensitisa- tion (eg, sleep and concentration difficulties, sensitivity to light and odours, spreading response to multimodal treatment for patients with chronic spinal pain support pain, stress as an aggravating factor, restless legs) in patients with chronic pain.1 For its responsiveness to change.11 clinical purposes, central sensitisation is defined asan amplification of neural signalling within the central nervous system that elicits pain hypersensitivity.2 In addition, cut-off scores (ie, scores 40/100 are suggestive of central sensitisation) and severity levels have been proposed to facilitate the interpre- The CSI is a two-part questionnaire that is freely available in 18 languages.3 Part A tation of the CSI.12,13 The following CSI severity levels were established: sub- measures a full array of 25 somatic and emotional symptoms associated with central sensitisation, scored on a 5-point Likert scale from 0 (never) to 4 (always), resulting clinical = 0 to 29; mild = 30 to 39; moderate = 40 to 49; severe = 50 to 59; and in a total possible score of 100.1 Higher scores denote a higher degree of self-reported extreme = 60 to 100.14 Cuestas Vargas et al recently developed a CSI cluster central sensitisation-related symptomatology. Part B (ie, registration of previous calculator15 that proposes to classify patients into three groups of central central sensitisation-related diagnoses) is often not considered in clinimetric studies and therefore not further addressed here. Moreover, clinicians are advised to only sensitisation-related symptom severity: low, medium and high levels of central use part A. The time needed to complete part A is estimated to be approximately 3 minutes. There is no specific training required to administer or score the CSI. sensitisation-related symptom severity. The psychometric properties of the CSI in patients who have nonspecific, non- In patients with musculoskeletal disorders, higher CSI scores were found to cancer pain are well-established.4 When examining the factor structure in a predict higher pain-related disability 3 months later.16 In patients undergoing pooled sample (n = 1,987), it was concluded that only total CSI scores should be used and reported.5 The internal consistency of the CSI is excellent, with Cron- total knee arthroplasty and lumbar fusion surgery, preoperative CSI scores of 40 bach’s a values ranging from 0.87 to 0.91.4,5 Test-retest reliability was also found were predictive of worse postoperative outcomes.4,17 These findings support the to be high.4,6 Scores obtained with the CSI showed concurrent validity with a predictive ability of the CSI and illustrate its clinical utility. range of relevant measures, including measures of resilience,7 negative affect,7 anxiety,7 pain catastrophising,8 duration and severity of pain,8 lateralisation of A shorter version of the CSI, called the CSI-9, has recently been developed using Rash analysis.18 Its scoring is similar to the original CSI (total score = 36). Although the CSI-9 has shown acceptable psychometric properties and adequately covers the factors identified in the original CSI,18 more validation work is needed before its use can be advocated. Commentary cResearch Foundation – Flanders (FWO), Brussels, Belgium dDepartment of Health and Rehabilitation, Unit of Physiotherapy, Insti- The CSI is practical to use in nearly every clinical setting, requires limited tute of Neuroscience and Physiology, Sahlgrenska Academy, University of resources, is available in a large number of languages, and a body of literature supports its psychometric features (in various languages). However, studies Gothenburg, Sweden specifically designed to examine the responsiveness of the CSI, including the eUniversity of Gothenburg Center for Person-Centred Care (GPCC), establishment of the threshold for clinically important differences, are still needed. Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden Also, due to the lack of a gold standard measure of central sensitisation, the content validity of the CSI cannot be directly verified. Caution is needed when interpreting References the CSI total score, and clinicians are advised to use the latter together with other clinical features of central sensitisation. Indeed, evidence-based recommendations 1. Mayer TG, et al. Pain Pract. 2012;12:276–285. are available for identifying a predominant presentation of central sensitisation in 2. Woolf CJ. Pain. 2011;152(3 Suppl):S2–S15. different patient populations,19–22 including the exclusion of neuropathic pain, 3. www.pridedallas.com/questionnaires/ assessment of disproportionate pain and examining whether the pain distribution 4. Scerbo T, et al. Pain Pract. 2018;18:544–554. is neuro-anatomically plausible in combination with the CSI total score.19 5. Cuesta-Vargas AI, et al. J Pain. 2018;19:317–329. 6. Bilika P, et al. Pain Pract. 2020;20:188–196. Funding: The article was unfunded. EH is a research fellow funded by the Research 7. Coronado RA, et al. Musculoskelet Sci Pract. 2018;36:61–67. Foundation Flanders (FWO), Belgium. JN is holder of a chair in oncological rehabil- 8. van Wilgen CP, et al. Pain Pract. 2018;18:239–246. itation funded by the Berekuyl Academy, Hierden, the Netherlands, and is part of the 9. Gervais-Hupé J, et al. Clin Rheumatol. 2018;37:3125–3132. guest professorship program of the University of Gothenburg, Sweden. 10. Zafereo J, et al. Pain Pract. 2020. 11. Malfliet A, et al. JAMA Neurol. 2018;75:808–817. Provenance: Invited. Not peer reviewed. 12. Neblett R, et al. Clin J Pain. 2014. 13. Neblett R, et al. J Pain. 2013;14:438–445. Jo Nijsa,b,d,e and Eva Huysmansa,b,c 14. Neblett R, et al. Pain Pract. 2017;17:166–175. aPain in Motion Research Group (PAIN), Department of Physiotherapy, 15. Cuesta-Vargas AI, et al. Pain Med. 2020;21:2430–2440. 16. Tanaka K, et al. Eur J Pain. 2019;23:1640–1648. Human Physiology and Anatomy, Faculty of Physical Education & 17. Koh IJ, et al. J Arthroplasty. 2020. Physiotherapy, Vrije Universiteit Brussel, Belgium 18. Nishigami T, et al. PloS One. 2018;13, e0200152. 19. Kosek E, et al. Pain. 2021;162:2629–2634. bChronic Pain Rehabilitation, Department of Physical Medicine and 20. Nijs J, et al. Pain Physician. 2014;17:447–457. Physiotherapy, University Hospital Brussels, Belgium 21. Lluch E, et al. Disabil Rehabil. 2018;40:2836–2845. 22. Nijs J, et al. Pain Physician. 2015;18:E333–E346. https://doi.org/10.1016/j.jphys.2021.10.004 1836-9553/© 2021 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/).
214 Correspondence References 5. Murphy KR, et al. Journal of Applied Psychology. 1999;84:234–248. https://doi.org/10. 1037/0021-9010.84.2.234 1. Elkins MR, et al. Journal of Physiotherapy. 2021. https://doi.org/10.1016/j.jphys.2021.12.001 2. Boos DD, et al. The American Statistician. 2011;65:213–221. https://doi.org/10.1198/ 6. Lakens D. Social Psychological and Personality Science. 2017;8:355–362. https://doi. org/10.1177/1948550617697177 tas.2011.10129 3. Miller J, et al. Psychological Methods. 2011;16:337–360. https://doi.org/10.1037/ 7. Lakens D. Perspectives on Psychological Science. 2021;16:639–648. https://doi.org/10. 1177/1745691620958012 a0023347 4. Lakens, D. 2022. Sample Size Justification. PsyArXiv. https://doi.org/10.31234/osf.io/9d3yf 8. Scheel AM, et al. Perspectives on Psychological Science. 2021;16:744–755. https://doi. org/10.1177/1745691620966795 32 Correspondence: Response to Lakens We thank Associate Professor Lakens for his interest in our therefore, effect sizes complement, but do not replace, hypothesis tests.’ editorial. We will address each of his five comments. We disagree for reasons explained in the previous paragraph. There is no reason to worry about authors claiming there is an effect ‘. only God knows the probability that the null hypothesis is true when there truly is exactly no effect, because there truly always is given the data observed, and no statistical method can provide it. Esti- at least some effect (although the effect may be microscopically mation will not tell you anything about the probability of hypotheses.’ small). Instead of being concerned with whether there is or is not Bayesian analyses can estimate the probability of a hypothesis given an effect we need to know if the effect is big enough to be of any the data. In any case, as the editorial explains, there is little point in substantive interest. p values convey no useful information on this knowing the probability that the null hypothesis is true. issue, and they convey no information that cannot be gleaned from a confidence interval. In contrast, confidence intervals contain much ‘A p-value does not constitute evidence. Neither do estimates, so their useful information that cannot be gleaned from a p value. Confi- proposed alternative suffers the same criticism.’ It is true that estimates, dence intervals can replace p values without any loss of useful like p values, are not evidence. However, proponents of null hy- information. pothesis testing imply that p values are useful or meaningful because they provide evidence that can be used to reject the null hypothesis A/Prof Lakens argues that the suggestion of how to interpret a (Fisherian significance testing) or that can be used to choose between confidence interval is not estimation but is ‘minimum effect testing’. the null and alternative hypotheses (Neyman-Pearson hypothesis In our opinion, the key feature of estimation is that it seeks to testing). In contrast estimates, unlike p values, are intrinsically estimate the value of a population parameter. That should be the meaningful. key objective of most inferential statistical analyses, and is the approach advocated in the editorial. Interpretation of the data from ‘It is not possible to determine the probability a study will replicate clinical trials inevitably involves consideration of the importance or based on a single value (Miller & Schwarz, 2011). Furthermore, well- clinical significance of the estimated average effect of the inter- designed replication studies do not use the same sample size as an vention. A/Prof Lakens points out that, if that is done formally using earlier study, but are designed to have high power for an effect size of the tools of significance or hypothesis testing then it becomes interest (Lakens, 2022).’ We don’t disagree with either assertion. minimum effect testing. And, as he points out, that requires formal Neither changes the substantive point: experimenters who obtain a enumeration of the smallest important effect. However, like Amr- significant test finding cannot expect that, if an exact replication of hein and Greenland,1 we do not see the need to use the machinery their study were possible, it too would obtain a significant finding. As of significance testing or hypothesis testing to rationally interpret Amhrein and Greenland1 state: ‘random variation alone can easily estimates of effect. In the absence of a well-established threshold lead to large disparities in P values, far beyond falling just to either for interpretation, authors can still interpret a confidence interval side of the 0.05 threshold. For example, even if researchers could by describing the practical implications of all values inside the conduct two perfect replication studies of some genuine effect, confidence interval.1 And there is another reason not to conduct each with 80% power (chance) of achieving P , 0.05, it would not minimum effect tests: researchers who supply confidence be very surprising for one to obtain P , 0.01 and the other P . intervals, rather than conducting minimum effect tests, devolve 0.30’ (p306). the responsibility of distinguishing between important and unimportant effects to their readers. Arguably that is where that ‘Fourth, the editors argue, without any empirical evidence, that in responsibility should lie. most clinical trials the null-hypothesis must be false.’ The assertion that the null hypothesis is false in most clinical trials does not require Mark R Elkins, Rafael Zambelli Pinto, Arianne Verhagen, empirical evidence, because it is self-evidently true. The null hy- Monika Grygorowicz, Anne Söderlund, Matthieu Guemann, pothesis is that there is exactly no effect – it is not, as A/Prof Lakens Antonia Gómez-Conesa, Sarah Blanton, Jean-Michel Brismée, implies, that the null is true within the bounds we can detect with Shabnam Agarwal, Alan Jette, Michele Harms, Geert Verheyden available resources. While the latter may often be true, the former never is. The null hypothesis may often be approximately true, but it and Umer Sheikh is rarely if ever exactly true. Moreover, empirical estimates of effects are always at least a little bit biased. So exactly null hypotheses must https://doi.org/10.1016/j.jphys.2022.06.003 always be false. The only reason they are not always found to be false is that almost all studies lack the precision to detect tiny effects. For References that reason, empirical evidence is unable to demonstrate that the null hypothesis is not always true. And that is why van der Laan and Rose2 1. Amrhein V, et al. Nature. 2019;567:305–307. state that ‘We know that for large enough sample sizes, every study, 2. van der Laan MJ, et al. Targeted Learning. Causal Inference for Observational and including one in which the null hypothesis of no effect is true, will declare a statistically significant effect’ (p xvi). Experimental Data. Springer; 2011. ‘Finally, the fifth point that “Researchers need to know more than just whether an effect does or does not exist” is correct, but the “more than” is crucial. It remains important to prevent authors from claiming there is an effect, when they are actually looking at random noise, and
Journal of Physiotherapy 68 (2022) 213-214 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 Correspondence: Reward, but do not yet require, interval hypothesis tests In a co-editorial by journal editors from the International Society of population.’ ‘If the estimate and the ends of its confidence interval are Physiotherapy Journal Editors1 state: ‘[This editorial] also advises all more favourable than the smallest worthwhile effect, then the researchers that some physiotherapy journals that are members of the treatment effect can be interpreted as typically considered worth- International Society of Physiotherapy Journal Editors (ISPJE) will be while by patients in that clinical population. If the effect and its expecting manuscripts to use estimation methods instead of null confidence interval are less favourable than the smallest worthwhile hypothesis statistical tests.’ In this comment, I highlight errors and effect, then the treatment effect can be interpreted as typically inconsistencies in the arguments the editors put forward for their considered trivial by patients in that clinical population.’ This is not a recommendation. description of an estimation approach. The editors are recommending a hypothesis testing approach against a smallest effect size of interest. The editors list five problems with p values. First, they state that When examining if the effect is more favorable than the smallest ‘p values do not equate to a probability that researchers need to effect size of interest, this is known as a minimum effect test.5 When know’ because ‘Researchers need to know the probability that the examining whether an effect is less favorable than the smallest effect null hypothesis is true given the data observed in their study.’ size of interest, this is known as an equivalence test.6 Both are Regrettably, the editors do not seem to realise that only God knows examples of interval hypothesis tests, where instead of comparing the probability that the null hypothesis is true given the data the observed effect against 0 (as in a null-hypothesis test) the effect observed, and no statistical method can provide it. Estimation will is compared against a range of values that are deemed theoretically or not tell you anything about the probability of hypotheses. Their practically interesting. If the title of the editorial had been ‘Statistical second point is that a p values does not constitute evidence. inference through interval hypothesis tests’ then the recommenda- Neither do estimates, so their proposed alternative suffers from the tions would have been coherent and ambitious. Minimum-effect tests same criticism. Third, the editors claim that significant results have and equivalence tests are an excellent improvement upon null a low probability of replicating, and that when a p value between hypothesis significance tests.7 But requiring every study to perform 0.005 and 0.05 is observed, repeating this study would only have a range predictions can be difficult because the justification for a 67% probability of observing a significant result. This is incorrect. smallest effect size of interest is not trivial, and often requires The citation to Boos and Stefanski2 is based on the assumption that dedicated research lines. Forcing researchers to prematurely test multiple p values are available to estimate the average power of the hypotheses against a smallest effect size of interest, before they set of studies, and that the studies will have 67% power. It is not have carefully established a smallest effect size of interest, will be possible to determine the probability a study will replicate based counterproductive.8 Requiring all researchers to perform hypothesis on a single p value.3 Furthermore, well-designed replication tests against a smallest effect size of interest runs the risk of forcing studies do not use the same sample size as an earlier study, but are researchers to perform meaningless tests. The editors can reward designed to have high power for an effect size of interest.4 Fourth, submissions that use interval hypothesis tests, but should allow the editors argue, without any empirical evidence, that in most submissions to use null hypothesis significance tests. All these tests clinical trials the null hypothesis must be false. The prevalence of need to justify why the null hypothesis, be it an effect of 0, or a null results make it doubtful this statement is true in any practical smallest effect size of interest, is an interesting value to reject. In sense. In an analysis of 11,852 meta-analyses from Cochrane re- research lines where a smallest effect size of interest has not been views, only 5,903 meta-analyses, or 49.8%, found a statistically established in a dedicated research line, and where the novelty of significant meta-analytic effect. Large registered replication re- the research question makes it difficult to make more than ports in psychology have found null effects with studies ranging directional predictions, the p values of a null hypothesis test might from 1,894 participants to 5,610 participants. Without infinite re- be the best researchers can do, and can provide the first indication sources, the claim that the null is always false can neither be that a question deserves further exploration, including work that proven nor falsified. It seems prudent to assume the null is true in accurately estimates the effect size, as well as research on what the some studies, at least within bounds we can detect with the re- smallest effect size of interest is. sources researchers typically have available. Finally, the fifth point that ‘Researchers need to know more than just whether an effect Acknowledgments: This work was supported by the Netherlands does or does not exist’ is correct, but the ‘more than’ is crucial. It Organization for Scientific Research 19 (NWO) VIDI grant 452-17-013. remains important to prevent authors from claiming there is an effect, when they are actually looking at random noise, and Daniël Lakens therefore, effect sizes complement, but do not replace, hypothesis Human-Technology Interaction Group, Eindhoven University of tests. Technology, Eindhoven, The Netherlands The editors recommend to use estimation, and report confidence intervals around estimates. But then the editors write ‘The estimate https://doi.org/10.1016/j.jphys.2022.06.004 and its confidence interval should be compared against the “smallest worthwhile effect” of the intervention on that outcome in that 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/).
214 Correspondence References 5. Murphy KR, et al. Journal of Applied Psychology. 1999;84:234–248. https://doi.org/10. 1037/0021-9010.84.2.234 1. Elkins MR, et al. Journal of Physiotherapy. 2021. https://doi.org/10.1016/j.jphys.2021.12.001 2. Boos DD, et al. The American Statistician. 2011;65:213–221. https://doi.org/10.1198/ 6. Lakens D. Social Psychological and Personality Science. 2017;8:355–362. https://doi. org/10.1177/1948550617697177 tas.2011.10129 3. Miller J, et al. Psychological Methods. 2011;16:337–360. https://doi.org/10.1037/ 7. Lakens D. Perspectives on Psychological Science. 2021;16:639–648. https://doi.org/10. 1177/1745691620958012 a0023347 4. Lakens, D. 2022. Sample Size Justification. PsyArXiv. https://doi.org/10.31234/osf.io/9d3yf 8. Scheel AM, et al. Perspectives on Psychological Science. 2021;16:744–755. https://doi. org/10.1177/1745691620966795 32 Correspondence: Response to Lakens We thank Associate Professor Lakens for his interest in our therefore, effect sizes complement, but do not replace, hypothesis tests.’ editorial. We will address each of his five comments. We disagree for reasons explained in the previous paragraph. There is no reason to worry about authors claiming there is an effect ‘. only God knows the probability that the null hypothesis is true when there truly is exactly no effect, because there truly always is given the data observed, and no statistical method can provide it. Esti- at least some effect (although the effect may be microscopically mation will not tell you anything about the probability of hypotheses.’ small). Instead of being concerned with whether there is or is not Bayesian analyses can estimate the probability of a hypothesis given an effect we need to know if the effect is big enough to be of any the data. In any case, as the editorial explains, there is little point in substantive interest. p values convey no useful information on this knowing the probability that the null hypothesis is true. issue, and they convey no information that cannot be gleaned from a confidence interval. In contrast, confidence intervals contain much ‘A p-value does not constitute evidence. Neither do estimates, so their useful information that cannot be gleaned from a p value. Confi- proposed alternative suffers the same criticism.’ It is true that estimates, dence intervals can replace p values without any loss of useful like p values, are not evidence. However, proponents of null hy- information. pothesis testing imply that p values are useful or meaningful because they provide evidence that can be used to reject the null hypothesis A/Prof Lakens argues that the suggestion of how to interpret a (Fisherian significance testing) or that can be used to choose between confidence interval is not estimation but is ‘minimum effect testing’. the null and alternative hypotheses (Neyman-Pearson hypothesis In our opinion, the key feature of estimation is that it seeks to testing). In contrast estimates, unlike p values, are intrinsically estimate the value of a population parameter. That should be the meaningful. key objective of most inferential statistical analyses, and is the approach advocated in the editorial. Interpretation of the data from ‘It is not possible to determine the probability a study will replicate clinical trials inevitably involves consideration of the importance or based on a single value (Miller & Schwarz, 2011). Furthermore, well- clinical significance of the estimated average effect of the inter- designed replication studies do not use the same sample size as an vention. A/Prof Lakens points out that, if that is done formally using earlier study, but are designed to have high power for an effect size of the tools of significance or hypothesis testing then it becomes interest (Lakens, 2022).’ We don’t disagree with either assertion. minimum effect testing. And, as he points out, that requires formal Neither changes the substantive point: experimenters who obtain a enumeration of the smallest important effect. However, like Amr- significant test finding cannot expect that, if an exact replication of hein and Greenland,1 we do not see the need to use the machinery their study were possible, it too would obtain a significant finding. As of significance testing or hypothesis testing to rationally interpret Amhrein and Greenland1 state: ‘random variation alone can easily estimates of effect. In the absence of a well-established threshold lead to large disparities in P values, far beyond falling just to either for interpretation, authors can still interpret a confidence interval side of the 0.05 threshold. For example, even if researchers could by describing the practical implications of all values inside the conduct two perfect replication studies of some genuine effect, confidence interval.1 And there is another reason not to conduct each with 80% power (chance) of achieving P , 0.05, it would not minimum effect tests: researchers who supply confidence be very surprising for one to obtain P , 0.01 and the other P . intervals, rather than conducting minimum effect tests, devolve 0.30’ (p306). the responsibility of distinguishing between important and unimportant effects to their readers. Arguably that is where that ‘Fourth, the editors argue, without any empirical evidence, that in responsibility should lie. most clinical trials the null-hypothesis must be false.’ The assertion that the null hypothesis is false in most clinical trials does not require Mark R Elkins, Rafael Zambelli Pinto, Arianne Verhagen, empirical evidence, because it is self-evidently true. The null hy- Monika Grygorowicz, Anne Söderlund, Matthieu Guemann, pothesis is that there is exactly no effect – it is not, as A/Prof Lakens Antonia Gómez-Conesa, Sarah Blanton, Jean-Michel Brismée, implies, that the null is true within the bounds we can detect with Shabnam Agarwal, Alan Jette, Michele Harms, Geert Verheyden available resources. While the latter may often be true, the former never is. The null hypothesis may often be approximately true, but it and Umer Sheikh is rarely if ever exactly true. Moreover, empirical estimates of effects are always at least a little bit biased. So exactly null hypotheses must https://doi.org/10.1016/j.jphys.2022.06.003 always be false. The only reason they are not always found to be false is that almost all studies lack the precision to detect tiny effects. For References that reason, empirical evidence is unable to demonstrate that the null hypothesis is not always true. And that is why van der Laan and Rose2 1. Amrhein V, et al. Nature. 2019;567:305–307. state that ‘We know that for large enough sample sizes, every study, 2. van der Laan MJ, et al. Targeted Learning. Causal Inference for Observational and including one in which the null hypothesis of no effect is true, will declare a statistically significant effect’ (p xvi). Experimental Data. Springer; 2011. ‘Finally, the fifth point that “Researchers need to know more than just whether an effect does or does not exist” is correct, but the “more than” is crucial. It remains important to prevent authors from claiming there is an effect, when they are actually looking at random noise, and
Journal of Physiotherapy 68 (2022) 203 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: Early surgery is not superior to exercise and education with the option of later surgery for meniscal tears in young adults Synopsis Summary of: Skou ST, Hölmich P, Lind M, Jensen HP, Jensen C, Garval baseline to 12 months (score range 0 to 100, 100 = best score). Secondary M, et al. Early surgery or exercise and education for meniscal tears in outcomes included KOOS subscales, Western Ontario Meniscal Evalua- young adults. NEJM Evidence. 2022;1:EVIDoa2100038. tion Tool, isometric leg press muscle strength using FysioMeter, maximum number of knee bends in 30 seconds, one-leg hop for dis- Question: Is early meniscal surgery superior to a strategy of exercise and tance, and the 6-m timed hop. Results: During follow-up, 16 participants education with the option of later surgery for young adults with mag- (26%) from the exercise group crossed over to surgery, while eight (13%) netic resonance imaging-verified meniscal tears? Design: Multicentre, from the surgery group did not undergo surgery; 107 (88%) completed pragmatic, superiority, parallel-group randomised controlled trial and the 12-month follow-up. The intention-to-treat analysis did not concealed allocation. Setting: Seven Danish hospitals. demonstrate any clear between-group difference in change from base- Participants: Young adults aged 18 to 40 years with magnetic resonance line to 12 months in KOOS4 (adjusted mean difference 5.4 (95% CI –0.7 imaging-verified meniscal tears eligible for meniscal surgery. Main to 11.4)). Both groups improved in all secondary outcomes. Per-protocol exclusion criteria were: prior surgery or fracture in the past 12 months and as-treated analyses yielded similar results with no clear between- in the affected knee, bucket handle meniscal tear, knee ligament rupture group differences in the primary outcome and most of the secondary or participation in supervised exercise therapy during the past 3 outcomes. Conclusion: A strategy of early meniscal surgery is not su- months. Randomisation (1 to 1) of 121 participants allocated 60 to early perior to a strategy of exercise and education with the option of later surgery and 61 to exercise and education. Interventions: The early surgery among young, active adults with meniscal tears. surgery group underwent arthroscopic partial meniscectomy or meniscal repair followed by postoperative rehabilitation for those un- Provenance: Invited. Not peer reviewed. dergoing meniscal repair. The other group received a 12-week individ- ualised and supervised exercise therapy and patient education program. Nina Østerås The exercise therapy group sessions took place twice weekly with 60 to Division of Rheumatology and Research, 90 minutes of neuromuscular and strengthening and 30 to 45 minutes of patient education by trained physiotherapists at 19 clinics. Outcome Diakonhjemmet Hospital, Norway measures: The primary outcome was between-group difference in change in Knee Injury and Osteoarthritis Outcome Score (KOOS4) from https://doi.org/10.1016/j.jphys.2022.05.010 Commentary The evidence refuting arthroscopic partial meniscectomy to treat Given the potential for healthcare cost savings with non-surgical interventions, future cost-effectiveness analysis will be important to meniscal tears and degeneration in older adults is well-established. At direct healthcare resources for young people with meniscal tears. Long-term follow-up will also elucidate any potential benefit or harm least 10 randomised controlled trials have similar conclusions: knee of meniscal surgery on knee osteoarthritis development – a key concern for patients and practitioners. arthroscopy confers no additional benefit to exercise therapy or pla- cebo surgery.1 However, the question has always remained for younger Provenance: Invited. Not peer reviewed. populations: Is arthroscopic meniscal surgery effective for meniscal tears Adam G Culvenor La Trobe Sport and Exercise Medicine Research Centre, in young adults? Little evidence was available to answer this question – until now. The DREAM trial by Skou et al2 provides evidence that what La Trobe University, Melbourne, Australia we have known for many years in older adults also appears to ring true https://doi.org/10.1016/j.jphys.2022.05.011 in young adults with meniscal tears mostly due to trauma. References The finding of no clinically important between-group difference 1. Thorlund JB, et al. BMJ. 2015;350:h2747. provides evidence that a progressive exercise program targeting lower- 2. Skou ST, et al. NEJM Evid. 2022;1:EVIDoa2100038. 3. Frobell R, et al. NEJM. 2010;363:331–342. limb strength and neuromuscular control should be exploited before 4. Roos EM, et al. Br J Sports Med. 2019;53:1474–1478. considering surgery, similar to other traumatic knee injuries (ie, anterior cruciate ligament rupture).3 Although one in four patients receiving exercise and education underwent surgery within the 12-month follow-up, 74% avoided surgery, with similar clinical outcomes. Impor- tantly, despite significant clinical improvements in both groups, most were not recovered – it is good to feel better, but better to feel good.4 Mean scores reported at final follow-up suggest that many patients likely remain unsatisfied with their current state.4 Approaches to promote further recovery following meniscal tear are required. https://doi.org/10.1016/j.jphys.2022.05.011 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 68 (2022) 203 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: Early surgery is not superior to exercise and education with the option of later surgery for meniscal tears in young adults Synopsis Summary of: Skou ST, Hölmich P, Lind M, Jensen HP, Jensen C, Garval baseline to 12 months (score range 0 to 100, 100 = best score). Secondary M, et al. Early surgery or exercise and education for meniscal tears in outcomes included KOOS subscales, Western Ontario Meniscal Evalua- young adults. NEJM Evidence. 2022;1:EVIDoa2100038. tion Tool, isometric leg press muscle strength using FysioMeter, maximum number of knee bends in 30 seconds, one-leg hop for dis- Question: Is early meniscal surgery superior to a strategy of exercise and tance, and the 6-m timed hop. Results: During follow-up, 16 participants education with the option of later surgery for young adults with mag- (26%) from the exercise group crossed over to surgery, while eight (13%) netic resonance imaging-verified meniscal tears? Design: Multicentre, from the surgery group did not undergo surgery; 107 (88%) completed pragmatic, superiority, parallel-group randomised controlled trial and the 12-month follow-up. The intention-to-treat analysis did not concealed allocation. Setting: Seven Danish hospitals. demonstrate any clear between-group difference in change from base- Participants: Young adults aged 18 to 40 years with magnetic resonance line to 12 months in KOOS4 (adjusted mean difference 5.4 (95% CI –0.7 imaging-verified meniscal tears eligible for meniscal surgery. Main to 11.4)). Both groups improved in all secondary outcomes. Per-protocol exclusion criteria were: prior surgery or fracture in the past 12 months and as-treated analyses yielded similar results with no clear between- in the affected knee, bucket handle meniscal tear, knee ligament rupture group differences in the primary outcome and most of the secondary or participation in supervised exercise therapy during the past 3 outcomes. Conclusion: A strategy of early meniscal surgery is not su- months. Randomisation (1 to 1) of 121 participants allocated 60 to early perior to a strategy of exercise and education with the option of later surgery and 61 to exercise and education. Interventions: The early surgery among young, active adults with meniscal tears. surgery group underwent arthroscopic partial meniscectomy or meniscal repair followed by postoperative rehabilitation for those un- Provenance: Invited. Not peer reviewed. dergoing meniscal repair. The other group received a 12-week individ- ualised and supervised exercise therapy and patient education program. Nina Østerås The exercise therapy group sessions took place twice weekly with 60 to Division of Rheumatology and Research, 90 minutes of neuromuscular and strengthening and 30 to 45 minutes of patient education by trained physiotherapists at 19 clinics. Outcome Diakonhjemmet Hospital, Norway measures: The primary outcome was between-group difference in change in Knee Injury and Osteoarthritis Outcome Score (KOOS4) from https://doi.org/10.1016/j.jphys.2022.05.010 Commentary The evidence refuting arthroscopic partial meniscectomy to treat Given the potential for healthcare cost savings with non-surgical interventions, future cost-effectiveness analysis will be important to meniscal tears and degeneration in older adults is well-established. At direct healthcare resources for young people with meniscal tears. Long-term follow-up will also elucidate any potential benefit or harm least 10 randomised controlled trials have similar conclusions: knee of meniscal surgery on knee osteoarthritis development – a key concern for patients and practitioners. arthroscopy confers no additional benefit to exercise therapy or pla- cebo surgery.1 However, the question has always remained for younger Provenance: Invited. Not peer reviewed. populations: Is arthroscopic meniscal surgery effective for meniscal tears Adam G Culvenor La Trobe Sport and Exercise Medicine Research Centre, in young adults? Little evidence was available to answer this question – until now. The DREAM trial by Skou et al2 provides evidence that what La Trobe University, Melbourne, Australia we have known for many years in older adults also appears to ring true https://doi.org/10.1016/j.jphys.2022.05.011 in young adults with meniscal tears mostly due to trauma. References The finding of no clinically important between-group difference 1. Thorlund JB, et al. BMJ. 2015;350:h2747. provides evidence that a progressive exercise program targeting lower- 2. Skou ST, et al. NEJM Evid. 2022;1:EVIDoa2100038. 3. Frobell R, et al. NEJM. 2010;363:331–342. limb strength and neuromuscular control should be exploited before 4. Roos EM, et al. Br J Sports Med. 2019;53:1474–1478. considering surgery, similar to other traumatic knee injuries (ie, anterior cruciate ligament rupture).3 Although one in four patients receiving exercise and education underwent surgery within the 12-month follow-up, 74% avoided surgery, with similar clinical outcomes. Impor- tantly, despite significant clinical improvements in both groups, most were not recovered – it is good to feel better, but better to feel good.4 Mean scores reported at final follow-up suggest that many patients likely remain unsatisfied with their current state.4 Approaches to promote further recovery following meniscal tear are required. https://doi.org/10.1016/j.jphys.2022.05.011 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 68 (2022) 205 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: For women at risk of breast cancer-related lymphoedema, prophylactic compression sleeve usage reduces and delays arm swelling Synopsis Summary of: Paramanandam VS, Dylke E, Clark GM, Daptardar AA, to be worn for a minimum of 8 hours/day during waking hours. Outcome Kulkarni AM, Nair NS, et al. Prophylactic Use of Compression Sleeves measures: The primary outcome was the incidence of arm swelling, Reduces the Incidence of Arm Swelling in Women at High Risk of measured using the interarm impedance ratio by bioimpedance spec- Breast Cancer-Related Lymphedema: A Randomized Controlled Trial. J troscopy. The secondary outcome measure was physical circumference Clin Oncol. 2022; Feb 2:JCO2102567. https://doi.org/10.1200/JCO.21. measurement of arm swelling. Patient-reported outcomes included four 02567. Epub ahead of print. PMID: 35108031. scales of the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire and the breast cancer-specific questionnaire. Question: Does prophylactic use of compression sleeves in combination Results: A total of 301 participants completed the study. At 1 year after with usual care reduce or delay swelling in women at risk of breast cancer- surgery, incidence of arm swelling was lower in the treatment group (58/ related lymphoedema in the first year after surgery compared with usual 152) compared with the control group (80/149), hazard ratio (HR) 0.61 care? Design: Randomised controlled trial with concealed allocation and (95% CI 0.43 to 0.85 units) and delayed, measured with bioimpedance blinded outcome assessments. Setting: Single centre in India. Partici- spectroscopy. Arm swelling using physical measurements was also lower pants: Women (aged 18 years) requiring unilateral breast cancer sur- and delayed for the intervention group. There were no between-group gery and who could complete questionnaires independently in English, differences for quality of life measures. Conclusion: The preventative Hindi, Marathi or Bengali were included. Exclusion criteria included use of compression sleeves for women at high risk of breast cancer-related preoperative arm swelling and conditions that restricted wearing a lymphoedema reduces and delays swelling at 1 year after surgery. compression sleeve. Randomisation of 307 participants allocated 155 to the intervention group and 152 to the usual care group. Interventions: Provenance: Invited. Not peer reviewed. Both groups received usual care, including one education session about arm, skin and drain care and daily shoulder exercises. Women were taught Alicia Spittle how to identify signs of lymphoedema, and further physiotherapy Department of Physiotherapy, University of Melbourne, Australia assessment and management were undertaken if noted. In addition, the intervention group was provided with two compression sleeves to wear https://doi.org/10.1016/j.jphys.2022.05.002 from the first postoperative day until 3 months after adjuvant treatments, Commentary Breast cancer-related lymphoedema is as a progressive condition Usual care in this study was group education, which may not be that can result in chronic morbidity and tissue composition changes.1 reflective of all health services, as more adopt prospective surveil- To potentially reverse this condition or reduce progression, regular lance. The authors noted that lymphoedema may develop between screening through prospective surveillance is recommended for screening appointments and a recent systematic review of prospec- early detection and intervention.2 tive surveillance2 reported that high-risk women were still devel- oping lymphoedema. Adding prophylactic compression may prove This important paper focuses a step earlier to evaluate whether key to improving outcomes. lymphoedema can be prevented in women at high risk due to axillary lymph node dissection. Participants were randomised to either usual Provenance: Invited. Not peer reviewed. care or usual care plus daily compression sleeve use until 3 months after completion of adjuvant treatment. Compression during this Sharon Czerniec period was supported by findings that some adjuvant treatments Physiotherapy, The University of Newcastle, Australia increase the risk of lymphoedema.3 https://doi.org/10.1016/j.jphys.2022.05.003 The key finding of this randomised trial was that prophylactic compression reduces the incidence and delays the occurrence of References lymphoedema in the first year after surgery. It is planned to follow-up this cohort for 5 years, given that the incidence continues to rise until 1. Dylke ES, et al. Lymphat Res Biol. 2013;11:211–218. 5 years after surgery.4 2. Rafn BS, et al. JCO. 2022;40:1009–1026. 3. Kilbreath SL, et al. Breast. 2016;28:29–36. When considering the clinical application of this study, it is 4. Bucci LK, et al. Ann Surg Oncol. 2021;28:8624–8633. important to note, as the authors did, that all participants were 5. Svensson BJ, et al. Support Care Cancer. 2020;28:3073–3080. educated to identify signs and symptoms of lymphoedema and to seek review as required. Prophylactic compression was therefore not ‘set and forget’. Women at risk of lymphoedema have previously been shown to reliably identify changes requiring review.5 https://doi.org/10.1016/j.jphys.2022.05.003 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 68 (2022) 205 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: For women at risk of breast cancer-related lymphoedema, prophylactic compression sleeve usage reduces and delays arm swelling Synopsis Summary of: Paramanandam VS, Dylke E, Clark GM, Daptardar AA, to be worn for a minimum of 8 hours/day during waking hours. Outcome Kulkarni AM, Nair NS, et al. Prophylactic Use of Compression Sleeves measures: The primary outcome was the incidence of arm swelling, Reduces the Incidence of Arm Swelling in Women at High Risk of measured using the interarm impedance ratio by bioimpedance spec- Breast Cancer-Related Lymphedema: A Randomized Controlled Trial. J troscopy. The secondary outcome measure was physical circumference Clin Oncol. 2022; Feb 2:JCO2102567. https://doi.org/10.1200/JCO.21. measurement of arm swelling. Patient-reported outcomes included four 02567. Epub ahead of print. PMID: 35108031. scales of the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire and the breast cancer-specific questionnaire. Question: Does prophylactic use of compression sleeves in combination Results: A total of 301 participants completed the study. At 1 year after with usual care reduce or delay swelling in women at risk of breast cancer- surgery, incidence of arm swelling was lower in the treatment group (58/ related lymphoedema in the first year after surgery compared with usual 152) compared with the control group (80/149), hazard ratio (HR) 0.61 care? Design: Randomised controlled trial with concealed allocation and (95% CI 0.43 to 0.85 units) and delayed, measured with bioimpedance blinded outcome assessments. Setting: Single centre in India. Partici- spectroscopy. Arm swelling using physical measurements was also lower pants: Women (aged 18 years) requiring unilateral breast cancer sur- and delayed for the intervention group. There were no between-group gery and who could complete questionnaires independently in English, differences for quality of life measures. Conclusion: The preventative Hindi, Marathi or Bengali were included. Exclusion criteria included use of compression sleeves for women at high risk of breast cancer-related preoperative arm swelling and conditions that restricted wearing a lymphoedema reduces and delays swelling at 1 year after surgery. compression sleeve. Randomisation of 307 participants allocated 155 to the intervention group and 152 to the usual care group. Interventions: Provenance: Invited. Not peer reviewed. Both groups received usual care, including one education session about arm, skin and drain care and daily shoulder exercises. Women were taught Alicia Spittle how to identify signs of lymphoedema, and further physiotherapy Department of Physiotherapy, University of Melbourne, Australia assessment and management were undertaken if noted. In addition, the intervention group was provided with two compression sleeves to wear https://doi.org/10.1016/j.jphys.2022.05.002 from the first postoperative day until 3 months after adjuvant treatments, Commentary Breast cancer-related lymphoedema is as a progressive condition Usual care in this study was group education, which may not be that can result in chronic morbidity and tissue composition changes.1 reflective of all health services, as more adopt prospective surveil- To potentially reverse this condition or reduce progression, regular lance. The authors noted that lymphoedema may develop between screening through prospective surveillance is recommended for screening appointments and a recent systematic review of prospec- early detection and intervention.2 tive surveillance2 reported that high-risk women were still devel- oping lymphoedema. Adding prophylactic compression may prove This important paper focuses a step earlier to evaluate whether key to improving outcomes. lymphoedema can be prevented in women at high risk due to axillary lymph node dissection. Participants were randomised to either usual Provenance: Invited. Not peer reviewed. care or usual care plus daily compression sleeve use until 3 months after completion of adjuvant treatment. Compression during this Sharon Czerniec period was supported by findings that some adjuvant treatments Physiotherapy, The University of Newcastle, Australia increase the risk of lymphoedema.3 https://doi.org/10.1016/j.jphys.2022.05.003 The key finding of this randomised trial was that prophylactic compression reduces the incidence and delays the occurrence of References lymphoedema in the first year after surgery. It is planned to follow-up this cohort for 5 years, given that the incidence continues to rise until 1. Dylke ES, et al. Lymphat Res Biol. 2013;11:211–218. 5 years after surgery.4 2. Rafn BS, et al. JCO. 2022;40:1009–1026. 3. Kilbreath SL, et al. Breast. 2016;28:29–36. When considering the clinical application of this study, it is 4. Bucci LK, et al. Ann Surg Oncol. 2021;28:8624–8633. important to note, as the authors did, that all participants were 5. Svensson BJ, et al. Support Care Cancer. 2020;28:3073–3080. educated to identify signs and symptoms of lymphoedema and to seek review as required. Prophylactic compression was therefore not ‘set and forget’. Women at risk of lymphoedema have previously been shown to reliably identify changes requiring review.5 https://doi.org/10.1016/j.jphys.2022.05.003 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 68 (2022) 204 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: High-intensity interval training after stroke improves some aspects of physical function, but benefits are not sustained Synopsis Summary of: Gjellesvik TI, Becker F, Tjønna AE, Indredavik B, Anxiety and Depression Scale, Stroke Impact Scale and Functional Lundgaard E, Solbakken H, et al. Effects of high-intensity interval Independence Measure. Outcomes were assessed at baseline, after training after stroke (The HIIT Stroke Study) on physical and cognitive intervention and at 12 months. Results: Fifty-six participants (n = 28 function: A multicenter randomized controlled trial. Arch Phys Med intervention, n = 28 control) completed the 12-month assessment. Rehabil. 2021;102:1683–1691. Immediately after the intervention, improvements favoured the intervention group in the 6-minute walk test (MD 28 m, 95% CI 3 to Question: Does high-intensity interval training improve physical, 54), Berg Balance Scale (MD 1.27 points, 95% CI 0.17 to 2.38) and Trail mental and cognitive function after stroke? Design: Randomised Making Test-B (MD –24 s, 95% CI –46 to –2), with no differences in controlled trial. Setting: Three rehabilitation hospitals in Norway. other outcomes. The only between-group differences at 12 months Participants: Adults 3 months to 5 years after stroke, ambulating were for Trail Making Test-B (MD –25 s, 95% CI –49 to –2), which independently, with modified Rankin Scale score 0 to 3. Exclusion favoured intervention, and the Functional Independence Measure criteria included unstable cardiac conditions, high resting blood (MD –2.37 points, 95% CI –4.30 to –0.44), which favoured control. pressure and subarachnoid haemorrhage. Randomisation of 70 par- Conclusion: High-intensity interval training produces immediate ticipants allocated 36 to the experimental group and 34 to the control improvements in walking distance, balance and some aspects of group. Interventions: Both groups received standard care. In addition, cognition; however, the benefits are generally not sustained. the experimental group received high-intensity interval treadmill training three times per week for 8 weeks. Training comprised a Provenance: Invited. Not peer reviewed. 10-minute warm up followed by four 4-minute exercise intervals at 85 to 95% of peak heart rate, with 3-minute reduced intensity breaks Prudence Plummer at 50 to 70% peak heart rate. The control group received information Department of Physical Therapy, MGH Institute of Health Professions, about the benefits of physical activity. Outcome measures: Physical function was assessed using the 6-minute walk test, 10-m walk test, USA Berg Balance Scale and Timed Up and Go test. Cognition was assessed with the Montreal Cognitive Assessment and Trail Making https://doi.org/10.1016/j.jphys.2022.05.014 Test (Parts A and B). Additional outcomes included the Hospital Commentary Exercise after stroke is not only important to regain function, programs that people can perform independently or with little improve physical activity and cardiorespiratory fitness, but is a key ongoing direction after initial training. Given the prevalence of poor secondary prevention measure.1 Numerous interventions have been outcomes following stroke,4 clinicians need to know what works, but tested with some success.2,3 High-intensity interval training (HIIT) also what ongoing support is required for maintenance of an exercise may be another potentially effective intervention, as it is effective in regimen that confers ongoing health benefits. An understanding of healthy populations and those with cardiovascular disease. the barriers and facilitators to life-long engagement in activities that improve and maintain health and function is vital and remains This trial showed that HIIT produced greater improvements in the holy grail in long-term stroke management. walking distance compared to information about physical activity, providing preliminary evidence that HIIT may be a promising inter- Provenance: Invited. Not peer reviewed. vention for chronic stroke. However, prior to routine clinical imple- mentation, further information on the safety and feasibility of HIIT is Gavin Williams required, particularly in the subacute phase of stroke. Additional in- Physiotherapy, University of Melbourne/Epworth Healthcare, formation on patient acceptability of HIIT is necessary, along with adherence and retention, given that the intervention dropout rate Melbourne, Australia was close to 30%. Further, this study was implemented in a relatively young and high-functioning group, so clinicians also need to know https://doi.org/10.1016/j.jphys.2022.05.015 the implications for other patients with stroke who are typically older, less mobile and have multiple comorbidities. References Knowledge of efficacious interventions is a high priority for cli- 1. Billinger SA, et al. Stroke. 2014;45:2532–2553. nicians, but successful implementation is required for meaningful 2. Pogrebnoy D, et al. Arch Phys Med Rehabil. 2020;101:154–165. impact. The challenge for researchers and clinicians is the adoption of 3. Billinger SA, et al. Stroke Res Treat. 2012;959120. an effective intervention into a sustainable lifestyle, particularly 4. Sennfält S, et al. Stroke. 2019;50:53–61. https://doi.org/10.1016/j.jphys.2022.05.015 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 68 (2022) 204 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: High-intensity interval training after stroke improves some aspects of physical function, but benefits are not sustained Synopsis Summary of: Gjellesvik TI, Becker F, Tjønna AE, Indredavik B, Anxiety and Depression Scale, Stroke Impact Scale and Functional Lundgaard E, Solbakken H, et al. Effects of high-intensity interval Independence Measure. Outcomes were assessed at baseline, after training after stroke (The HIIT Stroke Study) on physical and cognitive intervention and at 12 months. Results: Fifty-six participants (n = 28 function: A multicenter randomized controlled trial. Arch Phys Med intervention, n = 28 control) completed the 12-month assessment. Rehabil. 2021;102:1683–1691. Immediately after the intervention, improvements favoured the intervention group in the 6-minute walk test (MD 28 m, 95% CI 3 to Question: Does high-intensity interval training improve physical, 54), Berg Balance Scale (MD 1.27 points, 95% CI 0.17 to 2.38) and Trail mental and cognitive function after stroke? Design: Randomised Making Test-B (MD –24 s, 95% CI –46 to –2), with no differences in controlled trial. Setting: Three rehabilitation hospitals in Norway. other outcomes. The only between-group differences at 12 months Participants: Adults 3 months to 5 years after stroke, ambulating were for Trail Making Test-B (MD –25 s, 95% CI –49 to –2), which independently, with modified Rankin Scale score 0 to 3. Exclusion favoured intervention, and the Functional Independence Measure criteria included unstable cardiac conditions, high resting blood (MD –2.37 points, 95% CI –4.30 to –0.44), which favoured control. pressure and subarachnoid haemorrhage. Randomisation of 70 par- Conclusion: High-intensity interval training produces immediate ticipants allocated 36 to the experimental group and 34 to the control improvements in walking distance, balance and some aspects of group. Interventions: Both groups received standard care. In addition, cognition; however, the benefits are generally not sustained. the experimental group received high-intensity interval treadmill training three times per week for 8 weeks. Training comprised a Provenance: Invited. Not peer reviewed. 10-minute warm up followed by four 4-minute exercise intervals at 85 to 95% of peak heart rate, with 3-minute reduced intensity breaks Prudence Plummer at 50 to 70% peak heart rate. The control group received information Department of Physical Therapy, MGH Institute of Health Professions, about the benefits of physical activity. Outcome measures: Physical function was assessed using the 6-minute walk test, 10-m walk test, USA Berg Balance Scale and Timed Up and Go test. Cognition was assessed with the Montreal Cognitive Assessment and Trail Making https://doi.org/10.1016/j.jphys.2022.05.014 Test (Parts A and B). Additional outcomes included the Hospital Commentary Exercise after stroke is not only important to regain function, programs that people can perform independently or with little improve physical activity and cardiorespiratory fitness, but is a key ongoing direction after initial training. Given the prevalence of poor secondary prevention measure.1 Numerous interventions have been outcomes following stroke,4 clinicians need to know what works, but tested with some success.2,3 High-intensity interval training (HIIT) also what ongoing support is required for maintenance of an exercise may be another potentially effective intervention, as it is effective in regimen that confers ongoing health benefits. An understanding of healthy populations and those with cardiovascular disease. the barriers and facilitators to life-long engagement in activities that improve and maintain health and function is vital and remains This trial showed that HIIT produced greater improvements in the holy grail in long-term stroke management. walking distance compared to information about physical activity, providing preliminary evidence that HIIT may be a promising inter- Provenance: Invited. Not peer reviewed. vention for chronic stroke. However, prior to routine clinical imple- mentation, further information on the safety and feasibility of HIIT is Gavin Williams required, particularly in the subacute phase of stroke. Additional in- Physiotherapy, University of Melbourne/Epworth Healthcare, formation on patient acceptability of HIIT is necessary, along with adherence and retention, given that the intervention dropout rate Melbourne, Australia was close to 30%. Further, this study was implemented in a relatively young and high-functioning group, so clinicians also need to know https://doi.org/10.1016/j.jphys.2022.05.015 the implications for other patients with stroke who are typically older, less mobile and have multiple comorbidities. References Knowledge of efficacious interventions is a high priority for cli- 1. Billinger SA, et al. Stroke. 2014;45:2532–2553. nicians, but successful implementation is required for meaningful 2. Pogrebnoy D, et al. Arch Phys Med Rehabil. 2020;101:154–165. impact. The challenge for researchers and clinicians is the adoption of 3. Billinger SA, et al. Stroke Res Treat. 2012;959120. an effective intervention into a sustainable lifestyle, particularly 4. Sennfält S, et al. Stroke. 2019;50:53–61. https://doi.org/10.1016/j.jphys.2022.05.015 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 68 (2022) 206 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 people with chronic obstructive pulmonary disease, a 6-week behaviour change intervention was not superior to a sham intervention at reducing time spent in sedentary behaviour Synopsis Summary of: Cheng SWM, Alison J, Stamatakis E, et al. Six-week The primary outcome was time spent in sedentary behaviour behaviour change intervention to reduce sedentary behaviour in measured via a thigh-worn accelerometer. Secondary outcome people with chronic obstructive pulmonary disease: a randomised measures included time spent in prolonged sedentary behaviour (ie, controlled trial. Thorax. 2022;77:231–238. bouts of 30 minutes) and usual bout duration. Results: Sixty-five participants completed the study (31 in the experimental group Question: In people with chronic obstructive pulmonary disease, is a and 34 in the control group). At 6 weeks, no between-group differ- 6-week behaviour change intervention superior to a sham interven- ences in time spent in sedentary behaviour (MD 5 minutes/day, 95% tion at reducing time spent in sedentary behaviour? Design: Rando- CI 238 to 48), time spent in prolonged sedentary behaviour (MD 22 mised controlled trial with concealed allocation and blinded assessor. minutes/day, 95% CI 250 to 46) or usual bout duration (MD 21.2 Setting: Five hospitals in Sydney. Participants: Adults with chronic minutes/day, 95% CI 26.9 to 4.5) were observed. Conclusion: In obstructive pulmonary disease who were clinically stable and on the people with chronic obstructive pulmonary disease, a 6-week waitlist for pulmonary rehabilitation. Supervised exercise training in behaviour change intervention was not superior to a sham inter- the last 6 months and limited English were exclusion criteria. Ran- vention at reducing time spent in sedentary behaviour or at changing domisation of 70 participants allocated 34 to the experimental group the pattern in which sedentary behaviour is accumulated. and 36 to the control group. Interventions: The control group received a sham intervention, which included weekly phone calls for Provenance: Invited. Not peer reviewed. 6 weeks to enquire about and monitor health status. No instruction regarding physical activity or exercise was provided. The experi- Vinicius Cavalheri mental group received 6 weeks of weekly sessions with a physio- Curtin School of Allied Health, Curtin University, Australia therapist aimed at reducing sedentary behaviour via three target behaviours: replace sedentary behaviour with stepping, replace https://doi.org/10.1016/j.jphys.2022.05.004 sedentary behaviour with standing, and break up prolonged bouts of sedentary behaviour. Three of these sessions were face-to-face and the other three were delivered over the phone. Outcome measures: Commentary In chronic obstructive pulmonary disease, low levels of physical sitting for periods . 30 minutes) and evaluating measures specific to activity and prolonged sedentary time during daily life are well- this message (eg, usual bout duration during 3 pm and 6 pm) may be established. Although these adaptations serve to reduce dyspnoea more sensitive. Third, embedding this intervention within a pulmo- during daily life, they increase the risk of cardiovascular disease. To nary rehabilitation program, when people are most likely to have date, studies in this population that have attempted to increase mastery/vicarious experiences, may have increased their self-efficacy participation in physical activity have had little success.1 This study and success with reducing sedentary time. chose a potentially more feasible intervention message: reducing sedentary time. The methods had several features that reduced risk Provenance: Invited. Not peer reviewed. of bias and strengths were the use of the Behaviour Change Wheel and the Behaviour Change Technique Taxonomy.2,3 Nevertheless, Kylie Hill the analyses were unable to demonstrate any between-group differ- Curtin School of Allied Health, Curtin University, Perth, Australia ences. There are several learnings from this study. First, people with chronic obstructive pulmonary disease frequently experience exac- https://doi.org/10.1016/j.jphys.2022.05.005 erbations; in this study, 52% and 32% of the intervention and control groups, respectively, reported worsening during the intervention References period. Interventions that aim to reduce sedentary time need to allow time for recovery and embed action and coping strategies to facilitate 1. Burge AT, et al. Cochrane Database Syst Rev. 2020;4:CD012626. re-engagement with the intervention message. Perhaps a 6-week 2. Michie S, et al. Implement Sci. 2011;6:42. intervention period was insufficient. Second, total sedentary time is 3. Michie S, et al. Ann Behav Med. 2013;46:81–95. likely to be blunt to small changes. Setting more specific parameters for the intervention message (eg, between 3 pm and 6 pm, avoid https://doi.org/10.1016/j.jphys.2022.05.005 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 68 (2022) 206 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 people with chronic obstructive pulmonary disease, a 6-week behaviour change intervention was not superior to a sham intervention at reducing time spent in sedentary behaviour Synopsis Summary of: Cheng SWM, Alison J, Stamatakis E, et al. Six-week The primary outcome was time spent in sedentary behaviour behaviour change intervention to reduce sedentary behaviour in measured via a thigh-worn accelerometer. Secondary outcome people with chronic obstructive pulmonary disease: a randomised measures included time spent in prolonged sedentary behaviour (ie, controlled trial. Thorax. 2022;77:231–238. bouts of 30 minutes) and usual bout duration. Results: Sixty-five participants completed the study (31 in the experimental group Question: In people with chronic obstructive pulmonary disease, is a and 34 in the control group). At 6 weeks, no between-group differ- 6-week behaviour change intervention superior to a sham interven- ences in time spent in sedentary behaviour (MD 5 minutes/day, 95% tion at reducing time spent in sedentary behaviour? Design: Rando- CI 238 to 48), time spent in prolonged sedentary behaviour (MD 22 mised controlled trial with concealed allocation and blinded assessor. minutes/day, 95% CI 250 to 46) or usual bout duration (MD 21.2 Setting: Five hospitals in Sydney. Participants: Adults with chronic minutes/day, 95% CI 26.9 to 4.5) were observed. Conclusion: In obstructive pulmonary disease who were clinically stable and on the people with chronic obstructive pulmonary disease, a 6-week waitlist for pulmonary rehabilitation. Supervised exercise training in behaviour change intervention was not superior to a sham inter- the last 6 months and limited English were exclusion criteria. Ran- vention at reducing time spent in sedentary behaviour or at changing domisation of 70 participants allocated 34 to the experimental group the pattern in which sedentary behaviour is accumulated. and 36 to the control group. Interventions: The control group received a sham intervention, which included weekly phone calls for Provenance: Invited. Not peer reviewed. 6 weeks to enquire about and monitor health status. No instruction regarding physical activity or exercise was provided. The experi- Vinicius Cavalheri mental group received 6 weeks of weekly sessions with a physio- Curtin School of Allied Health, Curtin University, Australia therapist aimed at reducing sedentary behaviour via three target behaviours: replace sedentary behaviour with stepping, replace https://doi.org/10.1016/j.jphys.2022.05.004 sedentary behaviour with standing, and break up prolonged bouts of sedentary behaviour. Three of these sessions were face-to-face and the other three were delivered over the phone. Outcome measures: Commentary In chronic obstructive pulmonary disease, low levels of physical sitting for periods . 30 minutes) and evaluating measures specific to activity and prolonged sedentary time during daily life are well- this message (eg, usual bout duration during 3 pm and 6 pm) may be established. Although these adaptations serve to reduce dyspnoea more sensitive. Third, embedding this intervention within a pulmo- during daily life, they increase the risk of cardiovascular disease. To nary rehabilitation program, when people are most likely to have date, studies in this population that have attempted to increase mastery/vicarious experiences, may have increased their self-efficacy participation in physical activity have had little success.1 This study and success with reducing sedentary time. chose a potentially more feasible intervention message: reducing sedentary time. The methods had several features that reduced risk Provenance: Invited. Not peer reviewed. of bias and strengths were the use of the Behaviour Change Wheel and the Behaviour Change Technique Taxonomy.2,3 Nevertheless, Kylie Hill the analyses were unable to demonstrate any between-group differ- Curtin School of Allied Health, Curtin University, Perth, Australia ences. There are several learnings from this study. First, people with chronic obstructive pulmonary disease frequently experience exac- https://doi.org/10.1016/j.jphys.2022.05.005 erbations; in this study, 52% and 32% of the intervention and control groups, respectively, reported worsening during the intervention References period. Interventions that aim to reduce sedentary time need to allow time for recovery and embed action and coping strategies to facilitate 1. Burge AT, et al. Cochrane Database Syst Rev. 2020;4:CD012626. re-engagement with the intervention message. Perhaps a 6-week 2. Michie S, et al. Implement Sci. 2011;6:42. intervention period was insufficient. Second, total sedentary time is 3. Michie S, et al. Ann Behav Med. 2013;46:81–95. likely to be blunt to small changes. Setting more specific parameters for the intervention message (eg, between 3 pm and 6 pm, avoid https://doi.org/10.1016/j.jphys.2022.05.005 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 68 (2022) 191–196 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 Group-based pelvic floor muscle training is a more cost-effective approach to treat urinary incontinence in older women: economic analysis of a randomised trial Licia P Cacciari a, Christian RC Kouakou b, Thomas G Poder c,d, Luke Vale e, Mélanie Morin f, Marie-Hélène Mayrand g, Michel Tousignant f, Chantale Dumoulin h aSchool of Rehabilitation, Faculty of Medicine, Université de Montréal and Research Center of the Institut universitaire de gériatrie de Montréal, Montréal, Canada; bDépartement d’Économique, Université de Sherbrooke, Sherbrooke, Canada; c École de santé publique–Département de gestion, d’évaluation et de politique de santé, Université de Montréal, Montréal, Canada; dCentre de recherche de l’Institut universitaire en santé mentale de Montréal, CIUSSS de l’Est de l’île de Montréal, Montréal, Canada; eInstitute of Health & Society, Newcastle University, Newcastle upon Tyne, UK; fSchool of Rehabilitation, Faculty of Medicine and Health Sciences, Université de Sherbrooke and Research Center of the Centre hospitalier de l’Université de Sherbrooke, Sherbrooke, Canada; gDepartments of Obstetrics and Gynecology and Social and Preventive Medicine, Université de Montréal and Research Center of the Centre hospitalier de l’Université de Montréal, Montréal, Canada; hSchool of Rehabilitation, Faculty of Medicine, Université de Montréal and Research Center of the Institut universitaire de gériatrie de Montréal, Montréal, Canada KEY WORDS ABSTRACT Conservative treatment Question(s): How cost-effective is group-based pelvic floor muscle training (PFMT) for treating urinary Aged incontinence in older women? Design: Economic evaluation conducted alongside an assessor-blinded, Exercise therapy multicentre randomised non-inferiority trial with 1-year follow-up. Participants: A total of 362 women Urinary incontinence aged ! 60 years with stress or mixed urinary incontinence. Intervention: Twelve weekly 1-hour PFMT Economics sessions delivered individually (one physiotherapist per woman) or in groups (one physiotherapist per eight women). Outcome measures: Urinary incontinence-related costs per woman were estimated from a participant and provider perspective over 1 year in Canadian dollars, 2019. Effectiveness was based on reduction in leakage episodes and quality-adjusted life years. Incremental cost-effectiveness ratios and net monetary benefit were calculated for each of the effectiveness outcomes and perspectives. Results: Both group-based and individual PFMT were effective in reducing leakage and promoting gains in quality-adjusted life years. Furthermore, group-based PFMT was ! 60% less costly than individual treatment, regardless of the perspective studied: –$914 (95% CI –970 to –863) from the participant’s perspective and –$509 (95% CI –523 to –496) from the provider’s perspective. Differences in effects between study arms were minor and negligible. Adherence to treatment was high, with low loss to follow-up and no between-group differences. Conclusion: Compared with standard individual PFMT, group-based PFMT was less costly and as clinically effective and widely accepted. These results indicate that patients and healthcare decision- makers should consider group-based PFMT to be a cost-effective first-line treatment option for urinary incontinence. Trial registration: ClinicalTrials.gov NCT02039830. [Cacciari LP, Kouakou CRC, Poder TG, Vale L, Morin M, Mayrand M-H, Tousignant M, Dumoulin C (2022) Group-based pelvic floor muscle training is a more cost-effective approach to treat urinary incontinence in older women: economic analysis of a randomised trial. Journal of Physiotherapy 68:191–196] © 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/). Introduction Canada (2014),3 US$19.5 billion in the USA (2000)9 or £818 million in the UK (2004),10 which in most cases are covered out of pocket Urinary incontinence (UI) is an important and under-recognised problem affecting the lives of many adults.1,2 More frequent in and not reimbursed by third-party payers. Although coverage for women and increasing with age, UI is one of the most prevalent health concerns in older women.2,3 Up to 55% of older women suffer treatment varies across countries, the combined high prevalence from UI and 20 to 25% of them regularly experience severe symptoms (more than 10 episodes/week).2 UI is not only a frequent and unde- and hidden nature of incontinence leads to a large direct economic niable social problem that engenders embarrassment, negative self- burden on individuals, most often older women.11 perception and social isolation,4–7 it also leads to considerable per- sonal and societal expenditures.3,8 Clinical practice guidelines recommend supervised pelvic floor muscle training (PFMT) as the first-line treatment for women with Combined direct and indirect annual UI-related costs for both the most common UI subtypes, stress or mixed UI (Level A evi- individuals and society were estimated to exceed CA$5 billion in dence).12,13 Most often provided in individual physiotherapy sessions, PFMT reduces the number of leakage episodes and quantity of leakage, while improving UI-related quality of life.14 Despite this recommendation, PFMT is currently not consistently offered. Services https://doi.org/10.1016/j.jphys.2022.06.001 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/).
192 Cacciari et al: Cost-effectiveness of group-based PFMT are scattered and inconsistent; they are frequently not covered by PFMT sessions delivered once per week, either individually (one-on- public healthcare systems, which leads to under-treatment one) or in groups of six to eight women. The training protocol for worldwide.15,16 both interventions comprised the same standardised educational and exercise components, including home pelvic floor muscle exercises Group-based PFMT is proposed as a more affordable intervention assigned for 1 year. Only those who attended 10 or more of the 12 option for women with UI.17–19 However, economic evaluations of sessions were included in the analysis. group-based PFMT programs are scarce, and evidence is lacking on its cost-effectiveness compared with standard individual PFMT. In addition to the standard protocol, the individual arm used electromyography biofeedback as per usual practice. Participants in Therefore, the study question for this economic evaluation the group-based arm were offered up to three short one-on-one conducted alongside a randomised clinical trial20 was: assessment sessions with a physiotherapist to confirm correct PFM contractions, if necessary. How cost-effective is group-based PFMT for treating urinary in- continence in older women? Outcome measures Method Cost outcomes UI-related annual costs per woman were estimated from partici- Design pant and provider perspectives in each trial arm and reported in 2019 This was a planned secondary analysis of the Group Rehabilitation Canadian dollars. Therefore, all dollar values reported in this manu- Or IndividUal Physiotherapy (GROUP) trial, which was an assessor- script are in Canadian dollars, unless specifically stated otherwise. For blinded, multicentre, randomised non-inferiority trial comparing each perspective, total cost included the estimated intervention cost standard individual (one-on-one) with group-based PFMT (one plus UI-related additional costs to participants or providers collected physiotherapist for eight patients) for the treatment of stress and over the course of the study. Details of the cost are presented in mixed UI in older women.19,20 Figure 1. Participants, therapists, centres Participant perspective: Intervention cost was based on values customarily applied in private physiotherapy clinics in the study area. Eligible participants were women aged ! 60 years, who reported For the individual arm in particular, each participant’s intervention at least three UI episodes per week during the preceding 3 months.21 cost was based on the average cost per initial individual assessment Stress/mixed UI was confirmed with the validated Questionnaire for ($100) and PFMT sessions ($100 each) multiplied by their attendance. Incontinence Diagnosis.22 Women were excluded if they had reduced Group-based intervention cost was considered as a treatment pack- mobility or comorbidities that would interfere with the study. Addi- age ($260), including the initial individual assessment and the twelve tional details of the trial’s eligibility criteria are provided in the trial treatment sessions (independent of their attendance to each session). protocol.20 Additional costs from both arms were derived from participants’ re- ports on the Dowell Bryant Incontinence Cost Index, which were Women who met the eligibility criteria and consented to partici- acquired at baseline, after treatment and at the 1-year follow-up. pate were randomly assigned (1:1) to either group-based PFMT or These costs included UI-related expenses from incontinence care individual PFMT, with random block sizes (four to six), stratified by products and other UI treatments undertaken during the study centre (Montreal and Sherbrooke) and by UI type (stress and mixed). period. Annual cost estimates per participant were based on reported Participants were each assigned a random computer-generated event frequency and mean unit costs applied to the study area. When sequence, which was provided by a statistician external to data sources from prior to 2019 were used to derive costs, inflation was collection (allocation concealed). Outcome assessors remained blin- calculated for an equivalent price in 2019. ded to the participants’ intervention allocation. Intervention Provider perspective: Provider intervention cost was estimated from the current mean labour hour of a specialised physiotherapist (mar- Participants from both study arms attended an individual session ket value),23 in addition to room cost (including staff time, consum- with a physiotherapist to learn how to effectively contract their pelvic ables and all equipment necessary for PFMT clinical practice, and floor muscles using vaginal palpation. This was followed by 12 1-hour specific to each trial arm). For group-based PFMT, intervention cost Figure 1. Outline of cost calculation, including primary clinical costs and extra costs from the participant and provider perspectives. Values are presented in Canadian dollars, based on the current (2019) practice in the study area. Room cost includes staff time, consumables and all reusable equipment, comprising biofeedback-related expenses for the individual sessions and extra room space and reusable equipment for the group sessions. pps = number of participants per session.
Research 193 Table 1 based on previous assumptions that generic health-state measures Baseline demographic characteristics and effectiveness outcomes. usually used in economic evaluations (such as the EQ-5D-5L) would not be sensitive enough to capture clinically important differences Characteristic Group Individual related to UI on quality of life.25,26 Furthermore, the ICIQ-LUTSqol is a (n = 154) (n = 165) recommended questionnaire with which to assess the impact of UI on quality of life (Grade A1).27 For the cost-effectiveness analysis, full Age (y), mean (SD) 68.4 (5.9) 68.1 (5.9) scores were converted into a utility index (from 0 [worst imaginable BMI (kg/m2), mean (SD) 27.1 (4.5) 27.1 (4.7) health state] to 1 [best imaginable health state]) using a published Parity, median (IQR) 2 (1 to 3) 2 (1 to 2) algorithm.25 Type of UI, n (%) 31 (20) 24 (15) Cost-effectiveness outcomes Stress UI 123 (80) 141 (85) Incremental cost-effectiveness ratios were calculated for each of Mixed UI 8.8 (8.7) 10.3 (10.3) Duration of symptoms (y), (n = 151) (n = 162) the effectiveness outcomes and the two perspectives (ie, participant mean (SD) and provider). The incremental cost-effectiveness ratio was defined Leakage (episodes/d), 1.43 (0.86 to 2.00) 1.57 (0.86 to 2.79) by the difference in cost, divided by the difference in effect for the 0.43 (0.14 to 1.07) 0.43 (0.14 to 1.00) group-based versus individual PFMT. Incremental net benefits of each median (IQR) a intervention were calculated using the usual willingness-to-pay Baseline (n = 153) 32 (28 to 38) threshold of $50,000 per quality-adjusted life years (QALY)28: Follow-up (n = 163) 32 (27 to 40) ð½HRQ oLgroup 2 HRQ oLindividual à $50;000 2 ½Costgroup 2 CostindividualÞ HRQoL (ICIQ-LUTSqol), (n = 154) 23 (21 to 27) median (IQR) b (n = 163) Data analysis 23 (21 to 28) Baseline (n = 151) Analyses were focused on outcomes at 1 year following guideline recommendations,12 and only women who completed the 1-year Follow-up assessment were included. Baseline characteristics of the study arms were summarised with descriptive statistics. Spearman’s cor- BMI = body-mass index, HRQoL = health-related quality of life, n = number used in the relation was carried out between cost and effectiveness in order to choose an estimation method. If cost and effectiveness were corre- analysis, UI = urinary incontinence. lated, then a seemingly unrelated regression was used. If there was no a As reported on the 7-day bladder diary. evidence of a correlation, ordinary least squares was used. Cost and b As reported on International Consultation on Incontinence Modular Questionnaire effectiveness data were resampled 1,000 times with non-parametric and univariate parametric bootstrapping to estimate 95% CIs around - Lower Urinary Tract Symptoms Quality of Life (ICIQ-LUTSqol); scores ranging 19 to 76, the mean difference in costs and effects between study arms. greater values indicate an increased impact on quality of life (minimum clinically Results important difference, 3.71 points; incremental improvement, 6.63 points).29 A total of 362 women were randomised to either individual (n = was based on the number of participants attending per session. Oc- 184) or group-based PFMT (n = 178). Overall, 337 of 362 (93%) par- casional short one-on-one vaginal palpation assessments were esti- ticipants completed the intervention and 319 of 362 (88%) completed mated to cost one-third of the hour-long individual assessments the follow-up assessment and were included in this study. Table 1 based on their mean duration of 20 minutes. For both arms, addi- provides clinical and demographic characteristics of the partici- tional costs included follow-up calls to the participants undertaken at pants. No important baseline imbalances were found between study 3 and 6 months after intervention. These calls were estimated to cost arms. Adherence to treatment sessions was high (98% for one-sixth of the physiotherapist labour hour, based on their mean group-based and 95% for individual PFMT). Loss to follow-up until the duration of 10 minutes. Repeated attempts to contact participants end of the intervention was low, similar between arms and unrelated were also considered. to treatment allocation: 12 of 178 (7%) of the group-based and 13 of 184 (7%) of the individual PFMT participants. Details are presented in Effectiveness outcomes Appendix 1 on the eAddenda. Effectiveness of the two PFMT interventions were acquired at 1 Estimates for the average UI-related costs per woman in each year and included leakage reduction and health-related quality of life study arm and perspective are presented in Table 2. The main cost (HRQoL). Leakage reduction: Leakage reduction data were derived from par- ticipants’ self-reports on a 7-day bladder diary. The number of par- ticipants with a minimum clinically important difference in UI episodes (! 50%)24 at the 1-year follow-up relative to the pre- treatment baseline was considered for analysis. Health-related quality of life: HRQoL was derived from a condition- specific measure, the International Consultation on Incontinence Modular Questionnaire - Lower Urinary Tract Symptoms Quality of Life (ICIQ-LUTSqol) assessed at the 1-year follow-up. This choice was Table 2 Estimated average urinary incontinence-related annual cost per woman from participant and provider perspectives. Cost categories Participant perspective mean (SD) Mean difference (95% CI)a Provider perspective mean (SD) Mean difference (95% CI)a Group (n = 154) Individual (n = 165) Group (n = 154) Individual (n = 165) Initial assessment (A) 100 100 – 64 64 – 261 (6) 1,173 (68) –911 (–921 to –900) 238 (72) 747 (43) –509 (–523 to –496) Treatment sessions (B) 361 (6) 1,273 (68) –911 (–921 to –900) 302 (72) 811 (43) –509 (–523 to –496) Primary treatment cost (C) (A1B) 172 (230) 177 (240) –5 (–60 to 43) – – – Incontinence productsb (D) – – – 19 19 – Follow-up contact calls (E) – – – Other treatmentsc (F) 6 (34) 3 (26) 2 (–4 to 9) 19 19 – Additional costs (G)(D/E 1 F) 177 (234) 180 (241) –3 (–57 to 48) 321 (72) 830 (43) –509 (–523 to –496) Total cost (C1G) 539 (233) 1,453 (246) –914 (–970 to –863) All costs are reported in 2019 Canadian dollars. a Bootstrapping was used to estimate the 95% CI around the mean difference. b Annual costs were estimated using self-reported data acquired at baseline (T0), post treatment (T1) and 1 year (T2) as T0*31T1*31T2*6; costs were no different between study arms at either time point (T0: $24 [SD 33] group versus $25 [SD 32] individual, mean difference [MD][95% CI]: –0.3 [–7.1 to 6.4]; T1: $10 [SD 16] group versus $11 [SD 19] individual, MD [95% CI]: –1.2 [–5.0 to 2.6]; T2: $12 [SD 21] group versus $12 [SD 21] individual, MD [95% CI]: –0.3 [–4.9 to 4.3]). c Other treatment costs were based on the participants’ reported frequency and their mean current unitary cost (specific to this population in the study area).
194 Cacciari et al: Cost-effectiveness of group-based PFMT Table 3 Cost-effectiveness outcomes. Outcome Intervention Total cost Effect D cost D effect ICER (95% CI) mean (SD) mean (SD) mean (95% CI) mean (95% CI) NPB UPB Participant perspective Reduction in UI Group (n = 153) 536 (232) 74 (44) –917 2 (1.8 to 2.4) – – episodes (%) Individual 1,453 (246) 72 (45) (–918 to –915) (n = 165) –0.00006 1,958,228 1,571,525 QALY Group 537 (230) 0.9815 (0.018) –916 (–0.00080 to –0.00056) (–2,381,108 to 6,297,564) (–630,539 to 3,773,589) (n = 151) (–918 to –915) Individual 1,453 (248) 0.9820 (0.018) (n = 163) Provider perspective Reduction in UI Group 321 (72) 74 (44) –509 2 (1.8 to 2.4) – – episodes (%) (n = 153) 830 (43) 72 (45) (–507 to 512) Individual 322 (72) 855,423 QALY (n = 165) 830 (43) 0.9815 (0.018) –508 –0.0006 1,098,535 (–239,281 to 1,950,128) Group 0.9820 (0.018) (–510 to –507) (n = 151) (–0.00080 to –0.00056) (–1,339,830 to 3,536,899) Individual (n = 163) All costs are reported in 2019 Canadian dollars. D = incremental (group – individual) cost and effect, ICER = incremental cost-effectiveness ratio (D cost/ D effect), NPB = nonparametric bootstrapping (1,000 iterations), QALY = quality-adjusted life years, UI = urinary incontinence, UPB = univariate parametric bootstrapping (1,000 iterations). driver from both perspectives was the intervention cost, which was at episodes. On the other hand, when average QALY was considered, least 60% less costly in the group-based PFMT. Regardless of the individual PFMT was slightly more effective. However the incre- perspective, additional costs were similar between study arms. mental cost-effectiveness ratios were in excess of $800,000 per additional QALY gained for individual compared with group-based No important differences were found in effectiveness between the PFMT, which was much higher than the pre-established $50,000 study arms (Table 1). At the 1-year follow-up, . 70% of participants threshold for willingness to pay.28 from both group-based and individual PFMT demonstrated minimum clinically important differences (! 50% reduction) in the number of UI Figure 2 shows the probability of cost-effectiveness for various episodes. HRQoL also reached incremental clinically important dif- cost thresholds per unit of effectiveness gained, while distribution of ferences in both study arms: ICIQ-LUTSqol . 6.63 points;29 mean costs and effects are illustrated in Figure 3. Over the range of society’s difference 9.0 (95% CI 7.6 to 10.3) and 8.9 (95% CI 7.7 to 10.2) for the willingness to pay, group-based PFMT had no more than 87% and 94% group-based and individual PFMT, respectively. probability of being considered cost-effective for a 50% reduction in UI episodes, or 38% and 40% for a QALY gain from participant and The estimated average UI-related annual cost per woman from provider perspectives, respectively. Table 4 shows the net monetary participant and provider perspectives are shown in Table 2, while benefit of each intervention based on a society willingness to pay cost-effectiveness outcomes are shown in Table 3. Total cost per threshold of $50,000 per QALY gain, again favouring group-based woman was on average lower for group-based PFMT, with an average PFMT in both perspectives. gap of $914 from participant and $509 from provider perspectives. Bootstrapped incremental cost-effectiveness ratios showed conver- Discussion gent results for each perspective and effect. From the participant’s perspective, a variable correlation between participant costs and ef- Group-based PFMT is a cost-effective approach for treating UI: it fects was found, and therefore a seemingly unrelated regression incurred low costs, sustained reduction in UI episodes over 1 year and model estimation was used, while an ordinary least squares was used improved HRQoL. When compared with the standard individual for the provider’s perspective. When the effectiveness outcome was treatment, group-based PFMT provided at least 60% in cost savings, the reduction in UI episodes, the incremental costs and effects indi- regardless of who paid for the treatment, while differences in effects cated that individual PFMT was inferior, as the group-based arm had lower costs and slightly more people with a 50% reduction in UI Figure 2. Cost-effectiveness acceptability curve of maximum willingness to pay for group-based versus individual pelvic floor muscle training.
Research 195 Figure 3. Joint density of the incremental costs and benefits of each of the 1,000 simulated cost-effectiveness ratios. were negligible.25 Adherence to treatment sessions was high and loss participants who completed the study (88%) and the lack of baseline to follow-up low, indicating acceptability of both interventions. Of differences between completers and non-completers support the assumption that data were missing at random, which is unlikely to interest, participants in the group-based treatment reported benefits cause bias. Although the QALY gains (0.021) that were obtained were from peer support, and perceived the group classes as a safe space to relatively small, they reached incremental clinical differences and share their experiences.30,31 were comparable or higher than those found in other conservative management studies treating adult women with UI (0.01 to From the provider perspective, with budget constraints in 0.02).35,36 Finally, the results were consistent between both partici- pant and provider perspectives, and also considering both UI leakage healthcare worldwide, innovative ways of managing and stretching reduction and HRQoL. budgets are greatly needed.26 Even when available, PFMT currently involves long wait times, leading to under-treatment,3,15,16 or the Limitations of this study included disregarding costs related to frequent use of surgery as first-line therapy, despite serious adverse accessing care (such as laundry costs or time and travel costs). effects.3 Group-based PFMT could be a cost-effective strategy for Although all relevant costs should be accounted for in a complete reducing costs and optimising rehabilitation services. For instance, analysis, this exclusion should not impact the study results, as both study arms followed the same training program delivered in the the cost for treating three women individually would be equivalent to same location. In addition, the 1-year follow-up assessment revealed eight women treated in a group-based PFMT approach. This is of excellent adherence to treatment and improvement in symptoms. However, a longer-term follow-up should be considered in future particular importance when considering the combination of an aging research. Further, although generic quality of life measures (such as world population and the tendency for UI to become more prevalent the EQ-5D) are the recommended instrument with which to calcu- with age.32 Addressing UI earlier and more comprehensively amongst late QALY, they have often been considered not sensitive enough to seniors could go a long way to reduce related downstream health assess UI-specific outcomes.26 Here, the choice for a recommended outcomes.3,7 UI-specific quality of life measure (ICIQ-LUTSqol)27 proved to be ac- curate to capture small but clinically important differences related to From the participant perspective, the burden of UI has been PFMT. Finally, since effect variables (ie, UI episodes, QALY) were correlated with cost variables, it was decided to run the two re- compared with many other chronic diseases in women; however, gressions simultaneously using the seemingly unrelated regression unlike most chronic diseases, UI-related healthcare costs are often method; this enabled error terms correlation to be taken into ac- fully covered out-of-pocket.3,33 The average cost gap for choosing count and bias to be avoided. group-based over the standard individual PFMT ($914) may seem In conclusion, compared with standard individual PFMT, group- small, but represents as much as 3% of the average annual income for based PFMT was less costly and as clinically effective and widely an older woman living in the community (Canada, 2018).34 With the accepted. The implementation of group-based interventions to treat UI could optimise the use of common resources, help unburden benefits of peer support and lower costs, group-based PFMT could be a way of promoting access to care and encouraging women to pursue and adhere to treatment. It is believed that this is the first cost-effectiveness analysis conducted alongside a robust clinical trial involving supervised group-based versus standard individual PFMT over 1 year for the treatment of UI in older women. Furthermore, the high rate of Table 4 Net monetary benefit. Intervention Total cost QALYs Net monetary benefit Incremental net monetary benefit mean (SD) mean (SD) (95% CI) (95% CI) NPB UPB Participant perspective Group (n = 151) $537 (230) 0.9815 (0.018) $48,537 (48,376 to 48,698) $828 (821 to 834) $887 (880 to 894) Individual (n = 163) $1,453 (248) 0.9820 (0.018) $47,649 (47,498 to 47,799) Provider perspective Group (n = 151) $322 (72) 0.9815 (0.018) $48,752 (48,606 to 48,899) $458 (452 to 464) $451 (444 to 457) Individual (n = 163) $830 (43) 0.9820 (0.018) $48,272 (48,136 to 48,408) All costs are reported in 2019 Canadian dollars. Willingness to pay established at $50,000 per QALY. Net monetary benefit = HRQoL * $50,000 – cost; Incremental net benefit = (HRQoLgroup PFMT – HRQoLindividual PFMT) * $50,000 – (costgroup PFMT – costindividual PFMT). HRQoL = health-related quality of life, NPB = non-parametric bootstrapping (1,000 iterations), QALY = quality-adjusted life years, UPB = univariate parametric bootstrapping (1,000 iterations).
196 Cacciari et al: Cost-effectiveness of group-based PFMT primary care, and facilitate access to care for women seeking an incontinence: A cost-utility and value of information analysis. BMJ Open. effective and more affordable UI treatment. 2020;10:1–8. https://doi.org/10.1136/bmjopen-2019-035555 11. Subak LL, Marinilli Pinto A, Wing RR, Nakagawa S, Kusek JW, Herman WH, et al. What was already known on this topic: The combined high Decrease in urinary incontinence management costs in women enrolled in a prevalence and hidden nature of incontinence leads to a large clinical trial of weight loss to treat urinary incontinence. Obstet Gynecol. direct economic burden on individuals and high healthcare costs. 2012;120:277–283. https://doi.org/10.1097/AOG.0b013e31825dd268 Clinical practice guidelines recommend supervised pelvic floor 12. Dumoulin C, Adewuyi T, Booth J, Bradley C, Burgio K, Hagen S, et al. Adult con- muscle training, which is usually delivered individually but group- servative management. In: Abrams P, Cardozo L, Wagg A, Wein A, eds. Incontinence. based programs are effective. 6th ed. Bristol, UK: ICI-ICS International Continence Society; 2017:1443–1628. What this study adds: Compared with standard individual 13. Syan R, Brucker BM. Guideline of guidelines: urinary incontinence. BJU Int. PFMT, group-based PFMT was less costly and as clinically 2016;117:20–33. https://doi.org/10.1111/bju.13187 effective and widely accepted. 14. Dumoulin C, Cacciari LP, Hay-Smith EJC. Pelvic floor muscle training versus no treatment, or inactive control treatments, for urinary incontinence in women. Ethics approval: The research centre of the Institut universitaire Cochrane Database Syst Rev. 2018;10:CD005654. https://doi.org/10.1002/14651858. de gériatrie de Montréal and the research centre of the Centre hos- CD005654.pub4 pitalier universitaire de Sherbrooke, Canada, Ethics Committees 15. Milne JL, Moore KN. An exploratory study of continence care services worldwide. approved this study. All participants gave written informed consent Int J Nurs Stud. 2003;40:235–247. https://doi.org/10.1016/S0020-7489(02) before data collection began. 00082-2 16. Monz B, Hampel C, Porkess S, Wagg A, Pons ME, Samsioe G, et al. A description of Competing interests: Nil. health care provision and access to treatment for women with urinary inconti- eAddenda: Appendix 1 can be found online at https://doi.org/10. nence in Europe — A five-country comparison. Maturitas. 2005;52(SUPPL. 2):3–12. 1016/j.jphys.2022.06.001 https://doi.org/10.1016/j.maturitas.2005.09.007 Source(s) of support: This study was supported by the Canadian 17. Cook T. Group treatment of female urinary incontinence. Physiotherapy. Institutes of Health Research (grant MSH-258993). Dr Cacciari was 2001;87:226–234. https://doi.org/10.1016/S0031-9406(05)60783-3 supported by the Centre de recherche de l’Institut universitaire de 18. Lamb SE, Pepper J, Lall R, Jørstad-Stein EC, Clark MD, Hill L, et al. Group treatments gériatrie de Montréal and the Fonds de la recherche du Québec – for sensitive health care problems: a randomised controlled trial of group versus Santé. Dr Dumoulin received a salary award from the Canadian individual physiotherapy sessions for female urinary incontinence. BMC Womens Research Chair Tier II program. Drs Morin and Mayrand received Health. 2009;9:26. https://doi.org/10.1186/1472-6874-9-26 salary awards from the Fonds de la Recherche du Québec–Santé. The 19. Dumoulin C, Morin M, Danieli C, Cacciari L, Mayrand MH, Tousignant M, et al. laboratory infrastructures were funded by the Canada Foundation for Group-Based vs Individual Pelvic Floor Muscle Training to Treat Urinary Inconti- Innovation. The funding source had no role in the design or conduct nence in Older Women. JAMA Intern Med. 2020;180:1284. https://doi.org/10.1001/ of the study; the collection, management, analysis or interpretation of jamainternmed.2020.2993 the data; the preparation, review or approval of the manuscript; or 20. Dumoulin C, Morin M, Mayrand M-H, Tousignant M, Abrahamowicz M. Group the decision to submit the paper for publication. physiotherapy compared to individual physiotherapy to treat urinary incontinence Acknowledgements: We thank data assessors and analysts, in aging women: study protocol for a randomized controlled trial. Trials. physiotherapists involved in treatments and assessments, and study 2017;18:544. https://doi.org/10.1186/s13063-017-2261-4 participants for their support and dedication to this research project. 21. Burgio KL, Goode PS, Locher JL, Umlauf MG, Roth DL, Richter HE, et al. Behavioral Provenance: Not invited. Peer reviewed. training with and without biofeedback in the treatment of urge incontinence in Correspondence: Chantale Dumoulin, School of Rehabilitation, older women: a randomized controlled trial. JAMA. 2002;288:2293–2299. https:// Faculty of Medicine, Université de Montréal and Research Center of doi.org/10.1001/jama.288.18.2293 the Institut universitaire de gériatrie de Montréal, Canada. Email: 22. Bradley CS, Rahn DD, Nygaard IE, Barber MD, Nager CW, Kenton KS, et al. The [email protected] questionnaire for urinary incontinence diagnosis (QUID): Validity and respon- siveness to change in women undergoing non–surgical therapies for treatment of References stress predominant urinary incontinence. Neurourol Urodyn. 2010;29(June 2009): 726–733. https://doi.org/10.1002/nau.20818 1. DeLancey JOL. The hidden epidemic of pelvic floor dysfunction: achievable goals 23. Government of Canada. Job Bank - wage report. https://www.jobbank.gc.ca/ for improved prevention and treatment. Am J Obstet Gynecol. 2005;192:1488–1495. wagereport/occupation/18207 https://doi.org/10.1016/j.ajog.2005.02.028 24. Yalcin I, Peng G, Viktrup L, Bump RC. Reductions in stress urinary incontinence episodes: what is clinically important for women? Neurourol Urodyn. 2. Milsom I, Altman D, Cartwright R, Lapitan MC, Nelson R, Sillén U, et al. 2010;29:344–347. https://doi.org/10.1002/nau.20744 Epidemiology of urinary incontinence (UI) and other lower urinary tract 25. Brazier J, Czoski-Murray C, Roberts J, Brown M, Symonds T, Kelleher CJ. Estimation symptoms (LUTS), pelvic organ prolapse (POP) and anal Incontinence (AI). In: of a Preference-Based Index from a Condition-Specific Measure: The King’s Health Abrams P, Cardozo L, Wagg A, Wein A, eds. Incontinence. 6th ed. Bristol, UK: ICI- Questionnaire. Med Decis Mak. 2008;28:113–126. https://doi.org/10.1177/ ICS International Continence Society; 2017:1–141. 0272989X07301820 26. Wagner TH, Moore KH, Subak LL, Wachter S de, Dudding T. Economics of urinary & 3. Cameron Institute. Incontinence: The Canadian Perspective. The Canadian Conti- faecal incontinence, and prolapse. In: Abrams P, Cardozo L, Wagg A, Wein A, eds. nence Foundation; 2014. Incontinence. 6th ed. Bristol, UK: ICI-ICS International Continence Society; 2017:2479–2511. 4. Johnson T, Kincade JE, Bernard SL, Busby-Whitehead J, Hertz-Picciotto I, DeFriese GH. 27. Kelleher CJ, Staskin DR, Cherian P, et al. Patient-Reported Outcome Assessment. In: The association of urinary incontinence with poor self-rated health. J Am Geriatr Soc. Incontinence. 5th edition. Paris: Health Publication Ltd; 2013:389–428. 1998;46:693–699. https://doi.org/10.1111/j.1532-5415.1998.tb03802.x 28. Poder TG. Challenges to make cost-effectiveness studies usable by decision makers. J Thorac Cardiovasc Surg. 2018;156:1931–1932. https://doi.org/10.1016/j.jtcvs.2018. 5. Sims J, Browning C, Lundgren-Lindquist B, Kendig H. Urinary incontinence in a 05.062 community sample of older adults: Prevalence and impact on quality of life. Disabil 29. Nyström E, Sjöström M, Stenlund H, Samuelsson E. ICIQ symptom and quality of Rehabil. 2011;33:1389–1398. https://doi.org/10.3109/09638288.2010.532284 life instruments measure clinically relevant improvements in women with stress urinary incontinence. Neurourol Urodyn. 2015;34:747–751. https://doi.org/10.1002/ 6. Coyne KS, Kvasz M, Ireland AM, Milsom I, Kopp ZS, Chapple CR. Urinary Inconti- nau.22657 nence and its Relationship to Mental Health and Health-Related Quality of Life in 30. Saint-Onge K, Fraser S, Southall K, Frechette-Chaine C, Morin M, Dumoulin C. Men and Women in Sweden, the United Kingdom, and the United States. Eur Urol. Beyond the training: The benefits of peer support and improved self-perceptions 2012;61:88–95. https://doi.org/10.1016/j.eururo.2011.07.049 experienced by women completing a 12 week PFM training program. Neurourol Urodyn. 2018;37:S77–S78. 7. Ramage-Morin PL, Gilmour H. Urinary incontinence and loneliness in Canadian se- 31. Fréchette-Chaîné É, Mercier J, Fraser S, Southall K, Morin M, Dumoulin C. Psychosocial niors. Health Rep. 2013;24:3–10. https://doi.org/10.1007/s12468-015-0007-4 factors influencing physiotherapeutic adherence to group-based or individualized pelvic floor rehabilitation: perceptions of older women with urinary incontinence. 8. Subak LL, Brubaker L, Chai TC, et al. High costs of urinary incontinence among Neurourol Urodyn. 2018;37:S273–S274. https://doi.org/10.1002/nau.23757 women electing surgery to treat stress incontinence. Obstet Gynecol. 32. WHO. Global Health and Aging. Natl institutes Heal; 2020. 2008;111:899–907. https://doi.org/10.1097/AOG.0b013e31816a1e12 33. Wilson L, Brown JS, Shind GP, Luca KO, Subak LL. Annual direct cost of urinary incontinence. Obstet Gynecol. 2001;98:398–406. https://doi.org/10.1016/S0029- 9. Hu T-W, Wagner TH, Bentkover JD, Leblanc K, Zhou SZ, Hunt T. Costs of urinary 7844(01)01464-8 incontinence and overactive bladder in the United States: a comparative study. 34. Canada Statistics. Table 11-10-0239-01 Income of individuals by age group, sex and Urology. 2004;63:461–465. https://doi.org/10.1016/j.urology.2003.10.037 income source, Canada, provinces and selected census metropolitan areas. https:// doi.org/10.25318/1110023901-eng 10. Javanbakht M, Moloney E, Brazzelli M, Wallace S, Ternent L, Omar MI, et al. Eco- 35. Albers-Heitner CP, Joore MA, Winkens RAG, Lagro-Janssen ALM, Severens JL, nomic evaluation of surgical treatments for women with stress urinary Berghmans LCM. Cost-effectiveness of involving nurse specialists for adult patients with urinary incontinence in primary care compared to care-as- usual: an economic evaluation alongside a pragmatic randomized controlled trial. Neurourol Urodyn. 2012;31:526–534. https://doi.org/10.1002/ nau.21204 36. Sjöström M, Lindholm L, Samuelsson E. Mobile App for Treatment of Stress Urinary Incontinence: A Cost-Effectiveness Analysis. J Med Internet Res. 2017;19:e154. https://doi.org/10.2196/jmir.7383
Journal of Physiotherapy 68 (2022) 174–181 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 Home-based exercises are as effective as equivalent doses of centre-based exercises for improving walking speed and balance after stroke: a systematic review Lucas R Nascimento, Rafaela JS Rocha, Augusto Boening, Gabriel P Ferreira, Mikaella C Perovano Center of Health Sciences, Discipline of Physiotherapy, Universidade Federal do Espírito Santo, Brazil KEY WORDS ABSTRACT Stroke Questions: In people who have had a stroke, how comparable are the effects of home-based exercises with Mobility those of equivalent centre-based exercises for improving walking speed, balance, mobility and participation? Gait Is the comparability of the effects of these two types of exercise maintained beyond the intervention period? Balance Design: Systematic review of randomised controlled trials. Search strategy: Searches were conducted on Rehabilitation MEDLINE, AMED, EMBASE, Cochrane, PsycINFO and PEDro databases, without date or language restrictions. Participants: Participants in the reviewed studies were ambulatory adults at any time after stroke. Interventions: The experimental intervention consisted of home-based exercises, which was compared with equivalent doses of centre-based exercises. Outcome measures: Walking speed, balance, mobility and participation. Data analysis: The quality of included trials was assessed using the PEDro scores. Outcome data were extracted from the eligible trials and combined in random-effects meta-analyses. The quality of evidence was determined according to the Grading of Recommendations Assessment, Development and Evaluation (GRADE) system. Results: Nine trials involving 609 participants were included. Random-effects meta-analyses provided high-quality evidence that home-based and centre-based exercises provide similar effects on walking speed (MD –0.03 m/s, 95% CI –0.07 to 0.02) and balance (MD 0 points, 95% CI –1 to 2). Results regarding mobility (SMD –0.4, 95% CI –1.3 to 0.4) and participation (MD –5 points, 95% CI –19 to 10) were imprecise. For most outcomes, the effects of home-based exercises and centre-based exercises remained similar beyond the intervention period. Conclusion: Effects of home-based prescribed exercises on walking speed, balance, mobility and participation are likely to be similar to improvements obtained by equivalent doses of centre-based exercises after stroke. Review registration: PROSPERO (CRD42021254642). [Nascimento LR, Rocha RJS, Boening A, Ferreira GP, Perovano MC (2022) Home-based exercises are as effective as equivalent doses of centre-based exercises for improving walking speed and balance after stroke: a systematic review. Journal of Physiotherapy 68:174–181] © 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/). Introduction these interventions are typically delivered in rehabilitation centres Stroke is one of the leading causes of disability worldwide.1 with direct supervision from physiotherapists. This may contribute to Approximately 60% of people who survive stroke cannot walk inde- pendently early after stroke2,3 and those who regain independence non-adherence to treatment due to lack of motivation, time may still walk slowly or be unable to cover long distances.4,5 Balance impairments typically contribute to this reduced walking ability; for pressures, transport issues or inability to maintain exercises over a example, balance measured by the Berg Balance Scale (0 to 56 points, long period of time.10 Ultimately, non-adherence to prescribed in- where a higher score means better balance) is typically 23 points (SD 22) 14 days after stroke and 40 points (SD 18) 90 days after stroke.6 terventions reduces the amount of rehabilitation performed, whereas Those values are correlated with reduced walking and increased large amounts are needed to provide a beneficial effect.11 The delivery risk of falls.7 In addition, mobility limitations after stroke are not of semi-supervised practice (ie, the amount of supervision differs restricted to walking and may include difficulties in standing, turning and sitting.5 from the amount of prescribed intervention delivered) or remote Many interventions that improve balance, mobility and walking practice with patients in their homes has emerged as a feasible and can be implemented broadly.8,9 However, after hospital discharge, safe approach for increasing the amount of rehabilitation and adherence.12 Exercise that takes place in an informal and flexible setting, typically in patients’ homes, can be defined as home-based intervention.13 A home-based intervention can provide a feeling of familiarity with the location, which can be very comforting, as well as being https://doi.org/10.1016/j.jphys.2022.05.018 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/).
Research 175 more accessible, as it reduces the costs, environmental difficulties and retrieved papers was extracted and reviewed independently by two time needed to travel to a rehabilitation centre.14–16 In addition, reviewers (GPF and MCP) using predetermined criteria (Box 1). Both home-based intervention has been found to induce active participa- reviewers were blinded to titles, dates, authors, journals and results. tion of families and caregivers, and compensates for the lack of in- Disagreement or ambiguities were resolved by consensus after dis- teractions between patients and therapists.17 Both physiotherapists cussion with a third reviewer (LRN). and patients have reported very positive perceptions of the safety, effectiveness and technological manageability of home-based in- Assessment of characteristics of trials terventions using videoconferencing.18 In addition, home-based in- terventions have been effective for increasing completion rates in Quality clinical trials,19 and may be useful in periods that require social The methodological quality of the included trials was assessed by isolation, such as during the current Covid-19 pandemic. While high- income countries may easily convert centre-based rehabilitation into extracting the PEDro scores from the Physiotherapy Evidence Data- telerehabilitation, low-to-middle-income countries may have to base (www.pedro.org.au). The PEDro scale is an 11-item scale convert centre-based rehabilitation into in-person home-based designed for rating the methodological quality (internal validity and intervention. In low-to-middle-income countries, the use of tele- statistical information) of randomised trials. Each item, except for the rehabilitation is limited by: barriers related to the costs of consulta- Item 1, contributes 1 point to the total score (range 0 to 10 points). tions, low education and health literacy; and lack of appropriate Where a trial was not included on the database, it was scored by a infrastructure for access (ie, a smartphone or computer and minimum reviewer who had completed the PEDro scale training tutorial. broadband speed).20 Regardless of the mode of implementation, home-based interventions must be as effective or superior to centre- Participants based interventions in order to be worthwhile. Trials involving adults at any time following stroke were included. Previous systematic reviews21,22 have suggested that home-based The number of participants, age, time since stroke and baseline exercises improve overall motor function after stroke, but have walking speed were recorded to assess the similarity of the studies. provided no information about whether location impacts the effec- tiveness of prescribed exercise by comparing home-based and centre- Intervention based interventions. More recently, a Cochrane review23 based on Trials were included if the experimental intervention was home- two randomised trials of telerehabilitation (ie, remotely supervised home-based intervention) suggested that home-based and centre- based and consisted of structured and repetitive exercises targeting based interventions may have similar beneficial effects on activities the paretic lower limb for improving standing and/or walking. Home- of daily living (SMD 0, 95% CI –0.2 to 0.2) and mobility (MD 0, 95% CI based was defined as two-thirds of the exercise being conducted at –0.1 to 0.1). Given that that analysis was constrained to tele- home. To be included, a minimum dose of four sessions over 2 rehabilitation, a rigorous systematic review with meta-analysis of the weeks, prescribed by a physiotherapist or health professional with a current high-quality evidence is warranted. degree-level qualification in exercise prescription was required.25 Trials were excluded if the home-based exercises were designed as This systematic review was designed to estimate the effects of a sham or control with no potential therapeutic benefit. Trials were home-based exercises relative to centre-based exercises for included when the control group received an equivalent dose of improving walking and participation after stroke. Any type of home- centre-based exercises, being provided at a centre, such as hospital, based exercises were considered, regardless of whether supervision outpatient department, private practice, medical centre or commu- was provided remotely or in-person. Only trials that compared home- nity centre. Session duration, session frequency and program dura- based and centre-based exercises of similar doses were included. The tion were recorded to assess the similarity of the studies. outcomes that were examined included effects on clinically relevant walking outcomes and their carryover effects to participation. Outcome measures Four outcomes were of interest: walking speed, balance, mobility Therefore, the specific research questions for this systematic re- view were: and participation. The measurement of walking speed (typically ob- tained using a timed walk test) had to be reported as a relation of 1. In people who have had a stroke, how comparable are the ef- distance and time. The measurement of balance had to be repre- fects of home-based exercises to those of equivalent centre- sentative of the ability to maintain a controlled body position during based exercises for improving walking speed, balance, an activity (eg, Berg Balance Scale).26 The measurement of mobility mobility and participation? had to be representative of the ability to change body position or location and move (eg, Timed Up and Go test). The measurement of 2. Is the comparability of the effects of these two types of exercise participation had to be reported by questionnaires that included maintained beyond the intervention period? questions regarding the individual’s ability to perform activities in real-life situations (eg, Stroke Impact Scale or Assessment of Life Method Box 1. Inclusion criteria. This systematic review is reported according to the guidelines on Design the Preferred Reporting Items for Systematic Reviews and Meta- Analyses (PRISMA) statement.24 Randomised controlled trials Participants Identification and selection of trials Adults (. 18 years) Stroke Searches were conducted on MEDLINE, AMED, EMBASE, Cochrane, Intervention PsycINFO and PEDro databases, until March 2022, for relevant studies Home-based exercise without date or language restrictions. The search strategy was four sessions over 2 weeks registered at PubMed/Medline, so the authors received monthly no- two-thirds of the exercise is performed at home tifications with potential papers related to this systematic review. Prescribed by physiotherapist or health professional Search terms included words related to stroke, home-based in- Outcomes measures terventions and randomised trials (see Appendix 1 on the eAddenda Measures of walking speed, balance, mobility or participation for the full search strategy). Titles and abstracts were displayed and Comparison screened to identify relevant studies. The method section of the Home-based exercise versus equivalent dose of centre-based exercise
176 Nascimento et al: Home-based exercises after stroke Habits – LIFE-H or Reintegration to Normal Living Index).27 The Titles and abstracts screened (n = 3,856) timing of the measurements and the procedure used to measure the x AMED, MEDLINE, EMBASE, Cochrane, PsycINFO (n = 3,604) outcomes were recorded to assess the appropriateness of combining x PEDro (n = 252) studies in a meta-analysis. Duplicates removed (n = 88) Data analysis Titles and abstracts screened (n = 3,768) Information about the method (ie, design, participants, interven- tion and measures) and results (ie, number of participants and means Excluded based on title and abstract (n = 3,721) (SD) of outcomes related to walking, balance, mobility and partici- pation) were extracted by two reviewers (MCP and RJSR) and checked Full-text articles assessed for eligibility (n = 47) by two reviewers (LRN and GPF). Where information was unavailable in the published trials, details were requested from the corresponding Excluded after evaluation of full text (n = 38)a author. x intervention was not walking/balance training (n = 8) x comparator was not centre-based exercise (n = 7) The post-intervention or change scores were used to obtain the x dissimilar amount of exercise in the two groups (n = 6) pooled estimate of the effect of the intervention, using a random- x no eligible outcomes (n = 5) effects model. A visual inspection of the distribution of effect sizes x intervention was not home-based (n = 4) in the forest plots was performed and the I2 value was calculated to x intervention was not detailed (n = 4) indicate the proportion of variance that was due to heterogeneity. x home-based exercise was used in both groups (n = 4) Values of I2 . 50% are indicative of important heterogeneity.28,29 The x not a randomised trial (n = 2) analyses were performed using Review Manager softwarea. The x participants did not have stroke (n = 1) pooled data for each outcome were reported as weighted or stand- ardised mean differences between the groups and their 95% CI. Studies included in qualitative synthesis (n = 9) Where data from a trial could not be included in a pooled analysis, the between-group difference and, where possible, its 95% CI were Figure 1. Flow of trials through the review. reported. aTrials may have been excluded for failing to meet more than one inclusion criterion. The Grading of Recommendations Assessment, Development and allocation and no trials blinded the participants and therapists, which Evaluation (GRADE) system was used to summarise the overall is difficult for complex interventions. quality of evidence for each outcome. The GRADE system ranges from high to very low quality.30 This review rated evidence starting at the Participants high-quality level and downgraded it 1 point whenever one of the The mean age of participants ranged from 54 to 70 years old across following prespecified criteria was present: low methodological quality (defined as . 50% of trials with PEDro score , 6); inconsis- trials. Four trials31,32,34,38 included participants in the acute post-stroke tency of estimates among pooled studies (I2 . 50%) or when esti- phase, four trials33,34,36,39 in the chronic phase and one trial37 in both mation was not possible (no pooling); indirectness of participants acute and chronic phases on admission to the trial. The mean baseline (defined as . 50% of trials not reporting time since stroke or baseline walking speed ranged from 0.3 to 0.9 m/s across trials, but three tri- walking speed); and imprecision (pooling , 300 participants per als37–39 did not report the participants’ baseline walking speed. outcome).4 Two reviewers (RJSR and MCP) assessed the quality of the evidence using the GRADE system, with potential disagreements resolved by discussion with a third reviewer (LRN). Results Intervention Flow of trials through the review The experimental intervention in all trials was home-based. Par- ticipants undertook training for 30 to 120 minutes, three times per The electronic search strategy identified 3,856 papers. After screening titles and abstracts, 47 potentially relevant, full-text papers week (SD 1), for 9 weeks (SD 3). Three trials delivered interventions were retrieved. Thirty-eight papers failed to meet the inclusion to improve impairments (eg, strength and balance training),31,35,37 criteria (see Appendix 2 on the eAddenda for a summary of the two trials delivered task-oriented training,32,33 one trial delivered excluded papers); therefore, nine papers were included in the review. Bobath therapy combined with task-oriented training and electrical Flow of studies through the review is presented in Figure 1. stimulation,38 one trial delivered virtual-reality training39 and two trials delivered interventions to improve impairments combined with Characteristics of included trials task-oriented training.34,36 Seven trials provided supervised interventions31–34,36,38,39 and two trials provided unsupervised in- The nine trials involved 609 participants and investigated the ef- terventions combined with phone or written instructions.35,37 Among fects of home-based exercises for improving walking speed (n = trials that provided supervised interventions, the participants had in- 6),31–36 balance (n = 6),29,32,34,35–37 mobility (n = 4)30,31,34,37 and person supervision in five trials,31–34,36 remote supervision in one participation (n = 1)34 after stroke. Detailed information is provided trial39 and both in-person and remote supervision in one trial.38 in Table 1. Additional information was requested from the authors of three papers31,32,36 and received from one author.32 Supervision was predominantly provided by physiotherapists. The control groups received equivalent doses of centre-based exercises, Quality The mean PEDro score of the trials was 6.4 (range 4 to 8). PEDro predominantly delivered in rehabilitation clinics. Modes of exercise delivered to control groups were the same as those delivered to the criteria and scores for the included trials are presented in Table 2. All experimental groups; however, two trials31,36 used robotic devices trials randomly allocated the participants, had similar groups at unavailable for home-based interventions. baseline and reported between-group differences. Eight trials re- ported a point estimate and variability, seven had blinded assessors, Outcome measures and six had , 15% dropouts and reported whether an intention-to- Six trials31–36 measured walking speed using a timed walk mea- treat analysis was undertaken. Four trials reported use of concealed sure, reported in m/s. Six trials measured balance using a stand- ardised scale: four trials31,37–39 used the Berg Balance Scale, one trial32 used the Postural Assessment Scale for Stroke and one trial34 used the Short-form Assessment Scale for Stroke. Three trials32,33,36 measured mobility using a timed test (ie, Timed Up and Go or Sit to
Research 177 Table 1 Characteristics of the included trials (n = 9). Study Participants Intervention Outcome measures Altin et al (2009)37 Balance = BBS (0 to 56) Chen et al (2017)38 n = 20 Frequency and duration Characteristics Mobility = RMI (0 to 15) Age (y) = 62 (10) Timing: 0, 12, 48 wk Duncan et al (2011)31 Time since stroke (mth) = 3 to 24 Exp = home-based exercises Intervention = strength, balance, Walking speed (m/s) = NR 60 min x 3/wk x 12 wk mobility and coordination Balance = BBS (0 to 56) Gjelsvik et al (2014)32 Acute and chronic phase training Timing: 0, 12, 24 wk Hsieh et al (2018)33 Con = centre-based exercises Lloréns et al (2015)39 n = 54 60 min x 3/wk x 12 wk Amount of supervision (%) = 0 Walking speed = 10MWT (m/s) Olaleye et al (2014)34 Age (y) = 66 (12) Supervisor = not provided Balance = BBS (0 to 56) Olney et al (2006)35 Time since stroke (mth) = 1 (5) Type of supervision = not provided Timing: 0, 12, 40 wk Walking speed (m/s) = NR Encouragement = weekly phone call Acute phase Progression = increased repetitions Walking speed = 5MWT (m/s) Balance = PASS (0 to 36) n = 265 Exp = home-based exercises Intervention = Bobath therapy, Mobility = TUG (s) Age (y) = 61 (13) 80 min x 2/wk x 12 wk proprioceptive neuromuscular Timing: 0, 12 wk Time since stroke (mth) = 2 (0.2) facilitation, balance and walking Walking speed (m/s) = 0.38 (0.22) Con = centre-based rehabilitation training and electrical stimulation Walking speed = 10MWT (m/s) Acute phase 80 min x 2/wk x 12 wk Mobility = Sit-To-Stand Test Amount of supervision (%) = 100 (repetitions in 30 s) n = 70 Supervisor = physiotherapist and Timing: 0, 4 wk Age (y) = 70 (13) Time since stroke (mth) = NR caregiver Balance = BBS (0 to 56) Walking speed (m/s) = 0.90 (0.40) Encouragement = NR Timing: 0, 8, 12 wk Acute phase Type of supervision = in-person and Walking speed = 6MWT (m/s) n = 24 remote Balance = SF-PASS (0 to 15) Age (y) = 54 (18) Progression = NR Participation = RNLI (0 to 110) Time since stroke (mth) = 14 (12) Timing: 0, 10 wk Walking speed (m/s) = 0.59 (0.23) Exp = home-based exercises Intervention = strength, balance, Chronic phase 90 min x 3/wk x 12 wk coordination and flexibility Walking speed = 6mWT (m/s) training Timing: 0, 10, 24 wk n = 30 Con = centre-based treadmill Age (y) = 55 (8) training Amount of supervision (%) = 100 Time since stroke (mth) = 27 90 min x 3/wk x 12 wk Supervisor = physiotherapist Walking speed (m/s) = NR Type of supervision = in-person Chronic phase Encouragement = orientation to walk n = 52 daily Age (y) = 61 (9) Progression = increased velocity in Time since stroke (mth) = 1 Walking speed (m/s) = 0.3 (0.35) treadmill training Acute phase Exp = home-based exercises Intervention = task-oriented training n = 72 120 min x 2/wk x 5 wk Amount of supervision (%) = 100 Age (y) = 64 (12) Supervisor = physiotherapist and Time since stroke (mth) = 43 (49) Con = centre-based exercises Walking speed (m/s) = 0.74 (0.34) 120 min x 2/wk x 5 wk occupational therapist Chronic phase Type of supervision = in-person Encouragement = NR Progression = NR Exp = home-based exercises Intervention = mirror therapy and 75 to 105 min x 3/wk x 4 wk task-oriented training Con = centre-based exercises Amount of supervision (%) = 100 75 to 105 min x 3/wk x 4 wk Supervisor = therapist (not specified) Type of supervision = in-person Encouragement = verbal instructions Progression = NR Exp = home-based exercises (virtual Intervention = virtual reality-based reality) exercises for walking and balance 45 min x 3/wk x 6 wk Amount of supervision (%) = 100 Con = centre-based rehabilitation Supervisor = physiotherapist 45 min x 3/wk x 6 wk Type of supervision = remote Encouragement = NR Both = conventional rehabilitation Progression = increased difficulty Exp = home-based rehabilitation Intervention = strength, balance, 45 to 60 min x 2/wk x 10 wk task-oriented training Con = centre-based exercises Amount of supervision (%) = 100 45 to 60 min x 2/wk x 10 wk Supervisor = physiotherapist Type of supervision = in-person Encouragement = NR Progression = increased resistance and intensity Exp = home-based exercises Intervention = aerobic and strength 90 min x 3/wk x 9 wk training Con = centre-based exercises Amount of supervision (%) = 0 90 min x 3/wk x 9 wk Supervisor = not provided Type of supervision = not provided Both = supervised physical Encouragement = written and verbal conditioning program instructions for progression Progression = increased walking intensity and duration wkly
178 Nascimento et al: Home-based exercises after stroke Table 1 (Continued) Study Participants Intervention Outcome measures Uçar et al (2014)36 n = 22 Frequency and duration Characteristics Walking speed = 10MWT (m/s) Age (y) = 57 (8) Mobility = TUG (s) Time since stroke (mth) = 12 Exp = home-based exercises Intervention = Strength, balance, Walking speed (m/s) = 0.64 30 min x 5/wk x 2 wk task-oriented training Timing: 0, 2, 8 wk Chronic phase Con = centre-based robotic exercises Amount of supervision (%) = 100 30 min x 5/wk x 2 wk Supervisor = physiotherapist Type of supervision = in-person Encouragement = biofeedback of performance Progression = decreased amount of assistance Listed groups and outcome measures are those that were analysed in this systematic review; there may have been other groups or measures in the paper. Participant characteristics are presented as mean (SD) or range. BBS = Berg Balance Scale, Con = control group, CT = controlled trial, Exp = experimental group, RCT = randomised controlled trial, RMI = Rivermead Mobility Index, NR = not reported, RNLI = Reintegration to Normal Living Index, PASS = Postural Assessment Scale for Stroke, SF-PASS = Short Form-Postural Assessment Scale for Stroke, TUG = Timed Up and Go Test, 10MWT = 10-Metre Walk Test, 5MWT = 5-Metre Walk Test, 6mWT = 6-Minute Walk Test, 6MWT = 6-Metre Walk Test. Table 2 PEDro scores for the included trials (n = 9). Study Random Concealed Groups Participant Therapist Assessor , 15% lost Intention-to- Between-group Point estimate and Total allocation allocation similar at blinding blinding blinding to follow-up treat analysis difference reported variability reported (0 to 10) baseline Altin et al (2009)37 Y N Y N NN Y Y Y Y6 Y Y Y8 Chen et al (2017)38 Y Y Y N NY Y Y Y Y7 Y Y Y7 Duncan et al (2011)31 Y N Y N NY Y N Y Y5 Y Y Y8 Gjelsvik et al (2014)32 Y Y Y N NY N N Y Y6 Y Y Y7 Hsieh et al (2018)33 Y N Y N NY N N Y N4 Lloréns et al (2015)39 Y Y Y N NY Y Olaleye et al (2014)34 Y N Y N NY Y Olney et al (2006)35 Y Y Y N NN Y Uçar et al (2014)36 Y N Y N NY N N = no, Y = yes. Stand) and one trial37 used the Rivermead Mobility Index. One trial34 Subgroup MD (95% CI) measured participation using the Reintegration to Normal Living Study Random Index. Acute Effect of home-based exercises relative to centre-based exercises Duncan Gjelsvik Walking speed Olaleye The effect of home-based exercises compared with centre-based Subtotal exercises on walking speed was examined by pooling outcomes Chronic from six trials31–36 involving 499 participants. The mean difference Hsieh was –0.03 m/s (95% CI –0.07 to 0.02, I2 = 0), which indicates that Olney home-based exercises and centre-based exercises provided similar Uçar effects on walking speed (Figure 2). The quality of the evidence was Subtotal rated as high. The effects of home-based exercises and centre-based exercises remained similar beyond the intervention period (two tri- Total als, MD 0.02 m/s, 95% CI –0.02 to 0.07, I2 = 0) (Figure 3). For more detailed forest plots, see Figures 4 and 5 on the eAddenda. –1.0 –0.5 0 0.5 1.0 Favours home Effect on balance Favours centre The effect of home-based exercises compared with centre-based Figure 2. Mean difference (95% CI) of home-based versus centre-based exercises for exercises on balance was examined by pooling outcomes from four walking speed (m/s) immediately after the intervention period. trials31,37-39 involving 369 participants. The mean difference on Berg Balance Scale (0 to 56 points) was 0 (95% CI –1 to 2, I2 = 40%), which Effect on mobility indicates that home-based exercises and centre-based exercises The effect of home-based exercises compared with centre-based provide similar effects on balance (Figure 6). The quality of the evidence was rated as high. Two additional trials32,34 involving exercises on mobility was examined by pooling outcomes from three 122 participants measured balance using other scales and also trials33,36,37 involving 66 participants. The standardised mean differ- reported no clear difference between home-based exercises and ence was –0.4 in favour of centre-based exercises; however, the esti- centre-based exercises (SMD –0.1, 95% CI –0.4 to 0.3, I2 = 0, mate was imprecise (95% CI –1.3 to 0.4, I2 = 66%), which made it meta-analysis not shown). The effects of home-based exercises and unclear whether home-based exercises and centre-based exercises centre-based exercises remained similar beyond the intervention provide similar effects on mobility (Figure 10). For a more detailed period (four trials, MD 0 points on Berg Balance Scale, 95% CI –1 to 2, forest plot, see Figure 11 on the eAddenda. The quality of the evidence I2 = 0) (Figure 7). For more detailed forest plots, see Figures 8 and 9 on was rated as very low. One additional trial32 involving 78 participants the eAddenda. provided change scores for mobility and also reported a negligible difference between home-based exercises and centre-based exercises (MD 0 seconds, 95% CI 0 to 0.1). A negligible effect on mobility was observed beyond the intervention period in favour of centre-based
Research 179 Study MD (95% CI) Study MD (95% CI) Random Random Duncan Altin Olney Chen Duncan Total Llorens –1.0 –0.5 0 0.5 1.0 Total Favours home Favours centre –20 –10 0 10 20 Favours home Figure 3. Mean difference (95% CI) of home-based versus centre-based exercises for Favours centre walking speed (m/s) beyond the intervention period. Figure 7. Mean difference (95% CI) of home-based versus centre-based exercises on Berg Balance Scale (0 to 56 points) beyond the intervention period. Study MD (95% CI) SMD (95% CI) Random Random Altin Chen Study Duncan Llorens Altin Hsieh Total Uçar Total –20 –10 0 10 20 –4 –2 0 24 Favours home Favours home Favours centre Favours centre Figure 6. Mean difference (95% CI) of home-based versus centre-based exercises on Figure 10. Standardised mean difference (95% CI) of home-based versus centre-based Berg Balance Scale (0 to 56 points) immediately after the intervention period. exercises for mobility immediately after the intervention period. exercises (one trial, MD –1 point on Rivermead Mobility Index, 95% CI However, the current review did not perform subgroup analyses –2 to 0). based on types of supervision and exercises, due to the small number of trials included.44 Effect on participation The effect of home-based exercises compared with centre-based The experimental and control groups were homogeneous regarding the doses of intervention. Trials were only included when exercises on participation was examined by one trial34 involving 52 the centre-based group practised equivalent doses of exercise participants. The mean difference was –5 points on the Reintegration compared with the home-based group. Therefore, the results sug- to Normal Living Index (95% CI –19 to 10), which indicated that gest that it is not the location of the intervention that is important, home-based exercises and centre-based exercises may have similar but the amount of practice. These results are in accordance with a effects on participation. The quality of the evidence was rated as low. systematic review that found no differences in home-based and No trials compared the long-term effects on participation. centre-based interventions in Parkinson’s disease.25 In addition, the GRADE system of assessing the evidence suggested that two out- Discussion comes examined in the current review were credible (ie, they pro- vided high-quality evidence). The main reason that mobility and This systematic review provided high-quality evidence that home- participation were rated as (very) low-quality evidence was the low based prescribed exercises are as effective as centre-based exercises number of trials, the low number of participants included in the for improving walking speed and balance after stroke. Very low to pooled analyses and the statistical heterogeneity among trials. low-quality evidence suggested that effects on mobility and partici- However, because the mean difference and the confidence intervals pation are also similar. The review included nine trials that directly were close to zero, which indicated no clear evidence in favour of compared home-based exercises and centre-based exercises of either intervention, it is unlikely that further trials would change the equivalent dose, ensuring that the results can be attributed primarily overall results. Therefore, home-based prescribed exercises may be to location. The similarity in effects was maintained beyond the useful for patients who need to increase the amount of practice over intervention period, which should be interpreted cautiously due to a long period of time or are confined to their homes in the ongoing the small number of trials. pandemic due to COVID-19. It is important to highlight that the home-based interventions were predominantly monitored either This is the first review to compare the effects of home-based remotely or in person, which suggests that regular contact with a and centre-based exercises on clinically relevant walking out- physiotherapist is important for ensuring the success of the home- comes and participation after stroke; therefore, comparison with based treatment. previous results is challenging. The results from a Cochrane re- view23 suggested that home-based and centre-based interventions This review had both strengths and limitations. The external val- provide similar benefits, but the conclusions were based on two idity of the review was improved by the included trials having par- trials of telerehabilitation. This review strengthens previous evi- ticipants in both acute and chronic phases after stroke, and the level dence because the conclusions were based on nine randomised of walking speed ranging from 0.3 to 0.9 m/s, which covers the trials of moderate quality, the inclusion of home-based in- spectrum of walking disability. The preliminary observational anal- terventions regardless of the type of supervision, and the mea- ysis suggested that both acute and chronic participants may benefit surement of walking speed. Measurements of walking speed are from home-based or centre-based exercises similarly; however, ac- recommended in all stroke recovery trials40 as the results predict cording to Cochrane recommendations, more trials are required for independence41 and community ambulation.42 There was some robust subgroup analyses based on time since stroke. The experi- clinical heterogeneity related to the characteristics of intervention: mental interventions were mostly reported according to the Template the amount and type of supervision varied among trials; however, for Intervention Description and Replication (TIDIeR)45 in terms of it did not appear to influence the results. A previous systematic session duration, session frequency and program duration; however, review suggested that benefits from home-based rehabilitation are some trials failed to describe the type and progression of the exer- not affected by slight changes in intervention such as the use of cises. Because only trials of equivalent doses of home-based and technology and/or assistive devices in providing motivation.43 centre-based interventions were compared, the results support the
180 Nascimento et al: Home-based exercises after stroke rationale that location does not interfere with the effects on walking 6. Chen KL, Chou YT, Yu WH, Chen CT, Shih CL, Hsieh CL. A prospective study of the and balance abilities. This allows clinicians to consider the best responsiveness of the original and the short form Berg Balance Scale in people with location for intervention based on the patient’s preference and stroke. Clin Rehabil. 2015;29:468–476. available resources. On the other hand, for the measurements of mobility, the pooled effect was calculated using the standardised 7. Faria CD, Teixeira-Salmela LF, Nadeau S. Effects of the direction of turning on the mean difference, which is less clinically meaningful than a mean timed up & go test with stroke subjects. Top Stroke Rehabil. 2009;16:196–206. difference. In addition, five of the six trials included in the meta- analysis of the primary outcome provided full in-person supervi- 8. Stroke Foundation. Clinical Guidelines for Stroke Management. Published online sion, which precluded conclusions based on type or amount of 2017. https://informme.org.au/Guidelines/Clinical-Guidelines-for-Stroke-Manage supervision. Further studies are warranted to clarify if partially su- ment. Accessed 29 May, 2022. pervised home-based interventions are as effective as fully super- vised interventions, as this may have implications for the economies 9. Winstein CJ, Stein J, Arena R, Bates B, Cherney LR, Cramer SC, et al. Guidelines for and practicalities of service delivery. Lastly, few trials measured Adult Stroke Rehabilitation and Recovery: A Guideline for Healthcare Professionals participation or reported outcomes in the longer term, leading to From the American Heart Association/American Stroke Association. Stroke. inconclusive results. 2016;47:e98–e169. In conclusion, this review provides evidence that the effects of 10. Khoshbakht Pishkhani M, Dalvandi A, Ebadi A, Hosseini M. Factors affecting home-based prescribed exercises on walking speed and balance are adherence to rehabilitation in Iranian stroke patients: A qualitative study. J Vasc likely to be similar to improvements obtained by equivalent doses of Nurs. 2019;37:264–271. centre-based exercises. This suggests that home-based prescribed exercises may be an effective strategy for delivering high-quality 11. Schneider EJ, Lannin NA, Ada L, Schmidt J. Increasing the amount of usual reha- exercise to people after stroke in health services where adherence bilitation improves activity after stroke: a systematic review. J Physiother. to centre-based exercises is unsuccessful or in periods that require 2016;62:182–187. social isolation. 12. Dorsch S, Weeks K, King L, Polman E. In inpatient rehabilitation, large amounts of What was already known on this topic: Mobility impair- practice can occur safely without direct therapist supervision: an observational ment is common in people who have had a stroke and balance study. J Physiother. 2019;65:23–27. impairment may contribute to this reduced walking ability. Large amounts of rehabilitation are required to provide benefits and 13. Ashworth NL, Chad KE, Harrison EL, Reeder BA, Marshall SC. Home versus center home-based semi-supervised practice is a feasible and safe op- based physical activity programs in older adults. Cochrane Database Syst Rev. tion to facilitate this. 2005;1:CD004017. What this study adds: Home-based prescribed exercises pro- duced improvements in walking speed, balance, mobility and 14. Allen L, John-Baptiste A, Meyer M, Richardson M, Speechley M, Ure D, et al. participation that were similar to those obtained by equivalent Assessing the impact of a home-based stroke rehabilitation programme: a cost- doses of centre-based exercises after stroke. Beyond the inter- effectiveness study. Disabil Rehabil. 2019;41:2060–2065. vention period, home-based and centre-based exercises had similar effects. 15. Gaikwad R, Warren J. The role of home-based information and communications technology interventions in chronic disease management: a systematic literature Footnotes: a Review Manager V.5.3, The Nordic Cochrane Centre, review. Health Informatics J. 2009;15:122–146. Copenhagen, Denmark 16. Franco MR, Tong A, Howard K, Sherrington C, Ferreira PH, Pinto RZ, et al. eAddenda: Figures 4, 5, 8, 9 and 11, and Appendices 1 and 2 can be Older people’s perspectives on participation in physical activity: a systematic found online at https://doi.org/10.1016/j.jphys.2022.05.018 review and thematic synthesis of qualitative literature. Br J Sports Med. 2015;49:1268–1276. Ethics approval: Nil. Competing interests: The authors declare that they have no 17. van der Veen DJ, Döpp CME, Siemonsma PC, Nijhuis-van der Sanden MWG, de competing interests. Swart BJM, Steultjens EM. Factors influencing the implementation of home-based Sources of support: Fundação de Amparo à Pesquisa e Inovação do stroke rehabilitation: professionals’ perspective. PLoS One. 2019 Jul 25;14: Espírito Santo (FAPES). e0220226. https://doi.org/10.1371/journal.pone.0220226 Acknowledgements: Nil. Provenance: Not invited. Peer reviewed. 18. Bennell KL, Lawford BJ, Metcalf B, Mackenzie D, Russell T, van den Berg M, et al. Correspondence: Lucas R Nascimento, Discipline of Physiotherapy, Physiotherapists and patients report positive experiences overall with telehealth Universidade Federal do Espírito Santo, Brazil. Email: [email protected] during the COVID-19 pandemic: a mixed-methods study. J Physiother. 2021;67:201–209. References 19. De Menezes K, Ada L, Teixeira-Salmela LF, Scianni AA, Avelino PR, Faria CDCM, et al. 1. Virani SS, Alonso A, Aparicio HJ, Benjamin EJ, Bittencourt MS, Callaway CW, et al, Home-based interventions may increase recruitment, adherence, and measure- American Heart Association Council on Epidemiology and Prevention Statistics ment of outcomes in clinical trials of stroke rehabilitation. J Stroke Cerebrovasc Dis. Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke 2021;30:106022. Statistics-2021 Update: A Report From the American Heart Association. Circulation. 2021;143:e254–e743. 20. Fernandes LG, Saragiotto BT. To what extent can telerehabilitation help patients in low- and middle-income countries? Braz J Phys Ther. 2021;25:481–483. https://doi. 2. Jørgensen HS, Nakayama H, Raaschou HO, Olsen TS. Recovery of Walking Function org/10.1016/j.bjpt.2020.11.004 in Stroke Patients: The Copenhagen Stroke Study. Arch Phys Med Rehabil. 1995;76:27–32. 21. Chi NF, Huang YC, Chiu HY, Chang HJ, Huang HC. Systematic review and meta- analysis of home-based rehabilitation on improving physical function among 3. Rosa MC, Marques A, Demain S, Metcalf CD. Fast gait speed and self-perceived home-dwelling patients with a stroke. Arch Phys Med Rehabil. 2020;101:359–373. balance as valid predictors and discriminators of independent community https://doi.org/10.1016/j.apmr.2019.10.181 walking at 6 months post-stroke – a preliminary study. Disabil Rehabil. 2015;37: 129–134. 22. Gelaw AY, Janakiraman B, Gebremeskel BF, Ravichandran H. Effectiveness of home- based rehabilitation in improving physical function of persons with stroke and 4. Nascimento LR, Boening A, Galli A, Polese JC, Ada L. Treadmill walking improves other physical disability: a systematic review of randomized controlled trials. walking speed and distance in ambulatory people after stroke and is not inferior to J Stroke Cerebrovasc Dis. 2020;29:104800. https://doi.org/10.1016/j.jstrokecere- overground walking: a systematic review. J Physiother. 2021;67:95–104. brovasdis.2020.104800 5. Faria CD, Teixeira-Salmela LF, Nadeau S. Predicting levels of basic functional 23. Laver KE, Adey-Wakeling Z, Crotty M, Lannin NA, George S, Sherrington C. Tele- mobility, as assessed by the Timed “Up and Go” test, for individuals with stroke: rehabilitation services for stroke. Cochrane Database Syst Rev. 2020;1:CD010255. discriminant analyses. Disabil Rehabil. 2013;35:146–152. https://doi.org/10.1002/14651858.CD010255.pub3 24. Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;29:372:n71. 25. Flynn A, Allen NE, Dennis S, Canning CG, Preston E. Home-based prescribed ex- ercise improves balance-related activities in people with Parkinson’s disease and has benefits similar to centre-based exercise: a systematic review. J Physiother. 2019;65:189–199. 26. Nascimento LR, da Silva LA, Araújo Barcellos JVM, Teixeira-Salmela LF. Ankle-foot orthoses and continuous functional electrical stimulation improve walking speed after stroke: a systematic review and meta-analyses of randomized controlled trials. Physiotherapy. 2020;109:43–53. 27. WHO ICF. International Classification of Functioning, Disability and Health. Geneva: World Health Organization; 2001. ISBN: 92 4 154542 9. 28. Higgins JPT, Green S. Cochrane Handbook for Systematic Reviews of Interventions. Version 5.1.0. Cochrane Collaboration. Published online 2011. https://training. cochrane.org/handbook. Accessed 29 May, 2022. 29. Borenstein M, Hedges LV, Higgins JPT, Rothstein H. Introduction to Meta-Analysis. John Wiley Sons. Published online 2011. https://onlinelibrary.wiley.com/doi/ book/10.1002/9780470743386. Accessed 29 May, 2022. 30. Balshem H, Helfand M, Schünemann HJ, Oxman AD, Kunz R, Brozek J, et al. GRADE guidelines: 3. Rating the quality of evidence. J Clin Epidemiol. 2011;64:401–406. 31. Duncan PW, Sullivan KJ, Behrman AL, Azen SP, Wu SS, Nadeau SE, et al. Body- weight–supported treadmill rehabilitation after stroke. N Engl J Med. 2011;364:2026–2036. 32. Gjelsvik BEB, Hofstad H, Smedal T, Eide GE, Næss H, Skouen JS, et al. Balance and walking after three different models of stroke rehabilitation: Early supported discharge in a day unit or at home, and traditional treatment (control). BMJ Open. 2014;4:5. https://doi.org/10.1136/bmjopen-2013-004358 33. Hsieh YW, Chang KC, Hung JW, Wu CY, Fu MH, Chen C. Effects of home-based versus clinic-based rehabilitation combining mirror therapy and task-specific
Research 181 training for patients with stroke: a randomized crossover trial. Arch Phys Med 40. Kwakkel G, Lannin NA, Borschmann K, English C, Ali M, Churilov L, et al. Rehabil. 2018;99:2399–2407. https://doi.org/10.1016/j.apmr.2018.03.017 Standardized measurement of sensorimotor recovery in stroke trials: 34. Olaleye OA, Hamzat TK, Owolabi MO. Stroke rehabilitation: Should physiotherapy Consensus-based core recommendations from the Stroke Recovery and Reha- intervention be provided at a primary health care centre or the patients’ place of bilitation Roundtable. Int J Stroke. 2017;12:451–461. https://doi.org/10.1177/ domicile? Disabil Rehabil. 2014;36:49–54. https://doi.org/10.3109/09638288.2013. 1747493017711813 777804 35. Olney SJ, Nymark J, Brouwer B, Culham E, Day A, Heard J, et al. A randomized 41. Torres JL, Andrade FB, Lima-Costa MF, Nascimento LR. Walking speed and home controlled trial of supervised versus unsupervised exercise programs for ambula- adaptations are associated with independence after stroke: A population-based tory stroke survivors. Stroke. 2006;37:476–481. https://doi.org/10.1161/01.STR. prevalence study. Cien Saude Colet. 2022;27:2153–2162. 0000199061.85897.b7 36. Uçar DE, Paker N, Buǧdayci D. Lokomat: A therapeutic chance for patients with 42. Fulk GD, He Y, Boyne P, Dunning K. Predicting Home and Community Walking chronic hemiplegia. NeuroRehabilitation. 2014;34:447–453. https://doi.org/10. Activity Poststroke. Stroke. 2017;48:406–411. https://doi.org/10.1161/STROKEAHA. 3233/NRE-141054 116.015309 37. Ertekin Ö Altin, Gelecek N, Yildirim Y, Akdal G. Supervised versus home physio- therapy outcomes in stroke patients with unilateral visual neglect: A randomized 43. Wong Y, Ada L, Wang R, Månum G, Langhammer B. Self-administered, home- controlled follow-up study. J Neurol Sci. 2009;26:325–334. based, upper limb practice in stroke patients: A systematic review. J Rehabil Med. 38. Chen J, Jin W, Dong WS, Jin Y, Qiao FL, Zhou YF, et al. Effects of Home-based Tel- 2020;52:jrm00118. esupervising Rehabilitation on Physical Function for Stroke Survivors with Hemi- plegia: A Randomized Controlled Trial. Am J Phys Med Rehabil. 2017;96:152–160. 44. Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, et al. Cochrane https://doi.org/10.1097/PHM.0000000000000559 Handbook for Systematic Reviews of Interventions Version 6.2.; 2021. 39. Lloréns R, Noé E, Colomer C, Alcañiz M. Effectiveness, usability, and cost-benefit of a virtual reality-based telerehabilitation program for balance recovery after stroke: 45. Hoffmann TC, Glasziou PP, Boutron I, Milne R, Perera R, Moher D, et al. Better A randomized controlled trial. Arch Phys Med Rehabil. 2015;96:418–425.e2. https:// reporting of interventions: template for intervention description and replication doi.org/10.1016/j.apmr.2014.10.019 (TIDieR) checklist and guide. BMJ. 2014;348:g1687. https://doi.org/10.1136/bmj. g1687 Websites www.pedro.org.au
Journal of Physiotherapy 68 (2022) 165–173 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 Injury prevention programs that include balance training exercises reduce ankle injury rates among soccer players: a systematic review Wesam Saleh A Al Attar a,b,c, Ehdaa H Khaledi a,d, Jumana M Bakhsh a, Oliver Faude c, Hussain Ghulam e, Ross H Sanders b a Department of Physical Therapy, Faculty of Applied Medical Science, Umm Al Qura University, Makkah, Saudi Arabia; b Discipline of Exercise and Sport Science, Faculty of Medicine and Health Sciences, The University of Sydney, Sydney, Australia; c Department of Sport, Exercise and Health, Faculty of Medicine, University of Basel, Basel, Switzerland; d Department of Physical Therapy, King Abdullah Medical City, Makkah, Saudi Arabia; e Department of Physical Therapy, Faculty of Applied Medical Sciences, Najran University, Najran, Saudi Arabia KEY WORDS ABSTRACT Injury prevention programs Question: What is the effect of injury prevention programs that include balance training exercises on the Balance exercises incidence of ankle injuries among soccer players? Design: Systematic review of randomised trials with meta- Ankle injury analysis. Participants: Soccer players of any age, sex or competition level. Interventions: The experimental FIFA intervention was an injury prevention program that included balance training exercises. The control inter- Soccer vention was the soccer team’s usual warm-up program. Outcome measures: Exposure-based ankle injury Sports injury rates. Results: Nine articles met the inclusion criteria. The pooled results of injury prevention programs that included balance training exercises among 4,959 soccer players showed a 36% reduction in ankle injury per 1,000 hours of exposure compared to the control group with an injury risk ratio (IRR) of 0.64 (95% CI 0.54 to 0.77). The pooled results of the Fédération Internationale de Football Association (FIFA) injury prevention programs caused a 37% reduction in ankle injury (IRR 0.63, 95% CI 0.48 to 0.84) and balance-training exercises alone cause a 42% reduction in ankle injury (IRR 0.58, 95% CI 0.41 to 0.84). Conclusions: This meta-analysis demonstrates that balance exercises alone or as part of an injury prevention program decrease the risk of ankle injuries. PROSPERO CRD42017054450. [Al Attar WSA, Khaledi EH, Bakhsh JM, Faude O, Ghulam H, Sanders RH (2022) Injury prevention programs that include balance training exercises reduce ankle injury rates among soccer players: a systematic review. Journal of Physiotherapy 68:165–173] © 2022 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/). Introduction and social costs;9 for example, Marshall et al10 found that a neuro- Ankle injuries represent the second most common category of muscular training prevention program can reduce ankle injuries by injuries after knee injuries in sports such as rugby, soccer, volleyball, handball and basketball.1 They can occur at training and at all levels 43% and healthcare costs by Canadian $2.7 million among the Calgary of competition, representing nearly 15% of high school and university athletics injuries.2,3 In soccer, the most popular world sport,4 players soccer clubs during the season. are exposed to many types of injuries such as sprains, strains, con- tusions and fractures,5 and the ankle is one of the most common sites Non-contact ankle injuries represent 33% to 64% of all ankle in- of injury.5 Furthermore, a history of ankle injury is associated with juries in soccer players.11–14 Balance, neuromuscular control and increased risk of developing osteoarthritis, joint instability and a low physical activity level.6 People who have had an ankle injury are more proprioception have been proposed as intrinsic risk factors for non- than three times more likely to have an ankle injury in the future than contact ankle injuries.15 Among professional basketball and soccer those with no previous ankle injury.3 Ankle injuries incur healthcare- related expenses and time lost due to injury.7 In a typical soccer club players, poor single leg balance, laxity of the ankle joint and of around 28 players, there is an average of seven ankle injuries per season.8 Additionally, almost 87% of ankle injuries lead to time off due decreased ankle plantar flexion were more prevalent in players with a to injury and the mean time off per ankle sprain is around 15 days.8 history of an acute or recurrent lateral ankle sprain,16 but in that Therefore, prevention of ankle injuries among soccer players has the potential to make large reductions in complications and healthcare study it was not possible to know whether these deficits preceded the injury. Stronger evidence comes from a prospective cohort study,17 which showed that amateur soccer players with poor balance and lower limb strength are at increased risk of sustaining non-contact ankle injuries. Some individual studies have reported that balance training is an effective method of improving ankle stability and reducing ankle injuries.18,19 Furthermore, balance exercises greatly contribute to the improvement in proprioception and balance,20,21 and may also https://doi.org/10.1016/j.jphys.2022.05.019 1836-9553/© 2022 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/).
166 Al Attar et al: Balance training for soccer ankle injury prevention improve performance and prevent lower limb injuries.21 In a rando- systematic search covered publications from 1985 to 2020 using mised controlled trial with professional soccer players, 20 minutes of electronic databases: the Cochrane Central Register of Controlled balance training for at least 18 training sessions increased proprio- Trials (CENTRAL), PubMed, Web of Science and the Physiotherapy ceptive ability and improved body control, although specific balance Evidence Database (PEDro). The following keywords combination was measures improved similarly in both groups.22 used to perform electronic searches: (balance training) OR (proprio- ceptive training) OR (neuromuscular training) OR (injury prevention Several systematic reviews have provided some indication of the programs) OR (FIFA 111) OR (sensorimotor) OR (stability training) effect of balance training on ankle injury prevention in sport gener- AND (ankle sprain) OR (ankle injury) OR (inversion injury) OR (ankle ally and in soccer specifically.23–27 In sport generally, Hübscher et al24 instability) AND (soccer) OR (football) OR (athlete). assessed the effect of neuromuscular training programs on injury incidence. The pooled result of multi-intervention programs that Eligibility criteria included balance training showed reductions in lower limb injuries by 39% (RR 0.61, 95% CI 0.49 to 0.77) and ankle sprain injuries by 50% The inclusion criteria are shown in Box 1. There were no re- (RR 0.50, 95% CI 0.31 to 0.79).24 Furthermore, the pooled result of strictions on the age, sex or skill level of the soccer players in the balance training alone reduced the risk of ankle sprain injuries (RR eligible studies. Studies were excluded if the compliance of the par- 0.64, 95% CI 0.46 to 0.90).24 However, these meta-analyses were each ticipants with the randomised interventions was not reported. Studies based on only two studies. A later systematic review25 confirmed that were excluded if they had insufficient information and the corre- balance training reduced ankle sprains across various sports (RR 0.62, sponding author did not respond when contacted for missing data. 95% CI 0.43 to 0.90, five studies) and also improved joint position sense, postural sway and dynamic neuromuscular control. The records retrieved by the searches were pooled and duplicates were removed. The titles and abstracts of the remaining articles were In soccer specifically, the systematic review by Ojeda et al26 screened for potential eligibility by two investigators working inde- investigated which interventions are used to prevent lower limb in- pendently. The full texts of all potentially eligible studies were then juries in soccer players; the interventions included proprioceptive obtained. Articles that did not meet the eligibility criteria were training, neuromuscular training, balance training and postural con- excluded. In the event of any discrepancies, a third reviewer was trol training.26 Al Attar et al23 published a systematic overview of the consulted to reach a consensus. The remaining list of included studies systematic reviews and meta-analyses that have investigated the underwent reference tracking. Previous literature reviews were also preventive effect of the Fédération Internationale de Football Asso- screened for any further eligible studies. Commercial reference ciation (FIFA) injury prevention programs in soccer. The FIFA pro- management softwarea was used for collecting studies, screening, grams, which include the FIFA11 and the FIFA111, include balance eliminating duplicates and managing references. exercise, among other components. That overview found four sys- tematic reviews with meta-analyses that examined the effect of the Assessment of characteristics of studies FIFA programs; there were consistently positive results among these reviews when they examined the effect of the FIFA111 or mixed FIFA Risk of bias injury prevention programs on overall injury risk or lower limb injury The internal validity of all included studies was assessed by two risk. The effects on ankle injuries specifically were not reported. reviewers working independently using the revised Cochrane Risk of Several randomised controlled trials have evaluated the effec- Bias tool (RoB 2) for randomised trials,31 as recommended by Armijo- tiveness of balance exercises on the incidence and severity of ankle Olivo et al.32 Any discrepancy was resolved by a third reviewer. injuries in soccer players.27–29 One study indicated that the incidence of ankle injuries was reduced in the experimental group compared Participants with the control group (RR 0.5, 95% CI 0.26 to 0.97),28 while two studies showed the possibility of clinically relevant effects in either The age and sex of the participants and their compliance with the direction for the incidence rate (RR 0.63, 95% CI 0.31 to 1.27)27 and study interventions were extracted for each included trial to char- (RR 0.59, 95% CI 0.21 to 1.67).29 acterise the experimental and control groups. The level of soccer competition was also extracted. Given the inconsistent findings among these studies, there is a need to evaluate the role of injury preventon programs that include Intervention balance training exercises in preventing soccer-related ankle injury. However, none of the previous reviews assessed in isolation the effect The content of the injury prevention program was extracted from of injury prevention programs that include balance training on each included trial, along with the frequency of use prescribed and reducing ankle injuries. the total duration of the intervention period. The aim of this systematic review was to investigate how much ankle injury rates (pooling initial ankle injuries and re-injuries) are influenced by injury prevention programs that include balance training in soccer players. Therefore, the research question for this systematic review was: What is the effect of injury prevention programs that include Box 1. Inclusion criteria. balance training exercises on the incidence of ankle injuries among soccer players? Design Methods (Cluster) randomised controlled trials Published in English The systematic review was prospectively registered and is re- Participants ported according to the Preferred Reporting Items for Systematic Soccer players Reviews and Meta-Analysis (PRISMA) guidelines.30 Intervention Injury prevention programs that included balance training Identification and selection of studies exercises Search strategy Outcome measures Two researchers independently searched for relevant articles. The Reported at least two of: number of ankle injuries, ankle injury full search strategy is presented in Appendix 1 on the eAddenda. The rate and exposure hours Comparisons An injury prevention program that included balance training exercises versus usual/standard warm-up program
Research 167 Outcome measures Records identified through Records identified through database searching (n = 5,372) other sources (n = 0) Number of ankle injuries, ankle injury rates, exposure hours, follow-up duration and compliance rate were the outcome data ele- Records after duplicates removed (n = 3,689) ments that were extracted from the included trials. Titles and abstracts screened (n = 3,689) Data analysis Excluded based on title and abstract (n = 3,632) Two researchers independently extracted data from the full-text versions of the eligible articles using a data extraction form. The Full-text articles assessed for eligibility (n = 57) main outcome results were extracted for each included study and collected in commercial spreadsheet softwareb prior to analysing Excluded after evaluation of full text (n = 48) them by commercial meta-analysis software.c • balance was not included in the injury The meta-analysis software was used to enter and analyse prevention program (n = 31) extracted data for the meta-analyses, including subgroup analyses. • ankle injury rates and exposure hours were not The main meta-analysis was performed based on the total exposure hours. Subgroup meta-analysis was conducted for prespecified sub- reported (n = 9) sets of studies based on specific types of injury prevention program • author did not respond with data necessary for (balance exercise only, and the FIFA111 program) and sex (male and female). The random-effects model was used, assuming that the inclusion of the study (n = 6) studies incorporated a variety of populations and contexts as well as • not published in English (n = 1) variation in the delivered balance training doses and procedures. The • compliance was not reported (n = 1) incidence of injuries represents the injury rate; it is favoured in sports research for estimating injury incidence, because it adjusts for the Studies included in qualitative synthesis (n = 9) variation in athletes’ exposure hours among the included studies.33 This injury rate is determined by dividing the injury incidence Figure 1. Flow of trials through the review. number by the total risk time and multiplying by 1,000. Injury risk ratio (IRR) was the injury rate of the experimental group divided by Risk of bias the injury rate of the control group. A positive intervention effect is The results of the risk of bias assessment are presented in Figure 2 indicated by an IRR , 1; for example, an IRR of 0.80 indicates a 20% reduction in the injury rate relative to the control group. Each IRR was and Appendix 2 (see eAddenda for Appendix 2). The randomisation reported with a 95% CI. The heterogeneity was measured by con- process was assessed as low risk of bias in eight of the included ducting an I2 test for each meta-analysis; I2 values of 25%, 50% and studies,28,29,40–45 with some concerns of bias in the other study.27 75% were interpreted as low, moderate and high heterogeneity, Eight studies were at low risk of bias from deviations from inten- respectively.34 ded interventions,27,28,40–45 with some concerns of bias in the other study.29 The risk of bias due to missing outcome data was classified as If 10 studies were available, it was intended to create a funnel plot low in seven studies27-29,40-42,44 and high in two studies.43,45 Bias in to assess the risk of potential publication bias. Egger’s test35 and the measurement of the outcome was assessed as low in three Begg’s test36 would then be performed to assess the funnel plot studies28,41,44 and high in five studies,27,40,42,43,45 with some concerns asymmetry. The Duval and Tweedie’s Trim and Fill method37 would of bias in one study.29 Risk of bias in the selection of the reported be applied to determine whether the overall IRR estimate required result was low in all of the included studies.27-29,40–45 any adjustments for the publication bias based on the filled studies. Participants Definitions of injury and athlete workload Four studies included males only,27,42–44 two included females According to the consensus statement of sports epidemiologists, only29,41 and three included both male and female participants.28,40,45 ankle injuries in the included trials were required to meet the defi- Four studies included soccer players aged 13 to 19 years;28,40–42 one nition that the injury caused the player to be completely incapable of study included collegiate soccer players aged 18 to 25 years;43 one participating in the following game or training session.38 Athlete study included amateur soccer players aged 14 to 35 years;44 one workload was defined as the number of active hours spent by athletes study included soccer players participating in the Norwegian First, in either training or competition during the study period.39 Second, and Third Division, aged 17 to 35 years;27 one study included middle and high school soccer players aged 12 to 16 years;29 and one Results study included child soccer players aged 7 to 13 years.45 Flow of studies through the review Interventions Four studies used the FIFA 111 injury prevention program as a The initial database searches retrieved 5,372 records. After removal of duplicates, 3,689 records were screened based on warm-up for the experimental groups,41–44 of which one study per- assessment of the titles and abstracts, leaving 57 full-text articles to formed the FIFA 111 program before and after training.44 The ‘111’ be assessed. Forty-eight articles did not meet the eligibility criteria. program includes three levels of single-leg stance exercise (level one: Thus, nine articles were included in this meta-analysis. Figure 1 hold the ball; level two: throwing the ball with a partner; level three: shows the flow of articles through the search, screening and inclu- test your partner), two sets for each, 30 seconds for each leg. One sion processes. study used the FIFA 111 program for kids,45 including five levels of single-leg stance exercise (level one: throwing the ball, one set on Characteristics of studies each leg, five throws per player; level two: throwing the ball and move it around the free leg, one set on each leg, five throws per Eight studies were cluster randomised controlled trials28,29,40–45 player; level three: passing game, one set on each leg, five passes per and one was an individual randomised controlled trial.27 Three player; level four: throwing the ball and passing back without studies were conducted in the USA,29,40,43 two in Norway,27,41 one in touching the ground, one set on each leg, five throws per player; and Canada,28 one in Australia44 and one in Nigeria,42 with one multi- centre study conducted in four countries (Germany, Switzerland, the Netherlands and Czech Republic).45
168 Al Attar et al: Balance training for soccer ankle injury prevention Randomisation process Reduction of ankle injuries based on total exposure hours Deviations from intended interventions The pooled results showed a 36% reduction in overall ankle in- Missing outcome data Measurement of the outcome juries per 1,000 hours of exposure in the group using injury pre- Selection of the reported result vention programs that included balance training exercises compared Overall with control (IRR 0.64, 95% CI 0.54 to 0.77). The inconsistency statistic indicated no heterogeneity between studies (I2 = 0%) (Figure 3a). See Figure 4a on the eAddenda for a detailed forest plot. Low risk of bias Subgroup analyses Some concerns The pooled estimate of the effect of the FIFA 111 injury prevention High risk of bias program was a 36% reduction in ankle injuries per 1,000 hours of Al Attar et al44 exposure (IRR 0.64, 95% CI 0.48 to 0.84) compared with control Emery et al28 (Figure 3b). The inconsistency statistic indicated moderate hetero- Engebretsen et al27 geneity among FIFA 111 injury prevention program’s studies (I2 = 36%). See Figure 4b on the eAddenda for a detailed forest plot. Foss et al29 McGuine et al40 The pooled estimate of the effect of balance training exercises Owoeye et al42 alone was a 41% reduction in ankle injuries per 1,000 hours of exposure (IRR 0.59, 95% CI 0.41 to 0.84) when compared with control Rössler et al45 (Figure 3c). The inconsistency statistic indicated that these studies Silver-Granelli et al43 were homogeneous (I2 = 0%). See Figure 4c on the eAddenda for a detailed forest plot. Soligard et al41 Injury prevention programs that include balance training exer- Figure 2. Risk of bias assessment of the included studies. cises reduced the rate of ankle injuries per 1,000 hours of exposure by 42% in the trials with male participants (IRR 0.58, 95% CI 0.45 to 0.76), by 15% in the trials with female participants (IRR 0.85, 95% CI 0.59 to 1.22) and by 41% in the trials with some male and some female participants (IRR 0.59, 95% CI 0.42 to 0.83). The inconsistency statistic indicated negligible or no heterogeneity in these analyses (Figure 5). See Figure 6 on the eAddenda for a detailed forest plot. Publication bias No analysis of publication bias was undertaken because this re- view was unable to include a minimum of 10 trials in the review and meta-analysis. level five: testing the partner’s balance, one set on each leg, 20 sec- Discussion onds for each). One study used 5 minutes of warm-up routine, which included aerobic exercise and dynamic stretching in addition to 10 This is the first systematic review and meta-analysis that has minutes of a specific neuromuscular training program including evaluated the effectiveness of injury prevention programs that strength, balance and agility, with a further 15-minute home-based included balance training exercises for reducing the incidence of balance training program using a balance board.28 One study ankle injuries among soccer players. The nine included randomised applied 20 to 25 minutes of neuromuscular training including controlled trials27-29,40–45 yielded strong evidence that balance exer- strength, plyometric and balance exercises such as single-leg Roma- cises undertaken alone or as part of an injury prevention program are nian deadlift.29 Two studies used only balance training programs but very effective in reducing the risk of ankle injuries in soccer players. in different ways:27,40 the experimental group of one study partici- The main outcome of the current meta-analysis was the exposure- pated in a balance training program for the ankle joint using a bal- based ankle incidence rates, and the analysis revealed that injury ance board and balance pad,27 whereas the experimental group of the prevention programs that include balance training exercises reduced other study performed a balance training program comprising five the risk of ankle injuries in soccer players by 36% based on total phases on different surfaces, performed with open and closed eyes.40 exposure hours. The duration of the injury prevention programs in the included studies ranged from 10 weeks to 12 months. The trials of injury prevention programs that include balance training exercises were divided into two subgroups: the trials of the For the control intervention, five studies used usual warm-up pro- FIFA 111 injury prevention program formed one group, while the grams,27,41-43,45 one study used pre-training of the FIFA 111 injury studies that included neuromuscular training or balance training prevention program only (where the experimental group had used it alone were included in the specific balance training subgroup. The before and after training),44 one study used a protocol consisting of FIFA 111 injury prevention program and the specific balance training resisted running using elastic bands,29 one study used standard con- programs demonstrated ankle injury reductions of 36% and 41%, ditioning exercises only,40 and one study used a standardised warm-up respectively. This is consistent with the meta-analysis conducted by (include static and dynamic stretching, and aerobic components) and a Al Attar et al46 in 2015, which investigated the effectiveness of the home-based stretching program for control groups.28 Table 1 sum- FIFA injury prevention programs in reducing soccer-related injuries. marises the characteristics of the nine included studies. The findings of that review demonstrated a reduction in overall in- juries by 23% (IRR 0.77, 95% CI 0.65 to 0.92) and a 24% reduction in Effect of the intervention lower extremity injuries (IRR 0.76, 95% CI 0.62 to 0.94).46 Pooled injury estimates Several meta-analyses have investigated the effectiveness of pro- Table 2 presents the rates of ankle injury and exposure hours for prioceptive and balance training exercises in reducing the incidence of ankle injuries among different sports, not only soccer.25,47,48 A intervention and control groups. From the nine included studies, systematic review with meta-analysis by Bellows et al48 included pooled data from 9,633 participants with 775,606 exposure hours eight randomised controlled trials to assess the effect of ankle bracing identified 529 ankle injuries. and balance training in reducing the incidence of ankle injuries in competitive athletes across different sports. Bellows et al48 reported that athletes who performed balance training showed a reduction in
Research 169 Table 1 Characteristics of the included trials (n = 9). Study Participants Intervention Outcome Design Exp Con measures Country N = 280 Ankle injuries Al Attar 2017 44 Age (yr) = 14 to 35 FIFA 111 program performed before and FIFA 111 program performed before training Ankle injuries Cluster RCT Sex = M after training only Australia Compliance (%) = 83 2 to 3/week 3 6 months 2 to 3/week 3 6 months Ankle injuries Amateur soccer Emery 2010 28 Soccer-specific neuromuscular training Standardised warm-up (static and dynamic Ankle injuries Cluster RCT N = 744 program including dynamic stretching, stretching and aerobic components) and a Canada Age (yr) = 13 to 18 eccentric strengthening, agility, jumping and home-based stretching program Ankle injuries Sex = M, F balance exercises (including home-based 3/week 3 12 months Ankle injuries Engebretsen 2008 27 Compliance (%) = 85 balance training using a wobble board) Ankle injuries RCT Youth soccer 3/week 3 12 months Neuromuscular training, Nordic hamstring Norway lowers and groin strength training Ankle injuries N = 209 Targeted exercise program including balance 2 to 3/week 3 2.5 months Ankle injuries Foss 2018 29 Age (yr) = 17 to 35 exercise using a balance board and balance RCT Sex = M pad, and bouncing with both legs and single USA Compliance (%) = 28 leg First, Second and Third 2 to 3/week 3 2.5 months McGuine 2006 40 Division soccer RCT Neuromuscular training program including Resisted running using elastic bands USA N = 142 strengthening, plyometric exercises and 2 to 3/week 3 6 months Age (yr) = 12 to 16 balance exercises (eg, variations in jumping Owoeye 2014 42 Sex = F techniques, single-leg exercises, balance on Standard conditioning exercises, without any Cluster RCT Compliance (%) = 95 an inflated hemisphere and Romanian balance training exercises Nigeria Middle school and high deadlift) 3 to 5/week 3 5 months school soccer 2 to 3/week 3 6 months Rössler 2017 45 Cluster RCT N = 530 Balance training program comprising five Switzerland, Age (yr) = 15 to 18 phases on different surfaces, performed with Germany, Sex = M, F eyes open and closed Czech Republic, Compliance (%) = 85 3 to 5/week 3 5 months Netherlands Youth soccer FIFA 111 program including three levels of Aerobic warm-up (eg, jogging), static Silvers-Granelli 2015 43 N = 416 balance exercises in single-leg stance: L.1 stretches and soccer skills practice (eg, Cluster RCT Age (yr) = 14 to 19 hold the ball, L.2 throwing the ball with a running/cutting drills) USA Sex = M partner and L.3 test your partner 2/week 3 6 months Compliance (%) = 60 2/week 3 6 months Soligard 2008 41 Youth soccer Standard warm-up, typically including Cluster RCT FIFA 111 Kids program including five levels aerobic exercise (eg, running laps of the Norway N = 3,895 of balance exercise in single-leg stance: L.1 pitch), static and dynamic stretching, soccer Age (yr) = 7 to 13 throw the ball, L.2 throw the ball and move it skills practice (eg, dribbling and passing) and Sex = M, F around the free leg, L.3 passing game, L.4 small-sided games Compliance (%) = 50 to 100 throw the ball and pass back without 2/week 3 12 months Children’s soccer touching the ground, L.5 testing your partner 2/week 3 12 months Aerobic warm-up (eg, running exercises), N = 1,525 static and/or dynamic stretching and soccer Age (yr) = 18 to 25 FIFA 111 program including three levels of skills practice (eg, cutting and short passing Sex = M balance exercises in single-leg stance: L.1 drills) Compliance (%) = 73 hold the ball, L.2 throwing the ball with a 3/week 3 6 months Collegiate soccer partner and L.3 test your partner 3/week 3 6 months Standard warm-up, typically including N = 1,892 running exercises to warm-up and static Age (yr) = 13 to 17 FIFA 111 program including three levels of stretches Sex = F balance exercises in single-leg stance: L.1 3/week 3 8 months Compliance (%) = 77 hold the ball, L.2 throwing the ball with a Youth soccer partner and L.3 test your partner 3/week 3 8 months Con = control group, Exp = experimental group, F = female, FIFA 111 = Fédération Internationale de Football Association 111 injury prevention program, L = level, M = male, RCT = randomised controlled trial. ankle sprains of 46% compared with control groups that did not participated in proprioceptive training had a 35% reduction in the receive any intervention (RR 0.54, 95% CI 0.29 to 0.90), based on the analysis of 3,577 participants. Furthermore, de Vasconcelos et al25 incidence of ankle sprains (RR 0.65, 95% CI 0.55 to 0.77); results were conducted a systematic review and meta-analysis that included 12 randomised controlled trials in different languages, five of which similar in a subgroup analysis of participants with and without ankle were included in the quantitative analysis to investigate the effec- tiveness of balance training in reducing ankle sprains among athletes. sprain history: RR 0.64 (95% CI 0.51 to 0.81) and RR 0.57 (95% CI 0.34 They found that balance training increased the reduction of ankle to 0.97), respectively.47 sprain incidence by 38% compared with the control group (RR 0.62, 95% CI 0.43 to 0.90), based on the analysis of 1,606 participants.25 All of the other systematic reviews on similar topics relied on the Another meta-analysis by Schiftan et al47 of seven randomised number of ankle injuries rather than incidence rates,25,47,48 with controlled trials investigated the effectiveness of proprioceptive training in reducing the incidence of ankle sprains in different sports analyses based on the number of participants who became injured, populations, with or without ankle injury history. They found that regardless of the previous history of ankle injury, athletes who without considering the athletes’ exposure time. The incidence rate is the preferable and more meaningful measure of injury incidence in sports research studies because it can accommodate the variations in the athletes’ exposure time and provide an estimate of the injury risk;33,49,50 moreover, it provides an important foundation to study possible predictive factors and the efficacy of preventive measures.33 Soomro et al51 and Al Attar et al46 proposed in their meta-analyses,
170 Al Attar et al: Balance training for soccer ankle injury prevention Table 2 Injury rates per 1,000 hours of exposure in the experimental and control groups of the included studies. Study Exp Con Injury risk ratio (95% CI) N Ankle Exposure hours Ankle N Ankle injuries Exposure hours Ankle injuries injuries/1,000 hours injuries/1,000 hours Al Attar 2017 44 144 4 35,802 0.112 136 12 31,616 0.380 0.29 (0.10 to 0.91) Emery 2010 28 380 14 24,051 0.582 364 27 23,597 1.144 0.51 (0.27 to 0.97) Engebretsen 2008 27 102 13 21,666 0.600 107 20 22,222 0.900 0.67 (0.33 to 1.34) Foss 2018 29 74 6 6,060 0.990 68 9 5,409 1.664 0.60 (0.21 to 1.67) McGuine 2006 40 251 13 12,173 1.068 279 24 13,434 1.787 0.60 (0.31 to 1.17) Owoeye 2014 42 212 10 51,017 0.196 204 30 61,045 0.491 0.40 (0.20 to 0.82) Rössler 2017 45 2,066 26 140,716 0.185 1829 44 152,033 0.289 0.64 (0.39 to 1.04) Silvers-Granelli 2015 43 675 59 35,226 1.675 850 115 44,212 2.601 0.64 (0.47 to 0.88) Soligard 2008 41 1,055 51 49,899 1.022 837 52 45,428 1.145 0.89 (0.61 to 1.31) 4,959 196 376,610 4,674 333 398,996 0.64 (0.54 to 0.77) Pooled data which investigated the effectiveness of several injury prevention reducing ankle injuries in male soccer players than female soccer programs, that the athlete’s injury risk can be influenced by the players. When comparing studies including only male partici- exposure time of the player. It has been found that the athletes’ risk of pants27,42–44 with studies including only female participants,29,41 the injury was associated with an increase in the amount of exposure to pooled IRRs were 0.58 (ie, a 42% reduction) and 0.85 (ie, a 15% sports.52,53 Therefore, using athletes’ exposure-based incidence rate reduction), respectively. However, these subgroup analyses were in evaluating the efficacy of injury prevention programs is more ac- based on few trials, so there is considerable uncertainty in their es- curate and powerful. timates: IRR 0.58 (95% CI 0.45 to 0.75) and IRR 0.85 (95% CI 0.59 to 1.22), respectively. The 15% reduction in ankle injuries detected in the The subgroup analyses suggested that injury prevention programs pooled studies with only female participants should be interpreted that include balance training exercises might be more effective at with caution because the confidence interval does not exclude the possibility that the intervention is ineffective. At any rate, the pooled a Study Rate Ratio (95% CI) results of the three studies that included both males and females showed a 41% reduction in the risk of ankle injury.28,40,45 Conse- Al Attar⁴⁴ quently, further research studies are required to investigate the effect Emery²⁸ of injury prevention programs that include balance training exercises Engebretsen²⁷ for reducing ankle injuries in female soccer players, especially Foss²⁹ because female athletes are at higher risk of sustaining ankle injuries McGuine⁴º than males.50 Owoeye⁴² Rössler⁴⁵ It was difficult to comment on the most appropriate balance Silvers-Granelli⁴³ training protocol, as the programs varied in multiple ways including Soligard⁴¹ duration, frequency and the specific exercises used. Therefore, more trials are needed to determine the dose-response relationship and the Total optimal strategy of the balance training program to prevent ankle injuries. However, it is assumed that a valuable outcome is achieved 0.01 0.1 1 10 100 Favours exp Favours con Subgroup Study b Study Rate Ratio (95% CI) Males Emery²⁸ Rate Ratio (95% CI) Al Attar⁴⁴ Engebretsen²⁷ Engebretsen²⁷ Foss²⁹ Owoeye⁴² McGuine⁴º Silvers-Granelli⁴³ Subtotal Total 0.01 0.1 1 10 100 Favours exp Favours con Females and males Emery²⁸ c Study Rate Ratio (95% CI) McGuine⁴º Rössler⁴⁵ Al Attar⁴⁴ Subtotal Owoeye⁴² Rössler⁴⁵ Females Silvers-Granelli⁴³ Foss²⁹ Soligard⁴¹ Soligard⁴¹ Subtotal Total 0.01 0.1 1 10 100 Favours exp Favours con 0.01 0.1 1 10 100 Favours exp Favours con Figure 3. Forest plot of the effect on ankle injury rate ratio of (a) injury prevention Figure 5. Detailed forest plot of the effect of injury prevention programs that included programs that included balance training versus control; (b) balance training exercises balance training versus control on ankle injury rate ratio, subgrouped according to the alone versus control; and (c) Fédération Internationale de Football Association 111 sex of the study participants. (FIFA 111) injury prevention program versus control.
Research 171 when performing training sessions two to three times per week for 6 authors were contacted to provide specific data of soccer players if to 12 months.28,42–44 It has been hypothesised that to achieve a they reported only general results such as ankle injury rate for mul- tiple sports. preventive effect it is necessary to carry out the training session for at least 10 minutes,24 more than once per week for 3 to 12 months,24,54 This review also had some limitations. First, only studies published in English language were included; however, language restrictions do taking into account that a higher number of sessions will result in not necessarily influence or bias the systematic review results.63 extra benefits.46 Second, there were no specific criteria for the level of competition, and so the analysis included data across ages, sex, and playing grades; The level of compliance with injury prevention programs affects however, this variety does improve the external validity of the review. injury rates.55 The findings of two studies indicated that when Third, a reduction in the rate of ankle injuries could be influenced by compliance increased, the risk of injury decreased;56,57 so compliance other training elements of the preventive program and this might can influence the efficacy of the prevention program.58 Therefore, one confound the preventive effect of the balance training exercises; to study was excluded from this review because the authors mentioned address this, this review analysed the subgroup of trials that evaluated the preventive effect of balance exercises in isolation. that they had no information about participants’ compliance with the injury prevention programs that they used;18 however, their findings Further research evaluating the implementation of balance showed that proprioceptive training reduced the ankle incidence rate training exercises as a preventive strategy for ankle injuries in soccer players is recommended: to investigate the effectiveness of balance (RR 0.13, 95% CI 0.003 to 0.93). The studies included in the current exercises among players with and without a history of an ankle injury; and to assess whether balance training exercises are effective meta-analysis reported a moderate to excellent degree of compliance in reducing ankle injuries among athletic populations other than (60% to 100%) with the preventive programs used,28,29,40–45 except for soccer, especially those who are at high risk of sustaining ankle in- juries in sports such as basketball, rugby, volleyball and handball.1 one study, which reported low participant compliance of 28% with the targeted prevention program.27 All studies with moderate to high In conclusion, this is the first level 1 meta-analysis to show that balance exercises alone or combined with an injury prevention pro- compliance revealed a higher reduction in injuries in the experi- gram lead to a significant reduction in ankle injuries in soccer players. In addition, it is the first meta-analysis to set out the effectiveness of mental group than the control, regardless of the prevention program injury prevention programs that included balance training exercises used.28,29,40–45 Nevertheless, only four included studies with moder- in preventing ankle injuries among male and female soccer players. The results showed that performing injury prevention programs that ate to high compliance of 60% to 85% showed a substantial reduction include balance training exercises could reduce ankle injury rates by in ankle injuries in soccer players.28,42-44 Therefore, due to the dif- 36% compared with teams that did not apply. These findings can help stakeholders regarding the implementation of balance training as a ferences between previous literature and the current findings preventive measure of ankle injuries for soccer players. regarding the relationship between athletes’ compliance and the reduction in injury risk, further studies are needed to determine how What was already known on this topic: Soccer players commonly sustain ankle injuries, leading to healthcare expenses the compliance influences the efficacy of injury prevention programs and lost playing time. that include balance training exercises for reducing ankle injury risk What this study adds: Among a diverse population of soccer players, injury prevention programs that include balance training among soccer players. exercises reduced ankle injury risk by 36%. Balance training exercises alone produced a 42% reduction in ankle injury risk. It was difficult to run an additional analysis based on the history of ankle injury, as most of the studies did not distinguish between Footnotes: a Endnote version X8, Thomson Reuters, Philadelphia, USA. whether participants had a history of ankle injury or not. However, b Microsoft Excel for Mac 2011, Microsoft Corporation, Redmond, one included study examined the preventive effect of balance exer- USA. cises on soccer players with a history of ankle injury, and the results c Comprehensive Meta-Analysis software V.3, Biostat Inc, Engle- wood, USA. showed a reduction in ankle injuries by 33% (IRR 0.67, 95% CI 0.33 to 1.34).27 Another study investigated the effectiveness of a balance eAddenda: Figures 4 and 6, and Appendices 1 and 2 can be found online at https://doi.org/10.1016/j.jphys.2022.05.019 training program in reducing the risk of ankle injuries among soccer Ethics approval: Nil. and basketball players, and the second objective of that study was to Competing interests: The author(s) declare that they have no competing interests. compare the effectiveness of balance training between athletes with Source(s) of support: This research did not receive any specific and without a history of a previous ankle injury.40 The findings grant from funding agencies in the public, commercial or not-for- showed no clear reduction in ankle injuries among athletes without a profit sectors. Acknowledgements: The authors thank the Department of prior ankle injury, while a strong effect was identified among those Physical Therapy at Umm Al Qura University, Department of Physical with a history of ankle injury (RR 2.14, 95% CI 1.25 to 3.65).40 Therapy, King Abdullah Medical City; Department of Sport, Exercise and Health at the University of Basel; the Discipline of Exercise and Furthermore, previous clinical trials showed a reduction in ankle Sport Science, Faculty of Medicine and Health Sciences, The Univer- sity of Sydney; and the Department of Physical Therapy at Najran injuries when performing a proprioceptive training program in ath- University for their support. Wesam Saleh A. Al Attar would like to letes with a history of ankle injury only.59–61 Moreover, the previously thank the Deanship of Scientific Research at Umm Al Qura University mentioned meta-analysis by Schiftan et al47 revealed a marked for supporting his work by Grant Code: (22UQU4350385DSR01). Author contributions: All listed authors substantially contributed reduction in ankle sprains in sporting populations with and without a to preparing this review: protocol design (WA), data collection and extraction (WA, JB, EK), risk of bias assessment (WA, JB, EK), data history of ankle injury: RR 0.64 (95% CI 0.51 to 0.81) and RR 0.57 (95% CI 0.34 to 0.97), respectively. This indicates that proprioceptive training is an effective element in preventing secondary ankle sprains in different sporting populations. Furthermore, a meta-analysis by Vriend et al62 assessed the preventive effect of neuromuscular training that included balance exercises on the primary and sec- ondary ankle injury among different sports. The findings demon- strated that neuromuscular training is an effective strategy for reducing ankle injuries in athletes with a history of ankle sprain (RR 0.69, 95% CI 0.49 to 0.98), but the evidence remained inconclusive for preventing primary ankle sprains.62 Therefore, there is a need for more cluster-randomised controlled trials comparing the prophylac- tic effect of balance training in soccer players with and without a history of ankle injury. This review included subgroup analyses of sex (male, female) and type of intervention (FIFA 111 injury prevention program, specific balance training) to provide a better understanding of the factors that would influence the ankle injury risk. The uniqueness of this meta- analysis lay in the high methodological standards used in this re- view, including using exposure-based incidence rates. In addition, all eligible studies with missing data were initially included, as their
172 Al Attar et al: Balance training for soccer ankle injury prevention analysis (WA), data interpretation (WA, JB, EK); all six authors 27. Engebretsen AH, Myklebust G, Holme I, Engebretsen L, Bahr R. Prevention of in- juries among male soccer players: a prospective, randomized intervention study contributed to writing, drafting and critical revision for the important targeting players with previous injuries or reduced function. Am J Sports Med. intellectual content (WA, JB, EK, RS, HG, OF); all six authors approved 2008;36:1052–1060. the final version for publication (WA, JB, EK, RS, HG, OF). 28. Emery C, Meeuwisse W. The effectiveness of a neuromuscular prevention strategy Data sharing: All data generated or analysed during this study are to reduce injuries in youth soccer: a cluster-randomised controlled trial. Br J Sports Med. 2010;44:555–562. included in this published article. Provenance: Not invited. Peer reviewed. 29. Foss KDB, Thomas S, Khoury JC, Myer GD, Hewett TE. A school-based neuromus- cular training program and sport-related injury incidence: a prospective ran- Correspondence: Wesam Saleh A Al Attar, Department of Physical domized controlled clinical trial. J Athl Train. 2018;53:20–28. Therapy, Umm Al Qura University, Saudi Arabia. Email: 30. Moher D, Shamseer L, Clarke M, Ghersi D, Liberati A, Petticrew M, et al. Preferred [email protected] reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Syst Rev. 2015;4:1. References 31. Sterne JA, Savovic J, Page MJ, Elbers RG, Blencowe NS, Boutron I, et al. RoB 2: a 1. Fong DT-P, Hong Y, Chan L-K, Yung PS-H, Chan K-M. A systematic review on ankle revised tool for assessing risk of bias in randomised trials. BMJ. 2019;366, injury and ankle sprain in sports. Sports Med. 2007;37:73–94. 14898. 2. Hootman JM, Dick R, Agel J. Epidemiology of collegiate injuries for 15 sports: 32. Armijo-Olivo S, da Costa BR, Cummings GG, Ha C, Fuentes J, Saltaji H, et al. PEDro or summary and recommendations for injury prevention initiatives. J Athl Train. Cochrane to assess the quality of clinical trials? A meta-epidemiological study. PloS 2007;42:311–319. one. 2015;10:e0132634. 3. Kucera KL, Marshall SW, Wolf SH, Padua DA, Cameron KL, Beutler AI. Association of 33. Caine D, Maffulli N, Caine C. Epidemiology of injury in child and adolescent sports: injury history and incident injury in cadet basic military training. Med Sci Sports injury rates, risk factors, and prevention. Clin Sports Med. 2008;27:19–50. Exerc. 2016;48:1053–1061. 34. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta- 4. Giulianotti R. Football. The Wiley-Blackwell Encyclopedia of Globalization. 2012. analyses. BMJ. 2003;327:557–560. 5. Fernandez WG, Yard EE, Comstock RD. Epidemiology of lower extremity injuries 35. Egger M, Smith GD, Schneider M, Minder C. Bias in meta-analysis detected by a among US high school athletes. Acad Emerg Med. 2007;14:641–645. simple, graphical test. BMJ. 1997;315:629–634. 6. Gribble PA, Bleakley CM, Caulfield BM, Docherty CL, Fourchet F, Fong DT, et al. 36. Begg CB, Mazumdar M. Operating characteristics of a rank correlation test for Evidence review for the 2016 International Ankle Consortium consensus statement publication bias. Biometrics. 1994;50:1088–1101. on the prevalence, impact and long-term consequences of lateral ankle sprains. Br J Sports Med. 2016;50:1496–1505. 37. Duval S, Tweedie R. Trim and fill: a simple funnel-plot-based method of 7. Kemler E, van de Port I, Backx F, van Dijk CN. A systematic review on the treatment testing and adjusting for publication bias in meta-analysis. Biometrics. 2000;56: of acute ankle sprain. Sports Med. 2011;41:185–197. 455–463. 8. Waldén M, Hägglund M, Ekstrand J. Time-trends and circumstances surrounding ankle injuries in men’s professional football: an 11-year follow-up of the UEFA 38. Fuller CW, Ekstrand J, Junge A, Andersen TE, Bahr R, Dvorak J, et al. Consensus Champions League injury study. Br J Sports Med. 2013;47:748–753. statement on injury definitions and data collection procedures in studies of foot- 9. Barelds I, van den Broek AG, Huisstede BM. Ankle bracing is effective for primary ball (soccer) injuries. Scand J Med Sci Sports. 2006;16:83–92. and secondary prevention of acute ankle injuries in athletes: a systematic review and meta-analyses. Sports Med. 2018;48:2775–2784. 39. Al Attar WSA, Soomro N, Sinclair PJ, Pappas E, Sanders RH. Effect of injury pre- 10. Marshall DA, Lopatina E, Lacny S, Emery CA. Economic impact study: neuromus- vention programs that include the Nordic hamstring exercise on hamstring injury cular training reduces the burden of injuries and costs compared to standard rates in soccer players: a systematic review and meta-analysis. Sports Med. warm-up in youth soccer. Br J Sports Med. 2016;50:1388–1393. 2017;47:907–916. 11. Peterson L, Junge A, Chomiak J, Graf-Baumann T, Dvorak J. Incidence of football injuries and complaints in different age groups and skill-level groups. Am J Sports 40. McGuine TA, Keene JS. The effect of a balance training program on the risk of ankle Med. 2000;28(5 suppl):51–57. sprains in high school athletes. Am J Sports Med. 2006;34:1103–1111. 12. Mauntel TC, Wikstrom EA, Roos KG, Djoko A, Dompier TP, Kerr ZY. The epidemi- ology of high ankle sprains in National Collegiate Athletic Association sports. Am J 41. Soligard T, Myklebust G, Steffen K, Holme I, Silvers H, Bizzini M, et al. Compre- Sports Med. 2017;45:2156–2163. hensive warm-up programme to prevent injuries in young female footballers: 13. Kopec TJ, Hibberd EE, Roos KG, Djoko A, Dompier TP, Kerr ZY. The epidemiology of cluster randomised controlled trial. BMJ. 2008;337:a2469. deltoid ligament sprains in 25 National Collegiate Athletic Association Sports, 2009–2010 through 2014–2015 academic years. J Athl Train. 2017;52:350–359. 42. Owoeye OB, Akinbo SR, Tella BA, Olawale OA. Efficacy of the FIFA 111 warm-up 14. Roos KG, Kerr ZY, Mauntel TC, Djoko A, Dompier TP, Wikstrom EA. The epidemi- programme in male youth football: a cluster randomised controlled trial. J Sports ology of lateral ligament complex ankle sprains in National Collegiate Athletic Sci Med. 2014;13:321–328. Association sports. Am J Sports Med. 2017;45:201–209. 15. Correia MÂdC, Torres J. Intrinsic and extrinsic risk factors for lateral ankle sprains: 43. Silvers-Granelli H, Mandelbaum B, Adeniji O, Insler S, Bizzini M, Pohlig R, et al. a literature review. Arch Sports Med. 2019;3:172–177. Efficacy of the FIFA 111 injury prevention program in the collegiate male soccer 16. Halabchi F, Angoorani H, Mirshahi M, Shahi MHP, Mansournia MA. The prevalence player. Am J Sports Med. 2015;43:2628–2637. of selected intrinsic risk factors for ankle sprain among elite football and basketball players. Asian J Sports Med. 2016;7:e35287. 44. Al Attar WSA, Soomro N, Pappas E, Sinclair PJ, Sanders RH. Adding a post-training 17. Henry T, Evans K, Snodgrass SJ, Miller A, Callister R. Risk factors for noncontact FIFA 111 exercise program to the pre-training FIFA 111 injury prevention program ankle injuries in amateur male soccer players: a prospective cohort study. Clin J reduces injury rates among male amateur soccer players: a cluster-randomised Sport Med. 2016;26:251–258. trial. J Physiother. 2017;63:235–242. 18. Mohammadi F. Comparison of 3 preventive methods to reduce the recurrence of ankle inversion sprains in male soccer players. Am J Sports Med. 2007;35:922–926. 45. Rössler R, Junge A, Bizzini M, Verhagen E, Chomiak J, Meyer T, et al. A multinational 19. Cruz-Diaz D, Lomas-Vega R, Osuna-Pérez M, Contreras F, Martínez-Amat A. Effects cluster randomised controlled trial to assess the efficacy of ‘111 Kids’: a warm-up of 6 weeks of balance training on chronic ankle instability in athletes: a ran- programme to prevent injuries in children’s football. Sports Med. 2018;48:1493–1504. domized controlled trial. Int J Sports Med. 2015;36:754–760. 20. Hanney WJ. Proprioceptive training for ankle instability. Strength Cond J. 46. Al Attar WSA, Soomro N, Pappas E, Sinclair PJ, Sanders RH. How effective are F- 2000;22:63–68. MARC injury prevention programs for soccer players? A systematic review and 21. Daneshjoo A, Mokhtar AH, Rahnama N, Yusof A. The effects of comprehensive meta-analysis. Sports Med. 2016;46:205–217. warm-up programs on proprioception, static and dynamic balance on male soccer players. PloS one. 2012;7:e51568. 47. Schiftan GS, Ross LA, Hahne AJ. The effectiveness of proprioceptive training in 22. Gioftsidou A, Malliou P, Pafis G, Beneka A, Tsapralis K, Sofokleous P, et al. Balance preventing ankle sprains in sporting populations: a systematic review and meta- training programs for soccer injuries prevention. J Hum Sport Exerc. 2012;7:639– analysis. J Sci Med Sport. 2015;18:238–244. 647. 23. Al Attar WSA, Alshehri MA. A meta-analysis of meta-analyses of the effectiveness 48. Bellows R, Wong CK. The effect of bracing and balance training on ankle sprain of FIFA injury prevention programs in soccer. Scand J Med Sci Sports. 2019;29:1846– incidence among athletes: a systematic review with meta-analysis. Int J Sports Phys 1855. Ther. 2018;13:379–388. 24. Hübscher M, Zech A, Pfeifer K, Hänsel F, Vogt L, Banzer W. Neuromuscular training for sports injury prevention: a systematic review. Med Sci Sports Exerc. 49. Caine D, Caine C, Maffulli N. Incidence and distribution of pediatric sport-related 2010;42:413–421. injuries. Clin J Sport Med. 2006;16:500–513. 25. de Vasconcelos GS, Cini A, Sbruzzi G, Lima CS. Effects of proprioceptive training on the incidence of ankle sprain in athletes: systematic review and meta-analysis. Clin 50. Doherty C, Delahunt E, Caulfield B, Hertel J, Ryan J, Bleakley C. The incidence and Rehabil. 2018;32:1581–1590. prevalence of ankle sprain injury: a systematic review and meta-analysis of pro- 26. Ojeda ÁCH, Sandoval DAC, Barahona-Fuentes GD. Proprioceptive training methods spective epidemiological studies. Sports Med. 2014;44:123–140. as a tool for the prevention of injuries in football players: a systematic review. Arch Med Deporte. 2019;36:173–180. 51. Soomro N, Sanders R, Hackett D, Hubka T, Ebrahimi S, Freeston J, et al. The efficacy of injury prevention programs in adolescent team sports: a meta-analysis. Am J Sports Med. 2016;44:2415–2424. 52. Emery C, Tyreman H. Sport participation, sport injury, risk factors and sport safety practices in Calgary and area junior high schools. Paediatr Child Health. 2009;14:439–444. 53. Junge A, Cheung K, Edwards T, Dvorak J. Injuries in youth amateur soccer and rugby players—comparison of incidence and characteristics. Br J Sports Med. 2004;38:168–172. 54. Baltaci G, Kohl HW. Does proprioceptive training during knee and ankle rehabili- tation improve outcome? Phys Ther Rev. 2003;8:5–16. 55. van Reijen M, Vriend I, Van Mechelen W, Finch CF, Verhagen EA. Compliance with sport injury prevention interventions in randomised controlled trials: a systematic review. Sports Med. 2016;46:1125–1139. 56. Hägglund M, Atroshi I, Wagner P, Waldén M. Superior compliance with a neuro- muscular training programme is associated with fewer ACL injuries and fewer acute knee injuries in female adolescent football players: secondary analysis of an RCT. Br J Sports Med. 2013;47:974–979. 57. Soligard T, Nilstad A, Steffen K, Myklebust G, Holme I, Dvorak J, et al. Compliance with a comprehensive warm-up programme to prevent injuries in youth football. Br J Sports Med. 2010;44:787–793.
Research 173 58. Finch C. A new framework for research leading to sports injury prevention. J Sci ankle sprains: a prospective controlled trial. Am J Sports Med. 2004;32:1385– Med Sport. 2006;9:3–9. 1393. 62. Vriend I, Gouttebarge V, Van Mechelen W, Verhagen E. Neuromuscular 59. Stasinopoulos D. Comparison of three preventive methods in order to reduce the training is effective to prevent ankle sprains in a sporting population: a meta- incidence of ankle inversion sprains among female volleyball players. Br J Sports analysis translating evidence into optimal prevention strategies. JISAKOS. Med. 2004;38:182–185. 2016;1:202–213. 63. Moher D, Pham B, Lawson M, Klassen T. The inclusion of reports of randomised 60. Verhagen EA, van Mechelen W, de Vente W. The effect of preventive measures on trials published in languages other than English in systematic reviews. Health the incidence of ankle sprains. Clin J Sport Med. 2000;10:291–296. Technol Assess. 2003;7:1–90. 61. Verhagen E, Van der Beek A, Twisk J, Bouter L, Bahr R, Van Mechelen W. The effect of a proprioceptive balance board training program for the prevention of
Journal of Physiotherapy 68 (2022) 151–152 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 Nature prescriptions for community and planetary health: unrealised potential to improve compliance and outcomes in physiotherapy Thomas Astell-Burt a,b,c, Evangelos Pappas d,e, Julie Redfern f,g,h, Xiaoqi Feng a,b,h,i a Population Wellbeing and Environment Research Lab (PowerLab), Australia; b School of Health and Society, University of Wollongong, Wollongong, Australia; c Menzies Centre for Health Policy and Economics, The University of Sydney, Sydney, Australia; d School of Medicine and Illawarra Health and Medical Research Institute, University of Wollongong, Wollongong, Australia; e Sydney School of Health Sciences, Discipline of Physiotherapy, The University of Sydney, Sydney, Australia; f Engagement and Co-design Research Hub, School of Health Sciences, Faculty of Medicine and Health, The University of Sydney, Sydney, Australia; g Research Education Network, Western Sydney Local Health District, Sydney, Australia; h George Institute for Global Health, University of New South Wales, Sydney, Australia; i School of Population Health, Faculty of Medicine and Health, University of New South Wales, Sydney, Australia Despite enormous and prolonged investment into preventing and Urban green and blue spaces are often attractive and accessible treating cardiovascular diseases, ischaemic heart disease and stroke remain the leading causes of disease, disability and death world- settings to visit regularly for a wide range of social and physical wide.1 At the same time, according to some reports, the high global burden of musculoskeletal conditions is the second largest cause of activities. They provide natural outdoor fora, with cooler, fresher air; years lived with disability,2 while also contributing to chronic pain, physical inactivity, obesity and cardiometabolic conditions.3 Physio- this can mitigate the harms of heat and air pollution, which are therapists, who are at the forefront of tackling these diseases and supporting people with disability, are calling for healthcare delivery common in cities. Urban green and blue spaces also promote recre- to be more preventative, equitable and pro-environmental.4 This is ational experiences that relieve stress by helping us to feel ‘away’.13 due to the need to promote physical activity as a key evidence-based They can also help to restore depleted cognitive capacities; nature treatment across all of the aforementioned health conditions, despite their varied aetiologies. is appealing enough to attract and hold a person’s attention, thereby enabling rest of the neurocognitive mechanism by not using it for When attending for a specific clinical complaint, patients expect effortful directed attention.13 In this way, a person may reap some physiotherapists to also provide general health and physical activity advice to increase fitness levels.5 This typically involves the mental and cognitive health benefits simply from having views of prescription of an unsupervised exercise program, which (in addition trees and other green spaces from their window. to targeting overall fitness) may also address specific signs and symptoms. The success of these exercise programs is contingent on Qualitative research indicates that people like to engage with adherence, which is commonly suboptimal for many reasons. Adherence to treatment is poor when the home exercise programs nature for multiple reasons. Some point to the non-judgemental, are perceived as being complex, time consuming and interrupting ego-free and dependable support that nature provides.14 Others daily routine.6 emphasise parks and other green or blue areas as free-to-enter People are more likely to implement and maintain lifestyle change if the prescribed change aligns with their intrinsic motivations7 and settings, invested with individual and shared meaning that sup- can be conducted in settings that empower them to adopt the new port (re)connection with community.15 Exploratory survey-based behaviour. Nature’s green spaces (eg, parks and hills) and blue spaces (eg, lakes, rivers and beaches) offer a largely underutilised, low- evidence indicates that these experiences and the biopsychosocial (or no) cost opportunity in physiotherapy that can both attend to processes they support may ameliorate chronic pain.16 Cohort people’s interests and provide attractive settings for physical activity8 studies drawing upon the domains of pathways conceptual model13 to aid recovery and resilience in those with musculoskeletal, neuro- logical or cardiovascular health issues. The flocking of individuals to indicate that having 30% green space within a 1.6 km walk from green and blue spaces as a means of coping with the seismic impacts home may help to catalyse healthier sleep durations17 and enable of the COVID-19 pandemic9 has highlighted the size of the opportu- nity that exists. people to connect with their neighbours in ways that reduce chronic loneliness.18 The benefits of nature are not only for physical activity. John Muir, the Scottish-American naturalist and conservationist, wrote: This work has been considered compelling by some city admin- istrators around the world, with several cities including Barcelona,19 in every walk with Nature one receives far more than [s]he seeks.10 Seattle,20 Sydney21 and Vancouver22 setting 30% greening targets (ie, the percentage of total land use reserved for green space or Much research supports this; for instance, cohort studies report that populations with 30% tree canopy cover within 1.6 km of home, as particular types of green space, such as tree canopy cover). Such compared to peers with , 10%, have lower risks of cardiovascular diseases, hypertension, diabetes and dementia onset.11,12 actions are helping to fix the supply side of the equation; it may be less simple to ensure that those with the highest potential to benefit do so, given that it is likely that those in better health and wealth spend more of their discretionary time in nature. To address this, contemporary and national programs in Canada (BC Parks Foundation ‘PaRx’) and the US (‘ParkRx’) operate to support family doctors who recommend to their patients to spend 2 hours a week in nature. These ‘nature prescriptions’ (or ‘nature-based social prescriptions’) typically involve goal setting and provision of educa- tional materials, and may include a referral to a designated nature space(s) facilitated by an organisation or community volunteer group. Preliminary findings from randomised trials indicate that such pro- grams can increase physical activity and reduce depression, anxiety and blood pressure.23 https://doi.org/10.1016/j.jphys.2022.05.016 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/).
152 Editorial Physiotherapists may consider integrating nature prescriptions thumb.31 However, caution is needed in how such prescriptions are into standard care and/or as an adjunct therapy. For instance, the New written and delivered, given the evidence that perceived social pressure South Wales health department is offering the free ‘Get Healthy’ to visit nature may reduce the level of benefits accrued by increasing service, which uses health coaching to increase participants’ physical anxiety and attenuating intrinsic motivation.32 activity levels. Although this service is not specific to any setting, it could encourage visits to nature settings through its application of A robust, evidence-based system is needed to connect people with motivational interviewing. Nature prescriptions might also be inte- nature who currently have little time in it. This needs to be achieved grated within face-to-face therapies offered by physiotherapists, with in accessible, affordable, acceptable and effective ways that attend to emerging work highlighting nature’s role as a non-pharmaceutical individual differences in motivations and needs. Physiotherapists can means of supporting rehabilitation from acute myocardial infarc- and should play a key role in making contact with nature the fourth tion24 and heart surgery.25 pillar of health, alongside physical activity, diet and sleep. Despite the promising findings and supportive initiatives outlined Ethics approval: Not applicable. above, nature remains grossly underutilised as a motivator for Competing interests: Nil. behaviour change, particularly in relation to physical activity. A recent Source of support: Nil. scoping review identified only eight studies where healthcare pro- Acknowledgements: The authors thank the peer reviewers for fessionals included a ‘nature prescription’ for supporting self- providing thoughtful and constructive reviews. management of people living with long-term health conditions.26 In Provenance: Not invited. Peer reviewed. studies of interventions involving people with long-term health Correspondence: Thomas Astell-Burt, Population Wellbeing and conditions spending time in parks, gardens, forest/woodlands and Environment Research Lab (PowerLab), Australia. Email: wetlands, the authors reported improvements in the participants’ [email protected] psychological, physical and social wellbeing. References Such innovation within physiotherapy should ideally be led by a dedicated new stream of co-designed and practice-based experi- 1. Roth GA, et al. J Am Coll Cardiol. 2020;76:2982–3021. mental research to define what nature prescriptions are 2. Sebbag E, et al. Ann Rheum Dis. 2019;78:844–848. cost-effective, sustainable and acceptable to both consumers and 3. de Oliveira FCL, et al. Knee Surg Sports Traumatol Arthrosc. 2020;28:667–669. providers addressing specific health conditions. This work might bind 4. Maric F, Nicholls D. Physiother Theory Pract. 2009;35:905–907. onto existing interventions that already show some level of benefit 5. Kunstler B, et al. J Physiother. 2019;65:230–236. for the purposes of improving retention and strengthening health. 6. Escolar-Reina P, et al. BMC Health Serv Res. 2010;10:1–8. Other research avenues might focus on testing customised therapies 7. Ryan RM, Deci EL. Contemp Educ Psychol. 2000;25:54–67. for different conditions aligned with current ‘forest bathing’ practices 8. Feng X, et al. Urban For Urban Green. 2021:127349. common to parts of East Asia (eg, ‘shinrin yoku’).27 9. Astell-Burt T, Feng X. Int J Environ Res Public Health. 2021;18:2757. 10. Muir J. Chapter 9: Mormon Lilies (Letter dated “Salt Lake, July, 1877”). 1877. https:// Harnessing nature may help to address some common challenges in physiotherapy. For instance, recent studies indicate that only 19% vault.sierraclub.org/john_muir_exhibit/writings/steep_trails/chapter_9.aspx. Accessed and 30% of female and male survivors of acute coronary syndrome 6 May, 2022. seek cardiac rehabilitation, respectively.28 Barriers to physical activity 11. Astell-Burt T, Feng X. Int J Epidemiol. 2020;49:926–933. for cardiac rehabilitation are multidimensional and range from low 12. Astell-Burt T, et al. Environ Int. 2020;145:106102. motivation to negative prior experiences.29 For many people, inte- 13. Markevych I, et al. Environ Res. 2017;158:301–317. grating nature time into their rehabilitation may resonate with their 14. Birch J, et al. Health & Place. 2020:102296. intrinsic motivations for being connected with the ‘more than human 15. Graham TM, Glover TD. Leis Sci. 2014;36:217–234. world’, making physical activity an adjunct benefit of something they 16. Selby S, et al. Ir J Med Sci. 2019;188(3):973–978. enjoy doing without the need for extrinsic incentive to fulfil. 17. Astell-Burt T, Feng X. SSM-Popul Health. 2020:100497. 18. Astell-Burt T, et al. Int J Epidemiol. 2022;51:99–110. Enabling and empowering people to visit green and blue spaces 19. City of Barcelona. Trees for Life: Master Plan for Barcelona’s Trees 2017 – 2037; regularly after coronary artery bypass surgery, for example, has the 2021. https://www.c40knowledgehub.org/s/article/Trees-for-Life-Master-Plan-for- potential to provide differentially greater benefit to those from lower Barcelona-s-Trees-2017-2037?language=en_US. Accessed 13 March, 2021. socioeconomic circumstances.25 Harnessing nature in this way can 20. City of Seattle. Trees For Seattle; 2021. https://www.seattle.gov/trees/ lay foundations for more health equity and sustained behavioural management/canopy-cover. Accessed 13 March, 2021. change that extends beyond the clinical environment to everyday life 21. City of Sydney. Greening Sydney Strategy; 2021. https://www.cityofsydney.nsw. and routine, which might not otherwise be achieved in tightly gov.au/strategies-action-plans/greening-sydney-strategy. Accessed 12 October, controlled indoor settings. 2021. 22. City of Vancouver. Park Board achieves target to plant 150,000 trees by 2020. 2021. In a recent nationally representative survey of Australian adults https://vancouver.ca/news-calendar/park-board-achieves-target-to-plant-150000- aged . 18 years,30 an estimated 81% (95% CI 79 to 83) indicated being trees-by-2020.aspx. Accessed 13 March, 2021. likely to accept a ‘nature prescription’ from a professional if it was 23. Nguyen P-Y, et al. medRxiv. 2022. recommended in support of their health. Such prescriptions might 24. Patel DM, et al. Indoor Built Environ. 2019;28:1431–1440. recommend spending 2 hours a week in nature as a general rule of 25. Sadeh M, et al. Environ Res. 2022:113364. 26. Buckley A, et al. Phys Ther Rev. 2020;25:399–410. 27. Wen Y, et al. Environ Health Prev Med. 2019;24:1–21. 28. Redfern J, et al. Eur Heart J. 2020;41(Suppl 2):ehaa946. 3108. 29. Bäck M, et al. BMC Cardiovasc Disord. 2017;17:1–10. 30. Astell-Burt T, Feng X. Int J Epidemiol. 2021;51:1–5. 31. White MP, et al. Sci Rep. 2019;9:1–11. 32. Tester-Jones M, et al. Sci Rep. 2020;10:19408.
Journal of Physiotherapy 68 (2022) 158–164 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 Physiotherapy management of interstitial lung disease Anne E Holland Departments of Physiotherapy and Respiratory Medicine, Alfred Health, Melbourne, Australia; Central Clinical School, Monash University, Melbourne, Australia; Institute for Breathing and Sleep, Melbourne, Australia KEY WORDS Interstitial lung disease Physical therapy Restrictive lung disease Pulmonary rehabilitation Introduction Burden of interstitial lung disease The interstitial lung diseases (ILDs) are a group of over 200 chronic Although each of the ILDs is relatively rare, collectively the ILDs lung conditions characterised by dyspnoea on exertion, troublesome cough, exercise intolerance, and poor health-related quality of life. affect a large number of people across the world. The Global Burden Although ILDs vary in underlying diagnoses and clinical course, of Disease Study estimated the prevalence of ILD at 82/100,000 comprehensive supportive care is considered critical across all ILD people.5 The most common fibrotic ILD is idiopathic pulmonary subtypes to optimise clinical outcomes and patient wellbeing. Key fibrosis (IPF), which is a progressive condition that occurs most elements of supportive care for ILD include pulmonary rehabilitation, commonly in men aged . 65 years and confers a poor prognosis. supplemental oxygen, education, psychosocial support, symptom Five-year survival for people with IPF has been estimated at 46% (95% management, and end-of life care (Figure 1).1 This review summa- CI 42 to 50),6 contributing 0.26% of global all-cause mortality.5 rises: the classification and causes of ILD; the burden of this condition Recently it has been recognised that a substantial proportion of for individuals and the health system; the clinical features of ILD; the key elements of comprehensive ILD care focusing on interventions those with non-IPF fibrotic ILDs also exhibit a progressive phenotype delivered by physiotherapists; and future directions for research and known as progressive-fibrosing ILD (PF-ILD) with a similar mortality practice. risk to IPF.7 Frequent diagnoses in those with PF-ILD include chronic What is interstitial lung disease? hypersensitivity pneumonitis, autoimmune ILDs, sarcoidosis and idiopathic nonspecific interstitial pneumonia. Other ILDs in which The ILDs are a group of restrictive lung conditions characterised by inflammation is more prominent than fibrosis (eg, acute hypersen- inflammation and/or fibrosis of the lung tissue. In most of these sitivity pneumonitis, cryptogenic organising pneumonia, some con- conditions the pathologic abnormalities occur predominantly in the lung interstitium, which is the connective tissue framework sur- nective tissue diseases) have a less progressive course and better rounding the alveoli, airways and blood vessels. The ILDs can be survival.8 grouped into five broad clinical categories (Figure 2), depending on whether there is a known cause such as an underlying disease pro- The ILDs are associated with a high symptom burden, including cess; exposure to environmental toxins; exposure to radiation or dyspnoea (54 to 98% of patients) and a chronic dry cough (59 to 100% drugs known to cause ILD; presence of a connective tissue disease; or of patients).9 Fatigue and exhaustion may be more bothersome than if the cause is unknown.2 Recently, development of pulmonary dyspnoea for some patients.10 Impairments in health-related quality fibrosis after coronavirus 2019 (COVID-19) infection has been re- ported, most commonly in those who have undergone mechanical of life are substantial, including levels of distress related to breath- ventilation.3 Persistent radiographic changes in the lungs have also lessness that are frequently higher than in people with chronic been demonstrated in survivors of previous coronavirus pandemics, the severe acute respiratory distress syndrome (SARS) and Middle obstructive pulmonary disease (COPD), fatigue worse than in heart East respiratory syndrome (MERS), although these were generally failure, depression comparable to people with a major depressive mild and stable over time.4 Little is known about the clinical course of pulmonary fibrosis in COVID-19 survivors. disorder, and sleep disturbance similar to those with obstructive sleep apnoea.11 The physical, emotional, social and financial burdens experienced by caregivers of people with ILD are increasingly being recognised, particularly as their loved one’s disease becomes more severe.12 The direct costs of ILD to the health system are substantial and rising. A recent systematic review reported an annual median cost of US$32,384 per patient (range $1,824 to $116,927 per patient), with significantly increased costs since 2014 after new anti-fibrotic https://doi.org/10.1016/j.jphys.2022.06.006 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/).
Research 159 End-of-life care Supportive measures carbon monoxide. Reduced functional exercise capacity is often • Timely end-of-life • Supplemental oxygen demonstrated on 6-minute walk test, and marked exercise-induced • Pulmonary rehabilitation desaturation may also be present. Measures of respiratory function conversation • Education and exercise capacity are repeated at regular intervals to track pro- • Treatment limitations • Psychological support gression of disease over time. Development of hypoxaemia, either at • Preferred place of death rest or on exertion, confers a poor prognosis.16 Comprehensive supportive care Comorbidities are common in people with ILD, including cardio- through partnership between the vascular disease, lung cancer, obstructive sleep apnoea, gastro- patient, caregiver and healthcare oesophageal reflux, pulmonary hypertension and depression. Comorbidities adversely affect mortality and health-related quality of team life in ILD, and thus it is important that these are identified and optimally treated. Musculoskeletal dysfunction is also prevalent in Symptom relief Disease-modifying treatment ILD. Quadriceps strength and endurance were found to be reduced by • Dyspnoea • Antifibrotic therapies 20 to 30% in people with ILD compared with healthy controls17 and • Fatigue • Immunomodulatory strongly related to impaired exercise capacity. People with ILD took • Cough 65% fewer daily steps compared with their healthy age-matched • Anxiety and depression therapies peers,18 and those with the lowest levels of physical activity had • Trial options the worst health-related quality of life, independent of respiratory function.19 There was a threefold increase in the hazard of death for Figure 1. Comprehensive supportive care for patients with interstitial lung disease. patients with ILD who were highly sedentary (, 3,300 steps/day).19 Adapted from Wijsenbeek et al.1 Arthropathy is common in people with ILDs that are related to con- nective tissue disease (eg, rheumatoid arthritis, systemic sclerosis) medications for IPF were introduced into practice.13 In the USA, and may have a profound impact on functional mobility and physical annual direct medical costs for patients with IPF were estimated to be activity levels. twofold higher than age-matched and gender-matched controls in the year following IPF diagnosis.14 The main contributors to health- An acute exacerbation of IPF is defined as an acute, clinically care costs were inpatient (55%), outpatient (22%) and medication significant respiratory deterioration with evidence of new wide- costs (18%); costs varied substantially across healthcare systems. spread alveolar abnormality on high-resolution computed tomogra- There are few data on indirect costs, although work productivity loss phy. Acute exacerbations occur in 5 to 10% of patients with IPF in fibrotic ILD has been estimated at US$9,313 to $10,902 per patient annually.20 An acute exacerbation of IPF is a life-threatening event, per annum.13 Costs for work productivity loss amongst caregivers are with in-hospital mortality of . 50%, and median survival following an unavailable, but are likely to be significant. exacerbation of approximately 4 months.20 It is now recognised that acute exacerbations also occur in non-IPF ILDs, although the prog- Clinical features of interstitial lung disease nosis may be more favourable. Amongst 102 patients with an acute exacerbation, 90-day mortality was 55% in those with IPF, compared People with ILD generally present with breathlessness and/or with 31% in those with non-IPF ILDs.21 Care needs in survivors are cough, often resulting in reduced exercise tolerance. A high- substantial, with 63% of those with IPF and 41% of those with non-IPF resolution computed tomography of the chest reveals radiological ILDs requiring long-term oxygen therapy at hospital discharge. features of ILD, which vary according to the ILD subtype. The diag- nosis of an ILD is a complex process that requires integration of Physiotherapy management of ILD clinical, radiological and histopathological data, frequently in a specialist multidisciplinary meeting.15 Achieving an accurate diag- The approval of anti-fibrotic drugs for people with IPF (pirfeni- nosis of the ILD subtype is important, as it determines eligibility for done and nintedanib) was a watershed moment, when IPF became a specific pharmacotherapies. Respiratory function tests typically treatable disease for the first time.22,23 The recent identification of the reveal reductions in forced vital capacity and diffusing capacity for PF-ILD phenotype is also highly significant, as clinical trials have demonstrated that these patients have a similar response to anti- ILD related to an underlying disease fibrotic treatments as is seen in IPF.7 However, anti-fibrotic thera- pies only slow the progression of lung disease; they do not reverse eg, sarcoidosis, lymphangioleiomyomatosis, pulmonary alveolar the changes that have already occurred. In addition, there is no proteinosis convincing evidence that these therapies improve symptoms or quality of life.22,23 As a result, people with IPF and PF-ILD will likely ILD related to environmental exposures live longer with a chronic respiratory disease that has a high burden of symptoms and disability. Physiotherapists make a critical contri- eg, pneumoconiosis including asbestosis and silicosis; bution to the comprehensive management of people with ILD, with hypersensitivity pneumonitis related to inhalation of organic particles key treatment goals including reducing symptoms, improving exer- cise capacity, enhancing self-management and improving health- ILD from exposure to drugs or radiation related quality of life. eg, some cancer and rheumatology drugs, radiotherapy ILD associated with connective tissue disease Pulmonary rehabilitation eg, rheumatoid arthritis and systemic sclerosis There is robust evidence that pulmonary rehabilitation delivers important benefits for people with ILD. A recent Cochrane review ILD without a known cause included 21 randomised controlled trials, of which 16 were included in a meta-analysis.24 Outcomes of functional exercise capacity, eg, idiopathic pulmonary fibrosis (IPF), nonspecific interstitial pneumonia dyspnoea and health-related quality of life are in Table 1. Improve- (NSIP) ment in 6-minute walk distance (6MWD) following pulmonary rehabilitation was clinically significant (mean difference to usual care Figure 2. Clinical categories of interstitial lung diseases (ILDs). Adapted from 40 m), with the lower end of the confidence interval exceeding the Wijsenbeek et al.2 minimum important difference. Results were similar in the subgroup of participants with IPF, providing reassurance that gains can be made
160 Holland: Physiotherapy management of interstitial lung disease Table 1 Changes in functional exercise capacity, dyspnoea and health-related quality of life following pulmonary rehabilitation. From Dowman et al.24 Clinical measures All interstitial lung disease Idiopathic pulmonary fibrosis only n mean (95% CI) n mean (95% CI) 6-minute walk distance immediately after PR (m) 585 40 (33 to 47) 278 37 (26 to 48) 6-minute walk distance at 6 to 12-month follow-up (m) 297 32 (16 to 49) 123 2 (–25 to 28) Dyspnoea immediately after PR (MMRC scale, SMD) 348 –0.36 (–0.58 to –0.14) 155 –0.41 (–0.74 to –0.09) Dyspnoea at 6 to 12-month follow-up (MMRC scale, SMD) 335 –0.29 (–0.49 to –0.10) 123 –0.38 (–0.72 to –0.05) SGRQ total score immediately after PR 478 –9 (–11 to –8) 194 –8 (–11 to –5) SGRQ total score at 6 to 12-month follow-up 240 –5 (–8 to –2) 89 –3 (–7 to 1) Data are change in outcome in randomised controlled trials of pulmonary rehabilitation compared to usual care. ILD = interstitial lung disease, IPF = idiopathic pulmonary fibrosis, MMRC = modified Medical Research Council, PR = pulmonary rehabilitation, SGRQ = St George’s Respiratory Questionnaire, SMD = standardised mean difference. in this group of patients who often have progressive disease. Delivery of an effective dose of exercise training can be particularly Important improvements in dyspnoea and health-related quality of challenging in those who experience profound exercise-induced life were also evident, both for all participants with ILD and in the IPF desaturation. Interval training may be a useful strategy with which subgroup. Importantly, the updated Cochrane review reported to attenuate exertional desaturation in ILD. In nine people with ILD persistent benefits of pulmonary rehabilitation for the first time. (mean forced vital capacity 66% predicted), oxyhaemoglobin satura- Those who undertook pulmonary rehabilitation had a higher 6MWD, tion (SpO2) was higher after 10 minutes of high-intensity interval lower dyspnoea and better health-related quality of life 6 to 12 training (30-second intervals, 100% peak work and passive rest) months following program completion. Within-group changes compared with moderate-intensity interval training (2-minute in- showed that these benefits were not necessarily ‘maintained’ at the tervals, 80% peak work and 40% peak work) and moderate-intensity same level in rehabilitation participants, but the control group had a continuous training (60% peak work); however, exertional desatura- more substantial deterioration over time, resulting in a sustained tion was not abolished (mean 87% versus 85% and 84%, respec- between-group difference. tively).28 Oxygen supplementation during training is also critical for many patients with ILD, but standard methods may be insufficient to Only four randomised trials have examined the impact of pul- correct SpO2 during exercise. Recent studies have shown that high- monary rehabilitation on survival, with follow-up periods ranging flow nasal cannula oxygen therapy increases exercise time, im- from 6 to 12 months. Whilst the results tended to favour pulmonary proves SpO2 and reduces leg fatigue during exercise in IPF.29 It is not rehabilitation, this estimate came with substantial uncertainty (OR yet known whether these strategies improve pulmonary rehabilita- for mortality 0.40, 95% CI 0.14 to 1.12, 291 participants).24 A retro- tion outcomes. spective cohort study including 701 participants from five countries reported that a larger improvement in 6MWD following pulmonary Non-exercise components of pulmonary rehabilitation have rehabilitation was associated with a lower hazard ratio for death or included education, nutritional advice, stress management, breathing lung transplant for both inpatient pulmonary rehabilitation (hazard exercises, occupational therapy and social support.24 Key educational ratio (HR) per 10 m = 0.94, 95% CI 0.91 to 0.97) and outpatient pul- topics for inclusion in pulmonary rehabilitation for ILD have been monary rehabilitation (HR 0.97, 95% CI 0.95 to 1.00).25 Participation in identified by patients and health professionals, using a consensus 80% of sessions was associated with a 33% lower risk of death (HR approach (Box 1).30 Some of these topics require clinicians to have 0.67, 95% CI 0.49 to 0.92). Causality cannot be established by this ILD-specific knowledge and expertise (eg, managing medication side- uncontrolled study. effects, managing cough). People with ILD were found to be generally comfortable attending pulmonary rehabilitation programs that There is no evidence that those with more severe IPF cannot included other people with a variety of lung conditions, but expected benefit from pulmonary rehabilitation, but it is apparent that more ILD-specific education to be provided.31 sustained benefits are observed earlier in the disease. Pulmonary function at enrolment (forced vital capacity, diffusion capacity for Despite the known benefits of pulmonary rehabilitation, many carbon monoxide) was found to not predict response to pulmonary patients do not have the opportunity to undertake a program. rehabilitation.26 Those with lower baseline 6MWD and worse dysp- Referral rates have been reported as ranging from 20 to 40% of eligible noea had the greatest improvement at the end of the program, patients32,33 which, although higher than rates commonly reported in consistent with the critical role of pulmonary rehabilitation in other lung diseases, is far from optimal, considering that this treat- improving functional capacity and reducing symptoms. However, ment is recommended in clinical practice guidelines across the world. those with better physiology were more likely to have sustained People with ILD face similar barriers to attending pulmonary reha- benefits 6 months after program completion, both for pulmonary bilitation, as reported in COPD, including fear of exercise, debilitating function (each 100-ml increase in forced vital capacity was associated symptoms and the burden of travel to the pulmonary rehabilitation with 2 m greater 6MWD) and pulmonary hypertension (each 10- centre.33 Delivery of pulmonary rehabilitation directly into the home mmHg decrease in pulmonary artery systolic pressure was associ- using technology appears to be safe and may be beneficial to improve ated with 15 m greater 6MWD).26 These results suggest that referral pulmonary rehabilitation access and outcomes for people with ILD, to pulmonary rehabilitation early in the disease course should be although few patients with ILD have been included in randomised encouraged. trials to date.34 Components of pulmonary rehabilitation programs for people Musculoskeletal care with ILD have been found to be similar to those for people with other lung diseases, including COPD.24 Inpatient and outpatient pulmonary People with ILD related to underlying connective tissue disease rehabilitation programs have been found to be effective,25 with the frequently have arthropathy, which may impact physical function. model often determined by healthcare system organisation. Program Lung involvement is the most common extra-articular manifestation length varied but was most commonly 8 to 12 weeks of outpatient of rheumatoid arthritis, occurring in up to 60% of patients across the sessions (two to three sessions per week). Exercise components disease course.35 As a result, a subgroup of patients with ILD who included aerobic and resistance training. Unsurprisingly, greater present for rehabilitation may have limitations to whole-body exer- benefits were achieved by participants in whom consistent progres- cise related to joint disease, with associated reductions in range of sion of exercise intensity and duration occurred over the course of the motion, pain and disability. It is critical that optimal musculoskeletal program.26 This is easier to achieve in those who attend a larger number of sessions and those who have less progressive disease.27
Research 161 Box 1. Education topics for pulmonary rehabilitation in ambulatory oxygen therapy, including cumbersome and complicated interstitial lung disease. From Holland et al.30 equipment, perceived stigma, unmet expectations for symptom relief (particularly dyspnoea), reduced independence, and increased care- Core education topics giver burden.36 As a result of these potential burdens some patients Staying well with ILD: managing flare ups, regular may choose not to use ambulatory oxygen therapy, and shared decision-making between patients and healthcare professionals is vaccinations, importance of exercise, nutrition required. The availability of ambulatory oxygen and the criteria for its Keeping fit and strong after pulmonary rehabilitation prescription vary widely across jurisdictions, partly reflecting the Using oxygen therapy limited evidence of benefit underpinning its use. Managing breathlessness and cough Managing fatigue Long-term oxygen therapy is typically delivered using a stationary Managing anxiety, depression and panic oxygen concentrator in the home. Ambulatory oxygen therapy can be Optional education topicsa delivered using a variety of portable devices with different charac- End-of-life care and advance directives teristics and costs. Metal oxygen cylinders are available in multiple Accessing support for patients and carers sizes and are relatively low in cost. A typical oxygen tank that is Managing medications and side effects pulled on a trolley (E tank in the USA) lasts less than 2 hours if high Tuning up the whole system – managing co-existing medical flow rates (6 l/minute) are required, which will limit ambulation outside the home. Oxygen conserving devices may be used to prolong conditions the duration of supply, but often do not deliver a sufficient oxygen dose for those with higher requirements such as in ILD. Multiple a To be delivered if local resources allow. tanks may be required for patients requiring flow rates . 3 l/minute in order to spend more than 2 to 4 hours away from home. Portable management is delivered, to facilitate participation in rehabilitation concentrators are battery powered (and thus do not require refilling) and regular physical activity. This may include specific muscle and generally lighter than metal cylinders. However, portable con- strengthening exercises, exercise to address functional limitations, centrators are expensive (up to US$4,000); the pulsed dose of oxygen modifications to aerobic training to reduce joint loading, and edu- delivered may be insufficient for patients with higher oxygen re- cation regarding self-management of fatigue, pain and stress. Close quirements; and the battery runs out more quickly at higher settings collaboration between physiotherapists with expertise in rheuma- and respiratory rates. Neither delivery system fully corrects oxy- tology, orthopaedics and respiratory management may be required. haemoglobin saturation on exertion in ILD, even on maximal set- tings.38 Portable liquid oxygen allows delivery of continuous flow Supplemental oxygen therapy oxygen up to 15 l/minute, and enables a longer duration of use. However portable liquid oxygen is not available in many jurisdictions Hypoxaemia occurring during rest, sleep or exercise is a cardinal due to cost (up to four times higher per patient compared with cyl- feature of the ILDs. With the exception of lung transplantation, sup- inders or portable concentrators). The increased cost for liquid oxy- plemental oxygen is the only treatment that improves hypoxaemia gen is primarily attributable to the need for specialised delivery that persists despite optimal medical management.36 The principles vehicles and frequent refilling of home reservoirs. underlying delivery of supplemental oxygen are similar in ILD and other chronic lung diseases; however, the greater magnitude of Optimisation of oxygen therapy is a key role for physiotherapists hypoxaemia (and hence greater oxygen requirements) experienced in ILD care. Patients should be prescribed the ambulatory oxygen by people with ILD often poses unique challenges for delivery. delivery device that best meets their physical, physiological and lifestyle needs, and this should be re-evaluated as disease progresses. The American Thoracic Society (ATS) clinical practice guideline on For instance, some patients who have lower oxygen requirements or home oxygen therapy for adults with chronic lung disease36 provides are very active may be best served by small, lightweight cylinders or a a strong recommendation for long-term oxygen therapy for at least portable concentrator that can be carried in a backpack. Disease 15 hours/day in people with ILD who have severe chronic resting progression and increased oxygen requirements may require patients room air hypoxaemia (PaO2 55 mmHg). The recommendation is to transition from portable concentrators to metal cylinders, or even based on indirect evidence from clinical trials in COPD demonstrating multiple cylinders. Consideration of how the ambulatory oxygen a mortality benefit, and ethical concerns about withholding oxygen device can best be transported by patients and caregivers is critical, from patients with ILD who may be profoundly hypoxaemic and including use of trolleys and wheeled walkers. Multiple stationary dyspnoeic at rest. Long-term oxygen therapy is a well-accepted concentrators are sometimes needed to meet the oxygen re- treatment for people with advanced ILD and is commonly available quirements of patients on long-term oxygen therapy with very in jurisdictions with supplemental oxygen programs. There are few advanced disease. data on the clinical implications of hypoxaemia occurring only during sleep, or the impact of its treatment; as a result, isolated sleep- Patients frequently report a lack of information about how to use induced hypoxaemia is not generally considered to be an indication their oxygen equipment and have low confidence in their skills.39 All for oxygen therapy in ILD. patients and caregivers should receive instruction and training in the safe and effective use of their oxygen equipment.36 Important con- For patients who experience isolated exertional hypoxemia (SpO2 siderations for safety, education, training and monitoring in patients 88%) the ATS guideline makes a conditional recommendation for prescribed oxygen therapy are presented in Table 2. treatment with ambulatory oxygen, based on low-quality evidence.36 Most studies that underpin the recommendation evaluated the acute Education, self-management and support effects of oxygen on exercise performance during a single session in the laboratory, showing consistent improvements in 6MWD and ex- Supportive measures, including education and psychological ercise endurance, and reduced dyspnoea during exercise testing. support, are key components of comprehensive ILD care (Figure 1)1 However, it is unclear whether these acute effects of ambulatory and are a high priority for patients. In a systematic review of sup- oxygen are translated into beneficial effects during daily life. A portive care needs in pulmonary fibrosis,40 including data from 2,621 crossover randomised trial with a 2-week treatment period participants and 590 caregivers, the need for more information and compared ambulatory oxygen with no treatment in participants with education was reported in 26 of 35 studies. Specific information fibrotic ILD and showed a significant improvement in the primary needs reported by participants included understanding disease pro- outcome of health-related quality of life.37 However, participants and gression and prognosis; oxygen therapy, including travel with oxy- assessors were not blinded to treatment allocation, and longer-term gen; managing side-effects of drug therapies; planning for end-of-life effects could not be evaluated. The conditional recommendation for care; coping strategies; and managing symptoms of breathlessness this treatment acknowledges the potential for negative impacts of
162 Holland: Physiotherapy management of interstitial lung disease Table 2 Considerations for safety, education, training and monitoring in patients prescribed oxygen therapy. Adapted from Jacobs et al.36 Category Considerations Safety Education regarding avoidance of trips and falls; decreasing fire risk by not smoking or allowing smoking in the home; avoidance of open flames or sparks; use of nonpetroleum nasal products Smoking Education and Instruct liquid-oxygen users on avoidance of skin burns from contact with frosted parts on liquid-oxygen-device connectors training Provide guidance on transporting and travelling safely with oxygen Monitoring Confirm the presence of back-up devices for emergencies or power loss Instruct current smokers or caregiver smokers on smoking cessation and treatment of tobacco dependence; refer to appropriate resources Alert patients and caregivers that use of e-cigarettes, or vaping, is associated with burn accidents in people receiving home oxygen therapy Tailor patients’ education to their health literacy and cultural contexts Incorporate effective evaluation and return demonstration of their ability to use their prescribed devices both in the home and in ambulatory settings Instruct patients and caregivers on troubleshooting equipment problems Consider access to appropriate equipment on the basis of patients’ physical, physiologic, and lifestyle/mobility needs Reassess patients’ oxygen needs, with a frequency varying according to disease characteristics, such as rate of progression. Reassess oxygen needs for patients who are newly prescribed oxygen after hospital discharge, to confirm ongoing oxygen requirements Advise patients to bring their portable device to healthcare visits to assess its effectiveness and to reinforce self-management and cough. Physiotherapists may address some of these informational included in recent clinical trials are in Box 2, many of which are needs in pulmonary rehabilitation programs, but there may be other within the scope of practice of physiotherapists. Many of these opportunities including outpatient clinics, oxygen therapy clinics, components aim to reduce breathlessness, but strategies to manage support groups, patient seminars, and in the inpatient setting. fatigue, cough and anxiety are also important. Self-management is a relatively new concept in ILD care, driven by Cough is a particularly distressing symptom for many people with changes to patient and healthcare professional expectations of ILD. Whilst there is little research to guide physiotherapy manage- treatments and outcomes in the anti-fibrotic era. Whereas previously ment of cough, strategies that may be helpful include avoidance of it was assumed that patients with PF-ILD and IPF would face inexo- cough triggers (eg, dusty environments, animal fur, cleaning products, rable disease progression and death, with few treatment options or perfumes); regularly sipping water and maintaining good hydration; opportunities to be actively involved in their care, this is no longer using an air humidifier at home; use of non-medicated lozenges; nose the case. The focus of care has shifted to supporting individuals to live breathing to warm and moisten inspired air; controlled breathing well with ILD, including strategies that can be employed by patients exercises, including during exercise if this is a trigger; mindfulness to maintain their health and wellbeing over time. Key elements of and distraction techniques; regular swallowing and avoidance of self-management for ILD identified by patients and healthcare pro- throat clearing; and airway clearance techniques if a productive fessionals include exercise, physical activity, nutrition, weight man- cough is present. agement, regular vaccinations, avoiding infections, recognising deterioration, seeking help, managing symptoms, managing treat- Inpatient care ments, managing treatment side effects, and maintaining mental health.41 The impact of self-management on clinical outcomes in ILD Physiotherapy management for inpatients with ILD may be is not yet known. required if the admission is for management of an acute exacerbation, although treatment options are limited. There are no proven thera- In addition to the support provided by healthcare professionals, pies for acute exacerbations of ILD, with most centres delivering high- patients with ILD value opportunities for peer support. The ILDs are dose corticosteroids and antibiotics, as well as respiratory support relatively rare conditions, so many individuals will not have met (oxygen therapy, non-invasive or invasive mechanical ventilation).47 another person with the same diagnosis. Peer support programs are Marked exercise limitation and increased long-term oxygen re- most commonly offered by patient organisations, often online or via quirements are common in those who survive acute exacerbations, telephone, and provide unique opportunities to share experiences and these patients may benefit from physiotherapy management to and offer mutual support.42 Linking individuals with ILD with patient optimise function and independence. organisations also provides a mechanism by which they can access patient-centred, up-to-date information to improve disease knowl- Transplant preparation and rehabilitation edge and self-management. Symptom management and palliative care Lung transplantation is a life-saving and life-prolonging procedure for people with end-stage fibrosing ILD. Recent years have seen an Symptom management and palliative care aim to improve health- increase in the proportion of lung transplants performed for ILD,48 with related quality of life and support individuals to live well with ILD. a concomitant increase in the number of recipients requiring long-term These interventions should not be restricted to end-of-life care, but management. Physiotherapists play a key role in optimal management are relevant across the disease course to relieve suffering and of patients with ILD before and after lung transplantation. enhance wellbeing.43 Whilst the evidence underpinning symptom management approaches is limited, a small number of randomised Physical rehabilitation is considered a core component of prepara- trials suggest that there may be important benefits for people with tion for lung transplantation and often continues throughout the ILD. In 105 people with refractory breathlessness, 19 of whom had waiting period. Standard approaches to pulmonary rehabilitation can ILD, a multidisciplinary breathlessness support service significantly be applied, with modifications for disease severity, disease progression, improved survival at 6 months compared with usual care.44 In 53 functional deterioration and comorbidities.49 There is some evidence patients with advanced fibrotic ILD, a community case conference to that candidates who underwent rehabilitation had improvements in address palliative care needs improved symptoms and health-related 6MWD and health-related quality of life, as well as reduced hospital quality of life after 4 weeks compared with usual care.45 An inter- length of stay and improved survival following transplantation in pa- vention designed to reduce symptom burden and improve health- tients with IPF;50 however, this is largely from uncontrolled studies. related quality of life in patient-caregiver dyads (n = 76) improved Rehabilitation also provides an opportunity for pre-transplant educa- knowledge, confidence in managing disease and completion of tion, including preparation for the perioperative period (secretion advance care plans.46 These interventions are individualised accord- management, controlled coughing, pain management, importance of ing to the needs and goals of patients, and may include both phar- early mobilisation) and optimisation of disease management (oxygen macological (eg, low-dose opioids) and non-pharmacological therapy, pacing, energy conservation, regular exercise).51 components. Non-pharmacological intervention components During the acute inpatient phase following transplantation, phys- iotherapy management (in intensive care and on the ward) comprises
Research 163 Box 2. Components of symptom management and palliative will be required to advance knowledge and expand treatment options care interventions for interstitial lung disease. Components for these debilitating conditions. are from randomised controlled trials of symptom manage- ment and palliative care interventions.44–46 In clinical practice, there are many opportunities to improve ac- cess to the best care for people with ILD. Whilst it is well accepted Relaxation and breathing control that a multidisciplinary team is needed for accurate ILD diagnosis15 Cough control strategies and efforts are well underway to ensure that patients have access Handheld fan to expert diagnostic services regardless of geography,53 access to the Oxygen therapy full range of services required for ILD care remains challenging, Crisis plan for breathlessness particularly for those in regional areas.54 Optimal models of specialist Energy conservation and pacing ILD care in the future should include access to multidisciplinary team Walking aids, home adaptations management, with capacity to deliver care via telehealth to ensure Exercise training, pulmonary rehabilitation that this expertise is available to all patients. Recent developments in Advance care planning telerehabilitation may facilitate access to this critical component of Education and information ILD care, and similar models could be explored for delivery of self- Counselling management, symptom management and palliative care in- Cognitive behavioural therapy terventions. Finally, there is a need to increase awareness amongst Education and support for caregivers physiotherapists and other healthcare professionals regarding the Multidisciplinary case conference and care planning burden of ILD, the availability of effective treatments and opportu- Multidisciplinary breathlessness support service nities to optimise outcomes for this patient group. respiratory management, including airway clearance and oxygen eAddenda: Nil. titration, early mobility, exercise training and education.52 Rehabilita- Ethics approval: Not applicable. tion in the early post-transplant phase (months 1 to 6) may occur in the Competing interests: Nil. inpatient or outpatient setting and includes aerobic, resistance and Source(s) of support: Nil. flexibility training.49 Clinical trials have demonstrated benefits of this Acknowledgements: Nil. approach, including improved 6MWD and quadriceps strength, Provenance: Invited. Peer reviewed. increased daily physical activity, and improved bone mineral density. Correspondence: Anne E Holland, Departments of Physiotherapy Rehabilitation may also be important in long-term management (. 6 and Respiratory Medicine, Alfred Health, Melbourne, Australia. Email: months), to improve exercise capacity, strength and physical activity [email protected] levels, and to aid in the management of transplant-related comorbid- ities such as hypertension, hyperlipidaemia and diabetes.49 References Future directions for research and practice 1. Wijsenbeek MS, Holland AE, Swigris JJ, Renzoni EA. Comprehensive Supportive Care for Patients with Fibrosing Interstitial Lung Disease. Am J Respir Crit Care Med. It is an exciting time to be involved in ILD research and clinical 2019;200:152–159. practice, with advances in disease-modifying treatments for IPF and PF-ILD that bring hope for improved patient outcomes. In this 2. Wijsenbeek M, Cottin V. Spectrum of Fibrotic Lung Diseases. N Engl J Med. context, physiotherapy interventions to maximise long-term health 2020;383:958–968. and wellbeing are increasingly important and expanding in scope. 3. McGroder CF, Zhang D, Choudhury MA, Salvatore MM, D’Souza BM, Hoffman EA, Whilst pulmonary rehabilitation is a core component of ILD care et al. Pulmonary fibrosis 4 months after COVID-19 is associated with severity of that generally delivers excellent patient outcomes, it is clear that illness and blood leucocyte telomere length. Thorax. 2021;76:1242–1245. some patients do not benefit as expected.27 Future research should examine whether alternative training strategies (including high- 4. Zhang P, Li J, Liu H, Han N, Ju J, Kou Y, et al. Long-term bone and lung consequences intensity interval training and high-flow oxygen therapy) can opti- associated with hospital-acquired severe acute respiratory syndrome: a 15-year mise short-term and longer-term outcomes, and in whom such follow-up from a prospective cohort study. Bone Res. 2020;8:8. strategies should be applied. It will be important to quantify the benefits of repeating pulmonary rehabilitation, including health 5. Global Burden of Disease Chronic Respiratory Disease Collaborators. Prevalence economic outcomes, to inform future decisions on funding and policy. and attributable health burden of chronic respiratory diseases, 1990–2017: a sys- The emerging field of self-management in ILD provides significant tematic analysis for the Global Burden of Disease Study 2017. Lancet Resp Med. research opportunities to define its components, and test outcomes in 2020;8:585–596. clinical trials. Similarly, models of symptom management such as breathlessness clinics show great potential, but require further 6. Zheng Q, Cox IA, Campbell JA, Xia Q, Otahal P, de Graaff B, et al. Mortality and rigorous testing and a clearer understanding of the essential com- survival in idiopathic pulmonary fibrosis: a systematic review and meta-analysis. ponents. There are currently no evidence-based strategies for man- ERJ Open Res. 2022;8. aging cough in ILD, and there is a pressing need for research in this area. There is enormous scope for research to address knowledge 7. Cottin V, Teague R, Nicholson L, Langham S, Baldwin M. The Burden of Progressive- gaps regarding oxygen therapy for ILD, including an urgent need for Fibrosing Interstitial Lung Diseases. Front Med (Lausanne). 2022;9:799912. development and testing of improved oxygen delivery systems that can better meet the needs of this patient group who may have pro- 8. King TE, Jr., Lee JS. Cryptogenic Organizing Pneumonia. N Engl J Med. found hypoxaemia. As ILD is relatively uncommon, acceleration of 2022;386:1058–1069. clinical research across all fields will require innovative designs that enable participation by patients who are located away from major 9. Carvajalino S, Reigada C, Johnson MJ, Dzingina M, Bajwah S. Symptom prevalence centres (eg, teletrials) and master protocols that allow inclusion of of patients with fibrotic interstitial lung disease: a systematic literature review. well-phenotyped patients with a variety of ILD subtypes. Physio- BMC Pulm Med. 2018;18:78. therapists can support research in ILD by encouraging patients to participate in national ILD registries and clinical trials, both of which 10. Swigris JJ, Stewart AL, Gould MK, Wilson SR. Patients’ perspectives on how idio- pathic pulmonary fibrosis affects the quality of their lives. Health Qual Life Out- comes. 2005;3:61. 11. Yount SE, Beaumont JL, Chen SY, Kaiser K, Wortman K, Van Brunt DL, et al. Health- Related Quality of Life in Patients with Idiopathic Pulmonary Fibrosis. Lung. 2016;194:227–234. 12. Klein S, Logan A, Lindell KO. A scoping review of unmet needs of caregivers of patients with pulmonary fibrosis. Curr Opin Support Palliat Care. 2021;15:226–232. 13. Wong AW, Koo J, Ryerson CJ, Sadatsafavi M, Chen W. A systematic review on the economic burden of interstitial lung disease and the cost-effectiveness of current therapies. BMC Pulm Med. 2022;22:148. 14. Collard HR, Chen SY, Yeh WS, Li Q, Lee YC, Wang A, et al. Health care utilization and costs of idiopathic pulmonary fibrosis in U.S. Medicare beneficiaries aged 65 years and older. Ann Am Thorac Soc. 2015;12:981–987. 15. Prasad JD, Mahar A, Bleasel J, Ellis SJ, Chambers DC, Lake F, et al. The interstitial lung disease multidisciplinary meeting: A position statement from the Thoracic Society of Australia and New Zealand and the Lung Foundation Australia. Respir- ology. 2017;22:1459–1472. 16. Lama VN, Flaherty KR, Toews GB, Colby TV, Travis WD, Long Q, et al. Prognostic value of desaturation during a 6-minute walk test in idiopathic interstitial pneu- monia. Am J Respir Crit Care Med. 2003;168:1084–1090. 17. Mendoza L, Gogali A, Shrikrishna D, Cavada G, Kemp SV, Natanek SA, et al. Quadriceps strength and endurance in fibrotic idiopathic interstitial pneumonia. Respirology. 2014;19:138–143.
Search
