Journal of Physiotherapy 67 (2021) 223 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: Canadian 24-hour movement guidelines for children and youth: An integration of physical activity, sedentary behaviour, and sleep Date of latest update: June 2016. Date of next update: 2026. Patient group: Description: The Canadian 24-Hour Movement Guidelines for Children and Children and youth aged 5 to 17 years. Guidelines may be adapted, under the Youth is a concise two-page document. It includes a preamble and a clear written guidance of health professionals, for children and youth with disabilities. and visual representation of the amount of moderate to vigorous physical activity Intended audience: Youth, parents, teachers, paediatricians and exercise (. 60 minutes of ‘Sweat’), light physical activity (several hours of ‘Step’), sleep (9 professionals. Additional versions: Corresponding Tools and Resource to 11 hours) and sedentary behaviour (, 2 hours of ‘Sit’) that should be included Package. Updated integration of Canadian Physical Activity Guidelines for in a healthy 24-hour period for children and youth aged 5 to 17 years. The cor- Children (5 to 11 years) and Youth (12 to 17 years)1 and Canadian Sedentary responding ‘Tools and Resource Package’ includes valuable links to background Behaviour Guidelines for Children (5 to 11 years) and Youth (12 to 17 years).2 information (eg, hub of guidelines for different population groups), professional Expert working group: Consensus panel (27 members) of research and development opportunities (eg, webinars) and case scenarios for clinicians to practice experts and representatives of Canadian agencies in physical practise integrating the guidelines into the daily life of children and youth. activity, sedentary behaviour, sleep and health promotion; methodological experts; international content experts (Australia, USA and Wales); and Provenance: Invited. Not peer reviewed. parent and youth representatives. Funded by: Canadian Society for Exercise Physiology, Conference Board of Canada, Healthy Active Living Georgina L Clutterbuck a and Leanne M Johnston b and Obesity Research Group (Children’s Hospital of Eastern Ontario aThe University of Western Sydney Research Institute), Public Health Agency of Canada and ParticipACTION. Consultation with: External consultation was undertaken in French and bThe University of Queensland, Australia English with youth, parents, caregivers, teachers, exercise practitioners, paediatricians, government and non-government organisations, and na- https://doi.org/10.1016/j.jphys.2021.05.002 tional and international content experts via surveys (590 participants) and focus groups (28 groups of 104 participants). The consensus panel addressed References comments and concerns prior to final approval. Approved by: Complete consensus panel. Location: Online at: https://csepguidelines.ca/children- 1. Tremblay MS, et al. Appl Physiol Nutr Metab. 2011;36:36–46. and-youth-5-17/ 2. Tremblay MS, et al. Appl Physiol Nutr Metab. 2011;36:59–64. Appraisal of Clinical Practice Guideline: Developing clinical practice guidelines for physiotherapists working with people with inherited bleeding disorders Date of latest update: April 2021. Date of next update: Not stated. Patient healthy life style and continuing competence. Data extraction was obtained from group: People with inherited bleeding disorders. Intended audience: Physio- 63 articles, mostly (n = 30) expert opinion/review papers and 12 randomised therapists working with people with bleeding disorders (PWBD). Secondary controlled trials. Although the scope of the guidelines was extended to bleeding audience: Administrators in the management of PWBD, educators, consumers and disorders, all references pertained to only haemophilia. From the 39 practice researchers. Additional versions: This clinical practice guideline is an update of statements strong evidence was present for two statements (‘use of therapeutic the Clinical Guidelines for Physiotherapists working with Persons with Bleeding exercise to address hemophilia arthropathy’ and ‘use manual techniques, Disorders that was published in 2018 by the Canadian Hemophilia Society. Expert including traction, joint gliding, fascial therapy and manual stretching’), moderate working group: Members of the Canadian Physiotherapists in Hemophilia Care evidence for one (‘use of point of care ultrasound (POCUS) to confirm diagnosis or (CPHC). Funded by: not stated. Consultation with: All members (n = 37) of the monitor resolution’) and weak evidence for three (‘assessment of pain’, ‘must CPHC. Approved by: Canadian Hemophilia Society. Location: Journal article: actively and regularly watch for signs of bleeding or infection’, ‘use of hydro- https://onlinelibrary.wiley.com/doi/10.1111/hae.14327. The Canadian Hemophilia therapy after hip or knee arthroplasty’). The remaining statements were graded as Society, website in French and English: https://www.hemophilia.ca/physiotherapy theoretical (n = 30), best practice (n = 2) or research (n = 1). Despite the limitations in the available evidence, as well as the lack of consumer input, the authors feel Description: The CPHC published their updated Clinical Practice Guidelines for that the guidelines will assist with clinical decision-making. Physiotherapists Working with Persons with Bleeding Disorders in Haemophilia (2021, in press). The working group of the CPHC followed the eight steps outlined Provenance: Invited. Not peer reviewed. in the American Physical Therapy Association (APTA) Clinical Practice Guideline Process Manual (2018). The guidelines are comprised of practice statements in Rik Gosselink nine areas of practice: interdisciplinary care, assessment of acute musculoskeletal University of Leuven, Belgium bleeds, management of bleeds, annual assessment, physiotherapy treatment for musculoskeletal complications, physiotherapy treatment before and after https://doi.org/10.1016/j.jphys.2021.05.006 musculoskeletal surgery, consultation with other care providers, encouraging a 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/).
Journal of Physiotherapy 67 (2021) 223 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: Canadian 24-hour movement guidelines for children and youth: An integration of physical activity, sedentary behaviour, and sleep Date of latest update: June 2016. Date of next update: 2026. Patient group: Description: The Canadian 24-Hour Movement Guidelines for Children and Children and youth aged 5 to 17 years. Guidelines may be adapted, under the Youth is a concise two-page document. It includes a preamble and a clear written guidance of health professionals, for children and youth with disabilities. and visual representation of the amount of moderate to vigorous physical activity Intended audience: Youth, parents, teachers, paediatricians and exercise (. 60 minutes of ‘Sweat’), light physical activity (several hours of ‘Step’), sleep (9 professionals. Additional versions: Corresponding Tools and Resource to 11 hours) and sedentary behaviour (, 2 hours of ‘Sit’) that should be included Package. Updated integration of Canadian Physical Activity Guidelines for in a healthy 24-hour period for children and youth aged 5 to 17 years. The cor- Children (5 to 11 years) and Youth (12 to 17 years)1 and Canadian Sedentary responding ‘Tools and Resource Package’ includes valuable links to background Behaviour Guidelines for Children (5 to 11 years) and Youth (12 to 17 years).2 information (eg, hub of guidelines for different population groups), professional Expert working group: Consensus panel (27 members) of research and development opportunities (eg, webinars) and case scenarios for clinicians to practice experts and representatives of Canadian agencies in physical practise integrating the guidelines into the daily life of children and youth. activity, sedentary behaviour, sleep and health promotion; methodological experts; international content experts (Australia, USA and Wales); and Provenance: Invited. Not peer reviewed. parent and youth representatives. Funded by: Canadian Society for Exercise Physiology, Conference Board of Canada, Healthy Active Living Georgina L Clutterbuck a and Leanne M Johnston b and Obesity Research Group (Children’s Hospital of Eastern Ontario aThe University of Western Sydney Research Institute), Public Health Agency of Canada and ParticipACTION. Consultation with: External consultation was undertaken in French and bThe University of Queensland, Australia English with youth, parents, caregivers, teachers, exercise practitioners, paediatricians, government and non-government organisations, and na- https://doi.org/10.1016/j.jphys.2021.05.002 tional and international content experts via surveys (590 participants) and focus groups (28 groups of 104 participants). The consensus panel addressed References comments and concerns prior to final approval. Approved by: Complete consensus panel. Location: Online at: https://csepguidelines.ca/children- 1. Tremblay MS, et al. Appl Physiol Nutr Metab. 2011;36:36–46. and-youth-5-17/ 2. Tremblay MS, et al. Appl Physiol Nutr Metab. 2011;36:59–64. Appraisal of Clinical Practice Guideline: Developing clinical practice guidelines for physiotherapists working with people with inherited bleeding disorders Date of latest update: April 2021. Date of next update: Not stated. Patient healthy life style and continuing competence. Data extraction was obtained from group: People with inherited bleeding disorders. Intended audience: Physio- 63 articles, mostly (n = 30) expert opinion/review papers and 12 randomised therapists working with people with bleeding disorders (PWBD). Secondary controlled trials. Although the scope of the guidelines was extended to bleeding audience: Administrators in the management of PWBD, educators, consumers and disorders, all references pertained to only haemophilia. From the 39 practice researchers. Additional versions: This clinical practice guideline is an update of statements strong evidence was present for two statements (‘use of therapeutic the Clinical Guidelines for Physiotherapists working with Persons with Bleeding exercise to address hemophilia arthropathy’ and ‘use manual techniques, Disorders that was published in 2018 by the Canadian Hemophilia Society. Expert including traction, joint gliding, fascial therapy and manual stretching’), moderate working group: Members of the Canadian Physiotherapists in Hemophilia Care evidence for one (‘use of point of care ultrasound (POCUS) to confirm diagnosis or (CPHC). Funded by: not stated. Consultation with: All members (n = 37) of the monitor resolution’) and weak evidence for three (‘assessment of pain’, ‘must CPHC. Approved by: Canadian Hemophilia Society. Location: Journal article: actively and regularly watch for signs of bleeding or infection’, ‘use of hydro- https://onlinelibrary.wiley.com/doi/10.1111/hae.14327. The Canadian Hemophilia therapy after hip or knee arthroplasty’). The remaining statements were graded as Society, website in French and English: https://www.hemophilia.ca/physiotherapy theoretical (n = 30), best practice (n = 2) or research (n = 1). Despite the limitations in the available evidence, as well as the lack of consumer input, the authors feel Description: The CPHC published their updated Clinical Practice Guidelines for that the guidelines will assist with clinical decision-making. Physiotherapists Working with Persons with Bleeding Disorders in Haemophilia (2021, in press). The working group of the CPHC followed the eight steps outlined Provenance: Invited. Not peer reviewed. in the American Physical Therapy Association (APTA) Clinical Practice Guideline Process Manual (2018). The guidelines are comprised of practice statements in Rik Gosselink nine areas of practice: interdisciplinary care, assessment of acute musculoskeletal University of Leuven, Belgium bleeds, management of bleeds, annual assessment, physiotherapy treatment for musculoskeletal complications, physiotherapy treatment before and after https://doi.org/10.1016/j.jphys.2021.05.006 musculoskeletal surgery, consultation with other care providers, encouraging a 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/).
Journal of Physiotherapy 67 (2021) 162 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 for membership of the Editorial Board The Editorial Board currently consists of 13 members: nine local and four international. Applications are invited to fill the following Editorial Board vacancies beginning in 2022: three local. 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 2022 and expires on 31 December 2024. Editorial Board members are entitled to renominate for a further two successive terms. Knowledge and skills required: 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 To be eligible to apply, Australian applicants must: 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 Journal Editor as required 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, which includes a clear explanation of the impact of any career interruption(s) over the last 5 years and/or any relative to op- portunity considerations. Applicants will be assessed against the knowledge and skills listed above and potentially against other criteria. Applications close Friday, 1 October 2021 and should be directed to Marko Stechiwskyj at [email protected] The Journal of Physiotherapy endorses equal opportunity and encourages eligible applicants of all backgrounds to apply. https://doi.org/10.1016/j.jphys.2021.06.002 1836-9553/
Journal of Physiotherapy 67 (2021) 222 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: Photograph Series of Daily Activities – Short Electronic Version (PHODA-SeV) Summary The Photograph Series of Daily Activities (PHODA) is an electronic photographs were selected from the original version, which was tool that uses photographs to measure perceived harmfulness of daily activities in patients with low back pain. The photographs represent a developed using recognisable daily activities and the opinion of several set of specific movements in daily activities (eg, lifting, bending, experts on chronic low back pain.4 The PHODA-SeV has also shown turning, reaching, falling, intermittent load-unexpected movement, and long-lasting load instance or sitting with limited dynamics) across high internal consistency (Cronbach alpha . 0.75) and excellent test- four domains: activities of daily living, housekeeping, work, and sports retest reliability (ICC . 0.90).4,5 The standard error of measurement and leisure time. The original version of the PHODA was developed in and smallest detectable change were of 5.46 and 15.13 points on a scale 1999 and contained 100 photographs.1 A few adapted versions have subsequently been developed, such as the PHODA short-electronic from 0 to 100, respectively, indicating an adequate measurement er- version (PHODA-SeV) with 40 photographs for adults.2–4 The ror.5 In addition, exploratory factor analysis indicated one factor PHODA-SeV has the benefit of being easy and quick to administer (ie, less than 10 minutes) and is freely available to download on the structure, which indicated appropriate structural validity (ie, a factor following website: https://ppw.kuleuven.be/ogp/software/phodasev- that explained . 50% of the variance).4 With regards to hypothesis en.zip. Each photograph of the PHODA-SeV is scored using a vertical testing, the PHODA-SeV showed small to fair correlations (r , 0.4) with 100-point visual analogue scale from 0 (not harmful at all) to 100 fear of movement/injury/reinjury, pain catastrophising, functional (extremely harmful). The total score is obtained by calculating the disability, and current pain intensity.4,5 One of these studies confirmed average score summing the ratings for each photograph divided by the an a priori formulated hypothesis, as the PHODA-SeV showed a fair total number of photographs. correlation with one measure of pain-related fear.5 There was no Measurement properties: Some measurement properties of the evidence of ceiling and floor effects in one study investigating its PHODA-SeV have been investigated in adults with chronic low back interpretability.5 pain.4,5 With regards to the content validity, its relevance was inves- tigated in a previous study.4 Other aspects of content validity The responsiveness of the PHODA-SeV is conflicting. The PHODA- (comprehensiveness and comprehensibility) have not been evaluated. SeV was able to detect changes after a physiotherapy program5 and The relevance of the PHODA-SeV seems to be appropriate since the an a priori formulated hypothesis related to the changes in the in- strument over time was confirmed. Nevertheless, there was no rela- tionship between the changes in the PHODA-SeV and the changes in fear of re-injury and fear-avoidance beliefs after physiotherapy. Therefore, the responsiveness after a specific treatment for changing the patient’s perception of harmful activities remains unknown. Commentary Crystian B Oliveiraa,b and Rafael Z Pintoc aDepartment of Physiotherapy, Faculty of Science and Technology, The PHODA-SeV is a reliable tool, with appropriate internal Sao Paulo State University (UNESP), Presidente Prudente, Sao Paulo, Brazil consistency, measurement error and construct validity, which were bUniversity of Western São Paulo (Unoeste), Presidente Prudente, assessed with hypothesis testing to measure perceived harmfulness of activities. In addition, this tool is able to detect changes over time Sao Paulo, Brazil after a physiotherapy program. However, some measurement cDepartment of Physical Therapy, properties still need further investigation. Some aspects of the Universidade Federal de Minas Gerais (UFMG), content validity still need to be refined, such as the comprehen- Belo Horizonte, Minas Gerais, Brazil siveness and comprehensibility for patients and clinicians. In addition, the structural validity of the instrument should be inves- References tigated using confirmatory factor analysis. Finally, responsiveness of this tool should be determined after a course of graded exposure 1. Kugler K, et al. The Photograph series of Daily Activities (PHODA). 1999. program using an a priori formulated hypotheses. The identification 2. Verbunt JA, et al. Eur J Pain. 2015;19:695–705. of the perceived harmful activities with the PHODA-SeV can assist 3. Simons LE, et al. Pain. 2017;158:912–921. clinicians to target and monitor activities that patients perceive as 4. Leeuw M, et al. J Pain. 2007;8:840–849. harmful, encouraging them to gradually face them in order to 5. Oliveira CB, et al. J Orthop Sports Phys Ther. 2018;48:719–727. restore function.6 6. George SZ, et al. J Orthop Sports Phys Ther. 2009;39:496–505. https://doi.org/10.1016/j.jphys.2020.09.006 1836-9553/© 2020 Published by Elsevier B.V. on behalf of Australian Physiotherapy Association. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).
Journal of Physiotherapy 67 (2021) 220–221 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 Revised Fibromyalgia Impact Questionnaire Summary Description: The Revised Fibromyalgia Impact Questionnaire excellent test-retest reliability for the FIQR total score (ICC = 0.78 to (FIQR) is a self-reported questionnaire that measures functional ca- 0.99)3–6,8,11–13 and its function (ICC = 0.73 to 0.98),3,6,8,10,12,13 overall pacity and disease severity in patients with fibromyalgia.1 The FIQR is impact (ICC = 0.51 to 0.93)3,6,8,10,12,13 and symptoms domains (ICC = one of the most used functional rating instruments in clinical practice 0.76 to 0.98).3,6,8,10,12,13 However, the temporal stability for the and clinical trials for the evaluation of therapeutic efficacy in patients with fibromyalgia. The FIQR is easy to use and can be administered in overall impact domain shows higher variability (ICC = 0.51 to 0.93), under 3 minutes. It is applicable to both genders of all socioeconomic levels. Furthermore, the FIQR has been adapted for online adminis- which suggests possible improvement in this particular domain/ tration1,2 and is translated into several languages.3–11 The FIQR has also been validated for cross-cultural adaptation.2,8,12 subscale. This could also be due to the high variability demonstrated Instructions for completing and scoring: The FIQR consists of 21 by patients in achieving their goals (or their perception/beliefs in individual questions across three linked domains: function, overall impact and symptoms. All questions pertain to experiences during achieving their goals) and feeling generally overwhelmed (items 10 the past 7 days and are scored using an 11-point numeric rating scale (0 to 10, with 10 denoting the worst possible condition). The function and 11, respectively). The 21-item subscales show good to very good domain contains nine questions addressing activities of daily living internal consistency (Cronbach’s alpha = 0.84 to 0.964).1–13 Cron- and contributes to 30% of the total FIQR score weighting. The overall bach’s alpha coefficients for each of the three domains also range impact domain consists of two questions relating to the overall from good to very good: function domain = 0.83 to 0.92; overall impact of fibromyalgia on function and overall impact of symptom impact domain = 0.81 to 0.9; symptoms domain = 0.84 to 0.91.4,6–8 severity; 20% of the FIQR total score is attributed to this domain. The symptoms domain contains 10 questions and evaluates symptoms Furthermore, each of the three domains of the FIQR have demon- commonly reported by fibromyalgia patients such as tenderness, strated good correlation with the FIQR total score (r = 0.92 to 0.96).2 hyperalgesia, environmental sensitivity, balance disturbances and The dimensionality of the FIQR has also been investigated using memory problems; it accounts for 50% of the total FIQR score. The factorial analyses.7–9 The structure of the three domains of the FIQR total FIQR score (out of 100) is the sum of the three modified domain was confirmed for the Brazilian Portuguese8 and Spanish versions of scores, that is: the summed score for the function domain (range 0 to the FIQR.7 However, a two-factor structure (function and symptoms 90, divided by 3), the score for overall impact domain (range 0 to 20) domains) was reported for the Italian version of the FIQR.9 This and the score for the symptoms domain (range 0 to 100, divided by 2). discrepancy could be due to a different factor analysis used to assess Reliability and validity: The FIQR has excellent psychometric structural validity (confirmatory factor analysis was employed in the properties,1,9 and differentiates between fibromyalgia patients and Brazilian and Spanish studies, and exploratory factor analysis and those with rheumatoid arthritis, systemic lupus erythematosus and major depressive disorders.1 The FIQR has demonstrated good to Rasch Analysis were used in the Italian study). Nevertheless, this highlights the need for future studies to clarify model structure of the FIQR in different languages. One study has investigated the standard error of measurement and smallest detectable change of the FIQR items and scores.8 Lupi et al reported a standard error of measurement of 2.5 (3.46%) for the FIQR total score, and smallest detectable change of 6.91 (9.57%).8 These need to be further verified by future studies. There is currently no data on the minimum clinically important difference of the FIQR. Commentary Current evidence suggests that the FIQR is a valid and reliable clinical minutes, with minimal training required. Lastly, some of the questions in each of the FIQR sub-domains were also rephrased to make it more instrument for the assessment of functional capacity and disease applicable for both genders of all socioeconomic backgrounds.1 Previ- ous studies have also demonstrated that the FIQR has good correlation severity in patients with fibromyalgia. It has good temporal stability and with the original FIQ in the total score and each of the related domain is suitable for evaluation of symptoms and physical functioning in pa- scores;1,2,5,8,12,13 thus, is it feasible to compare studies that have used the original FIQ with studies using the FIQR. tients with fibromyalgia in both clinical and research settings. The FIQR was developed by Bennet et al in 20091 as a revision to the Given that there are currently no established objective measures of disease severity in fibromyalgia, a validated clinical questionnaire such original FIQ, which was first published in 1991.14 It has several advan- as the FIQR that can reliably measure patients’ subjective symptoms, tages over its predecessor: the FIQR includes the evaluation of symp- functional capacity and global impact is critical. Due to the multi- dimensionality of the FIQR, it is reasonable and perhaps essential to toms commonly reported by fibromyalgia patients such as tenderness, report not only the FIQR total score, but also the three domain/subscale hyperalgesia, environmental sensitivity, balance and memory prob- scores. Future studies should focus on more thoroughly exploring the structural validity of the three FIQR domains and verify the model lems, whilst maintaining the essential clinimetric properties of the original questionnaire.1 Second, the scoring system was substantially simplified compared with the original FIQ by adopting an 11-point (0 to 10) numeric scale for all questions, which significantly improved its usability. This user-friendly questionnaire can be completed in , 3 https://doi.org/10.1016/j.jphys.2020.09.002 1836-9553/© 2020 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/).
Appraisal 221 structure of different translated versions of the FIQR items. Further- References more, given that the overall impact domain has slightly more variable temporal stability, this domain may benefit from further development. 1. Bennett RM, et al. Arthritis Res Ther. 2009;11:R120. Lastly, there is a need to properly evaluate the responsiveness of the 2. Paiva ES, et al. Clin Rheumatol. 2013;32:1199–1206. FIQR in the context of intervention outcome studies. 3. Abu-Dahab S, et al. Clin Rheumatol. 2014;33:391–396. 4. Costa C, et al. Acta Reumatol Port. 2016;41:240–250. Provenance: Invited. Not peer reviewed. 5. Ediz L, et al. Clin Rheumatol. 2011;30:339–346. 6. Isomura T, et al. Int J Rheum Dis. 2017;20:1088–1094. Michael Lee 7. Luciano JV, et al. Arthritis Care Res. 2013;65:1682–1689. School of Health, Discipline of Physiotherapy, Federation University, 8. Lupi JB, et al. Disabil Rehabil. 2017;39:1650–1663. 9. Salaffi F, et al. Clin Exp Rheumatol. 2013;31:S41–S49. Australia 10. Salgueiro M, et al. Health Qual Life Outcomes. 2013;11:132. 11. Srifi N, et al. Rheumatol Int. 2013;33:179–183. 12. Seo SR, et al. Int J Rheum Dis. 2016;19:459–464. 13. Ghavidel Parsa B, et al. Rheumatol Int. 2014;34:175–180. 14. Burckhardt CS, et al. J Rheumatol. 1991;18:728–733.
Journal of Physiotherapy 67 (2021) 229 Appraisal j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j p hy s Correspondence: Author response to Cao We thank Dr Yubin Cao and colleagues for their interest in our used a fixed-effect model. The pooled estimate is imprecise regardless paper ‘Exercise programs may be effective in preventing a new of whether it is reported as OR or RR. episode of neck pain: a systematic review and meta-analysis’.1 Cao and colleagues also point out that we downgraded quality of Cao and colleagues have questioned why we downgraded the evidence due to inconsistency for the ergonomic intervention versus quality of evidence due to imprecision for one of the intervention control contrast. Our criterion was to downgrade one level due to contrasts (exercise versus control). In our study, the criteria for inconsistency of results if substantial heterogeneity (I2 . 50%) was downgrading quality of evidence due to imprecision was based on a present, which on review we now realise was not the case. Therefore, threshold of , 400 participants for each pooled outcome, and also we agree that this is an error and the overall quality of evidence for observation of the 95% confidence intervals (CIs). The pooled effect our finding of no significant preventative effect due to ergonomic in our meta-analysis for the exercise intervention versus control interventions should be moderate rather than low. contrast had a wide confidence interval (OR 0.32, 95% CI 0.12 to 0.86) and we therefore downgraded the quality of evidence. Tarcisio F de Camposa, Chris G Maherb, Daniel Steffensc,d, Joel T Fullera and Mark J Hancocka We agree that results presented as relative risk (RR) are generally easier to interpret than odds ratio (OR); however, one randomised aDepartment of Health Professions, Macquarie University controlled trial2 in this review only presented data as an OR, so we bInstitute for Musculoskeletal Health, The University of Sydney could not calculate the RR for all contrasts as we did not have raw data for this study. For that reason, we decided to present the effect cSurgical Outcomes Research Centre (SOuRCe), size estimates as ORs for consistency throughout the paper. We were Royal Prince Alfred Hospital able to calculate the RR for the exercise intervention versus control contrast and these data are presented in Figure 5 on the eAddenda. dFaculty of Medicine and Health, Central Clinical School, We found that exercise reduced the risk of a new episode of neck pain The University of Sydney, Sydney, Australia by 53% (RR 0.47, 95% CI 0.32 to 0.68), which is slightly different from data presented by Cao and colleagues (RR 0.43, 95% CI 0.30 to 0.63). References We believe this small difference is due to the fact that we used a random-effects model to pool estimates, whereas Cao and colleagues 1. de Campos TF, et al. J Physiother. 2018;64:159–165. 2. Driessen MT, et al. Scand J Work Environ Health. 2011;37:383–393. https://doi.org/10.1016/j.jphys.2021.06.008 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/).
Appraisal Correspondence 233 treadmill training and warned about the possibility of an augmented https://doi.org/10.1016/j.jphys.2021.06.014 fall risk after treadmill intervention.5 References Nascimento et al1 have shed some interesting light on an important rehabilitation topic, but further work is necessary. 1. Nascimento LR, et al. J Physiother. 2021;67:95–104. 2. Scrivener K, et al. J Physiother. 2020;66:225–235. Marco Godi, Ilaria Arcolin, Stefano Corna and Marica Giardini 3. Macko RF, et al. Top Stroke Rehabil. 2005;12:45–57. Division of Physical Medicine and Rehabilitation, Istituti Clinici Scientifici 4. Taylor JR. An introduction to error analysis. Suasalito: University Science Books; 1997. 5. Duncan PW, et al. N Engl J Med. 2011;364:2026–2036. Maugeri IRCCS, Scientific Institute of Veruno, Gattico-Veruno, Italy Correspondence: Author response to Godi et al We thank Dr Godi et al for their interest in our recently pub- Lucas R Nascimentoa,b, Augusto Boeninga, Abílio Gallia, lished systematic review that examined the effects of treadmill Janaine C Polesec and Louise Adad training in comparison with no/non-walking intervention or overground walking after stroke.1 In the comparison of treadmill aCenter of Health Sciences, Discipline of Physiotherapy, walking and overground walking, Dr Godi et al suggested that Universidade Federal do Espírito Santo two randomised trials2,3 should be excluded based on differences in training intensity between the experimental and control bNeuroGroup, Department of Physiotherapy, Universidade Federal de groups. The fact that omitting a few studies resulted in a slight Minas Gerais change in the estimated magnitude of the effect, suggests that there are still not enough trials to provide a robust estimate. On cDiscipline of Physiotherapy, Faculdade Ciências Médicas de Minas the other hand, the overall finding did not change, that is, there Gerais, Brazil is no compelling evidence that treadmill training should be chosen over overground walking. Ambulatory people with stroke dSydney School of Health Sciences, Discipline of Physiotherapy, may benefit from walking training performed either on a The University of Sydney, Australia treadmill or overground. The rationale for our inclusion criteria is outlined in the introduction, and the authors are free to https://doi.org/10.1016/j.jphys.2021.06.005 perform their own systematic review based on different inclusion criteria. References 1. Nascimento LR, et al. J Physiother. 2021;67:95–104. 2. Aguiar LT, et al. NeuroRehabilitation. 2020;46:1–11. 3. Eich HJ, et al. Clin Rehabil. 2004;18:640–651.
Journal of Physiotherapy 67 (2021) 231 Appraisal j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j p hy s Correspondence: Author response to Vaughan-Graham et al We acknowledge the letter received from Vaughan-Graham and paragraph on page 234 we discuss the challenge for us as researchers, colleagues that discusses aspects of our recent systematic review.1 We and for therapists generally, in understanding what Bobath therapy is believe it is essential that reviews are rigorously conducted and that within the included trials.1 authors are transparent in any decisions made whilst conducting the review so that, as the letter states, readers are able to interpret Complex physiotherapy interventions are often reported poorly in findings appropriately. randomised controlled trials;5 this is not a new challenge for authors of systematic reviews. We challenge physiotherapy researchers to Many of the points raised in the letter have been addressed within describe in detail the components of their intervention using the the review itself.1 As examples, we explain the rationale for conducting TIDier reporting guidelines,6 to help researchers and therapists un- derstand what was delivered. the review and meta-analyses in the third paragraph of the introduc- If, as suggested, an intervention such as Bobath therapy requires tion on page 226 and we discuss in detail the methodological quality of significant postgraduate training to be implemented successfully, the included studies in the second paragraph on page 234.1 However, then its application to a wider clinical setting is likely to be limited. We are fortunate to have high levels of evidence for many in- we would like to address the following points more specifically. terventions in stroke rehabilitation that do not require additional The review was submitted to PROSPERO (CRD42019112451) prior postgraduate training and can therefore be implemented broadly.7 to the commencement of the review. Katharine Scrivenera, Simone Dorschb,c, Annie McCluskeyc,d, Three publications from the same study were reported in this Karl Schurrc, Petra L Grahame, Zheng Caof, Roberta Shepherdd review. We are, however, confident that we have managed the data and Sarah Tysong appropriately, as outlined in the third paragraph on page 227, never aDepartment of Health Professions, Macquarie University, including different data from the same study in one meta-analysis.1 Sydney, Australia It was suggested that we might have omitted a study by Cooke bFaculty of Health Sciences, Australian Catholic University, et al (reference 6 in the letter). However, that study did not meet our inclusion criteria, because it is a review protocol involving elite Sydney, Australia athletes. cStrokeEd Collaboration, Sydney, Australia dFaculty of Health Sciences, The University of Sydney, Sydney, Australia We acknowledge that we devised a criterion to determine eDepartment of Mathematics and Statistics, Macquarie University, whether each trial’s intervention was, or was not, defined as Bobath Sydney, Australia (described under Intervention on page 226).1 We used a broad defi- fHammondcare, Sydney, Australia nition of Bobath therapy and contacted two authors to seek addi- gSchool of Health Sciences, University of Manchester, Manchester, UK tional information about the intervention within their trials. When References one author provided additional information it was determined that it 1. Scrivener K, et al. J Physiother. 2020;66:225–235. did not meet the inclusion criteria for this review.2 2. Kerr A, et al. Physiotherapy. 2017;103:259–265. 3. Vaughan-Graham J, et al. Physiother Res Int. 2020;25:e1832. Despite recent work by Vaughan-Graham herself and others that 4. Vaughan-Graham J, et al. Disabil Rehabil. 2015;37:1793–1807. attempts to define a more modern Bobath therapy,3,4 it remains very 5. Candy B, et al. Trials. 2018;19:1–9. difficult for most therapists and researchers to distinguish what the 6. Hoffmann TC, et al. BMJ. 2014;348:g1687. changes are and precisely when these changes in Bobath occurred. If 7. Stroke Foundation. Clinical Guidelines for Stroke Management. Melbourne Australia, Bobath therapy has changed markedly over time, it would be useful to rename this new therapy approach. The lack of clarity around what 2017. constitutes a specific named approach highlights the fact that as a profession we need to move away from named ‘approaches’ and describe the specific components of an intervention. In the second https://doi.org/10.1016/j.jphys.2021.06.016 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/).
Journal of Physiotherapy 67 (2021) 228 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: Exercise programs in preventing a new episode of neck pain We would like to comment on the systematic review by de In addition, due to the limited data, we found that the calculation Campos et al,1 which found moderate-quality evidence favouring of RR could not be realised for ergonomic programs, so we agree with the assessment that evidence was low quality. However, from the exercise in reducing the risk of a new episode of neck pain. forest plot it seemed that the statistical heterogeneity among studies One of the two included studies, by Shihawong et al,2 was a rand- was not large, so we think it might be inappropriate to downgrade the evidence due to inconsistency. The decisions would change across omised controlled trial with a large sample size and adequate follow- the range of the confidence interval, so we would downgrade it due to imprecision. up. When the sample size and number of events are large, the results are labelled as ‘precise’.3 However, the evidence was downgraded due To summarise, we appreciate the hard work by the authors, but to imprecision, which was contradictory to our expectations. we think that further discussion may be warranted about the meta- analysis and quality of evidence. For prevention of a new episode of According to the Cochrane Handbook,4 results presented by neck pain, we would like to recommend exercise programs as high- quality evidence, but we still need to be cautious about the applica- relative risk (RR) may be more suitable than odds ratios (OR) for tion of results. intervention reviews because they are easier to interpret. However, Yubin Cao, Yingyi Wu and Changhao Yu Sichuan University, Chengdu, China the authors only estimated the effect using an OR. Therefore, we independently extracted data and found that the data were suffi- cient to conduct a meta-analysis presented as RR. The results showed that exercise could reduce the incidence proportion of neck pain by 57% (RR = 0.43, 95% CI 0.30 to 0.63, p , 0.00001, I2 = 0) (Figure 1). In addition, we found that Shihawong et al2 reported time-to-event data (HR = 0.45, 95% CI 0.28 to 0.71, p = 0.0007) (Figure 2), indicating that exercise could reduce the possibility of Exercise Control Risk Ratio Risk Ratio M-H, Fixed, 95% CI Study Events Total Events Total Weight M-H, Fixed, 95% CI Sihawong 2014 1 12 8 17 8.5% 0.18 (0.03 to 1.24) Tveito 2009 32 264 72 270 91.5% 0.45 (0.31 to 0.66) Total (95% CI) 276 287 100.0% 0.43 (0.30 to 0.63) Total events 33 80 Heterogeneity: Chi² = 0.88, df = 1 (p = 0.35); I² = 0% 0.2 0.5 1 2 5 Test for overall effect: Z = 4.43 (p < 0.00001) Favours exercise Favours control Figure 1. Forest plot showing the estimate effect (relative risk) of exercise programs in preventing a new episode of neck pain. M-H = Mantel-Haenszel. Study Log(Hazard Ratio) SE Weight Hazard Ratio Hazard Ratio Sihawong 2014 100.0% IV, Fixed, 95% CI IV, Fixed, 95% CI –0.7985 0.2348 0.45 (0.28 to 0.71) Total (95% CI) 100.0% 0.45 (0.28 to 0.71) Heterogeneity: Not applicable 0.2 0.5 1 2 5 Test for overall effect: Z = 3.40 (p = 0.0007) Favours exercise Favours control Figure 2. Forest plot showing the estimate effect (hazard ratio) of exercise programs in preventing a new episode of neck pain. IV = inverse variance. neck pain occurring by 55% at any timepoint during follow-up.5 The References results from both relative measures of benefit were highly consis- tent. In the GRADE system, we consider that this result is consistent 1. de Campos TF, et al. J Physiother. 2018;64:159–165. and precise enough to warrant the quality of evidence being 2. Sihawong R, et al. Occup Environ Med. 2014;71:63–70. upgraded to high. Even so, we still suggest that readers remain 3. Murad MH, et al. JAMA. 2014;312:171–179. cautious about this conclusion because the analysis was dominated 4. Higgins JPT, Green S, eds. Cochrane Handbook for Systematic Reviews of Interventions by one study.2 Version 5.1.0. The Cochrane Collaboration; 2011. Available from www.handbook. cochrane.org 5. Sashegyi A, et al. Oncologist. 2017;22:484–486. https://doi.org/10.1016/j.jphys.2021.06.015 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/).
Journal of Physiotherapy 67 (2021) 230 Appraisal j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j p hy s Correspondence: Re: Scrivener et al Systematic reviews are important summaries of the evidence; comparator for improving gait speed. The Discussion section of this paper acknowledges that the comparator treatment was based on the however, significant flaws in the methodology of the review by Bobath concept. The same investigator evaluated strength training for Scrivener et al limit the usefulness of the results.1 sit to stand (n = 93),7 showing no between-group difference, with a near identical description of the comparator treatment to the first Protocol registration in PROSPERO was undertaken after formal study. However, when contacted by the systematic review group, the authors did not acknowledge this connection to the Bobath concept. screening of search results against eligibility criteria was completed. One wonders if Bobath may be explicitly stated when the results are unfavourable and implicitly when the results are more favourable or This order does not guarantee transparency of the search strategy, as equivocal. the criteria originally planned for including publications in the review Given the recent systematic review publication by Diaz-Arribas et al,[13] we question the need for a second Bobath systematic can be changed. review in less than a year, especially because the authors failed to address the above methodological flaws. We trust that the The inclusion criteria for Bobath interventions used outdated educated reader will interpret the findings of this review with caution. theoretical frameworks, so the review reports findings from 16 [Citation numbers in square brackets refer to the reference list in outdated studies of 22, which bear no resemblance to current Bobath the original publication.1] clinical practice. Eight studies referenced a Bobath publication from Julie Vaughan-Grahama, Camila Torriani-Pasinb, 1990,[27,29–31,42-45] two studies referenced a Bobath publication Miguel Benito-Garciac, Konstantinos Kypriglis-Kypriotisd and from 1978,[25,26] and one study referenced a Bobath publication from 1960.[24] Five studies provided no reference to Bobath.[28,32,35,39,41] Agnieszka Sliwkae aDepartment of Physical Therapy, University of Toronto, Canada Only two studies based their Bobath intervention on a Bobath refer- bSchool of Physical Education & Sport, University of Sao Paulo, Brazil ence published after 2000.[33,36] cFaculty of Physiotherapy and Nursing, Salus Infirmorum, Pontificia de Three publications from the same study were included[29–31] and Salamanca University, Spain two[29,31] were used separately for analysis, despite recommenda- dVioanadrasis Physiotherapy Center, Athens, Greece eFaculty of Health Sciences, Jagiellonian University Medical College, tions that multiple reports of the same study should be collated, and Krakow, Poland review authors should choose and justify which primary report they use as a source for study results.2 References Eight of 22 studies scored 4 on the PEDro scale,[25–27,30,34,36–38] 1. Scrivener K, et al. J Physiother. 2020;66:225–235. which is considered poor quality.3 Only six of the 22 primary studies 2. Higgins JPT, Green S. eds. Cochrane Handbook for Systematic Reviews of Interventions. met the criteria of concealed allocation of patients to treatment Version 5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011. Available groups and blinded outcome assessment.[33,39,41–44] Therefore, the from www.handbook.cochrane.org 3. van Tulder M, et al. Spine. 2003;28:1290–1299. meta-analyses of the remaining studies potentially overestimated the 4. Wood L, et al. BMJ. 2008;336:601–605. 5. Hart T, et al. Arch Phys Med Rehabil. 2012;93:S117–126. treatment effects, and the findings should be considered with 6. Cooke R, et al. BMJ Open. 2020;10, e042975. caution.4 Causality determinations depend on aspects of study fi- 7. Kerr A, et al. Physiotherapy. 2017;103:259–658. delity, including intervention description, therapist adherence and expertise.5 Only four studies identified individualisation of in- terventions,[33,35,37,44] seven did not describe the interven- tion[24,28,29–31,42,43] and 11 provided no description of the therapists’ skill level.[24–26,28,32,36,38,39,41,43,45] Only two studies explicitly stated that therapists delivering the Bobath intervention had formal training in the Bobath concept.[33,44] In four studies that identified the ther- apists’ skill level,[33,35,37,44] Bobath demonstrated superiority compared with other approaches. No studies reported clinical adherence. The search strategy did not identify a large randomised trial investigating strength training of the lower limb in stroke (n = 109),6 which demonstrates that strength training is less effective than the https://doi.org/10.1016/j.jphys.2021.06.017 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/).
Journal of Physiotherapy 67 (2021) 232–233 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: Treadmill walking after stroke We read with great interest the article by Nascimento et al.1 In this However, training intensity affects the VO2 and endurance of patients systematic review, the authors attempted to understand the with stroke and, consequently, the gait speed and distance walked.3 effectiveness of treadmill training compared to no/non-walking or over- To include these training groups in the meta-analysis, the intensity ground walking in people after stroke. We appreciate the authors’ results should have been the same. regarding the first part of the comparison, but find issue with the second part. Third, Olawale et al was excluded from the meta-analysis because the authors presented the mean and SD of the time taken to perform Nascimento et al concluded that there is moderate-quality evi- the 10m Walk Test instead of the gait speed. However, by application dence that the effect of treadmill walking on walking speed and of binary operation properties and propagation of uncertainty for- distance is the same as or somewhat better than overground mula,4 it is possible to calculate the mean and SD of gait speed. walking.1 In this regard, we have some concerns about the selection Therefore, the decision to exclude this study which is quite of the studies included in the meta-analysis. appropriate in terms of the review’s rationale is not justified – it should have been considered in the analysis. First, in the paper of Eich et al, while the experimental group walked on a graded treadmill with body weight support (15%, not Based on these considerations, we have performed a new meta- meeting the inclusion criterion), the control group underwent analysis. In contrast to Nascimento’s review1 where the same Bobath-oriented physiotherapy, in which overground walking was outcome measure was represented in different ways (ie, mean not exclusive and performed in addition to other activities. Moreover, value or change scores), for each included study we report in our Bobath therapy is known to be less effective than task-specific forest plots the mean and SD of gait speed and distance walked training in improving walking outcomes.2 Therefore, it can be after the intervention (Figure 1). From the new analysis, another deduced that the overground training performed by the control perspective emerges in which no data show a benefit of treadmill group was not strictly comparable with the treadmill training in training over overground walking in patients with stroke. terms of dosage and modality. Finally, in Nascimento’s review,1 studies with body weight support Second, in the study by Aguiar et al, the two groups were trained .10% were excluded without a detailed explanation. This led to the at different intensities: 60 to 80% of heart rate reserve (HRR) for exclusion of one of the largest trials on patients with stroke (about treadmill training and 40% of HRR for overground training. 400 subjects), in which the authors declared no superiority of a Treadmill Treadmill Mean difference Mean difference IV, Random, 95% CI Study Mean SD N Mean SD N Weight IV, Random, 95% CI Hollands 2015 0.6 0.27 14 0.59 0.26 12 13.3% 0.01 (–0.19 to 0.21) Langhammer 2010 1 0.4 21 0.9 0.4 18 8.7% 0.10 (–0.15 to 0.35) Olawale 2011 0.41 0.21 20 0.5 0.21 20 32.6% –0.09 (–0.22 to 0.04) Park 2013 0.6 0.31 20 0.6 0.31 20 15.0% 0.00 (–0.19 to 0.19) Park 2015 0.35 0.14 9 0.32 0.16 10 30.4% 0.03 (–0.10 to 0.16) Total (95% CI) 84 80 100.0% –0.01 (–0.08 to 0.16) Heterogeneity: Tau² = 0.00, Chi² = 2.57, df = 4 (p = 0.63); I² = 0% –0.50 –0.25 0 0.25 0.50 Test for overall effect: Z = 0.27 (p = 0.79) Favours overground Favours treadmill b Treadmill Treadmill Mean difference Mean difference Weight IV, Random, 95% CI IV, Random, 95% CI Study Mean SD N Mean SD N Langhammer 2010 321 154 21 310 164 18 7.0% 11 (–90 to 111) Olawale 2011 145 75 20 155 66 20 36.6% –10 (–54 to 34) Park 2013 234 117 20 225 118 20 13.3% 8 (–65 to 81) Park 2015 126 50 9 123 39 10 43.1% 3 (–38 to 43) Total (95% CI) 70 68 100.0% –1 (–27 to 26) Heterogeneity: Tau² = 0.00, Chi² = 0.31, df = 3 (p = 0.96); I² = 0% –100 –50 0 50 100 Test for overall effect: Z = 0.04 (p = 0.97) Favours overground Favours treadmill Figure 1. Detailed forest plot showing mean difference (95% CI) in the effect of treadmill walking versus overground walking on (a) walking speed (m/s) and (b) walking distance (m), immediately after the intervention period. No forest plot with follow-up data after the intervention period is presented, because there were insufficient studies included in the analysis, for the aforementioned reasons. 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/).
Appraisal Correspondence 233 treadmill training and warned about the possibility of an augmented https://doi.org/10.1016/j.jphys.2021.06.014 fall risk after treadmill intervention.5 References Nascimento et al1 have shed some interesting light on an important rehabilitation topic, but further work is necessary. 1. Nascimento LR, et al. J Physiother. 2021;67:95–104. 2. Scrivener K, et al. J Physiother. 2020;66:225–235. Marco Godi, Ilaria Arcolin, Stefano Corna and Marica Giardini 3. Macko RF, et al. Top Stroke Rehabil. 2005;12:45–57. Division of Physical Medicine and Rehabilitation, Istituti Clinici Scientifici 4. Taylor JR. An introduction to error analysis. Suasalito: University Science Books; 1997. 5. Duncan PW, et al. N Engl J Med. 2011;364:2026–2036. Maugeri IRCCS, Scientific Institute of Veruno, Gattico-Veruno, Italy Correspondence: Author response to Godi et al We thank Dr Godi et al for their interest in our recently pub- Lucas R Nascimentoa,b, Augusto Boeninga, Abílio Gallia, lished systematic review that examined the effects of treadmill Janaine C Polesec and Louise Adad training in comparison with no/non-walking intervention or overground walking after stroke.1 In the comparison of treadmill aCenter of Health Sciences, Discipline of Physiotherapy, walking and overground walking, Dr Godi et al suggested that Universidade Federal do Espírito Santo two randomised trials2,3 should be excluded based on differences in training intensity between the experimental and control bNeuroGroup, Department of Physiotherapy, Universidade Federal de groups. The fact that omitting a few studies resulted in a slight Minas Gerais change in the estimated magnitude of the effect, suggests that there are still not enough trials to provide a robust estimate. On cDiscipline of Physiotherapy, Faculdade Ciências Médicas de Minas the other hand, the overall finding did not change, that is, there Gerais, Brazil is no compelling evidence that treadmill training should be chosen over overground walking. Ambulatory people with stroke dSydney School of Health Sciences, Discipline of Physiotherapy, may benefit from walking training performed either on a The University of Sydney, Australia treadmill or overground. The rationale for our inclusion criteria is outlined in the introduction, and the authors are free to https://doi.org/10.1016/j.jphys.2021.06.005 perform their own systematic review based on different inclusion criteria. References 1. Nascimento LR, et al. J Physiother. 2021;67:95–104. 2. Aguiar LT, et al. NeuroRehabilitation. 2020;46:1–11. 3. Eich HJ, et al. Clin Rehabil. 2004;18:640–651.
Journal of Physiotherapy 67 (2021) 161 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 Corrigendum Corrigendum to ‘Group-based pelvic floor muscle training for all women during pregnancy is more cost-effective than postnatal training for women with urinary incontinence: cost-effectiveness analysis of a systematic review’ [J Physiother 2021;67:105–114] Robyn Brennen a,b, Helena C Frawley a, Jennifer Martin c, Terry P Haines a,d a School of Primary and Allied Health Care, Faculty of Medicine, Nursing and Health Sciences, Monash University, Melbourne, Australia; b Monash Health Community Continence Service, Specialist Clinics, Monash Health, Melbourne, Australia; c School of Arts, Social Science and Humanities, Faculty of Health, Arts & Design, Swinburne University of Technology, Melbourne, Australia; d Rehabilitation, Ageing and Independent Living Research Centre, National Centre for Healthy Ageing, Monash University, Melbourne, Australia In our recent systematic review1 under the Results subheading ‘Faecal incontinence’, we reported a meta-analysis of trials of individual pelvic floor muscle training. We incorrectly stated that the pelvic floor muscle training was ‘during pregnancy’ however the text should have stated ‘postnatal’. This mistake was reiterated in the immediate next paragraph. The intervention in this meta-analysis related to faecal incontinence is correctly described as ‘postnatal’ in all other parts of the paper, including the Abstract, the cost-effectiveness analyses, the Discussion and the lay summary. The authors apologise to readers for the error. Reference 1. Brennen R, Frawley HC, Martin J, Haines TP. Group-based pelvic floor muscle training for all women during pregnancy is more cost-effective than postnatal training for women with urinary incontinence: cost-effectiveness analysis of a systematic review. J Physiother. 2021;67:105–114. DOI of original article: https://doi.org/10.1016/j.jphys.2021.03.001. Correspondence: Robyn Brennen, Monash Health Community Continence Service, Monash Health, Melbourne, Australia. E-mail: [email protected] https://doi.org/10.1016/j.jphys.2021.06.013 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/).
Journal of Physiotherapy 67 (2021) 217 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: Additional rehabilitation following botulinum toxin-A does not improve goal attainment and upper limb activity in chronic stroke survivors Synopsis Summary of: Lannin NA, Ada L, English C, Ratcliffe J, Fauz SG, Palit M, et al received a handout with stretching and arm and hand exercises; they on behalf of the InTENSE Trial Group. Effect of additional rehabilitation also received one follow-up telephone call. Outcome measures: The after botulinum toxin-A on upper limb activity in chronic stroke: The primary outcome was the Goal Attainment Scale T-score and upper InTENSE Trial. Stroke. 2020;51:556–562. limb activity measured using the Box and Block test (number of blocks) immediately post intervention at 3 months. Secondary outcome mea- Question: In chronic stroke survivors who receive botulinum toxin-A in sures were spasticity, wrist extension range of motion, grip strength, muscles that cross the wrist, does the addition of 3 months of pain, burden of care and quality of life. Results: A total of 138 (99%) evidence-based movement training compared with usual care change participants completed the study. At the end of the 3-month inter- goal attainment and upper limb activity? Design: Randomised vention period there was no difference in Goal Attainment Scale T- controlled trial with concealed allocation and blinded outcome scores (MD 2, 95% –2 to 7) or upper limb activity (MD 0.00 blocks, 95% assessment. Setting: Seven spasticity clinics across three states in CI –0.02 to 0.01). The experimental group had greater grip strength Australia. Participants: Adults who were . 3 months post-stroke, compared with the control group (MD 1.4 kg, 95% CI 0.2 to 2.7). No scheduled to receive botulinum toxin-A injection to any muscle that other between-group differences were demonstrated. Conclusion: In crosses the wrist and not currently receiving upper limb rehabilitation. chronic stroke survivors who received botulinum toxin-A to muscles Key exclusion criteria were botulinum toxin-A and casting in the past 6 that cross the wrist, the addition of 3 months of movement training months and/or cognitive impairment. Randomisation of 140 partici- was no more effective than a handout in changing goal attainment and pants allocated 69 to the experimental group and 71 to the control upper limb activity. Greater increases in grip strength were seen in the group. Interventions: Both groups received botulinum toxin-A to one experimental group. or more muscle(s) that crossed the wrist. In addition, the experimental group received up to three serial casts to maximise wrist extension for Provenance: Invited. Not peer reviewed. 2 weeks, followed by 10 weeks of movement training, including elec- trical stimulation and progressive resistance exercises aimed at Alicia J Spittle decreasing weakness and improving active movement. Participants Department of Physiotherapy, University of Melbourne, Australia were recommended to practise for 60 minutes/day on 7 days/week and were supported by clinic-based sessions, home visits and phone calls https://doi.org/10.1016/j.jphys.2021.05.005 by physiotherapists or occupational therapists. The control group Commentary Botulinum toxin-A for the treatment of post-stroke upper limb spasticity is therapy is not warranted following botulinum toxin-A in those with both commonly administered in combination with rehabilitation therapies, and greater stroke severity and chronicity. Furthermore, the trial suggests supported by clinical guideline recommendations.1 Whilst the effectiveness of caution using botulinum toxin-A itself for this population, given the small botulinum toxin-A to reduce resistance to passive movement is strongly amount of change seen across outcomes. An intensive therapy regimen supported, the evidence for functional improvements remains equivocal.2,3 after injection may be beneficial earlier after stroke in those with some Research in this area has been complicated by variations in injection active movement,4 and this should be explored in future trials. The use of protocols, goals, concurrent therapy and outcome measurements. It is botulinum toxin-A for upper limb spasticity should continue to be person- particularly difficult to isolate the effect of adjunctive therapy compared to centred, with careful consideration of appropriate goals, adjunctive botulinum toxin-A itself. therapy approaches and outcome measurements. The InTENSE trial protocol was rigorous and interventions were evidence Provenance: Invited. Not peer reviewed. based. Exercise adherence in the experimental group was high. Outcome Kelly Bower measures spanned multiple domains and included the primary outcomes of the Goal Attainment Scale, appropriate for heterogenous goals, and the Box Department of Physiotherapy, University of Melbourne, Australia and Block Test, a measure of manual dexterity. https://doi.org/10.1016/j.jphys.2021.05.004 The intensive upper limb rehabilitation program was not effective in this study, with no between-group differences in the primary outcomes References and only grip strength significantly favouring the intervention group. Changes over time in outcomes were non-significant or of small magni- 1. Royal College of Physicians. https://www.rcplondon.ac.uk/guidelines-policy/spa tude. However, it is important to highlight that participants were on sticity-adults-management-using-botulinum-toxin. average 3 years post stroke and 78% were unable to move at least one block on the Box and Block Test at baseline. This may indicate very limited 2. Mills PB, et al. Clin Rehabil. 2016;30:537–548. potential for improvement. Clinically, this study implies that intensive 3. Levy J, et al. Ann Phys Rehabil Med. 2019;62:234–240. 4. Rosales RL, et al. J Neurol Sci. 2016;371:6–14. 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/).
Journal of Physiotherapy 67 (2021) 217 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: Additional rehabilitation following botulinum toxin-A does not improve goal attainment and upper limb activity in chronic stroke survivors Synopsis Summary of: Lannin NA, Ada L, English C, Ratcliffe J, Fauz SG, Palit M, et al received a handout with stretching and arm and hand exercises; they on behalf of the InTENSE Trial Group. Effect of additional rehabilitation also received one follow-up telephone call. Outcome measures: The after botulinum toxin-A on upper limb activity in chronic stroke: The primary outcome was the Goal Attainment Scale T-score and upper InTENSE Trial. Stroke. 2020;51:556–562. limb activity measured using the Box and Block test (number of blocks) immediately post intervention at 3 months. Secondary outcome mea- Question: In chronic stroke survivors who receive botulinum toxin-A in sures were spasticity, wrist extension range of motion, grip strength, muscles that cross the wrist, does the addition of 3 months of pain, burden of care and quality of life. Results: A total of 138 (99%) evidence-based movement training compared with usual care change participants completed the study. At the end of the 3-month inter- goal attainment and upper limb activity? Design: Randomised vention period there was no difference in Goal Attainment Scale T- controlled trial with concealed allocation and blinded outcome scores (MD 2, 95% –2 to 7) or upper limb activity (MD 0.00 blocks, 95% assessment. Setting: Seven spasticity clinics across three states in CI –0.02 to 0.01). The experimental group had greater grip strength Australia. Participants: Adults who were . 3 months post-stroke, compared with the control group (MD 1.4 kg, 95% CI 0.2 to 2.7). No scheduled to receive botulinum toxin-A injection to any muscle that other between-group differences were demonstrated. Conclusion: In crosses the wrist and not currently receiving upper limb rehabilitation. chronic stroke survivors who received botulinum toxin-A to muscles Key exclusion criteria were botulinum toxin-A and casting in the past 6 that cross the wrist, the addition of 3 months of movement training months and/or cognitive impairment. Randomisation of 140 partici- was no more effective than a handout in changing goal attainment and pants allocated 69 to the experimental group and 71 to the control upper limb activity. Greater increases in grip strength were seen in the group. Interventions: Both groups received botulinum toxin-A to one experimental group. or more muscle(s) that crossed the wrist. In addition, the experimental group received up to three serial casts to maximise wrist extension for Provenance: Invited. Not peer reviewed. 2 weeks, followed by 10 weeks of movement training, including elec- trical stimulation and progressive resistance exercises aimed at Alicia J Spittle decreasing weakness and improving active movement. Participants Department of Physiotherapy, University of Melbourne, Australia were recommended to practise for 60 minutes/day on 7 days/week and were supported by clinic-based sessions, home visits and phone calls https://doi.org/10.1016/j.jphys.2021.05.005 by physiotherapists or occupational therapists. The control group Commentary Botulinum toxin-A for the treatment of post-stroke upper limb spasticity is therapy is not warranted following botulinum toxin-A in those with both commonly administered in combination with rehabilitation therapies, and greater stroke severity and chronicity. Furthermore, the trial suggests supported by clinical guideline recommendations.1 Whilst the effectiveness of caution using botulinum toxin-A itself for this population, given the small botulinum toxin-A to reduce resistance to passive movement is strongly amount of change seen across outcomes. An intensive therapy regimen supported, the evidence for functional improvements remains equivocal.2,3 after injection may be beneficial earlier after stroke in those with some Research in this area has been complicated by variations in injection active movement,4 and this should be explored in future trials. The use of protocols, goals, concurrent therapy and outcome measurements. It is botulinum toxin-A for upper limb spasticity should continue to be person- particularly difficult to isolate the effect of adjunctive therapy compared to centred, with careful consideration of appropriate goals, adjunctive botulinum toxin-A itself. therapy approaches and outcome measurements. The InTENSE trial protocol was rigorous and interventions were evidence Provenance: Invited. Not peer reviewed. based. Exercise adherence in the experimental group was high. Outcome Kelly Bower measures spanned multiple domains and included the primary outcomes of the Goal Attainment Scale, appropriate for heterogenous goals, and the Box Department of Physiotherapy, University of Melbourne, Australia and Block Test, a measure of manual dexterity. https://doi.org/10.1016/j.jphys.2021.05.004 The intensive upper limb rehabilitation program was not effective in this study, with no between-group differences in the primary outcomes References and only grip strength significantly favouring the intervention group. Changes over time in outcomes were non-significant or of small magni- 1. Royal College of Physicians. https://www.rcplondon.ac.uk/guidelines-policy/spa tude. However, it is important to highlight that participants were on sticity-adults-management-using-botulinum-toxin. average 3 years post stroke and 78% were unable to move at least one block on the Box and Block Test at baseline. This may indicate very limited 2. Mills PB, et al. Clin Rehabil. 2016;30:537–548. potential for improvement. Clinically, this study implies that intensive 3. Levy J, et al. Ann Phys Rehabil Med. 2019;62:234–240. 4. Rosales RL, et al. J Neurol Sci. 2016;371:6–14. 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/).
Journal of Physiotherapy 67 (2021) 219 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: Group-based pelvic floor muscle training is not inferior to individual training for the treatment of urinary incontinence in older women Synopsis Summary of: Dumoulin C, Morin M, Danieli C, Cacciari L, Mayrand MH, Tousignant performance. Participants in both groups were encouraged to perform exercises at M, et al; for the Urinary Incontinence and Aging Study Group. Group-based vs home during the intervention and for the subsequent 9 months. Outcome individual pelvic floor muscle training to treat urinary incontinence in older measures: The primary outcome measure was percentage reduction in the women: a randomized clinical trial. JAMA Intern Med. 2020;180:1284–1293. number of urinary incontinence episodes at 1 year, reported in a 7-day bladder diary relative to the pre-treatment baseline. Secondary outcomes measured uri- Question: Is the effectiveness of group-based pelvic floor muscle training non- nary leakage, micturition, urinary incontinence-related symptoms, quality of life, inferior to individual training in women with urinary incontinence? Design: self-efficacy, patient global impression of improvement, and satisfaction with Multicentre, noninferiority, randomised controlled trial with concealed allocation. treatment. Results: A total of 319 (88%) participants completed the study. After 1 Setting: Two Canadian research centres. Participants: Women aged 60 years year, the median percentage reduction in incontinence episodes with the indi- with symptoms of stress or mixed incontinence reporting three or more episodes vidual intervention was 70% (95% CI 44 to 89) versus 74% (95% CI 46 to 86) with of involuntary urine loss per week during the preceding 3 months. The main the group-based intervention (difference 24%, 95% CI 210 to 7), confirming exclusion criteria were: body mass index 35 kg/m2, requiring a mobility aid, noninferiority within the prespecified margin of 10%. The groups did not differ chronic constipation, important pelvic organ prolapse, and physiotherapy/surgery on the secondary outcomes. Adverse events were minor and uncommon. for incontinence/pelvic organ prolapse in the past year. Randomisation of 362 Conclusion: Group-based pelvic floor muscle training is not inferior to individual participants allocated 184 to individual pelvic floor muscle training and 178 to pelvic floor muscle training for the treatment of stress and mixed urinary in- group-based training. Interventions: Both groups received an individual session continence in older women. Widespread use of group-based pelvic floor muscle to learn how to effectively contract pelvic floor muscles before receiving a 12- training may lead to increased treatment availability and affordability. week pelvic floor muscle training program under the direction of an experi- enced pelvic floor physiotherapist, either in individual or group sessions. For both Provenance: Invited. Not peer reviewed. interventions, each weekly session included 15 minutes of education and 45 minutes of exercise targeting pelvic floor muscle strength, power, endurance, Nina Østerås coordination, and integration into daily living activities. Participants allocated to Division of Rheumatology and Research, Diakonhjemmet Hospital, individual pelvic floor muscle training used intravaginal electromyographic biofeedback during each treatment session for 10 to 15 minutes. Participants Oslo, Norway allocated to group sessions who had difficulty with the exercises were offered short private sessions with the physiotherapist to ensure correct exercise https://doi.org/10.1016/j.jphys.2021.05.010 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/). Commentary Urinary incontinence is a prevalent condition and becomes more prevalent groups at one year after randomisation. The effectiveness of group training as women age. It contributes to reduced quality of life, shame and embar- compared with individual training was well within the pre-set 10% difference. rassment, and reduced socialisation and physical activity. Although supervised individual pelvic floor muscle training to treat urinary incontinence has the Physiotherapists currently use group training for women with urinary in- highest level of evidence for its effectiveness from multiple trials,1 there are continence, but the training protocols may vary. Since the group training in this inadequate resources available to meet patient needs in many healthcare study showed effectiveness, physiotherapists should familiarise themselves systems. Workforce shortages in public healthcare, constantly increasing with this protocol4 and incorporate it into their group training for older women healthcare costs and increasing demands associated with ageing populations with urinary incontinence. demand new ways of approaching old problems.2 Provenance: Invited. Not peer reviewed. Dumoulin et al undertook the GROUP trial, a high-quality randomised trial Margaret Sherburn of pelvic floor muscle training via group sessions and compared this with su- pervised individual pelvic floor muscle training. Their results across a wide Department of Physiotherapy, The University of Melbourne, range of outcomes indicated that group training is not inferior to individual Melbourne, Australia pelvic floor muscle training. Group training has healthcare cost benefits and has been shown to enhance patient self-efficacy and self-management of condi- https://doi.org/10.1016/j.jphys.2021.05.009 tions,3 and the authors conclude that group training could be widely implemented to improve accessibility to physiotherapy treatment. While References there have been other small studies investigating group training for urinary incontinence, none have been properly powered to test the noninferiority of 1. Dumoulin C, et al. Cochrane Database Syst Rev. 2018;10:CD005654. group training compared with individual training. The GROUP trial was 2. Brennen R, et al. ANZJOG. 2019;59:450–456. powered to test a small difference in incontinence episodes between the two 3. Wayment HA, McDonald RL. J Str Condit Res. 2017;31:3137–3145. 4. Dumoulin C, et al. Trials. 2017;18:544.d. 1836-9553/© 2021 Published by Elsevier B.V. on behalf of Australian Physiotherapy Association. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).
Journal of Physiotherapy 67 (2021) 219 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: Group-based pelvic floor muscle training is not inferior to individual training for the treatment of urinary incontinence in older women Synopsis Summary of: Dumoulin C, Morin M, Danieli C, Cacciari L, Mayrand MH, Tousignant performance. Participants in both groups were encouraged to perform exercises at M, et al; for the Urinary Incontinence and Aging Study Group. Group-based vs home during the intervention and for the subsequent 9 months. Outcome individual pelvic floor muscle training to treat urinary incontinence in older measures: The primary outcome measure was percentage reduction in the women: a randomized clinical trial. JAMA Intern Med. 2020;180:1284–1293. number of urinary incontinence episodes at 1 year, reported in a 7-day bladder diary relative to the pre-treatment baseline. Secondary outcomes measured uri- Question: Is the effectiveness of group-based pelvic floor muscle training non- nary leakage, micturition, urinary incontinence-related symptoms, quality of life, inferior to individual training in women with urinary incontinence? Design: self-efficacy, patient global impression of improvement, and satisfaction with Multicentre, noninferiority, randomised controlled trial with concealed allocation. treatment. Results: A total of 319 (88%) participants completed the study. After 1 Setting: Two Canadian research centres. Participants: Women aged 60 years year, the median percentage reduction in incontinence episodes with the indi- with symptoms of stress or mixed incontinence reporting three or more episodes vidual intervention was 70% (95% CI 44 to 89) versus 74% (95% CI 46 to 86) with of involuntary urine loss per week during the preceding 3 months. The main the group-based intervention (difference 24%, 95% CI 210 to 7), confirming exclusion criteria were: body mass index 35 kg/m2, requiring a mobility aid, noninferiority within the prespecified margin of 10%. The groups did not differ chronic constipation, important pelvic organ prolapse, and physiotherapy/surgery on the secondary outcomes. Adverse events were minor and uncommon. for incontinence/pelvic organ prolapse in the past year. Randomisation of 362 Conclusion: Group-based pelvic floor muscle training is not inferior to individual participants allocated 184 to individual pelvic floor muscle training and 178 to pelvic floor muscle training for the treatment of stress and mixed urinary in- group-based training. Interventions: Both groups received an individual session continence in older women. Widespread use of group-based pelvic floor muscle to learn how to effectively contract pelvic floor muscles before receiving a 12- training may lead to increased treatment availability and affordability. week pelvic floor muscle training program under the direction of an experi- enced pelvic floor physiotherapist, either in individual or group sessions. For both Provenance: Invited. Not peer reviewed. interventions, each weekly session included 15 minutes of education and 45 minutes of exercise targeting pelvic floor muscle strength, power, endurance, Nina Østerås coordination, and integration into daily living activities. Participants allocated to Division of Rheumatology and Research, Diakonhjemmet Hospital, individual pelvic floor muscle training used intravaginal electromyographic biofeedback during each treatment session for 10 to 15 minutes. Participants Oslo, Norway allocated to group sessions who had difficulty with the exercises were offered short private sessions with the physiotherapist to ensure correct exercise https://doi.org/10.1016/j.jphys.2021.05.010 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/). Commentary Urinary incontinence is a prevalent condition and becomes more prevalent groups at one year after randomisation. The effectiveness of group training as women age. It contributes to reduced quality of life, shame and embar- compared with individual training was well within the pre-set 10% difference. rassment, and reduced socialisation and physical activity. Although supervised individual pelvic floor muscle training to treat urinary incontinence has the Physiotherapists currently use group training for women with urinary in- highest level of evidence for its effectiveness from multiple trials,1 there are continence, but the training protocols may vary. Since the group training in this inadequate resources available to meet patient needs in many healthcare study showed effectiveness, physiotherapists should familiarise themselves systems. Workforce shortages in public healthcare, constantly increasing with this protocol4 and incorporate it into their group training for older women healthcare costs and increasing demands associated with ageing populations with urinary incontinence. demand new ways of approaching old problems.2 Provenance: Invited. Not peer reviewed. Dumoulin et al undertook the GROUP trial, a high-quality randomised trial Margaret Sherburn of pelvic floor muscle training via group sessions and compared this with su- pervised individual pelvic floor muscle training. Their results across a wide Department of Physiotherapy, The University of Melbourne, range of outcomes indicated that group training is not inferior to individual Melbourne, Australia pelvic floor muscle training. Group training has healthcare cost benefits and has been shown to enhance patient self-efficacy and self-management of condi- https://doi.org/10.1016/j.jphys.2021.05.009 tions,3 and the authors conclude that group training could be widely implemented to improve accessibility to physiotherapy treatment. While References there have been other small studies investigating group training for urinary incontinence, none have been properly powered to test the noninferiority of 1. Dumoulin C, et al. Cochrane Database Syst Rev. 2018;10:CD005654. group training compared with individual training. The GROUP trial was 2. Brennen R, et al. ANZJOG. 2019;59:450–456. powered to test a small difference in incontinence episodes between the two 3. Wayment HA, McDonald RL. J Str Condit Res. 2017;31:3137–3145. 4. Dumoulin C, et al. Trials. 2017;18:544.d. 1836-9553/© 2021 Published by Elsevier B.V. on behalf of Australian Physiotherapy Association. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).
Journal of Physiotherapy 67 (2021) 218 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: Non-invasive brain stimulation does not enhance the effect of robotic-assisted upper limb training on arm motor recovery after stroke Synopsis Summary of: Reis SB, Bernardo WM, Oshiro CA, Krebs HI, Conforto AB. Ef- high for both impairment-level and activity-level outcome measures. Based fects of robotic therapy associated with noninvasive brain stimulation on on the quantitative pooling of the available data, there was no effect of non- upper-limb rehabilitation after stroke: systematic review and meta-analysis invasive brain stimulation on upper limb performance on the Fugl-Meyer of randomized clinical trials. Neurorehabil Neural Repair. 2021;35:256–266. Assessment (seven studies, MD 0.15, 95% CI 23.10 to 3.40) or on upper limb activity limitation (five studies, SMD 0.03, 95% CI 20.28 to 0.33). Planned Objective: To review the evidence as to whether adding non-invasive brain subgroup analyses demonstrated similar results for both subacute and chronic stimulation enhances the effects of robotic-assisted upper limb training on stroke, robotic device characteristics (end-effector and exoskeleton), upper upper limb motor recovery in individuals with stroke. Data sources: MEDLINE, limb joints involved in training, and unimanual and bimanual training. There EMBASE, CENTRAL, LILACS, CINAHL, DORIS, and PEDro were searched up to was no evidence that non-invasive brain stimulation paradigms (increased or July 2019. This search was supplemented by searching online archives of decreased cortical excitability), timing of stimulation (before, after and during theses and trial registries. Study selection: Randomised controlled trials robotic-assisted therapy), or number of sessions influenced the results. (parallel or crossover) involving people with upper limb paresis due to stroke, Conclusion: At present, there is high-quality evidence to suggest that the ef- in which non-invasive brain stimulation before, during or after robotic- fects of robotic-assisted upper limb training on upper limb motor impairment assisted upper limb rehabilitation was compared with sham non-invasive or motor activity for individuals with stroke are not enhanced by existing non- brain stimulation or robotic-assisted upper limb rehabilitation without non- invasive brain stimulation approaches. invasive brain stimulation. Outcome measures were upper limb performance in either impairment-level and/or activity-level domains. Data extraction: Provenance: Invited. Not peer reviewed. Two reviewers extracted data and discrepancies were resolved by consensus. For crossover designs, only the first-phase intervention data were extracted. Prudence Plummer Risk of bias for individual studies was assessed according to specified criteria Department of Physical Therapy, MGH Institute of Health Professions, by two reviewers and quality of the body of evidence was rated according to GRADEpro. Data synthesis: Of 1,176 articles identified by the search, eight USA unique trials with a total of 324 participants (161 active, 163 control) met the selection criteria and were included in the review. The quality of evidence was https://doi.org/10.1016/j.jphys.2021.05.008 1836-9553/© 2021 Published by Elsevier B.V. on behalf of Australian Physiotherapy Association. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/). Commentary In people with hemiparesis after stroke, intensive upper limb motor training.2,3 One consideration here may be ceiling effects with robotic practice, such as robot-assisted training, can lead to clinically meaningful training, which by nature is more intensive than traditional motor improvement, yet the benefit typically falls far short of full functional therapies. Future trials may evaluate whether non-invasive brain stimula- recovery. Supplemental therapies have the potential to enhance training tion alters the recovery rate across a training regimen, in addition to the effects, and non-invasive brain stimulation as a candidate add-on therapy more commonly assessed effect magnitude or number of clinical responders. has previously been reported; this literature was systematically reviewed by Reis and colleagues. The review found that there is no beneficial effect This work makes an important and timely contribution to the litera- of non-invasive brain stimulation as a supplement to robot-assisted ture in an area of continued scientific and clinical interest. Understanding training. The eight reviewed trials used a variety of existing best non- intervention features that are ineffective helps limit premature clinical invasive brain stimulation practices and available robotic technology. adoption and shape future trials. The interpretation of these findings should consider the diversity of Provenance: Invited. Not peer reviewed. non-invasive brain stimulation methods in the reviewed studies. This included inhibitory or excitatory protocols, and stimulation by trans- Dylan J Edwards cranial electric or magnetic stimulators, on a background of varied uni- Moss Rehabilitation Research Institute, Philadelphia, USA lateral or bilateral robotic training of different doses and/or intensities. Taken together with the modest aggregate sample size, this challenges the Edith Cowan University, Joondalup, Australia ability to speak to all forms of non-invasive brain stimulation being ineffective. Modifying non-invasive brain stimulation parameters or https://doi.org/10.1016/j.jphys.2021.05.007 methods has been likened to manipulating a drug’s chemical composi- tion,1 fundamentally changing the action and effect. Thus, the forms of References non-invasive brain stimulation across trials were fundamentally different. 1. Bikson M, et al. Nature. 2013;501:167. A clinical benefit of supplemental non-invasive brain stimulation has 2. Allman C, et al. Sci Transl Med. 2016;8:330re1. been demonstrated in individual studies of non-robotic upper limb 3. Lindenberg R, et al. Neurology. 2010;75:2176–2184. 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/).
Journal of Physiotherapy 67 (2021) 218 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: Non-invasive brain stimulation does not enhance the effect of robotic-assisted upper limb training on arm motor recovery after stroke Synopsis Summary of: Reis SB, Bernardo WM, Oshiro CA, Krebs HI, Conforto AB. Ef- high for both impairment-level and activity-level outcome measures. Based fects of robotic therapy associated with noninvasive brain stimulation on on the quantitative pooling of the available data, there was no effect of non- upper-limb rehabilitation after stroke: systematic review and meta-analysis invasive brain stimulation on upper limb performance on the Fugl-Meyer of randomized clinical trials. Neurorehabil Neural Repair. 2021;35:256–266. Assessment (seven studies, MD 0.15, 95% CI 23.10 to 3.40) or on upper limb activity limitation (five studies, SMD 0.03, 95% CI 20.28 to 0.33). Planned Objective: To review the evidence as to whether adding non-invasive brain subgroup analyses demonstrated similar results for both subacute and chronic stimulation enhances the effects of robotic-assisted upper limb training on stroke, robotic device characteristics (end-effector and exoskeleton), upper upper limb motor recovery in individuals with stroke. Data sources: MEDLINE, limb joints involved in training, and unimanual and bimanual training. There EMBASE, CENTRAL, LILACS, CINAHL, DORIS, and PEDro were searched up to was no evidence that non-invasive brain stimulation paradigms (increased or July 2019. This search was supplemented by searching online archives of decreased cortical excitability), timing of stimulation (before, after and during theses and trial registries. Study selection: Randomised controlled trials robotic-assisted therapy), or number of sessions influenced the results. (parallel or crossover) involving people with upper limb paresis due to stroke, Conclusion: At present, there is high-quality evidence to suggest that the ef- in which non-invasive brain stimulation before, during or after robotic- fects of robotic-assisted upper limb training on upper limb motor impairment assisted upper limb rehabilitation was compared with sham non-invasive or motor activity for individuals with stroke are not enhanced by existing non- brain stimulation or robotic-assisted upper limb rehabilitation without non- invasive brain stimulation approaches. invasive brain stimulation. Outcome measures were upper limb performance in either impairment-level and/or activity-level domains. Data extraction: Provenance: Invited. Not peer reviewed. Two reviewers extracted data and discrepancies were resolved by consensus. For crossover designs, only the first-phase intervention data were extracted. Prudence Plummer Risk of bias for individual studies was assessed according to specified criteria Department of Physical Therapy, MGH Institute of Health Professions, by two reviewers and quality of the body of evidence was rated according to GRADEpro. Data synthesis: Of 1,176 articles identified by the search, eight USA unique trials with a total of 324 participants (161 active, 163 control) met the selection criteria and were included in the review. The quality of evidence was https://doi.org/10.1016/j.jphys.2021.05.008 1836-9553/© 2021 Published by Elsevier B.V. on behalf of Australian Physiotherapy Association. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/). Commentary In people with hemiparesis after stroke, intensive upper limb motor training.2,3 One consideration here may be ceiling effects with robotic practice, such as robot-assisted training, can lead to clinically meaningful training, which by nature is more intensive than traditional motor improvement, yet the benefit typically falls far short of full functional therapies. Future trials may evaluate whether non-invasive brain stimula- recovery. Supplemental therapies have the potential to enhance training tion alters the recovery rate across a training regimen, in addition to the effects, and non-invasive brain stimulation as a candidate add-on therapy more commonly assessed effect magnitude or number of clinical responders. has previously been reported; this literature was systematically reviewed by Reis and colleagues. The review found that there is no beneficial effect This work makes an important and timely contribution to the litera- of non-invasive brain stimulation as a supplement to robot-assisted ture in an area of continued scientific and clinical interest. Understanding training. The eight reviewed trials used a variety of existing best non- intervention features that are ineffective helps limit premature clinical invasive brain stimulation practices and available robotic technology. adoption and shape future trials. The interpretation of these findings should consider the diversity of Provenance: Invited. Not peer reviewed. non-invasive brain stimulation methods in the reviewed studies. This included inhibitory or excitatory protocols, and stimulation by trans- Dylan J Edwards cranial electric or magnetic stimulators, on a background of varied uni- Moss Rehabilitation Research Institute, Philadelphia, USA lateral or bilateral robotic training of different doses and/or intensities. Taken together with the modest aggregate sample size, this challenges the Edith Cowan University, Joondalup, Australia ability to speak to all forms of non-invasive brain stimulation being ineffective. Modifying non-invasive brain stimulation parameters or https://doi.org/10.1016/j.jphys.2021.05.007 methods has been likened to manipulating a drug’s chemical composi- tion,1 fundamentally changing the action and effect. Thus, the forms of References non-invasive brain stimulation across trials were fundamentally different. 1. Bikson M, et al. Nature. 2013;501:167. A clinical benefit of supplemental non-invasive brain stimulation has 2. Allman C, et al. Sci Transl Med. 2016;8:330re1. been demonstrated in individual studies of non-robotic upper limb 3. Lindenberg R, et al. Neurology. 2010;75:2176–2184. 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/).
Journal of Physiotherapy 67 (2021) 197–200 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 Feeling reassured after a consultation does not reduce disability or healthcare use in people with acute low back pain: a mediation analysis of a randomised trial Aidan G Cashin a,b, Hopin Lee c,d, Adrian C Traeger e, Markus Hübscher a, Ian W Skinner f, James H McAuley a,g a Centre for Pain IMPACT, Neuroscience Research Australia, Sydney, Australia; b Prince of Wales Clinical School, Faculty of Medicine, University of New South Wales, Sydney, Australia; c Oxford Clinical Trials Research Unit and Centre for Statistics in Medicine, Nuffield Department of Orthopaedics Rheumatology and Musculoskeletal Sciences (NDORMS), University of Oxford, Oxford, United Kingdom; d School of Medicine and Public Health, University of Newcastle, Newcastle; e Institute for Musculoskeletal Health, Faculty of Medicine and Health, University of Sydney, Sydney, Australia; f School of Allied Health, Exercise and Sports Science, Faculty of Science and Health, Charles Sturt University, Port Macquarie, Australia; g School of Health Sciences, Faculty of Medicine, University of New South Wales, Sydney, Australia KEY WORDS ABSTRACT Mediation analysis Question: Does feeling reassured after a consultation reduce future disability or healthcare use in people Reassurance with acute low back pain (LBP)? Design: Mediation analysis of a randomised, sham-controlled trial. Low back pain Participants: Two hundred and two people with acute LBP at above average risk (high risk) of developing Patient education chronic LBP. Intervention: All participants received guideline-based care from their usual clinician. Partic- Physical therapy ipants received two additional 1-hour sessions of patient education focused on emphasising the benign nature of LBP or sham patient education that included active listening only. Outcome measures: The two primary outcomes for this study were self-reported disability and healthcare use. The mediator was feeling reassured that LBP was not caused by serious illness. Results: Data from 194 (96%) participants and 178 (88%) participants were included in the analysis for disability and healthcare use outcome models, respectively. Feeling reassured did not mediate the effect of the intervention on disability (indirect effect 20.23, 95% CI 20.71 to 0.18) or healthcare use (indirect effect 0.00, 95% CI 20.04 to 0.04). Patient education intervention increased feeling reassured (1.14 points, 95% CI 0.43 to 1.85) compared with sham patient education. However, the mediator (ie, feeling reassured) did not reduce disability (20.20 points, 95% CI 20.58 to 0.19) or healthcare use (OR 1.09, 95% CI 0.98 to 1.21). Conclusion: Feeling reassured after a consultation did not lead to improvements in disability and healthcare use for people with acute LBP. Clinicians should reflect on the time that they allocate to reassuring their patients and consider reallocating time to other aspects of the consultation that could reduce disability and future healthcare use. Trial registration: ACTRN12612001180808, study protocol https://osf.io/4tfaz/. [Cashin AG, Lee H, Traeger AC, Hübscher M, Skinner IW, McAuley JH (2021) Feeling reassured after a consultation does not reduce disability or healthcare use in people with acute low back pain: a mediation analysis of a randomised trial. Journal of Physiotherapy 67:197–200] © 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/). Introduction Low back pain is the second most common symptom-related reason to consult with a primary care clinician4 and the majority of There are increasing demands on primary care clinicians and more people with low back pain seek care.5 For nearly three decades low than half report feeling stressed due to a lack of time.1 The efficient back pain has been the leading cause of disability worldwide.6 The use of the consultation is critical to provide guideline-based care that is individualised for the patient. Clinicians often spend time reas- healthcare burden associated with low back pain is increasing more suring patients, which is a core aspect of daily practice.2 Clinicians rapidly than for any other disease or condition.7 reassure patients to reduce their fears and worries about their health and to change future healthcare behaviours3 such as self-reported Primary care guidelines for the management of low back pain disability and healthcare use. Little is known about whether feeling reassured can change self-reported disability and healthcare use. It is advise clinicians to focus on providing reassurance that back pain is important to understand these effects so that clinicians can efficiently benign and on advice that encourages self-management.8 Feeling manage their consultation by focusing on aspects of care that are likely to improve patient outcomes. reassured (ie, the removal of fears about serious illness) can be effectively achieved by using patient education for those with low back pain.3 The role of feeling reassured following a patient education inter- vention on health outcomes can be investigated through causal https://doi.org/10.1016/j.jphys.2021.06.007 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/).
198 Cashin et al: Reassurance for acute low back pain inference methods such as mediation analysis.9 This method can be Statistical analysis used to investigate the mechanisms by which health interventions are expected to have their effects on health outcomes by partitioning We followed a preregistered analysis plan for the mediation the total effect of an intervention into an indirect effect, which analysis (https://osf.io/4tfaz/). We used a model-based inference operates through a selected mechanism of interest, and a direct effect, approach to mediation analyses developed by Imai, Keel and Ting- which operates through all other mechanisms.10,11 Well-conducted ley.21 All analyses were conducted in R softwarea using the ‘media- mediation analyses can guide the optimisation and implementation tion’ package.22 of health interventions into clinical practice.12 We constructed a single mediator model for each outcome Therefore, the study question for this mediation analysis was: (disability and healthcare use). For each mediator model, we esti- mated the intervention-mediator effect, the mediator-outcome effect, Does feeling reassured after a consultation reduce future disability the average causal mediation effect (ACME), the average direct effect or healthcare use in people with acute low back pain? (ADE) and the average total effect (ATE). The ACME is the average intervention effect through the mediator; the ADE is the average Method intervention effect that works through all other mechanisms, excluding the selected mediator; and the ATE is the average effect of This study was a mediation analysis10 of a randomised, the intervention on the outcome. The ATE is the sum of the ACME and sham-controlled clinical trial (the PREVENT trial, ADE on the additive scale. ACTRN12612001180808)13 that compared patient education with sham patient education. The methods and main results for the pri- For each causal model, we fitted two regression models: a medi- mary outcome (ie, pain) have been reported elsewhere.13 Briefly, 202 ator model and an outcome model. The mediator models used linear participants aged between 18 and 75 who all had low back pain with regression with treatment allocation as the independent variable and or without leg pain for , 6 weeks’ duration were recruited from the mediator as the dependent variable, with the baseline level of the primary care in Sydney, Australia, and enrolled in the study. Between mediator as a covariate. The outcome model for disability used linear 10 September 2013 and 2 December 2015, participants who were at regression and the outcome model for healthcare use used probit above average risk (high risk) of developing chronic low back pain regression. Each outcome model was constructed with the mediator based on the Predicting the Inception of Chronic Pain (PICKUP) Tool – as the independent variable, the outcome as the dependent variable, a validated prognostic model14 – were randomised in a 1:1 ratio to and the treatment allocation, baseline values of the meditator and receive either patient education or sham patient education. Partici- outcome variable (disability model only) in addition to the set of pants received two 1-hour treatment sessions of either patient edu- observed pre-treatment confounders as covariates. To improve model cation focused on emphasising the biopsychosocial and benign nature flexibility, we included an interaction term (treatment allocation 3 of low back pain adapted from the book Explain Pain15 for people with mediator) in the outcome models.11 We used the ‘mediate’ function22 acute low back pain at high risk of chronicity, or sham patient edu- to compute ATE, ACME and ADE. We computed 1,000 bootstrapped cation that only included active listening designed to control for time simulations to generate 95% confidence intervals. We interpreted the with an expert clinician. The protocol for this current study was average of the two conditional effects for the ACME and ADE as there preregistered prior to data analysis on the Open Science Framework at was no evidence for a significant (p , 0.05) intervention-mediator https://osf.io/4tfaz/. interaction. We conducted all analyses on complete cases (,15% missing in all models). Mediator and outcomes Results The two primary outcomes for this current study were the mean Data from 194 (96%) participants and 178 (88%) participants were rating of disability (as self-reported on the 24-item Roland-Morris included in the analysis for disability and healthcare use outcome Disability questionnaire,16 assessed 3 months after the onset of low models, respectively. Missing data was due to participants lost to back pain) and healthcare use (dichotomised as having sought follow-up (lost contact) at 3 and 12 months. The baseline de- healthcare for low back pain or not in the last 6 months, assessed 12 mographics of the 202 participants randomised for the trial are months after the onset of low back pain). Disability and healthcare shown in Table 1. use were selected as the primary outcomes because they have been proposed, at least in part, to be influenced by feeling reassured (the Patient education compared with sham education increased proposed mediator)17,18 and by patient education.13 Feeling reassured feeling reassured 1 week after the intervention by 1.14 points (95% CI that low back pain was not caused by serious illness was selected as 0.43 to 1.85, p = 0.002), as shown in Table 2. However, feeling reas- the mediator because one of the primary aims of patient education is sured was not associated with either disability (20.20 points, 95% to reassure the patient.3 Feeling reassured was assessed on a 0 to 10 CI 20.58 to 0.19, p = 0.32) or healthcare use (OR 1.09, 95% CI 0.98 to Likert scale (0 not reassured to 10 completely reassured) in response 1.21, p = 0.79), as shown in Table 2. The ACME for disability was 20.23 to the question ‘How reassured do you feel that there is no serious (95% CI 20.71 to 0.18, p = 0.23) and for healthcare use 0.00 (95% condition causing your back pain?’ adapted from Deyo et al,19 at CI 20.04 to 0.04, p = 0.82), as shown in Figure 3. 1 week after the two consultations. Mediator and outcome assess- ments were assessed by a blinded assessor and self-reported by the We did not conduct the pre-planned sensitivity analyses or the patient. secondary analyses investigating the magnitude of change required in the mediator reassurance to produce a minimum clinically important Causal assumptions difference in the outcomes disability and healthcare use. These sensitivity and secondary analyses were abandoned because the mediator was not found to be associated with the outcomes and there was no evidence for an indirect effect. We assumed no confounding of the intervention-mediator and Discussion intervention-outcome effects because treatment allocation was randomised. To identify possible confounders of the mediator- Reducing fears of serious illness in people with acute low back outcome effect, we constructed directed acyclic graphs,20 as pre- pain did not lead to improvements in disability or reduce healthcare sented in Figures 1 and 2 on the eAddenda. The minimum sufficient use. Although clinical guidelines recommend that clinicians spend set of possible confounders for the mediator-outcome effects time reassuring patients with acute low back pain, this study suggests included: healthcare setting, back beliefs, pain self-efficacy, pain that if the objective of the consultation is to improve patient catastrophising, and depressive symptoms.
Research 199 Table 1 Total Baseline characteristics of the participants randomised in the trial. (n = 202) Characteristic Patient Sham 45.1 (14.5) education education 103 (51) (n = 101) (n = 101) 6.2 (2.3) 11.4 (5.6) Age (y), mean (SD) 46.5 (14.7) 43.8 (14.1) 5.5 (2.7) 53 (52) 50 (50) Female, n (%) 6.3 (2.4) 6.1 (2.2) 167 (83) Pain intensity (0 to 10), mean (SD)a 11.0 (5.4) 11.7 (5.8) 28.0 (6.6) Disability (0 to 24), mean (SD)b 5.6 (2.7) 5.4 (2.7) 34.3 (13.0) Feeling reassured (0 to 10), mean (SD)c 19.1 (11.6) 82 (81) 85 (84) 4.6 (4.4) Possible confounders 27.7 (6.8) 28.3 (6.4) Clinician healthcare setting, n (%)d 35.5 (13.1) 33.2 (13.0) Back beliefs (9 to 45), mean (SD)e 18.4 (12.0) 19.9 (11.2) Pain self-efficacy (0 to 60), mean (SD)f 4.1 (3.7) 5.1 (5.0) Catastrophising (0 to 52), mean (SD)g Depressive symptoms (0 to 21), mean (SD)h a Numeric rating scale with range from 0 (no pain) to 10 (worst pain possible). b Roland Morris Disability Questionnaire with range from 0 (no disability) to 24 (high disability). c ‘How reassured do you feel that there is no serious condition causing your back pain?’ Range from 0 (not reassured at all) to 10 (completely reassured). d Referred by a physiotherapist. e Back Beliefs Questionnaire with range from 9 (maladaptive or pessimistic beliefs) to 45 (helpful or positive beliefs). f Pain Self-Efficacy Questionnaire with range from 0 (low pain self-efficacy) to 60 (high pain self-efficacy). g Pain Catastrophizing Scale with range from 0 (low catastrophising) to 52 (high catastrophising). h Depression severity scale of Depression, Anxiety and Stress Scale with range from 0 (no depressive symptoms) to 21 (high depressive symptoms). Table 2 Effect decomposition for the effect of patient education on disability and healthcare use, with feeling reassured as the hypothesised mediator. Analysis Intervention-mediator Mediator-outcome ATE ADE ACME effect (path a) effect (path b) 20.23 Feeling reassured 1.14 20.20 21.28 21.05 (20.71 to 0.18) Disability (0.43 to 1.85)a (20.58 to 0.19) (22.70 to 0.15) (22.49 to 0.42) 0.00 Healthcare use 1.14 1.09 20.06 20.06 (20.04 to 0.04) (0.43 to 1.85)a (0.98 to 1.21)b (20.20 to 0.09) (20.21 to 0.10) All effects unstandardised and presented with their 95% confidence intervals. ACME = average causal mediation effect, ADE = average direct effect, ATE = average total effect. a P value , 0.05. b The healthcare use mediator-outcome effect is presented as an odds ratio. A Patients commonly request, and clinicians routinely order, diag- nostic imaging of the spine to provide reassurance. Data now sug- ACME gest that imaging does not reduce fears23 or improve outcomes24 and is therefore not recommended for non-specific low back ADE pain.8 Clinicians may also consider patient education to reduce fears of serious illness.3 The findings from this study show that although Total effect 0 1 patient education can effectively reassure patients, this does not –4 –3 –2 –1 lead to improved healthcare behaviours. Considering the time re- straints within clinical practice, clinicians should be selective in B how much time they allocate to providing reassurance. Although it is natural for a clinician to reassure a patient who is distressed, ACME clinicians risk overestimating the impact of feeling reassured on outcomes. Clinicians could spend more time: listening to their ADE patient’s story;25 estimating and discussing their prognosis;14 and matching their preferences with recommended non- Total effect pharmacological treatment options and self-management strategies.8 –1.0 –0.5 0.0 0.5 1.0 The main limitation of this study is that the duration of patient Figure 3. Effect decomposition plots for each mediator model: (a) disability and education intervention delivered was greater than what is typically (b) healthcare use. provided in primary care. Although the interpretation of the effect For the ACME and ADE, solid dots and lines represent point estimates and 95% con- of feeling reassured on disability and healthcare use is in relation fidence intervals for the patient education group; the hollow dots and broken lines to this 2-hour intervention, it is unlikely that a shorter, more represent point estimates and confidence intervals for the sham patient education pragmatic patient education intervention would have shown a group. For the total effect, the solid dot and line represent the point estimate and 95% mediating effect. Finally, the study population excluded people with confidence interval for the marginal effect. All effects are reported unstandardised acute low back pain who were at lower than average risk of pain with their 95% confidence intervals. chronicity, which may have influenced the generalisability of ACME = average causal mediation effect, ADE = average direct effect. findings. outcomes such as disability or healthcare use, clinicians should pri- Comprehensive attempts to remove a patient’s fear of serious oritise other aspects of the clinical encounter. illness (eg, through extended consultations) are unlikely to improve outcomes and therefore may be unwarranted. Although reassurance During the consultation, clinicians may feel compelled to reas- is a core aspect of clinical practice, clinicians should consider reallo- sure their patient that no serious illness is causing their back pain. cating time to other aspects of the consultation that could reduce disability and future healthcare use.
200 Cashin et al: Reassurance for acute low back pain What was already known on this topic: Clinical guidelines population. Arch Intern Med. 2012;172:1377–1385. https://doi.org/10.1001/ for the management of low back pain advise clinicians to spend archinternmed.2012.3199. time reassuring patients to aid recovery. Patient education is an 2. Kessel N. Reassurance. Lancet. 1979;313:1128–1133. effective means of reassuring patients. 3. Traeger AC, Hübscher M, Henschke N, Moseley GL, Lee H, McAuley JH. Effect What this study adds: Patient reports of feeling reassured of primary care based education on reassurance in patients with acute low following patient education consultations did not lead to improve- back pain systematic review and meta-analysis. JAMA Intern Med. 2015;175: ments in disability and healthcare use. Clinicians should reconsider 733–743. the role of efforts to reassure their patients in consultations for 4. Deyo RA, Weinstein JN. Low back pain. N Engl J Med. 2001;344:363–370. acute low back pain. 5. Ferreira ML, Machado G, Latimer J, Maher C, Ferreira PH, Smeets RJ. Factors defining care-seeking in low back pain - A meta-analysis of population based Footnotes: a R software V3.6.1, R Core Team, Vienna, Austria. surveys. Eur J Pain. 2010;14:747.e1–747.e7. eAddenda: Figures 1 and 2 can be found online at https://doi.org/ 6. James SL, Abate D, Abate KH, Abay SM, Abbafati C, Abbasi N, et al. Global, 10.1016/j.jphys.2021.06.007. regional, and national incidence, prevalence, and years lived with disability for Ethics approval: The PREVENT trial was approved by the Uni- 354 diseases and injuries for 195 countries and territories, 1990–2017: a sys- versity of New South Wales Human Research Ethics Committee, tematic analysis for the Global Burden of Disease Study 2017. Lancet. Sydney, New South Wales, Australia on 05 February 2013 (reference 2018;392:1789–1858. number: HC12664). Written and informed consent was obtained 7. Dieleman JL, Baral R, Birger M, Bui AL, Bulchis A, Chapin A, et al. US spending from all participants before they enrolled in the trial. No additional on personal health care and public health, 1996-2013. JAMA. 2016;316:2627– approval was required for this mediation analysis of that trial. 2646. Competing interests: No disclosures were reported. 8. Oliveira CB, Maher CG, Pinto RZ, Traeger AC, Lin CW, Chenot JF, et al. Clinical Source(s) of support: The Australian National Health and Medical practice guidelines for the management of non-specific low back pain in primary Research Council funded this trial (project identification number: care: an updated overview. Eur Spine J. 2018;27:2791–2803. 1047827), which was investigator initiated (chief investigator, Prof 9. Herbert RD. Research note: causal inference. J Physiother. 2020;66:273–277. McAuley). Mr Cashin was supported by the University of New South 10. Lee H, Herbert RD, McAuley JH. Mediation analysis. JAMA. 2019;321:697–698. Wales Prince of Wales Clinical School Postgraduate Research Schol- 11. Vanderweele TJ. Explanation in Causal Inference. Oxford University Press; 2015. arship and a NeuRA PhD Candidature Supplementary Scholarship. Dr 12. Cashin AG, Lee H. An introduction to mediation analyses of randomised controlled Lee and Dr Traeger were supported by National Health and Medical trials. J Clin Epidemiol. 2021. https://doi.org/10.1016/j.jclinepi.2021.02.014. Research Council research fellowships. Dr Lee was also supported by 13. Traeger AC, Lee H, Hübscher M, Skinner IW, Moseley GL, Nicholas MK, et al. Effect the National Institute for Health Research (NIHR) Collaboration for of intensive patient education vs placebo patient education on outcomes in pa- Leadership in Applied Health Research and Care Oxford at Oxford tients with acute low back pain: a randomized clinical trial. JAMA Neurol. Health NHS Foundation Trust. The funders/sponsors had no role in 2019;76:161–169. the design and conduct of the study; collection, management, anal- 14. Traeger AC, Henschke N, Hübscher M, Williams CM, Kamper SJ, Maher CG, et al. ysis, and interpretation of the data; preparation, review, or approval Estimating the risk of chronic pain: development and validation of a prognostic of the manuscript; and decision to submit the manuscript for model (PICKUP) for patients with acute low back pain. PLoS Med. 2016;13: publication. e1002019. Acknowledgements: Nil. 15. Butler DS, Moseley GL. Explain Pain. 2nd ed. Adelaide: Noigroup Publications; 2013. Data sharing: Data and statistical/analytic code available on 16. Roland M, Morris R. A study of the natural history of back pain. Part I: development request from the corresponding author. of a reliable and sensitive measure of disability in low-back pain. Spine. Provenance: Not invited. Peer reviewed. 1983;8:141–144. Correspondence: Dr Hopin Lee, Nuffield Department of Ortho- 17. Pincus T, Holt N, Vogel S, Underwood M, Savage R, Walsh DA, et al. Cognitive and paedics Rheumatology and Musculoskeletal Sciences, University of affective reassurance and patient outcomes in primary care: a systematic review. Oxford, Oxford, United Kingdom. Email: [email protected] Pain. 2013;154:2407–2416. 18. Traeger AC, O’Hagan ET, Cashin A, McAuley JH. Reassurance for patients with non- References specific conditions – a user’s guide. Braz J Phys Ther. 2017;21:1–6. 19. Deyo RA, Diehl AK, Rosenthal M. Reducing roentgenography use: can patient ex- 1. Shanafelt TD, Boone S, Tan L, Dyrbye LN, Sotile W, Satele D, et al. Burnout and pectations be altered? Arch Intern Med. 1987;147:141–145. satisfaction with work-life balance among US physicians relative to the general US 20. Textor J, van der Zander B, Gilthorpe MS, Liskiewicz M, Ellison GT. Robust causal inference using directed acyclic graphs: the R package “dagitty”. Int J Epidemiol. 2016;45:1887–1894. 21. Imai K, Keele L, Tingley D. A general approach to causal mediation analysis. Psychol Methods. 2010;15:309–334. 22. Tingley D, Yamamoto T, Hirose K, Keele L, Imai K. Mediation: R package for causal mediation analysis. J Stat Softw. 2014;59. 23. Rolfe A, Burton C. Reassurance after diagnostic testing with a low pretest proba- bility of serious disease: systematic review and meta-analysis. JAMA Intern Med. 2013;173:407–416. 24. Van Ravesteijn H, Van Dijk I, Darmon D, van de Laar F, Lucassen P, Hartman TO, et al. The reassuring value of diagnostic tests: a systematic review. Patient Educ Couns. 2012;86:3–8. 25. Zulman DM, Haverfield MC, Shaw JG, Brown-Johnson CG, Schwartz R, Tierney AA, et al. Practices to foster physician presence and connection with patients in the clinical encounter. JAMA. 2020;323:70–81.
Journal of Physiotherapy 67 (2021) 177–189 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 Patient education improves pain and function in people with knee osteoarthritis with better effects when combined with exercise therapy: a systematic review Anthony J Goff a,b, Danilo De Oliveira Silva a, Mark Merolli c, Emily C Bell a, Kay M Crossley a, Christian J Barton a,d,e a La Trobe Sport and Exercise Medicine Research Centre, School of Allied Health, Human Services and Sport, La Trobe University, Melbourne, Australia; b Health and Social Sciences, Singapore Institute of Technology, Singapore; c Centre for Health, Exercise, and Sports Medicine, The University of Melbourne, Melbourne, Australia; d Department of Physiotherapy, Podiatry and Prosthetics and Orthotics, School of Allied Health, Human Services and Sport, La Trobe University, Melbourne, Australia; e Department of Surgery, St Vincent’s Hospital, University of Melbourne, Melbourne, Australia KEY WORDS ABSTRACT Patient education Question: Is patient education effective as a standalone intervention or combined with other interventions Knee for people with knee osteoarthritis? Design: Systematic review of randomised controlled trials. MEDLINE, Osteoarthritis EMBASE, SPORTDiscus, CINAHL and Web of Science were searched from inception to April 2020. The Physical therapy Cochrane Risk of Bias tool was used for included studies, and Grading of Recommendations, Assessment, Systematic review Development and Evaluations (GRADE) was used to interpret certainty of results. Participants: People with knee osteoarthritis. Intervention: Any patient education intervention compared with any non- pharmacological comparator. Outcome measures: Primary outcomes were self-reported pain and func- tion. Results: Twenty-nine trials involving 4,107 participants were included, informing low to very-low certainty evidence. Nineteen of 28 (68%) pooled comparisons were not statistically significant. Patient ed- ucation was superior to usual care for pain (SMD 20.35, 95% CI 20.56 to 20.14) and function in the short term (20.31, 95% CI 20.62 to 0.00), but inferior to exercise therapy for pain in the short term (0.77, 95% CI 0.07 to 1.47). Combining patient education with exercise therapy produced superior outcomes compared with patient education alone for pain in the short term (0.44, 95% CI 0.19 to 0.69) and function in the short (0.81, 95% CI 0.54 to 1.08) and medium term (0.39, 95% CI 0.15 to 0.62). When using the Western Ontario and McMaster Universities Osteoarthritis Index for these comparisons, clinically important differences indicated that patient education was inferior to exercise therapy for pain in the short term (MD 1.56, 95% CI 0.14 to 2.98) and the combination of patient education and exercise therapy for function in the short term (8.94, 95% CI 6.05 to 11.82). Conclusion: Although patient education produced statistically superior short-term pain and function outcomes compared with usual care, differences were small and may not be clinically important. Patient education should not be provided as a standalone treatment and should be combined with exercise therapy to provide statistically superior and clinically important short-term improvements in function compared with education alone. Registration: PROSPERO CRD42019122004. [Goff AJ, De Oliveira Silva D, Merolli M, Bell EC, Crossley KM, Barton CJ (2021) Patient education improves pain and function in people with knee osteoarthritis with better effects when combined with exercise therapy: a systematic review. Journal of Physiotherapy 67:177–189] © 2021 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 These recommendations are supported by compelling evidence of the effectiveness and cost-effectiveness of exercise therapy8–10 and Knee osteoarthritis (OA) is a leading cause of disability worldwide, weight management11–13 in people with knee OA. However, the in- affecting up to one in four people over the age of 50 years.1,2 The growing healthcare burden related to knee OA in many developed clusion of patient education as a first-line intervention for people countries is considered unsustainable. For example, A$905 million was spent in Australia in 2013 on knee replacement surgery alone, a with knee OA in clinical practice guidelines is often justified by evi- figure expected to rise to $1.38 billion by 2030.3 All major clinical practice guidelines recommend patient education, exercise therapy dence relating to people with OA elsewhere in the body, other forms and weight management as first-line interventions for knee OA.4–7 of arthritis or chronic pain.14–17 The search for the most recently published high-quality evidence synthesis evaluating the effectiveness of patient education on pain and function in people with OA was completed in 2012.18 This https://doi.org/10.1016/j.jphys.2021.06.011 1836-9553/© 2021 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/).
178 Goff et al: Patient education for knee osteoarthritis Cochrane review18 did not distinguish knee OA from other arthritic Box 1. Inclusion criteria. conditions, and reported little to no benefit of patient education programs for pain, function and quality of life compared with Design providing information only, usual care only or no treatment. Other Randomised controlled trials including cluster randomised more-recent reviews have focused on self-efficacy and quality of life outcomes in people with knee OA following education.19,20 The re- controlled trials view that examined self-efficacy reported no difference when patient Participants education was compared with a combination of patient education People with clinically or radiographically diagnosed knee and exercise therapy.19 The review that examined quality of life re- ported no difference when self-directed education was compared osteoarthritis with a combination of self-directed education and physical activity or Intervention therapist-facilitated patient education alone.20 No recent compre- Any form of patient education hensive review specifically evaluating pain and function outcomes in Outcome measures people with knee OA exists to guide clinical practice guidelines. Primary: self-reported joint-related pain or function scales Secondary: self-reported psychological outcomes The primary objectives of this review were to estimate the effects Comparisons of patient education on self-reported pain and function outcomes as a Any other non-pharmacological intervention including usual standalone intervention or in combination with other interventions for people with knee OA. The secondary aims of this review were to care or no treatment estimate the effects of patient education on psychological outcomes and to estimate the effects of therapist-facilitated education outcome data; selective reporting; and other bias (imbalances in compared with self-directed education on pain and function out- baseline characteristics and compliance with the intervention). comes for people with knee OA. Following consultation and agreement between three people in the research team (AJG, DOS and CJB), trials were classified as low risk of Therefore, the research question for this systematic review was: bias when they appropriately reported that they met at least four of these seven criteria, and high risk of bias otherwise. Is patient education effective as a standalone intervention or combined with other interventions for people with knee osteoarthritis? Methods Data analysis The protocol for this systematic review was prospectively regis- Participant and study characteristics and means and SDs for tered in January 2019. Design and reporting of this review followed primary and secondary outcomes were extracted by two reviewers the Preferred Reporting Items for Systematic Reviews and Meta- independently (AJG and ECB). Primary outcomes were self- Analyses (PRISMA) statement.21 reported joint-related pain and function measures, such as West- ern Ontario and McMaster Universities Osteoarthritis Index26 Identification and selection of studies (WOMAC) and a visual analogue scale for pain. Secondary out- comes were self-reported psychological measures, such as the A comprehensive search strategy was devised based on the arthritis self-efficacy scale,27 pain catastrophising scale28 and Cochrane Handbook22 and a previous review investigating patient Coping Strategies Questionnaire.29 Authors were contacted a education in patellofemoral pain,23 and applied to each of the maximum of two times via email to request necessary data when it following databases from inception to April 2020: MEDLINE via OVID, could not be extracted from a published manuscript (eg, data were EMBASE via OVID, SPORTDiscus via EBSCO, CINAHL via EBSCO, and pooled for hip and knee OA, or mean changes were reported) before Web of Science. The search strategy was developed using Medical the manuscript was excluded. When a trial reported data that Subject Headings and key words to identify randomised controlled required transformation to a different statistic for meta-analysis, trials in adults with knee OA. The search strategies can be found in appropriate calculations were made according to the Cochrane Appendix 1 on the eAddenda. Handbook22 and previous research.30 All references were imported into a reference management soft- Data analysis of primary and secondary outcomes was completed warea and duplicates were removed. Two reviewers (AJG and DOS) using Cochrane Collaboration softwareb. Data were pooled when independently reviewed the titles and abstracts, obtaining full-text trials investigated similar patient education interventions as a stand- copies of potentially eligible articles for review. Full texts were then alone intervention or in combination with other interventions. independently reviewed by the same two reviewers to determine Overall estimate of effect was calculated using a random-effects eligibility based upon the inclusion criteria shown in Box 1. In the model and reported as a SMD and 95% CI. Based upon consider- case of disagreements, a third reviewer (CJB) was consulted to reach ation of recommendations from Cohen31 and the Cochrane Hand- consensus. book,22 the effect sizes were categorised as small (, 0.3), moderate (0.3 to , 0.5), large (0.5 to , 0.8) or very large ( 0.80). Heterogeneity Randomised controlled trials, including cluster randomised trials, was quantified with the I2 statistic. delivering any form of patient education for people with either clinical or radiographically confirmed knee OA,4–7,24 compared with The certainty of evidence for pooled trials was assessed and any non-pharmacological intervention were considered for inclusion interpreted using Grading of Recommendations, Assessment, Devel- in this review, even if the patient educational intervention was the opment and Evaluations (GRADE)32,33 and summarised using GRADE control intervention. No date, setting or language restrictions were Pro Softwarec. Although it was planned to use a modified version of applied. Non-randomised controlled trials, cross-sectional studies, van Tulder’s criteria,34 it was decided to follow the Cochrane Hand- case series and case reports were excluded from this review. book’s recommendation to use GRADE. Full details of upgrade and downgrade criteria for all categories of GRADE, including heteroge- Assessment of characteristics of studies neity, can be found in Table 1 on the eAddenda. Risk of bias was assessed by two independent reviewers (AJG and Data that could not be pooled were presented in table format MM) using the Cochrane Risk of Bias Tool25 categories: random and pooled data were presented using forest plots and summar- sequence generation; allocation concealment; blinding of partici- ised as SMDs and 95% CIs. In addition to this planned analysis, pants and personnel; blinding of outcome assessment; incomplete when all trials containing the same intervention type used the same outcome measure, MD was also calculated using a random- effects model to aid clinical interpretation. Each was subsequently
Research 179 compared against suggested minimum clinically important dif- Unique records screened by title and ferences (MCID) in published literature.35,36 MCIDs of 1.5 points abstract (n = 4,528) for pain35 and 6 points for function36 were nominated on the respective subsections of the WOMAC outcome measure. When a Papers excluded after screening manuscript represented mean or MCID in a scaled format, it was titles/abstracts (n = 4,400) converted back into the outcome’s original raw form for pooling and interpretation. Possibly relevant papers retrieved for evaluation of full text (n = 128) Due to large variation in when outcome measures were assessed, we introduced subgrouping of short-term (, 6 months), medium- Papers excluded after evaluation of term (6 to , 12 months) and long-term ( 12 months) results full text (n = 99) where possible. These timelines are in line with Cochrane reviews x wrong study type to assess investigating patient education in OA and exercise therapy in knee OA.18,37 patient education (n = 47) x unable to acquire data to be A post hoc comparison between therapist-facilitated and self- directed education was deemed important, considering potential included in review (n = 25) differences in healthcare resources and outcomes between the two. x wrong outcomes (n = 13) Therapist-facilitated education was classified as any educational x wrong interventions (n = 8) intervention where the education was actively facilitated by a x wrong population (n = 6) healthcare professional, regardless of profession (eg, physiotherapist, dietician, doctor), including one-to-one consultations, group classes, Papers included in this review (n = 29) telephone consultations and telerehabilitation. Self-directed educa- tion was classified as any educational intervention that did not Figure 1. Flow of trials through the review. involve a healthcare professional explanation or opportunity for participants to ask questions related to the educational content (eg, leaflets, booklets, websites). Although sensitivity analyses of effect were not planned, they were deemed necessary due to included trials with: multiple groups with similar interventions, multiple data outcomes within the same pre-specified time point, or multiple outcome measures for the same construct. A planned mixed-methods analysis including a content evaluation of included trials and a cross-sectional analysis of general web con- tent will be published elsewhere. Splitting of these further evalua- tions from this systematic review was considered necessary to improve clarity and impact of each component. Results Flow of studies through the review Effects of intervention Flow of trials through the review process can be found in Figure 1. The results of data pooling for primary outcomes are shown in Following removal of duplicates, 4,528 records were screened and Figure 3 (ie, the effects of patient education as a standalone inter- 128 full-text articles were assessed for eligibility. This assessment led vention) and Figure 4 (ie, the effects of patient education in combi- to the exclusion of 99 articles, primarily due to an ineligible study nation with other interventions). The summarised results of pooled design (n = 47) or inability to acquire the necessary data from authors outcomes for secondary outcomes can be found in Table 4. The results (n = 25). Full details of all excluded trials can be found in Appendix 2 of data pooling for secondary outcomes can be found in Figure 5 on on the eAddenda. Twenty-nine trials involving 4,107 participants the eAddenda. Summary GRADE tables for all comparisons are shown were included for analysis. in Table 5 on the eAddenda. The summarised results of un-pooled data are shown in Table 6 on the eAddenda. The summarised re- Characteristics of trials sults of MD analysis for all possible comparisons are shown in Table 7. Forest plots for all MD comparisons can be found in Figure 6 on the Characteristics of the 29 included trials are provided in Table 2. eAddenda. Note that more detailed forest plots for Figures 3 and 4 are Twenty-eight trials included evaluation of patient education as a available as Figures 7 and 8 on the eAddenda. standalone intervention.38–65 When patient education was combined with other interventions, it was always combined with exercise Patient education as a standalone intervention therapy (n = 10).43–45,51,53,56–58,62,66 A total of 41 patient education in- terventions were identified across the 29 trials. Of the 41 patient Very-low certainty evidence indicated that patient education is education interventions, 14 were provided as a control, all of which superior to usual care for pain in the short term (SMD 20.35, 95% were provided as standalone interventions.38,42,44–47,49–52,56,58,62,63,65 CI 20.56 to 20.14) based on six trials41,48,54,55,60,61 (Figure 3a). Low More details can be found for the included trials and interventions in certainty evidence indicated that patient education produces similar Table 3 on the eAddenda. outcomes to usual care for pain in the medium term (SMD 20.10, 95% CI 20.26 to 0.05, four trials,40,41,48,60 Figure 3b). Very-low certainty Results from risk of bias can be found in Figure 2. Before final evidence indicated that patient education produces similar outcomes decisions were made, there was a 91% agreement rate between the two to usual care for pain in the long term (SMD 20.12, 95% CI 20.30 to independent reviewers. Eleven trials (38%) were classified as low risk 0.05, two trials,39,40 Figure 3c). Very-low certainty evidence indicated of bias according to the definition of 4/7 categories on the Cochrane that patient education is superior to usual care for function in the Risk of Bias Tool,39,42,43,45,47,48,51,53,55,56,60 with the remaining 18 (62%) short term (SMD 20.31, 95% CI 20.62 to 0.00, six trials,41,48,55,59,60,61 classified as having high risk of bias.38,40,41,44,46,49,50,52,54,57–59,61–66 All Figure 3d) but produces similar outcomes for function in the trials were downgraded for performance and detection bias.
Table 2 Characteristics of included trials. Study Participants Eligible population Education Comparator Ackerman et al (2012) Orthopaedic or rheumatology patients Age (y) = 64 (11) Age (y) = 67 (11) M/F (%) = 38/62 M/F (%) = 42/62 Allen et al (2010)a Primary care patients from a BMI = 30 (24 to 35) BMI = 29 (26 to 35 Allen et al (2016) Veterans’ medical centre Allen et al (2019) Age (y) = 60 (10) Age (y) = 60 (11) Ay et al (2013)a Overweight patients recruited from M/F (%) = 92/8 M/F (%) = 93/7 a Veterans’ medical centre BMI = 32.0 (7.0) BMI = 31.6 (6.5) African Americans recruited from a Age (y) = 60 (9) Age (y) = 62 (9) Veterans’ medical centre M/F (%) = 87/13 M/F (%) = 95/5 BMI = 34.3 (6.0) BMI = 33.4 (5.7) NR Age (y) = 59 (10) Age (y) = 59 (11) Baker et al (2001) Community-dwelling M/F (%) = 51/49 M/F (%) = 51/49 BMI = 35.6 (8.4) BMI = 34.8 (7.9) Bennell et al (2016) Community-dwelling Age (y) = 59 (12) Age (y) = 62 (11) M/F (%) = 3/17 M/F (%) = 25/75 BMI = NR BMI = NR Age (y) = 69 (6) Age (y) = 68 (6) M/F (%) = 15/85 M/F (%) = 83/17 BMI = 31 (4) BMI = 32 (5) Age (y) = 63 (8) Age (y) = 63 (8) M/F (%) = 39/61 M/F (%) = 41/59 BMI = 30.8 (20) BMI = 31.5 (6) Brosseau et al (2016)a Community-dwelling Age (y) = 62 (7) Age (y) = 65 (8) Chen et al (2019) Community-dwelling M/F (%) = 37/63 M/F (%) = 40/60 Cheung et al (2017)b Community-dwelling BMI = 29.9 (5.3) BMI = 31.0 (6) Cheung et al (2020) Community-dwelling Coleman et al (2012) Primary care Age (y) = 69 (7) Age (y) = 64 (10) De Rezende et al (2016)b Trauma and orthopaedic patients M/F (%) = 14/86 M/F (%) = 30/70 De Rezende et al (2017)b Trauma and orthopaedic patients BMI = 25.4 (3.5) BMI = 29.4 (5.4) Age (y) = 72 (8) Age (y) = 69 (8) M/F (%) = NR M/F (%) = 17/83 BMI = 27.8 (7.9) BMI = 25 (3.5) Age (y) = 623 (6) Age (y) = 74 (8) M/F (%) = 72/28 M/F (%) = NR BMI = 22.1 (2.1) BMI = 29.2 (7.1) Age (y) = 65 (8) Age (y) = 64 (6) M/F (%) = 20/80 M/F (%) = 82/18 BMI = NR BMI = 22.7 (1.3) Age (y) = NR Age (y) = 65 (9) M/F (%) = NR M/F (%) = 31/69 BMI = NR BMI = NR Age (y) = NR Age (y) = NR M/F (%) = NR M/F (%) = NR BMI = NR BMI = NR Age (y) = NR M/F (%) = NR BMI = NR
Education Intervention 180 Goff et al: Patient education for knee osteoarthritis Comparator Patient education (therapist-facilitated) (n = 58) Patient education 5) (self-directed) (n = 62) Patient education Usual care (n = 172) (therapist-facilitated) (n = 172) Patient education Usual care (n = 149) (therapist-facilitated) (n = 151) Patient education (therapist-facilitated) (n = 124) Usual care (n = 124) Patient education (therapist-facilitated) 1 exercise Exercise therapy (n = 20) therapy (n = 20) Exercise therapy (n = 23) Patient education (therapist-facilitated) (n = 23) Patient education (therapist-facilitated) (n = 74) Exercise therapy (n = 75) Patient education (self-directed) (n = 74) Patient education (therapist-facilitated) 1 Patient education (therapist-facilitated) (n = 70) exercise therapy (n = 73) Patient education (therapist-facilitated) (n = 23) Patient education (self-directed) 1 exercise therapy (n = 79) Patient education (therapist-facilitated) 1 exercise therapy (n = 70) Exercise therapy (n = 28) Patient education (therapist-facilitated) (n = 18) Acupressure (n = 17) Patient education (therapist-facilitated) (n = 71) Usual care (n = 75) Patient education (therapist-facilitated) (Group 1a) (n = 29) Patient education (self-directed) Patient education (therapist-facilitated) (Group 1a) (n = 29) (Group 4b) (n = 29) Patient education (self-directed) (Group 4b) (n = 29)
Table 2 (Continued) Eligible population Participants Comparator Study Community-dwelling Dias et al (2017) Education Age (y) = 71 (5) Ettinger et al (1997)b Community-dwelling M/F (%) = 0/100 Farr et al (2010) Age (y) = 71 (5) BMI = 30.5 (4.3) Community-dwelling M/F (%) = 0/100 Ganji et al (2018) BMI = 30.0 (5.2) Age (y) = 68 (6) Helminen et al (2015) M/F (%) = 27/73 Hinman et al (2020) Age (y) = 69 (6) BMI = NR Keefe et al (2004) M/F (%) = 36/64 BMI = NR Age (y) = 56 (7) Messier et al (2004) M/F (%) = 27/73 Age (y) = 56 (6) BMI = 27.5 (4.5) Murphy et al (2018) M/F (%) = 28/72 O’Brien et al (2018) BMI = 28 (4.0) Age (y) = 54 (7) O’Moore et al (2018) M/F (%) = 21/79 Patients referred to an elderly care clinic Age (y) = 65 (6) BMI = 27.2 (4.2) Recruited from primary healthcare M/F (%) = NR Community-dwelling BMI = NR Age (y) = 65 (5) Patients recruited from rheumatology clinics M/F (%) = NR Age (y) = 65 (7) BMI = NR M/F (%) = 29/71 BMI = 30.1 (6) Age (y) = 63 (7) M/F (%) = 32/68 Age (y) = 63 (8) BMI = 29.9 (6.3) M/F (%) = 38/62 BMI = 31.2 (7.6) Age (y) = 62 (9) M/F (%) = 37/63 Age (y) = 60 (12) BMI = 31.1 (6.8) M/F (%) = 50/50 BMI = NR Age (y) = 60 (9) M/F (%) = 66/34 Community-dwelling Age (y) = 69 (0.1)c BMI = NR M/F (%) = 32/68 Community-dwelling BMI = 34.2 (0.6)c Age (y) = 60 (9) Patients on orthopaedic consultation waitlist M/F (%) = 35/65 Recruited from health care organisations Age (y) = 68 (0.7)c BMI = NR M/F (%) = 28/72 BMI = 34.5 (0.6)c Age (y) = 58 (14) M/F (%) = 39/61 Age (y) = 65 (8) BMI = NR M/F (%) = 23/77 BMI = 32.9 (6.3) Age (y) = 69 (0.8)c M/F (%) = 26/74 Age (y) = 63 (11) BMI = 34.2 (0.6)c M/F (%) = 34/66 BMI = 33.4 (3.4) Age (y) = 69 (0.8)c M/F (%) = 26/74 Age (y) = 63 (7) BMI = 34.0 (0.7)c M/F (%) = 14/86 BMI = NR Age (y) = 61 (9) M/F (%) = 27/73 BMI = 29.8 (5.3) Age (y) = 60 (14) M/F (%) = 42/58 BMI = 32.1 (3.1) Age (y) = 60 (6) M/F (%) = 32/68 BMI = NR
Education Intervention Patient education (therapist-facilitated) (n = 32) Comparator Patient education (therapist-facilitated) 1 exercise therapy (n = 33) Patient education (therapist-facilitated) (n = 149) Exercise therapy (n = 146) Patient education (therapist-facilitated) (n = 57) Exercise therapy (n = 52) Patient education (therapist-facilitated) (n = 42) Patient education (therapist-facilitated) 1 exercise therapy (n = 62) Usual care (n = 41) Patient education (therapist-facilitated) (n = 55) Usual care (n = 56) Patient education (therapist-facilitated) (n = 88) Patient education (therapist-facilitated) 1 exercise Research 181 Patient education (therapist-facilitated) (n = 18) therapy (n = 87) Exercise therapy (n = 16) Patient education (therapist-facilitated) 1 exercise therapy (n = 20) Usual care (n = 18) c Patient education (therapist-facilitated, healthy Exercise therapy (n = 80) lifestyle) (n = 78) Patient education (therapist-facilitated, c Patient education (therapist-facilitated, weight weight loss focus) 1 exercise therapy (n = 76) loss focus) (n = 82) Usual care (n = 15) Patient education (therapist-facilitated) (n = 31) Patient education (therapist-facilitated) (n = 59) Usual care (n = 60) Patient education (self-directed) (n = 44) Usual care (n = 25)
182 Goff et al: Patient education for knee osteoarthritisComparatorPatient education (therapist-facilitated) 1 exerciseExercise therapy (n = 23)Exercise therapy (n = 31) Patient education (self-directed) (n = 73) therapy (n = 40) medium term (SMD 20.17, 95% CI 20.40 to 0.07, four trials,40,41,48,60 Figure 3e).Intervention Very-low certainty evidence indicated that patient education isEducationPatient education (therapist-facilitated) (n = 20)Patient education (therapist-facilitated) (n = 23)Patient education (therapist-facilitated) (n = 29)Patient education (therapist-facilitated) (n = 120) inferior to exercise therapy for pain in the short term (SMD 0.77, 95% CI 0.07 to 1.47, five trials,43,46,53,63,64 Figure 3f) but produces similarComparator results for pain in the medium term (SMD 0.12, 95% CI 20.11 to 0.36,Age (y) = 72 (6) four trials,42,43,53,58 Figure 3g) and long term (SMD 0.18, 95% CI 20.11M/F (%) = NR to 0.46, three trials,43,52,58 Figure 3h). Very-low certainty evidenceBMI = 24.6 (2.5) indicated that patient education produces similar outcomes forAge (y) = 65 (3) function in the short term (SMD 0.33, 95% CI 20.02 to 0.69, threeM/F (%) = 0/100 trials,43,46,63 Figure 3i) and medium term (SMD 0.23, 95% CI 20.08 toBMI = 25.2 (3.5) 0.54, two trials42,43 Figure 3j).Age (y) = 67 (6) M/F (%) = 26/74 Patient education in combination with other interventionsBMI = 30.4 (0.9) Age (y) = 65 (11) Very-low certainty evidence indicated that patient educationM/F (%) = 32/68 combined with exercise therapy produces similar outcomesBMI = NR compared with exercise therapy alone for pain in the short term (SMD 0.61, 95% CI 20.40 to 1.62, three trials,43,53,66 Figure 4a) andEducationAge (y) = 71 (5)Age (y) = 65 (3)Age (y) = 69 (7)Age (y) = 62 (11) Age and BMI data are mean (SD) or median (IQR), except where noted. medium term (SMD 20.10, 95% CI 20.30 to 0.50, two trials,43,53M/F (%) = NRM/F (%) = 0/100M/F (%) = 38/62M/F (%) = 25/75 BMI = body mass index, F = female, M = male, NR = not reported. Figure 4b), and for function in the short term (SMD 1.32, 95%ParticipantsBMI = 25.7 (3.8)BMI = 25.0 (3.4)BMI = 29.2 (0.8)BMI = NR CI 20.57 to 3.20, two trials,43,66 Figure 4c). a Trial included at least one other intervention group that was not used in analysis. Eligible population General community Recruited from community centres Recruited from primary healthcare facility Patients referred to rheumatology clinics b Trial included at least one other intervention group that was used for sensitivity analysis only. Very-low certainty evidence indicated that patient education c Standard error. Table 2 (Continued) combined with exercise therapy is superior to patient educationStudyOh et al (2020)Qingguang et al (2017)Taglietti et al (2018)Victor et al (2005) alone for pain in the short term (SMD 0.44, 95% CI 0.19 to 0.69, five trials,43,45,51,53,62 Figure 4d). Low certainty evidence indicated that patient education combined with exercise therapy produces similar outcomes to patient education alone for pain in the medium term (SMD 0.14, 95% CI 20.04 to 0.32, four trials,43,53,56,58 Figure 4e). Low certainty evidence indicated that patient education combined with exercise therapy produces similar outcomes to patient education alone for pain in the long term (SMD 0.17, 95% CI 20.13 to 0.33, two trials,43,56 Figure 4f). Low certainty evidence indicated that patient education combined with exercise therapy produces superior out- comes compared with patient education alone for function in the short term (0.81, 95% CI 0.54 to 1.08, three trials,43,51,62 Figure 4g) and medium term (SMD 0.39, 95% CI 0.15 to 0.62, two trials,43,56 Figure 4h). Very-low certainty evidence indicated that patient edu- cation combined with exercise therapy produces similar outcomes compared with patient education alone for function in the long term (SMD 0.24, 95% CI 20.06 to 0.54, two trials,43,56 Figure 4i). Secondary outcomes Full details of all comparisons can be found in Table 4 and Figure 5 (see eAddenda for Figure 5). Very-low certainty evidence indicated that patient education combined with exercise therapy is superior to patient education alone in the short term for self-efficacy (SMD 0.46, 95% CI 0.02 to 0.89, two trials,43,57 Figure 5g on the eAddenda). Very- low certainty evidence indicated that patient education is superior for pain coping compared with usual care (SMD 20.71, 95% CI 21.32 to 20.01) and exercise therapy (SMD 20.96, 95% CI 21.42 to 20.49, two trials,43,57 Figure 5f on the eAddenda) in the short term. Sensitivity analyses Multiple sensitivity analyses were performed during data analysis. There were five instances where performing a sensitivity analysis for alternate options would have changed outcome or size of effect. Details can be found in Appendix 3 on the eAddenda. The selection of comparisons used in this review was always based upon similarity of outcome, time points and interventions across pooled trials. Discussion This review provides a comprehensive synthesis of evidence related to patient education for knee OA, which can inform guide- lines, clinical practice and future research. It is important to note that recommendations are primarily informed by very-low certainty
Research 183 = Low risk of bias = Unclear risk of bias = High risk of bias A trial was classified as low risk of bias if ≥ 4/7 items were reported as low risk of bias Figure 2. Risk of bias for included trials.
184 Goff et al: Patient education for knee osteoarthritis Outcome, timepoint Education versus usual care Outcome, timepoint Education versus exercise therapy Study SMD (95% CI) Study SMD (95% CI) Random Random a Pain, short term f Pain, short term O’Moore 2018 Cheung 2017 Ganji 2018 Bennell 2016 Coleman 2012 Farr 2010 Helminen 2015 Qingguang 2017 O’Brien 2018 Taglietti 2018 Allen 2019 Pooled Pooled g Pain, medium term b Pain, medium term Bennell 2016 Coleman 2012 Messier 2004 Allen 2019 Farr 2010 Allen 2016 Baker 2001 O’Brien 2018 Pooled Pooled h Pain, long term c Pain, long term Messier 2004 Allen 2010 Bennell 2016 Allen 2018 Ettinger 1997 Pooled Pooled d Function, short term i Function, short term Coleman 2012 Cheung 2017 O’Moore 2018 Bennell 2016 Murphy 2018 Qingguang 2017 Allen 2019 Pooled O’Brien 2018 Helminen 2015 j Function, medium term Pooled Coleman 2012 Allen 2019 e Function, medium term Pooled Coleman 2012 Allen 2019 –2 –1 0 1 2 3 Allen 2016 O’Brien 2018 Favours education Favours exercise Pooled –2 –1 0 1 2 Favours education Favours usual care Figure 3. Effect of patient education: relative to usual care on pain in the (a) short, (b) medium and (c) long term and on function in the (d) short and (e) medium term; and relative to exercise on pain in the (f) short, (g) medium and (h) long term and on function in the (i) short and (j) medium term. evidence. Pooling of 19 comparisons was possible; however, only six Very-low certainty evidence indicated that patient education (32%) produced statistically significant findings. Of these six com- combined with exercise therapy produces moderate, statistically parisons, two were clinically important for pain (ie, . 1.5 points)36 or significant short-term improvements in pain, and very large, statis- function (ie, . 6 points)37 using the respective subsections of the tically and clinically important short-term improvements in function WOMAC. Exercise therapy produced statistically superior and clini- compared with patient education alone. Low-certainty evidence cally important improvements in pain compared with patient edu- indicated that functional improvements are maintained in the me- cation alone, and combining patient education with exercise therapy dium term; however, differences are not clinically important. resulted in statistically superior and clinically important short-term Although not statistically significant, very-low certainty evidence improvements in function compared with education alone. It is indicated that patient education combined with exercise therapy therefore recommended that patient education should be provided in produces clinically important improvements in short-term pain and combination with exercise therapy whenever possible. function compared with exercise therapy alone. Further research is warranted to explore the benefits of combining patient education Very-low certainty evidence indicated that patient education with exercise therapy. A possible explanation for enhanced outcomes produces a moderate beneficial effect at reducing pain and improving may be that providing patient education alongside exercise therapy function in the short term compared with usual care. However, im- improves exercise therapy adherence in people with knee OA.11,67 provements may not be clinically important for either pain or func- Additionally, research across a number of musculoskeletal condi- tion using the WOMAC, questioning its value in isolation. It is worth tions suggests that patient education may improve illness percep- noting that as patient education is a guideline-recommended first- tions,68 self-efficacy69 and fear-avoidance behaviours;70 and create line intervention for people with knee OA,4–7 education provided as positive attitudes towards,71 and a safer environment in which to part of usual care is unknown. Further research is warranted to attempt, exercise therapy. One way to potentially sustain improve- investigate the effectiveness of patient education versus a wait-and- ments in the medium to long term could be providing additional face- see approach.
Research 185 Outcome, timepoint Exercise versus education + exercise Outcome, timepoint Education versus education + exercise Study SMD (95% CI) Study SMD (95% CI) Random Random a Pain, short term d Pain, short term Farr 2010 Farr 2010 Bennell 2016 Chen 2019 Ay 2013 Bennell 2016 Pooled Dias 2017 Oh 2020 b Pain, medium term Pooled Farr 2010 Bennell 2016 e Pain, medium term Pooled Messier 2004 Farr 2010 c Function, short term Hinman 2020 Bennell 2016 Bennell 2018 Ay 2013 Pooled Pooled f Pain, long term –2 –1 0 1 2 3 Brosseau 2016 Hinman 2020 Bennell 2016 Pooled Favours exercise Favours education + exercise g Function, short term Dias 2017 Bennell 2016 Oh 2020 Pooled h Function, medium term Hinman 2020 Bennell 2016 Pooled i Function, long term Brosseau 2016 Hinman 2020 Bennell 2016 –2 –1 0 1 2 Favours education Favours education + exercise Figure 4. Effect of patient education in combination with exercise therapy: relative to exercise therapy alone on pain in the (a) short and (b) medium term and on function in the (c) short term; and relative to education alone on pain in the (d) short, (e) medium and (f) long term and on function in the (g) short, (h) medium and (i) long term. to-face education sessions following the immediate treatment period, Very-low certainty evidence indicated that patient education has a which unlike exercise therapy does not appear to have been explored. moderate effect at improving short-term pain coping compared with Booster exercise therapy sessions have been found to improve adher- usual care, and with a very large effect compared with exercise ence to treatment in people with OA and low back pain.72 therapy. This is likely due to the highly targeted nature of education interventions to specifically develop pain coping skills in the trials Very-low certainty evidence indicated that patient education is associated with this comparson.39,43,57 Combining patient education inferior to exercise therapy for pain outcomes, with a large and with exercise therapy did not appear to improve pain coping clinically important difference in the short-term, but these are not compared with patient education alone in the short term. However, sustained in the medium term or long term. Very-low certainty evi- very-low certainty evidence indicated that a combination of patient dence indicated that patient education produces similar outcomes education and exercise therapy is superior to patient education alone, compared with exercise therapy for function in the short term and with a large effect for self-efficacy in the short term. These findings medium term. These findings are in contrast to a recent review, are in contrast to a previous review,19 which reported that patient which identified that patient education produces similar pain and education programs combined with exercise therapy produced function outcomes compared with exercise therapy in younger peo- similar outcomes to patient education programs alone for self- ple with knee pain.27 The lack of improvements in pain in younger efficacy outcomes in people with knee OA. These contrasting find- people with knee pain may reflect a reduced need for exercise ther- ings may be explained by the inclusion of different trials in the apy in a less chronic condition. Additionally, it may also highlight the analysis of total arthritis self-efficacy scale27 score in this review, greater role of exercise therapy for people with knee OA due to rather than the pain, function and other subsections used in Brand associated high rates of comorbidities73 and systemic inflammation.74
186 Goff et al: Patient education for knee osteoarthritis Table 4 Summary of secondary outcomes (SMD) for all pooled data. Comparison Outcome Time n SMD (95% CI) Certainty Figure in point eAddenda very low Patient education versus usual care Self-efficacy short 2 20.41 (20.82 to 0.01) very low 5a Pain catastrophising short 3 20.02 (20.45 to 0.42) very low 5b very low 5c Pain coping short 2 20.71 (21.32 to 20.01) very low 5d very low 5e Patient education versus exercise therapy Self-efficacy short 2 0.09 (20.82 to 0.65) very low 5f Pain catastrophising short 2 20.16 (20.62 to 0.30) very low 5g very low 5h Pain coping short 2 20.96 (21.42 to 20.49) very low 5i very low 5j Patient education versus patient education 1 exercise therapy Self-efficacy short 2 0.46 (0.02 to 0.89) very low 5k Pain catastrophising short 2 0.15 (20.15 to 0.46) very low 5l 5m Pain coping short 2 0.04 (20.34 to 0.26) Therapist-facilitated education versus self-directed education Pain short 3 0.03 (20.29 to 0.23) long 3 20.04 (20.48 to 0.39) Function short 2 0.09 (20.21 to 0.40) long 2 20.05 (20.53 to 0.62) n = number of trials, SMD = standardised mean difference, 95% CI = 95% confidence interval. et al’s review.19 The current findings combined with Brand et al’s19 conditions,79,80 as well as in tertiary medical81 and healthcare82 ed- findings suggest that patient education with or without exercise ucation for people with knee OA. Identifying optimal mode and therapy has the potential to impact different domains of self-efficacy content of patient education will influence how healthcare providers for people with knee OA; however, the relationship is not well un- develop educational interventions and prioritise resources for people with knee OA. derstood and requires further investigation. Enhanced pain coping It is important to consider that the results in this review and self-efficacy is desirable due to known association with im- were informed by all patient education interventions. Education provements in pain, function and physical activity,69,75,76 which has interventions were included regardless of intervention devel- opment process (co-design, based on learning theory, etc) or the potential to enhance quality of life and reduce healthcare uti- whether they were used as a control. The decision to include all patient education interventions was chosen to reduce se- lisation for people with knee OA. The psychological benefits of patient lection bias and appropriately assess all patient education in- education identified in this review highlight the importance of terventions in published literature for people with knee OA. considering outcomes beyond pain and function when guiding Further analysis of interventions based on development process or whether the intervention was designed as a control or not treatment recommendations based on available evidence. may impact these results. Certainty of the findings is limited due to the low and very-low categorisation of evidence using Very-low certainty evidence indicated that therapist-facilitated GRADE, and findings may change in the future with updated reviews on this topic. The variation and nature of patient ed- education produces similar short-term and long-term pain and ucation interventions, combined with the self-reported outcome measures assessed in this review significantly impacted risk of function outcomes compared with self-directed education. An bias assessment, and the indirectness and imprecision measures of GRADE. The large heterogeneity between the included trials important consideration in interpreting these findings is the ap- also impacted the imprecision measure of GRADE. Assessment proaches used in therapist-facilitated education and the content of of publication bias29 was not possible for any comparison due each education intervention. Participants in De Rezendes’s49,50 self- to the small number of trials included within each comparison. directed education intervention received DVD recordings of the Caution should be taken when applying the recommendations to younger people with knee OA (eg, post-traumatic knee OA), face-to-face therapist-facilitated education lectures and workshops. as the typical mean age of participants in this review was in the 60s. Lastly, clinical interpretation of results was not Therefore, the delivery method was different between groups; how- possible for all comparisons, and significant variation exists for MCID values for WOMAC pain and function subsections in knee ever, the content was the same. In comparison, both Ackerman OA.83 Clinical interpretation of results may change depending et al’s38 and Victor et al’s65 delivery method and content were upon the comparative MCID chosen and the inclusion of future different between therapist-facilitated education and self-directed research. education interventions. The varied interventions used in these tri- Although patient education produced statistically superior short-term pain and function outcomes compared with usual als makes it challenging to draw any clear conclusions related to how care, the differences were small and may not be clinically important. Patient education should not be provided as a to provide patient education for people with knee OA. standalone treatment and should be combined with exercise therapy to provide statistically superior and clinically important Delivery of patient education interventions in this review varied short-term improvements in function compared with education from singular lectures66 and intensive group-based sessions over a alone. number of sessions38,43 to provision of self-directed education ma- terials with or without follow-up telephone calls39–41,45–47 or home visits.38,42 The content of patient education interventions was equally varied, ranging from interventions targeting basic knowledge acqui- sition45,48,51,66 to more complex psychologically informed self- management skill development.41,43,57 Combined, these variations reflect the lack of recommendations for delivery method and content in clinical practice guidelines4–7 and the lack of studies identifying how people with knee OA learn best. With the growing emphasis placed around patient-centred care for people with knee OA4–7,77 and other musculoskeletal conditions,78 further research evaluating ways to match delivery methods and/or content to the individual needs or preferences of people with knee OA is warranted. This could include consideration of blended learning approaches used for other chronic
Research 187 Table 7 Mean difference (95% CI) for all possible comparisons. Comparison n Outcome Time MD (95% CI) MD better Figure in point than MCID eAddenda Pain 4 WOMAC pain medium 20.40 (20.94 to 0.14) N 6a Patient education versus usual care 5 WOMAC pain short 1.56 (0.14 to 2.98) Y 6d Patient education versus exercise therapy 4 WOMAC pain 0.42 (20.39 to 1.23) N 6e Patient education versus exercise therapy 3 WOMAC pain medium 2.01 (21.16 to 5.18) Y 6h Exercise therapy versus patient education 1 exercise therapy 2 WOMAC pain short 0.31 (20.97 to 1.59) N 6i Exercise therapy versus patient education 1 exercise therapy 5 WOMAC pain 1.48 (0.48 to 2.49) N 6k Patient education versus patient education 1 exercise therapy 4 WOMAC pain medium 0.49 (20.07 to 1.06) N 6l Patient education versus patient education 1 exercise therapy 3 WOMAC pain short 0.45 (20.22 to 1.13) N 6m Patient education versus patient education 1 exercise therapy 3 WOMAC pain N 6q Therapist-facilitated education versus self-directed education 3 WOMAC pain medium 20.15 (20.99 to 0.68) N 6r Therapist-facilitated education versus self-directed education long 20.13 (22.00 to 1.74) short long Function Patient education versus usual care 6 WOMAC function short 22.43 (24.71 to 20.16) N 6b N 6c Patient education versus usual care 4 WOMAC function medium 22.12 (24.55 to 0.31) N 6f N 6g Patient education versus exercise therapy 3 WOMAC function short 3.36 (21.00 to 7.71) Y 6j Y 6n Patient education versus exercise therapy 2 WOMAC function medium 2.78 (21.11 to 6.67) N 6o N 6p Exercise therapy versus patient education 1 exercise therapy 2 WOMAC function short 13.08 (25.44 to 31.60) N 6s N 6t Patient education versus patient education 1 exercise therapy 3 WOMAC function short 8.94 (6.05 to 11.82) Patient education versus patient education 1 exercise therapy 2 WOMAC function medium 4.61 (1.88 to 7.33) Patient education versus patient education 1 exercise therapy 3 WOMAC function long 2.69 (20.26 to 5.65) Therapist-facilitated education versus self-directed educationa 2 WOMAC function short 1.02 (22.91 to 4.94) Therapist-facilitated education versus self-directed education 2 WOMAC function long 0.71 (27.98 to 9.40) Pain coping 2 CSQ short 219 (231 to 27) N/A 6u Patient education versus usual care 2 CSQ Patient education versus exercise therapy 2 CSQ short 225 (233 to 21) N/A 6v Patient education versus patient education 1 exercise therapy short 21 (29 to 7) N/A 6w CSQ = Coping Strategies Questionnaire, MD = mean difference, MCID = minimal clinically important difference, N = no, n = number of trials, N/A = not assessed, 95% CI = 95% confidence intervals, WOMAC = Western Ontario and McMaster Universities Osteoarthritis Index, Y = yes. MCID values are 1.5 for pain and 6 for function. Negative MD = supports first listed intervention in comparison. Positive MD = supports second listed intervention in comparison. a Assumed a typo in manuscript by Victor et al 2005: mean of 21 (SD 7) for control and 21 (SD 10) for experimental for the WOMAC function section at 1 month, rather than the reported 2.1 (SD 7) and 2.1 (SD 10), respectively. What was already known on this topic: Patient education, and Sport, La Trobe University, Melbourne, Australia. Email: exercise therapy and weight management are recommended by [email protected] all major guidelines as a first-line intervention for people with knee osteoarthritis. Evidence supporting the effectiveness and References cost-effectiveness for exercise therapy and weight management has been synthesised in recent systematic reviews. There is a 1. Cross M, Smith E, Hoy D, Nolte S, Ackerman I, Fransen M, et al. The global burden of lack of an up-to-date evidence synthesis for patient education to hip and knee osteoarthritis: estimates from the global burden of disease 2010 inform guidelines and practice. study. Ann Rheum Dis. 2014;73:1323–1330. What this study adds: The review findings indicate that pa- tient education may reduce pain and improve function compared 2. Litwic A, Edwards MH, Dennison EM, Cooper C. Epidemiology and burden of with usual care, although differences may not be clinically osteoarthritis. Br Med Bull. 2013;105:185–199. important, questioning its value in isolation. Combining patient education with exercise therapy should be encouraged consid- 3. Ackerman IN, Bohensky MA, Zomer E, Tacey M, Gorelik A, Brand CA, et al. The ering statistically superior and clinically important improvements projected burden of primary total knee and hip replacement for osteoarthritis in in function compared with patient education alone. Australia to the year 2030. BMC Musculoskelet Disord. 2019;20:90. Footnotes: a EndNote X8, Thomson Reuters, Carlsbad, USA. 4. Fernandes L, Hagen KB, Bijlsma JW, Andreassen O, Christensen P, Conaghan PG, b RevMan 5.3, The Nordic Cochrane Centre, Copenhagen, Denmark. et al. EULAR recommendations for the non-pharmacological core management of c GRADEpro GDT, McMaster University, Ontario, Canada. hip and knee osteoarthritis. Ann Rheum Dis. 2013;72:1125–1135. eAddenda: Figures 5 to 8, Tables 1, 3, 5 and 6, and Appendices 1 to 5. Conaghan PG, Dickson J, Grant RL. Care and management of osteoarthritis in 3 can be found online at https://doi.org/10.1016/j.jphys.2021.06.011 adults: summary of NICE guidance. BMJ. 2008;336:502–503. Ethics approval: Not applicable. 6. McAlindon TE, Bannuru RR, Sullivan M, Arden NK, Berenbaum F, Bierma- Competing interests: Nil. Zienstra ZM, et al. OARSI guidelines for the non-surgical management of knee Source(s) of support: Anthony Goff holds a sponsorship from osteoarthritis. Osteoarthritis Cartilage. 2014;22:363–388. Singapore Institute of Technology for the completion of his PhD at La Trobe. Christian J Barton was supported by an MRFF TRIP 7. Kolasinski SL, Neogi T, Hochberg MC, Oatis C, Guyatt G, Block J, et al. 2019 American Fellowship (APP1150439). The financial sponsors played no role in College of Rheumatology/Arthritis Foundation guideline for the management of the design, execution, analysis and interpretation of data, or writing osteoarthritis of the hand, hip, and knee. Arthritis Rheumatol. 2020;72:220–233. of the study. Acknowledgements: Nil. 8. Ackerman IN, Skou ST, Roos EM, Barton CJ, Kemp J, Crossley KM, et al. Imple- Provenance: Not invited. Peer reviewed. menting a national first-line management program for moderate-severe knee Correspondence: Dr Christian Barton, La Trobe Sport and Exercise osteoarthritis in Australia: a budget impact analysis focusing on knee replacement Medicine Research Centre, School of Allied Health, Human Services avoidance. Osteoarthr Cartil Open. 2020:100070. 9. Verhagen A, Ferreira M, Reijneveld-van de Vendel E, Teirlinck CH, Runhaar J, van Middlekoop M, et al. Do we need another trial on exercise in patients with knee osteoarthritis?: no new trials on exercise in knee OA. Osteoarthritis Cartilage. 2019;27:1266–1269. 10. Fransen M, McConnell S, Harmer AR, Van der Esch M, Simic M, Bennell KL. Exercise for osteoarthritis of the knee: a Cochrane systematic review. Br J Sports Med. 2015;49:1554. 11. Gay C, Chabaud A, Guilley E, Coudeyre E. Educating patients about the benefits of physical activity and exercise for their hip and knee osteoarthritis. Systematic literature review. Ann Phys Rehabil Med. 2016;59:174–183. 12. Hall M, Castelein B, Wittoek R, Calders P, Van Ginckel A. Diet-induced weight loss alone or combined with exercise in overweight or obese people with knee oste- oarthritis: a systematic review and meta-analysis. Semin Arthritis Rheum. 2019;48:765–777.
188 Goff et al: Patient education for knee osteoarthritis 13. Losina E, Smith KC, Paltiel AD, Collins JE, Suter LG, Hunter DJ, et al. Cost-effec- 42. Baker KR, Nelson ME, Felson DT, Layne JE, Sarno R, Roubenoff R. The efficacy of tiveness of diet and exercise for overweight and obese patients with knee osteo- home based progressive strength training in older adults with knee osteoarthritis: arthritis. Arthritis Care Res. 2019;71:855–864. a randomized controlled trial. J Rheumatol. 2001;28:1655–1665. 14. Hansson EE, Jönsson-Lundgren M, Ronnheden A-M, Sörensson E, Bjärnung Å, 43. Bennell KL, Ahamed Y, Jull G, Bryant C, Hunt MA, Forbes AB, et al. Physical Dahlberg LE. Effect of an education programme for patients with osteoarthritis in therapist-delivered pain coping skills training and exercise for knee osteoarthritis: primary care-a randomized controlled trial. BMC Musculoskelet Disord. 2010;11:244. randomized controlled trial. Arthritis Care Res. 2016;68:590–602. 15. Buszewicz M, Rait G, Griffin M, Nazareth I, Patel A, Atkinson A, et al. Self man- 44. Brosseau L, Wells GA, Kenny GP, Reid R, Maetzel A, Tugwell P, et al. The imple- agement of arthritis in primary care: randomised controlled trial. BMJ. mentation of a community-based aerobic walking program for mild to moderate 2006;333:879. knee osteoarthritis (OA): a knowledge translation (KT) randomized controlled trial (RCT): Part I: the uptake of the Ottawa Panel clinical practice guidelines (CPGs). 16. Broderick JE, Keefe FJ, Bruckenthal P, Junghaenel DU, Schneider S, Schwartz JE, et al. BMC Public Health. 2012;12:871. Nurse practitioners can effectively deliver pain coping skills training to osteoar- thritis patients with chronic pain: a randomized, controlled trial. Pain. 45. Chen H, Zheng X, Huang H, Liu C, Wan Q, Shang S. The effects of a home-based 2014;155:1743–1754. exercise intervention on elderly patients with knee osteoarthritis: a quasi- experimental study. BMC Musculoskelet Disord. 2019;20:160. 17. Du S, Yuan C, Xiao X, Chu J, Qiu Y, Qian H. Self-management programs for chronic musculoskeletal pain conditions: a systematic review and meta-analysis. Patient 46. Cheung C, Wyman JF, Bronas U, McCarthy T, Rudser K, Mathiason MA. Managing Educ Couns. 2011;85:e299–e310. knee osteoarthritis with yoga or aerobic/strengthening exercise programs in older adults: a pilot randomized controlled trial. Rheumatol Int. 2017;37:389– 18. Kroon FP, van der Burg LR, Buchbinder R, Osborne RH, Johnston RV, Pitt V. Self- 398. management education programmes for osteoarthritis. Cochrane Database Syst Rev. 2014:CD008963. 47. Cheung DST, Yeung W-F, Suen LK-P, Chung Chong T, Ho Y-S, Yu BY-M, et al. Self- administered acupressure for knee osteoarthritis in middle-aged and older adults: 19. Brand E, Nyland J, Henzman C, Mcginnis M. Arthritis self-efficacy scale scores in a pilot randomized controlled trial. Acupunct Med. 2020;38:75–85. knee osteoarthritis: a systematic review and meta-analysis comparing arthritis self-management education with or without exercise. J Orthop Sports Phys Ther. 48. Coleman S, Briffa NK, Carroll G, Inderjeeth C, Cook N, McQuade J. A randomised 2013;43:895–910. controlled trial of a self-management education program for osteoarthritis of the knee delivered by health care professionals. Arthritis Res Ther. 2012;14:R21. 20. Briani RV, Ferreira AS, Pazzinatto MF, Pappas E, Silva DDO, de Azevedo FM. What interventions can improve quality of life or psychosocial factors of in- 49. de Rezende MU, Hissadomi MI, de Campos GC, Frucchi R, Felicio Pailo A, dividuals with knee osteoarthritis? A systematic review with meta-analysis of Pasqualin T, et al. One-year results of an educational program on osteoarthritis: a primary outcomes from randomised controlled trials. Br J Sports Med. prospective randomized controlled trial in Brazil. Geriatr Orthop Surg Rehabil. 2018;52:1031–1038. 2016;7:86–94. 21. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche P, Ioannidis JPA, et al. The 50. de Rezende MU, Frucchi R, Pailo AF, Campos GCD, Pasqualin T, Hissadomi MI. PRISMA statement for reporting systematic reviews and meta-analyses of studies PARQVE: Project Arthritis recovering quality of life through education: two-year that evaluate health care interventions: explanation and elaboration. J Clin Epi- results. Acta Ortop Bras. 2017;25:18–24. demiol. 2009;62:e1–e34. 51. Dias JM, Cisneros L, Dias R, Fritsch C, Gomes W, Pereira L, et al. Hydrotherapy 22. Higgins JP, Thomas J, Chandler J, Cumpston M, Li T, Page M, et al. eds. Cochrane improves pain and function in older women with knee osteoarthritis: a random- handbook for systematic reviews of interventions. Chichester, UK: John Wiley & Sons; ized controlled trial. Braz J Phys Ther. 2017;21:449–456. 2019. 52. Ettinger WH, Burns R, Messier SP, Applegate W, Rejeski WJ, Morgan T, et al. 23. de Oliveira Silva D, Pazzinatto MF, Rathleff MS, Holden S, Bell E, Azevedo S, et al. A randomized trial comparing aerobic exercise and resistance exercise with a Patient education for patellofemoral pain: a systematic review. J Orthop Sports Phys health education program in older adults with knee osteoarthritis: the Fitness Ther. 2020;50:388–396. Arthritis and Seniors Trial (FAST). JAMA. 1997;277:25–31. 24. Kohn MD, Sassoon AA, Fernando ND. Classifications in brief: Kellgren-Lawrence 53. Farr JN, Going SB, McKnight PE, Kasle S, Cussler EC, Cornett M. Progressive resis- classification of osteoarthritis. Clin Orthop Relat Res. 2016;474:1886–1893. tance training improves overall physical activity levels in patients with early osteoarthritis of the knee: a randomized controlled trial. Phys Ther. 2010;90:356– 25. Higgins JP, Altman DG, Gøtzsche PC, Juni P, Moher D, Oxman A, et al. The Cochrane 366. Collaboration’s tool for assessing risk of bias in randomised trials. BMJ. 2011;343:d5928. 54. Ganji R, Pakniat A, Armat MR, Tabatabaeichehr M, Mortazavi H. The effect of self- management educational program on pain intensity in elderly patients with knee 26. Bellamy N, Buchanan W. A preliminary evaluation of the dimensionality and osteoarthritis: a randomized clinical trial. Open Access Maced J Med Sci. clinical importance of pain and disability in osteoarthritis of the hip and knee. Clin 2018;6:1062–1066. Rheumatol. 1986;5:231–241. 55. Helminen E-E, Sinikallio SH, Valjakka AL, Väisänen-Rouvali RH, Arokoski JP. 27. Lorig K, Chastain RL, Ung E, Shoor S, Holman HR. Development and evaluation of a Effectiveness of a cognitive-behavioural group intervention for knee osteoarthritis scale to measure perceived self-efficacy in people with arthritis. Arthritis Rheum. pain: a randomized controlled trial. Clin Rehabil. 2015;29:868–881. 1989;32:37–44. 56. Hinman RS, Campbell PK, Lawford BJ, Briggs AM, Gale J, Bills C, et al. Does 28. Sullivan MJ, Bishop SR, Pivik J. The pain catastrophizing scale: development and telephone-delivered exercise advice and support by physiotherapists improve pain validation. Psychol Assess. 1995;7:524. and/or function in people with knee osteoarthritis? Telecare randomised controlled trial. Br J Sports Med. 2020;54:790–797. 29. Monticone M, Ferrante S, Giorgi I, Galandra C, Rocca B, Foti C. The 27-item coping strategies questionnaire—revised: confirmatory factor analysis, reliability and 57. Keefe FJ, Blumenthal J, Baucom D, Affleck G, Waugh R, Caldwell DS, et al. Effects of validity in italian-speaking subjects with chronic pain. Pain Res Manag. 2014;19. spouse-assisted coping skills training and exercise training in patients with oste- oarthritic knee pain: a randomized controlled study. Pain. 2004;110:539–549. 30. Hozo SP, Djulbegovic B, Hozo I. Estimating the mean and variance from the median, range, and the size of a sample. BMC Med Res Methodol. 2005;5:13. 58. Messier SP, Loeser RF, Miller GD, Morgan TM, Rejeski WJ, Sevick MA, et al. Exercise and dietary weight loss in overweight and obese older adults with knee osteoar- 31. Cohen J. A power primer. Psychol Bull. 1992;112:155. thritis: the Arthritis, Diet, and Activity Promotion Trial. Arthritis Rheum. 32. Schünemann H, Broz_ ek J, Guyatt G, Oxman A. Handbook for grading the quality of 2004;50:1501–1510. evidence and the strength of recommendations using the GRADE approach (updated 59. Murphy SL, Janevic MR, Lee P, Williams DA. Occupational therapist-delivered October 2013). GRADE Working Group; 2013. cognitive-behavioral therapy for knee osteoarthritis: a randomized pilot study. 33. Guyatt GH, Oxman AD, Kunz R, Vist GE, Falck-Ytter Y, Schünemann HJ. GRADE: Am J Occup Ther. 2018;72. what is “quality of evidence” and why is it important to clinicians. BMJ. 2008;3:995–998. 60. O’Brien K, Wiggers J, Williams A, Campbell E, Hodder RK, Wolfenden L, et al. 34. Van Tulder M, Furlan A, Bombardier C, Bouter L, Editorial Board of the Cochrane Telephone-based weight loss support for patients with knee osteoarthritis: a Collaboration Back Review Group. Updated method guidelines for systematic re- pragmatic randomised controlled trial. Osteoarthritis Cartilage. 2018;26:485–494. views in the cochrane collaboration back review group. Spine. 2003;28:1290–1299. 35. Angst F, Aeschlimann A, Stucki G. Smallest detectable and minimal clinically 61. O’moore KA, Newby JM, Andrews G, Hunter DJ, Bennell K, Smith J, et al. Internet important differences of rehabilitation intervention with their implications for cognitive-behavioral therapy for depression in older adults with knee osteoar- required sample sizes using WOMAC and SF-36 quality of life measurement in- thritis: a randomized controlled trial. Arthritis Care Res. 2018;70:61–70. struments in patients with osteoarthritis of the lower extremities. Arthritis Care Res. 2001;45:384–391. 62. Oh S-L, Kim D-Y, Bae J-H, Lim J-Y. Effects of rural community-based integrated 36. Tubach F, Ravaud P, Baron G, Falissard B, Logeart I, Bellamy N, et al. Evaluation of exercise and health education programs on the mobility function of older adults clinically relevant changes in patient reported outcomes in knee and hip osteo- with knee osteoarthritis. Aging Clin Exp Res. 2020:1–10. arthritis: the minimal clinically important improvement. Ann Rheum Dis. 2005;64:29–33. 63. Qingguang Z, Lingyan H, Jingxian L, Zhang Y, Min F, Li J, et al. Effect of 37. Hurley M, Dickson K, Hallett R, Grant R, Hauari H, Walsh N, et al. Exercise Taijiquan practice versus wellness education on knee proprioception in interventions and patient beliefs for people with hip, knee or hip and knee patients with knee osteoarthritis: a randomized controlled trial. J Tradit osteoarthritis: a mixed methods review. Cochrane Database Syst Rev. Chin Med. 2017;37:774–781. 2018;4:CD010842. 38. Ackerman IN, Buchbinder R, Osborne RH. Challenges in evaluating an Arthritis Self- 64. Taglietti M, Facci LM, Trelha CS, de Melo FC, da Silva DW, Sawczuk G, et al. Effec- Management Program for people with hip and knee osteoarthritis in real-world tiveness of aquatic exercises compared to patient-education on health status in clinical settings. J Rheumatol. 2012;39:1047–1055. individuals with knee osteoarthritis: a randomized controlled trial. Clin Rehabil. 39. Allen KD, Oddone EZ, Coffman CJ, Grant R, Hauari H, Walsh N, et al. Telephone- 2018;32:766–776. based self-management of osteoarthritis: a randomized trial. Ann Intern Med. 2010;153:570–579. 65. Victor CR, Triggs E, Ross F, Lord J, Axford JS. Lack of benefit of a primary care-based 40. Allen KD, Yancy Jr WS, Bosworth HB, Coffman CJ, Jeffreys AS, Datta SK, et al. nurse-led education programme for people with osteoarthritis of the knee. Clin A combined patient and provider intervention for management of osteoarthritis in Rheumatol. 2005;24:358–364. veterans: a randomized clinical trial. Ann Intern Med. 2016;164:73–83. 41. Allen KD, Somers TJ, Campbell LC, Arbeeva L, Coffman CJ, Cené CW, et al. Pain 66. Ay S, Koldas¸ Dogan S¸ , Evcik D. Is there an effective way to prescribe a home-based coping skills training for African Americans with osteoarthritis: results of a ran- exercise program in patients with knee osteoarthritis? A randomized controlled domized controlled trial. Pain. 2019;160:1297. study. Turk J Phys Med Rehabil. 2013;59. 67. Greaves CJ, Sheppard KE, Abraham C, Hardeman W, Roden M, Evans PH, et al. Systematic review of reviews of intervention components associated with increased effectiveness in dietary and physical activity interventions. BMC Public Health. 2011;11:1–12.
Research 189 68. Claassen AA, Schers HJ, Koëter S, van der Laan WH, Kremers-van de Hei KC, 76. Marks R. Self-efficacy and arthritis disability: an updated synthesis of the evidence Botman J, et al. Preliminary effects of a regional approached multidisciplinary base and its relevance to optimal patient care. Health Psychol Open. 2014;1: educational program on healthcare utilization in patients with hip or knee 2055102914564582. osteoarthritis: an observational study. BMC Fam Pract. 2018;19:82. 77. Caneiro J, O’Sullivan PB, Roos EM, Smith AJ, Choong P, Dowsey M, et al. Three 69. Zhaoyang R, Martire LM, Sliwinski MJ. Morning self-efficacy predicts physical steps to changing the narrative about knee osteoarthritis care: a call to action. Br J activity throughout the day in knee osteoarthritis. Health Psychol. 2017;36:568. Sports Med. 2020;54:256–258. 70. Vlaeyen JW, Linton SJ. Fear-avoidance and its consequences in chronic musculo- 78. Lin I, Wiles L, Waller R, Goucke R, Nagree Y, Gibberd M, et al. What does best skeletal pain: a state of the art. Pain. 2000;85:317–332. practice care for musculoskeletal pain look like? Eleven consistent recommenda- tions from high-quality clinical practice guidelines: systematic review. Br J Sports 71. Nijs J, Roussel N, Van Wilgen CP, Köke A, Smeets R. Thinking beyond muscles Med. 2020;54:79–86. and joints: therapists’ and patients’ attitudes and beliefs regarding chronic musculoskeletal pain are key to applying effective treatment. Man Ther. 79. Sangrar R, Docherty-Skippen SM, Beattie K. Blended face-to-face and 2013;18:96–102. online/computer-based education approaches in chronic disease self- management: a critical interpretive synthesis. Patient Educ Couns. 72. Nicolson PJ, Bennell KL, Dobson FL, Van Ginckel A, Holden MA, Hinman RS. In- 2019;102:1822–1832. terventions to increase adherence to therapeutic exercise in older adults with low back pain and/or hip/knee osteoarthritis: a systematic review and meta-analysis. 80. Talboom-Kamp EP, Verdijk NA, Kasteleyn MJ, Numans ME, Chavannes NH. From Br J Sports Med. 2017;51:791–799. chronic disease management to person-centered eHealth; a review on the necessity for blended care. Clinical eHealth. 2018;1:3–7. 73. Muckelt PE, Roos E, Stokes M, McDonough S, Gronne DT, Ewings S, et al. Comor- bidities and their link with individual health status: a cross-sectional analysis of 81. Vallée A, Blacher J, Cariou A, Sorbets E. Blended learning compared to traditional 23,892 people with knee and hip osteoarthritis from primary care. J Comorb. learning in medical education: systematic review and meta-analysis. J Med Internet 2020;10:2235042X20920456. Res. 2020;22:e16504. 74. Belluzzi E, El Hadi H, Granzotto M, Rossato M, Ramonda R, Macchi V, et al. 82. Ma˛ cznik AK, Ribeiro DC, Baxter GD. Online technology use in physiotherapy Systemic and local adipose tissue in knee osteoarthritis. J Cell Physiol. teaching and learning: a systematic review of effectiveness and users’ perceptions. 2017;232:1971–1978. BMC Med Educ. 2015;15:160. 75. Degerstedt Å, Alinaghizadeh H, Thorstensson CA, Olsson CB. High self-efficacy–a 83. Angst F, Benz T, Lehmann S, Aeschlimann A, Angst J. Multidimensional minimal predictor of reduced pain and higher levels of physical activity among patients clinically important differences in knee osteoarthritis after comprehensive reha- with osteoarthritis: an observational study. BMC Musculoskelet Disord. 2020;21: bilitation: a prospective evaluation from the Bad Zurzach Osteoarthritis Study. 1–13. RMD Open. 2018;4:e000685.
Journal of Physiotherapy 67 (2021) 201–209 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 Research Physiotherapists and patients report positive experiences overall with telehealth during the COVID-19 pandemic: a mixed-methods study Kim L Bennell a, Belinda J Lawford a, Ben Metcalf a, David Mackenzie a, Trevor Russell b, Maayken van den Berg c, Karen Finnin d, Shelley Crowther e, Jenny Aiken e, Jenine Fleming e,f, Rana S Hinman a a Centre for Health, Exercise and Sports Medicine, Department of Physiotherapy, University of Melbourne, Melbourne, Australia; b RECOVER Injury Research Centre, Faculty of Health and Behavioural Sciences, University of Queensland, Brisbane, Australia; c Clinical Rehabilitation, College of Nursing and Health Sciences, Flinders University, Adelaide, Australia; d Digital Practice; e Australian Physiotherapy Association, Melbourne, Australia; f Physiotherapy Research Foundation, Melbourne, Australia KEY WORDS ABSTRACT Telehealth Question: What were the experiences of physiotherapists and patients who consulted via videoconference COVID-19 during the COVID-19 pandemic and how was it implemented? Design: Mixed methods study with cross- Video sectional national online surveys and qualitative analysis of free-text responses. Participants: A total of Experiences 207 physiotherapists in private practice or community settings and 401 patients aged 18 years who Patient consulted (individual and/or group) via videoconference from April to November 2020. Methods: Separate Physical therapy customised online surveys were developed for physiotherapists and patients. Data were collected regarding the implementation of videoconferencing (cost, software used) and experience with videoconferencing (perceived effectiveness, safety, ease of use and comfort communicating, each scored on a 4-point ordinal scale). Qualitative content analysis was performed of physiotherapists’ free-text responses about perceived facilitators, barriers and safety issues. Results: Physiotherapists gave moderate-to-high ratings for the effectiveness of and their satisfaction with videoconferencing. Most intended to continue to offer individual consultations (81%) and group classes (60%) via videoconferencing beyond the pandemic. For individual consultations and group classes, respectively, most patients had moderately or extremely positive percep- tions about ease of technology use (94%, 91%), comfort communicating (96%, 86%), satisfaction with man- agement (92%, 93%), satisfaction with privacy/security (98%, 95%), safety (99% both) and effectiveness (83%, 89%). Compared with 68% for group classes, 47% of patients indicated they were moderately or extremely likely to choose videoconferencing for individual consultations in the future. Technology was predominant as both a facilitator and barrier. Falls risk was the main safety factor. Conclusion: Patients and physiotherapists had overall positive experiences using videoconferencing for individual consultations and group classes. The results suggest that videoconferencing is a viable option for the delivery of physiotherapy care in the future. [Bennell KL, Lawford BJ, Metcalf B, Mackenzie D, Russell T, van den Berg M, Finnin K, Crowther S, Aiken J, Fleming J, Hinman RS (2021) Physiotherapists and patients report positive experiences overall with telehealth during the COVID-19 pandemic: a mixed-methods study. Journal of Physiotherapy 67:201–209] © 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/). Introduction replacement,11 heart failure,12 and older patients with disability.13 However, research on the effectiveness and acceptability of tele- In the past decade, telehealth has emerged as a viable mode of health for other conditions and streams of physiotherapy is lacking. In service delivery that has the potential to increase healthcare acces- addition, outside of cardiac and pulmonary conditions,9,12,14,15 there is sibility. There is evidence that telehealth is an effective physiotherapy very limited evidence examining the efficacy and acceptability of service delivery mode for some conditions, with outcomes similar to, group physiotherapy classes via telehealth. With advances in tech- or even better than, those achieved with in-person care in muscu- nology and the availability of affordable videoconferencing software, loskeletal conditions,1–4 joint surgery,5 and cardiac6 and pulmonary7 telehealth has the potential to revolutionise the way in which rehabilitation. There is also some evidence that telehealth is healthcare is provided,16 and thus further research is warranted. perceived to be safe and effective by physiotherapists delivering the service8 and by patients with various conditions, including: osteoar- Although there is some evidence supporting telehealth’s effec- thritis,8 chronic obstructive pulmonary disease,9,10 following knee tiveness and acceptability for some conditions, uptake had previously been slow in Australia and around the world due to a range of factors, https://doi.org/10.1016/j.jphys.2021.06.009 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/).
202 Bennell et al: Physiotherapist and patient experiences with telehealth including: lack of reimbursement for services; inadequate physio- Telehealth Usability Questionnaire were also included.24 The ques- therapist knowledge, experience or confidence in telehealth; clinician tions mostly required check box answers, including use of 11-point resistance to changing clinical practice; and patient beliefs or pref- numeric rating scales from 0 to 10 for satisfaction and effectiveness erences for in-person care.3,17,18 In addition, most research has (ranging from 0 = not at all satisfied/effective to 10 = extremely examined physiotherapy via telehealth in the context of research satisfied/effective) and 4-point Likert scales for evaluating experi- settings, often as part of a clinical trial and often using sophisticated ences with the ease, safety and privacy of telehealth (rated as ‘not at and potentially inaccessible or expensive telehealth technologies. As all’, ‘somewhat’, ‘moderately’ or ‘extremely’). Free-text responses such, it is currently unclear whether the existing evidence reflects were sought for some questions from physiotherapists. Surveys were user experiences with telehealth in a ‘real-world’ setting. administered via a secure online platforma. In 2020, the COVID-19 pandemic had a dramatic impact on Patients and physiotherapists first completed online screening healthcare delivery worldwide,19,20 with many physiotherapy ser- and, if eligible, provided consent and proceeded to the corresponding vices rapidly transitioning to telehealth, often with limited prepara- survey. Both surveys ascertained respondent demographics and prior tion or staff training.21 This unprecedented uptake in telehealth experience with technology and telehealth, as well as experiences provided a unique opportunity to evaluate the perceived effective- with and perceptions about physiotherapy care via videoconference. ness, acceptability and implementation of such services in the com- In addition, the physiotherapist survey ascertained the respondents’ munity and across a wide range of users and patient populations.22,23 area of practice and the patient survey ascertained the respondents’ Such information will help identify factors that facilitate or impede clinical condition. The surveys contained separate questions for in- use of telehealth in ‘real-world’ settings, thus informing future dividual consultations and for group sessions. The physiotherapist development and implementation of such services, as well as phys- survey also contained free-text questions about facilitators and bar- iotherapy telehealth training programs and funding sources. riers to videoconferencing as well as any safety issues experienced. Physiotherapists and patients went into draws for $1,000 and $500 As such, this study aimed to investigate the implementation of prizes, respectively, if they completed the survey. and experiences with individual consultations and group classes delivered via videoconferencing during the COVID-19 pandemic, from Data analysis the perspective of patients who received and physiotherapists who delivered care. Data were exported from the secure online platforma into a spreadsheetb for analysis. Descriptive statistics, including frequencies Therefore, the research question for this mixed-methods study (percentages) and means and standard deviations, were used to was: summarise the data from those who completed the surveys. Geographic locations of respondents (residential location for patients What were the experiences of physiotherapists and patients who and practice location for physiotherapists) were categorised by consulted via videoconference during the COVID-19 pandemic and postcodes into: metropolitan, regional/rural and remote areas.25 how was it implemented? Responses to free-text questions underwent qualitative content Method analysis.26 This involved three researchers (BJL, BM, DM) indepen- dently reading through all responses and coding the data to identify Design topics and initial patterns of ideas. Codes were organised into cate- gories and combined with similar ideas to form larger themes. A mixed-methods study was conducted with descriptive, cross- Themes with the highest number of individual data points were sectional national online surveys of samples of physiotherapists and identified and reported. patients in Australia, and qualitative analysis of free-text responses. Results Participants A total of 380 physiotherapists underwent online screening, with Physiotherapists 162 excluded: 42 were ineligible (not currently providing care via Assisted by the Australian Physiotherapy Association, we recruited videoconferencing (n = 22), not registered to practise in Australia (n = 11) and not in private practice, community health or outpatient physiotherapists using advertisements in social media (Facebook, centre (n = 9)) and 120 chose not to participate. The survey was Twitter and LinkedIn), targeted emails, publications such as InMotion, commenced by 218 physiotherapists, of whom 207 (95%) completed and newsletters. To be eligible, physiotherapists had to: be registered the survey. with the Australian Health Practitioner Regulation Agency; be working in private practice or community settings; and have pro- In total, 671 patients underwent online screening, with 251 vided one or more patient consultations (individual and/or group) via excluded: 69 were ineligible (had not had physiotherapy by video- videoconference between April and October 2020. conferencing between April and October 2020 (n = 68), no email or internet access (n = 1)) and 182 chose not to participate. The survey Patients was commenced by 420 eligible patients, with 308 (77%) recruited via We recruited people aged 18 years who had consulted with a email invitation from their treating physiotherapist, 82 (20%) from social media and 11 (3%) by word of mouth. Of the eligible patients, physiotherapist via videoconferencing (individual and/or group) be- 401 (95%) completed the survey. At the time of survey completion, tween April and October 2020 for any health problem/condition and 315 (79%) patients had finished their episode of physiotherapy care who had an email address and access to the Internet. Participating via videoconference, 31 (8%) were still receiving care via videocon- physiotherapists were asked to email eligible patients on behalf of the ference, and this was unknown for 55 (14%). Collectively, patients had researchers inviting them to participate. Additional patients were consulted with 172 different physiotherapists across 148 clinics in all recruited through advertisements on social media (Facebook and eight states and territories of Australia. LinkedIn). Characteristics of participants Procedure The characteristics of physiotherapists and patients are shown in In order to capture information of relevance to key stakeholders Tables 1 and 2, respectively. Participants mostly resided in major and to ensure readability and clarity, customised separate surveys cities (77%) and identified as female (76%). (Appendix 1 on the eAddenda) for physiotherapists and patients were designed with input from researchers, physiotherapists, staff mem- bers of the Australian Physiotherapy Association, health insurers, compensable bodies, and consumers. Selected items from the
Research 203 Table 1 Table 2 Characteristics of physiotherapists. Characteristics of patients. Characteristics Physiotherapists Characteristics Patients (n = 207) (n = 401) Sex, n (%) 55 (27) Sex, n (%) 95 (24) male 152 (73) male 305 (76) female female 80 (39) undisclosed 1 (0) State, n (%) 45 (22) Victoria 40 (19) State, n (%) 188 (47) New South Wales 14 (7) Victoria 46 (11) Queensland New South Wales 55 (14) South Australia 7 (3) Queensland 47 (12) Tasmania 1 (0) South Australia 29 (7) Northern Territory 4 (2) Tasmania Australian Capital Territory 16 (8) Northern Territory 0 (0) Western Australia Australian Capital Territory 22 (5) 158 (76) Western Australia 14 (3) Geographical location, n (%) 45 (22) major city Geographical location, n (%) 307 (77) regional 4 (2) major city 91 (23) remote 19 (12) regional remote 3 (1) Clinical experience (yr), mean (SD) 16 (8) Postgraduate qualifications, n (%) 10 (5) Age (yr), n (%) 85 (21) 52 (25) 18 to 39 60 (15) PhD 25 (12) 40 to 49 88 (22) Masters by research 14 (7) 50 to 59 87 (22) Masters by coursework 90 (43) 60 to 69 40 (10) Postgraduate diploma 70 to 79 other 23 (11) 80 8 (2) none 8 (4) Prior training in telehealth, n (%) Confidence using technology, n (%) 1 (0) yes, online 176 (85) not at all confident 26 (6) yes, in person somewhat confident 158 (39) no 177 (86) moderately confident 216 (54) Clinical setting, n (%)a 20 (10) extremely confident private practice 23 (11) 24 (6) community health centre Predominant body part being treated, n (%) 47 (12) outpatient clinic 8 (4) head or neck 93 (23) other back/chest/abdomen 93 (23) Predominant clinical focus, n (%)a 130 (63) hip/pelvis 69 (17) musculoskeletal 71 (34) lower limb 66 (16) sports and exercise 32 (15) upper limb paediatrics 29 (14) whole body 9 (2) neurology other cardiorespiratory 4 (2) 235 (59) gerontology 14 (7) Main reasons for seeking treatment, n (%)a 177 (44) occupational health 5 (2) pain 135 (34) aquatic 3 (1) impaired function 100 (25) women’s, men’s and pelvic health 43 (21) stiffness 73 (18) cancer, palliative care 11 (5) weakness 72 (18) mental health 2 (1) difficulty walking 70 (17) Telehealth experience prior to COVID-19, n (%) rehabilitation following trauma/injury 43 (11) provided individual videoconference care 44 (21) rehabilitation following surgery provided group videoconference care 0 (0) balance/falls problems 37 (9) bladder/bowel control or prolapse 32 (8) a Percentages total . 100 as respondents could chose more than one answer. fatigue 28 (7) rehabilitation for a neurological condition 23 (6) Physiotherapists had a mean of 19 years (SD 12) of clinical expe- deconditioning 15 (4) rience, with the majority having postgraduate qualifications. Their reduced cardiovascular fitness 10 (2) predominant clinical focus was musculoskeletal (63%), sports/exer- breathlessness 3 (1) cise (34%), and women’s, men’s and pelvic health (21%); however, 11 frailty 52 (13) different clinical areas were represented. Few reported prior training other in telehealth (15%). Prior to the pandemic, physiotherapists had 38 (9) limited telehealth experience, with 21% having delivered individual Duration of problem, n (%) 52 (13) care and none having delivered group classes via videoconference. , 6 weeks 97 (24) 6 to 12 weeks 214 (53) Patients were aged from 18 to . 80 years and most reported being 3 to 12 months moderately (39%) or extremely (54%) confident using technology. . 12 months Patients sought treatment for a variety of reasons including pain (58%), impaired function (44%) and stiffness (34%), with most prob- a Percentages total . 100 as respondents could chose more than one answer. lems being chronic and longer than 12 months in duration (53%). group classes. Physiotherapists generally rated their level of experi- Physiotherapists’ implementation of and experiences with care via ence and confidence in providing such care as moderate to high videoconferencing (average . 7 out of 10). The most common videoconferencing plat- forms used for individual consultations were Physitrack (30%), Coviu Table 3 summarises physiotherapists’ implementation of and ex- (20%) and Zoom (16%). Zoom was used by the majority (94%) for periences with care delivered via videoconferencing. Individual care group classes. A range of supporting patient resources was used, the was provided by 204 (99%) physiotherapists, while 35 (17%) physio- most common being written instructions, diagrams or booklets (63%), therapists provided group classes. The mean (SD) duration of phys- educational material about the issue/condition (54%) and apps for iotherapist experience providing videoconferencing care was 11.9 smart phone or tablet (40%) for individual consultations, and text (16.2) months for individual consultations and 6.8 (1.4) months for message reminders (66%) and follow-up phone calls (31%) for group classes. Physiotherapists charged fees for individual consultations that were slightly lower than those they usually charged for face-to face care: mean 89% (SD 24) of the cost of an equivalent in-person visit for initial
204 Bennell et al: Physiotherapist and patient experiences with telehealth Table 3 Individual Group Physiotherapist implementation of and experiences with care provided by videoconference. (n = 204) (n = 35) Survey items 11.9 (16.2) 6.8 (1.4) 5.9 (2.3) 6.9 (2.7) Duration of providing VC consultations (mth), mean (SD) 7.3 (1.8) 8.1 (1.8) Experience with VC consultations (0 to 10), mean (SD)a 135 (66) Confidence providing VC consultations (0 to 10), mean (SD)a N/A Deemed some patients unsuitable for VC, n (%)b 77 (38) Main reasons patients deemed unsuitable, n (%)c 70 (34) N/A 75 (37) N/A patient unable to access technology 55 (27) N/A complexity of problem/condition 54 (26) N/A patient required hands-on treatment 51 (25) N/A unable to adequately diagnose/assess patient 32 (16) N/A complexity of patient 11 (5) N/A patient unable to use technology (eg, impairment) 167 (82) N/A safety concerns 29 (85) other 129 (63) Received positive patient feedback, n (%)b 110 (54) 9 (26) Patient resources used to support VC consultations, n (%)c 81 (40) 9 (26) written instructions, diagrams or booklets 81 (40) 10 (29) educational material about issue/condition 71 (35) 10 (29) apps for smart phone or tablet 67 (33) 6 (17) videos 66 (32) 11 (31) websites for further information 28 (14) 8 (23) follow-up phone calls 23 (11) 2 (6) provision/purchase of equipment/devices 7.0 (1.7) 23 (66) log books/diaries 7.1 (1.6) 7.7 (1.4) text message reminders 7.5 (1.7) Effectiveness of VC care (0 to 10), mean (SD)a 62 (30) Satisfaction with VC care (0 to 10), mean (SD)a 40 (20) 0 (0) VC platform used, n (%)c 33 (16) 2 (6) Physitrack 30 (15) 33 (94) Coviu 19 (9) 0 (0) Zoom 13 (6) 2 (6) Cliniko 12 (6) 0 (0) Facetime 46 (23) 2 (6) Health Direct 5 (14) Microsoft Teams 29 (14) other 68 (33) N/A Business costs of VC versus in-person consultations, n (%) 63 (31) N/A VC would cost the business more 44 (22) N/A VC and in-person would cost the same amount N/A in-person would cost the business more 166 (81) don’t know 8 (4) 21 (60) Intending to continue VC care after pandemic, n (%) 5 (14) yes 30 (15) 9 (26) no unsure N/A = not assessed; VC = videoconference. a 0 = not at all, 10 = extremely. b Number (%) of physiotherapists. c Percentages total 100 as respondents could chose more than one answer. consultation and 90% (SD 20) of the cost of an equivalent in-person preparing ahead of appointments, having patient resources available visit for review. However, opinions differed as to whether the busi- (particularly exercise apps) and patients being willing and engaged. ness costs of providing individual care via videoconference would be Other barriers included lack of physical touch, perceived inability to more than, less than, or the same as providing in-person care. assess the patient properly and room setup. One of the key safety issues mentioned for both individual and group care was falls risk. Over 24 different classes were reported, with the most common being Pilates (38%), Good Life with osteoArthritis:Denmark (GLA:D) Patient experiences with physiotherapy care via (20%), low back pain (18%), post-natal (13%) and pre-natal (11%). videoconferencing Thirty-three (70%) physiotherapists reported setting a limit on group class size, with the average maximum per class being 7.9 patients (SD Patient experiences with physiotherapy care via videoconfer- 5.4). encing are summarised in Table 5. Of the patient respondents, 341 (85%) reported receiving individual physiotherapy care via video- A total of 66% of physiotherapists deemed one or more patients conference, while 77 (19%) attended group classes. For the particular unsuitable for individual treatment via videoconference, with a range episode of care, the mean number of videoconferencing consultations of reasons given, including: patient unable to access the technology, was 3.9 (SD 5.5) for individual care and 17.9 (SD 26.4) for group complexity of the problem/patient, unable to adequately assess and sessions. Most patients had previously consulted the physiotherapist requiring hands-on treatment. Physiotherapists gave moderate-to- in-person for the same problem before switching to videoconfer- high ratings (7 to 8 out of 10) for effectiveness and satisfaction with encing. Most patients either fully (41% for individual and 49% for care for both individual treatment and group classes. Many physio- group) or partially (30% for individual and 35% for group) paid for therapists intended to continue offering videoconferencing care their care themselves. While the majority of patients considered individually (81%) or in groups (60%) after the pandemic ended, videoconferencing care to be the same or better quality compared although a proportion were unsure (15% and 26%, respectively). with in-person care, just under half (41% of patients receiving indi- vidual care and 43% receiving group care) rated it as lower quality. Table 4 summarises the main themes relating to physiotherapists’ Patients valued individual videoconferencing care for a number of perceived facilitators, barriers and safety issues of delivering care via videoconference. Technology was predominant as both a facilitator and barrier for individual and group care. Other facilitators included
Research 205 Table 4 Main themes relating to physiotherapists’ perceived facilitators, barriers and safety issues with delivery of care via videoconference. Questions Individual consultations Group classes What things helped you the most Good technology setup: to deliver physiotherapy care via Good technology setup: reliable internet connection; good telehealth telehealth? platform; good hardware setup (n = 98) reliable internet connection; good hardware setup (n = 25) What barriers did you experience Using patient resources: written/online information; exercise videos Already knowing the clients (n = 7) delivering physiotherapy care or instructions; follow-up email summaries; exercise apps (n = 65) Preparing ahead of time (n = 6) via telehealth? Preparing ahead of appointment: preparing for technology issues; Technology issues: poor What safety issues did you patient instructions before appointment; having resources/equipment internet quality (n = 20) experience delivering care ready (n = 31) via telehealth? Poor room setup: poor Patient willingness and engagement (n = 21) lighting; poor camera angles (n = 8) Technology issues: poor internet quality; issues with device; poor Lack of physical touch (n = 4) technology skills (n = 130) Falls risk (n = 4) Lack of physical touch: unable to facilitate movement or exercise; unable to physically examine; unable to use hands-on techniques; limits ability to do thorough assessment (n = 74) Poor room setup: noisy space; poor camera angles; poor lighting; limited space (n = 17) Falls risk (n = 17) Unsupervised exercise/incorrect technique (n = 8) Difficult to assess thoroughly (n = 6) n = number of responses that contributed to each theme. Table 5 Individual Group Patients’ experiences with physiotherapy care provided by videoconference. (n = 341) (n = 77) Survey questions 3.9 (5.5) 17.9 (26.4) 56 (34) 53 (39) VC consultations for this problem (n), mean (SD) 292 (86) 55 (71) Percentage of physiotherapy consultations delivered via VC (%), mean (SD) Had prior in-person consultations with the same physiotherapist for the same problem, n (%)a 140 (41) 38 (49) Payment for VC consultation, n (%) 104 (30) 27 (35) 97 (28) 12 (16) patient paid entire fee patient paid part fee 100 (50) 26 (67) fee paid by other 62 (31) 7 (14) Funding source for VC consultation, if part/all of fee paid by other, n (%)b 11 (5) 1 (3) private health insurance 19 (9) 6 (15) Medicare 0 (0) workers compensation scheme 6 (3) 1 (3) National Disability Insurance Scheme 8 (4) Department of Veterans’ Affairs 5 (6) Transport Accident Commission 20 (6) 22 (29) Expectations and experiences with VC, n (%) 112 (33) 50 (65) less than what I expected 209 (61) what I expected 19 (43) exceeded my expectations 97 (42) 16 (36) Quality compared to in-person, n (%)c 111 (48) 9 (20) VC lower quality 24 (10) VC same quality N/A VC better quality 299 (88) N/A Most valued about VC, n (%)d 183 (54) N/A convenience 134 (39) N/A access 110 (32) N/A less waiting time 79 (23) N/A undivided attention of physio 71 (21) N/A treatment effectiveness 67 (20) N/A privacy 51 (15) N/A cost savings COVID-19 safety/social distancing 18 (5) other Numbers do not sum to the total for some items due to missing data. N/A = not assessed, VC = videoconference. a Number (%) of patients. b Only includes patients who had part/all of fee paid by other: n = 201 for individual and n = 39 for group. c Only includes patients who had received previous in-person care: n = 232 for individual and n = 44 for group. d Percentages total . 100 as respondents could chose more than one answer. reasons, most commonly convenience (88%), access (54%), less wait- likely to choose to use videoconferencing for individual consultations ing time (39%) and undivided physiotherapist attention (32%). beyond the pandemic, with 28% not at all likely to do so. For group classes, 68% were moderately or extremely likely to choose to do so Patient ratings of their experiences are shown in Figure 1 for in- via videoconferencing beyond the pandemic, with 13% not at all likely dividual consultations and Figure 2 for group sessions. Most had to do so. Full numerical data used to generate Figures 1 and 2 are moderately or extremely positive perceptions about the ease of using available in Tables 6 and 7 on the eAddenda. the technology (94% individual consultations versus 91% group clas- ses), comfort communicating (96% versus 86%), satisfaction with Discussion management (92% versus 93%), satisfaction with privacy/security (98% versus 95%), safety during the consultation (99% versus 99%), This study found that patients and physiotherapists had overall safety doing prescribed activities (93% versus 99%), and effectiveness positive experiences with care delivered via videoconferencing (83% versus 89%). Around half (47%) were moderately or extremely
206 Bennell et al: Physiotherapist and patient experiences with telehealth Percentage of patients 0% 20% 40% 60% 80% 100% Ease of using technologya Aspect of videoconference experience Comfort communicatingb Satisfaction with managementc Satisfaction with privacy/securityc Safety during consultationd Safety doing prescribed activitiesd Effectivenesse Likely to choose in the futuref Not at all Somewhat Moderately Extremely Figure 1. Patient ratings of their experiences with individual consultations via videoconference with their physiotherapist (n = 341). a Rated on a 4-point scale ranging from ‘not at all easy’ to ‘extremely easy’. b Rated on 4-point scale ranging from ‘not at all comfortable’ to ‘extremely comfortable’. c Rated on 4-point scale ranging from ‘not at all satisfied’ to ‘extremely satisfied’. d Rated on 4-point scale ranging from ‘not at all safe’ to ‘extremely safe’. e Rated on 4-point scale ranging from ‘not at all effective’ to ‘extremely effective’. f Rated on 4-point scale ranging from ‘not at all likely’ to ‘extremely likely’. Percentage of patients 0% 20% 40% 60% 80% 100% Ease of using technologyᵃ Aspect of videoconference experience Comfort communicatingᵇ Satisfaction with managementc Satisfaction with privacy/securityc Safety during consultationᵈ Safety doing prescribed activitiesᵈ Effectivenessᵉ Likely to choose in the futuref Not at all Somewhat Moderately Extremely Figure 2. Patient ratings of their experiences with group classes via videoconference with their physiotherapist (n = 77). a Rated on a 4-point scale ranging from ‘not at all easy’ to ‘extremely easy’. b Rated on 4-point scale ranging from ‘not at all comfortable’ to ‘extremely comfortable’. c Rated on 4-point scale ranging from ‘not at all satisfied’ to ‘extremely satisfied’. d Rated on 4-point scale ranging from ‘not at all safe’ to ‘extremely safe’. e Rated on 4-point scale ranging from ‘not at all effective’ to ‘extremely effective’. f Rated on 4-point scale ranging from ‘not at all likely’ to ‘extremely likely’. during the COVID-19 pandemic. However, although most patients and of participating clinicians believed that telehealth was as effective as in-person care.28 The lack of physical contact during telehealth was physiotherapists indicated willingness to use telehealth in the future, also perceived to hamper accurate and effective diagnosis and man- agement.28 Similar experiences were reported by our cohort of almost one-third of patients were unlikely to choose to do so. Barriers physiotherapists, who were moderately-to-highly satisfied with and to telehealth delivery experienced by physiotherapists included confident using videoconferencing to provide care to patients. In addition, our cohort of physiotherapists was also mostly untrained in technology issues, a lack of physical touch and poor room setup. telehealth and reported that the lack of physical touch was a barrier The findings broadly reflect those of other studies investigating and limited their ability to conduct a thorough assessment. physiotherapist’s experiences with telehealth during COVID-19 in Our findings are also broadly similar to others investigating pa- other countries. Other surveys of allied healthcare clinicians in tient experiences using telehealth for physiotherapy during COVID- Australia, Europe and North America found that satisfaction with 19. Patients in the US and Italy were found to have high overall telehealth was high,27 physiotherapists believed that using telehealth satisfaction,27,29–31 being very satisfied with their communication via was part of their professional role28 and they felt confident using telehealth to treat patients.28 However, those studies also found that physiotherapist training in telehealth was lacking28 and less than half
Research 207 telehealth,29 the development and execution of their treatment including having reliable hardware and software (on both the phys- plan,29 and the vast majority indicated that they would use telehealth iotherapist and patient end) and use of written/online information, in the future.29,31 Technology issues (including setup and camera videos and apps. Similar facilitators to telehealth were also reported angles) and elements of hands-on care (lack of tactile feedback, by other studies investigating implementation of telehealth during inability to perform soft tissue work, absence of ‘healing touch’) were COVID-19,23,28,35 suggesting that future telehealth services should identified as limitations of telehealth.29 These findings broadly reflect consider these factors. Barriers to telehealth included technology is- ours, where, overall, patients had a positive experience using the sues (at the physiotherapist and patient end) and the lack of physical technology and communicating with the physiotherapist, and felt touch, which they perceived limited their ability to conduct a thor- safe. ough assessment and rendered them unable to use hands-on tech- niques. Such barriers have also been reported in other studies.23,28,35 Our findings suggest that a much higher proportion of patients Cottrell and Russell3 suggest that telehealth may be most appropriate as opposed to physiotherapists would not be willing to use video- for observational assessments, but not those requiring physical con- conferencing to do consultations in the future (28% versus 4%, tact. Blended models of service delivery, where a combination of in- respectively). This is somewhat surprising given that there is some person and telehealth consultations are offered, may be the most suggestion in the literature of poor acceptability of and resistance to suitable approach, where patients and physiotherapists use either, telehealth amongst physiotherapists.3,28 It is possible that first-hand depending on the patient preferences, circumstances and re- experience with telehealth contributed to a shift in physiothera- quirements. Technology and setup barriers may be overcome as tel- pists’ perceptions about such services.32 However, it is not imme- ehealth services become more mainstream within the community, diately clear why patients appear to be less willing to use telehealth and also with appropriate training in telehealth. This is particularly than physiotherapists, given that the majority reported positive relevant given that around one-third (31%) of physiotherapists used experiences using the technology and communicating, and believed platforms that are not specifically designed for telehealth (eg, Zoom, that the care they received via videoconferencing was effective. This FaceTime, Microsoft Teams). Both physiotherapists and patients in difference may partly be explained by the fact that patients were our cohort were relatively naïve with respect to telehealth and, given rating telehealth for themselves and their individual situation/ recent evidence that level of experience with telehealth was associ- condition, whereas physiotherapists answered in relation to their ated with more positive perceptions and greater physiotherapist entire caseload of patients, some of whom may not have been satisfaction,27 it is likely that further experience by both users and suitable for telehealth. In fact, 66% of physiotherapists deemed some providers will lead to higher quality and more acceptable services. patients as unsuitable for videoconferencing, suggesting that even though they indicated intentions to continue using telehealth, it Another important consideration is that most physiotherapists in appears unlikely that they would intend to use it with all of their our surveyed cohort treated mostly chronic, rather than acute, con- patients. ditions via telehealth. It is unclear whether this was because those with acute conditions were less likely to seek care during the We believe that no previous studies have examined patient or pandemic, whether they were less suited for telehealth, or whether physiotherapist experiences with group classes via videoconferencing funding/reimbursement was unavailable for acute conditions. In during COVID-19. Most existing research on group classes delivered addition, almost three-quarters of patients had already seen their via telehealth by a physiotherapist (before the COVID-19 pandemic) physiotherapist in person for the same problem prior to using tele- has been in cardiac and pulmonary conditions9,14,15,33 and in research health. Again, it is unclear whether this was because patients were settings, rather than a ‘real-world’ environment. Two of those studies less likely to see a physiotherapist for the first time via telehealth, or included a mixed methods exploration of patient experiences with whether physiotherapists were less willing to see new patients via home-based group exercise classes via videoconferencing for people telehealth. Further research is required to develop guidelines and with chronic obstructive pulmonary disorder34 and people with heart recommendations to help physiotherapists and service providers failure.33 Those studies reported high levels of patient satisfaction, better determine which patients may be unsuitable for telehealth. with patients enjoying exercising with others, feeling safe and appreciating the accessibility of care (ie, reduced burden and costs of One of the most commonly reported barriers to the imple- transportation). However, patients in both studies also experienced mentation of telehealth is lack of reimbursement by public or private technical difficulties and needed help operating the system, and health insurers, as well as the costs of implementing such services suggested that improvements to the audio and visual components of (such as obtaining necessary infrastructure).3,20,36 However, we found the software/hardware would be beneficial. Around three-quarters of that more than one-third of patients paid the entire fee of their their participants agreed or strongly agreed that they would continue consultation via videoconferencing, and around half paid the entire to participate in group classes via telehealth.34 These findings appear fee for a group class via videoconferencing, suggesting that patients to broadly reflect ours, in that our cohort had overall positive expe- are willing to pay for telehealth. In addition, most physiotherapists riences, with 68% being moderately or extremely likely to choose to believed that the business costs of delivering care via videoconfer- attend group classes via videoconferencing in the future. Although encing would be equal to or less than the costs of doing so in-person, 91% of our cohort found the technology moderately or extremely easy suggesting that implementing telehealth would not have a detri- to use, our physiotherapists reported technology issues as one of the mental financial impact on service providers. Collectively, these main barriers to group classes. Interestingly, patients in our cohort findings suggest that funding and cost barriers to the implementation appeared to be more willing to choose to use videoconferencing for of telehealth may not be as great as initially thought; further inves- group classes in the future (87%), compared with individual consul- tigation into the long-term costs and funding of telehealth is tations (72%). It is unclear why this is, particularly given that satis- required. faction with other elements (including ease of use, safety and effectiveness) were similar between individual consultations and Only 15% of physiotherapists in our cohort reported that they had group classes. In addition, our results suggest that fewer physio- prior training in telehealth. This also reflects other studies, which therapists intend to continue to offer group classes via videoconfer- found that only a minority of clinicians had been trained in tele- encing beyond the pandemic (60%), compared with individual health.28 Cottrell and Russell3 argue that many barriers to telehealth consultations (81%). Further research is required to determine why delivery amongst physiotherapists (such as resistance to changing patients and physiotherapists may be more or less willing to attend or practice, poor technological self-efficacy, perceived de- deliver group classes via telehealth, compared with individual personalisation of care, and privacy and safety concerns) reflect a consultations. lack of skills and confidence to safely and effectively deliver care via telehealth. As such, this, along with a number of previous publica- Physiotherapists identified numerous barriers and facilitators to tions,23,37 have highlighted the need for telehealth training programs. delivering care via telehealth that have implications for the future With the uptake of telehealth and the potential of services continuing design and delivery of such services. Technology setup and patient beyond the pandemic, telehealth training programs may become resources were the two most commonly mentioned facilitators, more common in undergraduate and postgraduate physiotherapist
208 Bennell et al: Physiotherapist and patient experiences with telehealth training programs, helping clinicians overcome some of these Correspondence: Kim L Bennell, Department of Physiotherapy, barriers. The University of Melbourne, Melbourne, Australia. Email: Some strengths of our study include relatively large numbers of physiotherapists and patients across all states and territories in [email protected] Australia and across numerous practices. Patients were clearly informed that their results would not be shared with their physio- References therapist, in order to facilitate more accurate responses. Our study also had limitations. The sampling approach contained an element of 1. Grona SL, Bath B, Busch A, Rotter T, Trask C, Harrison E. Use of videoconferencing convenience sampling, which may have introduced bias. Physio- for physical therapy in people with musculoskeletal conditions: a systematic re- therapists and patients in tertiary/public hospital settings were not view. J Telemed Telecare. 2018;24:341–355. directly sampled and, as such, no information about the full range of settings in which physiotherapy care was provided was available. 2. Cottrell MA, Galea OA, O’Leary SP, Hill AJ, Russell TG. Real-time telerehabilitation While the experiences of some physiotherapists whose predominant for the treatment of musculoskeletal conditions is effective and comparable to focus was paediatrics were captured, patient/carer experiences for standard practice: a systematic review and meta-analysis. Clin Rehabil. patients aged , 18 years were not captured. There was also a limited 2017;31:625–638. number of respondents who had delivered or undertaken group sessions via videoconference. Our results may not necessarily 3. Cottrell MA, Russell TG. Telehealth for musculoskeletal physiotherapy. Muscu- generalise to other countries where healthcare contexts and physio- loskelet Sci Pract. 2020;48:102193. therapy practice may differ. A proportion of people chose not to participate (22 to 30%), which may have been because of the 4. Dario AB, Moreti Cabral A, Almeida L, Ferreira ML, Refshauge K, Simic M, et al. burdensome nature of participation. This investigation was also Effectiveness of telehealth-based interventions in the management of non-specific confined to care via videoconferencing; it would have been inter- low back pain: a systematic review with meta-analysis. Spine J. 2017;17:1342–1351. esting to have examined and compared physiotherapy services pro- vided by telephone, given that, anecdotally, this delivery mode was 5. Shukla H, Nair SR, Thakker D. Role of telerehabilitation in patients following total also frequently used. knee arthroplasty: evidence from a systematic literature review and meta-analysis. J Telemed Telecare. 2017;23:339–346. In conclusion, this study found that patients and physiotherapists had overall positive experiences using videoconferencing for both 6. Rawstorn JC, Gant N, Direito A, Beckmann C, Maddison R. Telehealth exercise- individual consultations and group classes. The results suggest that based cardiac rehabilitation: a systematic review and meta-analysis. Heart. videoconferencing is a viable option for the delivery of physiotherapy 2016;102:1183–1192. care in the future. Attention to perceived barriers, facilitators and potential safety issues may enhance the implementation of and ex- 7. Chan C, Yamabayashi C, Syed N, Kirkham A, Camp PG. Exercise telemonitoring and periences with telehealth. telerehabilitation compared with traditional cardiac and pulmonary rehabilitation: a systematic review and meta-analysis. Physiother Can. 2016;68:242–251. What was already known on this topic: Telehealth is an effective physiotherapy service delivery mode, with outcomes 8. Hinman RS, Nelligan RK, Bennell KL, Delany C. “Sounds a bit crazy, but it was similar to or even better than those achieved with in-person care almost more personal”: a qualitative study of patient and clinician experiences of in some clinical conditions. During the COVID-19 pandemic, physical therapist-prescribed exercise for knee osteoarthritis via Skype. Arthritis many physiotherapy services are rapidly transitioning to Care Res. 2017;69:1834–1844. telehealth. What this study adds: Patients and physiotherapists had 9. Tsai LL, McNamara RJ, Moddel C, Alison JA, McKenzie DK, McKeough ZJ. Home- overall positive experiences using videoconferencing for indi- based telerehabilitation via real-time videoconferencing improves endurance ex- vidual consultations and group classes. The results suggest that ercise capacity in patients with COPD: the randomized controlled TeleR Study. videoconferencing is a viable option for the delivery of physio- Respirology. 2017;22:699–707. therapy care in the future. 10. Hoaas H, Andreassen HK, Lien LA, Hjalmarsen A, Zanaboni P. Adherence and factors Footnotes: aREDCap, Vanderbilt University, Nashville, USA. affecting satisfaction in long-term telerehabilitation for patients with chronic bExcel, Microsoft Corporation, Redmond, USA. obstructive pulmonary disease: a mixed methods study. BMC Med Inform Decis Mak. 2016;16:26. eAddenda: Tables 6 and 7, and Appendix 1 can be found online at DOI: https://doi.org/10.1016/j.jphys.2021.06.009 11. Kairy D, Tousignant M, Leclerc N, Cote AM, Levasseur M, Researchers TT. The pa- tient’s perspective of in-home telerehabilitation physiotherapy services following Ethics approval: The University of Melbourne Human Research total knee arthroplasty. Int J Environ Res Public Health. 2013;10:3998–4011. Ethics Committee approved this study (Ethics ID: 2056784). All par- ticipants gave written informed consent before data collection began. 12. Hwang R, Bruning J, Morris NR, Mandrusiak A, Russell T. Home-based tele- rehabilitation is not inferior to a centre-based program in patients with chronic Competing interests: Karen Finnin owns a telehealth physio- heart failure: a randomised trial. J Physiother. 2017;63:101–107. therapy practice. Trevor Russell is involved in revenue-generating telehealth educational programs. 13. Shulver W, Killington M, Morris C, Crotty M. ‘Well, if the kids can do it, I can do it’: older rehabilitation patients’ experiences of telerehabilitation. Health Expect. Source(s) of support: This work was supported by funding from 2017;20:120–129. the Physiotherapy Research Foundation. Professor Bennell is sup- ported by a National Health and Medical Research Council (NHMRC) 14. Ptomey LT, Willis EA, Lee J, Washburn RA, Gibson CA, Honas JJ, et al. The feasibility Investigator grant (#1174431). Professor Hinman is supported by a of using pedometers for self-report of steps and accelerometers for measuring National Health and Medical Research Council Senior Research physical activity in adults with intellectual and developmental disabilities across Fellowship (#1154217). The funders had input in the development of an 18-month intervention. J Intellect Disabil Res. 2017;61:792–801. the study method, interpretation of the results and reporting as collaborative partners. 15. Burkow TM, Vognild LK, Johnsen E, Risberg MJ, Bratvold A, Breivik E, et al. Comprehensive pulmonary rehabilitation in home-based online groups: a mixed Acknowledgements: Participant survey data were collected and method pilot study in COPD. BMC Res Notes. 2015;8:766. managed using the REDCap Survey Software hosted at the University of Melbourne. 16. Hinman RS, Campbell PK, Lawford BJ, Briggs AM, Gale J, Bills C, et al. Does telephone-delivered exercise advice and support by physiotherapists improve pain Provenance: Not invited. Peer reviewed. and/or function in people with knee osteoarthritis? Telecare randomised controlled trial. Br J Sports Med. 2020;54:790–797. 17. Lawford BJ, Bennell KL, Kasza J, Hinman RS. Physical therapists’ perceptions of telephone- and internet video-mediated service models for exercise management of people with osteoarthritis. Arthritis Care Res. 2018;70:398–408. 18. Hinman R, Lawford BJ, Bennell K. Harnessing technology to deliver care by physical therapists for people with persistent joint pain: telephone and video-conferencing service models. J Appl Biobehav Res. 2018;24:e12150. 19. Monaghesh E, Hajizadeh A. The role of telehealth during COVID-19 outbreak: a systematic review based on current evidence. BMC Public Health. 2020;20:1193. 20. Duckett S. What should primary care look like after the COVID-19 pandemic? Aust J Prim Health. 2020;26:207–211. 21. World Physiotherapy. Impact of the COVID-19 pandemic on physiotherapy services globally; 2021. https://world.physio/sites/default/files/2021-03/Covid-Report_ March2021_FINAL.pdf. Accessed 19 May, 2021. 22. Stanhope J, Weinstein P. Learning from COVID-19 to improve access to physio- therapy. Aust J Prim Health. 2020;26:271–272. 23. Signal N, Martin T, Leys A, Maloney R, Bright F. Implementation of tele- rehabilitation in response to COVID-19: lessons learnt from neurorehabilitation clinical practice and education. NZ J Physiother. 2020;48:117–126. 24. Parmanto B, Lewis AN, Graham KM, Bertolet MH. Development of the Telehealth Usability Questionnaire (TUQ). Int J Telerehabil. 2016;8:3–10. 25. Department of Health. Modified Monash Model. Australian Government Depart- ment of Health; 2015. https://www.health.gov.au/health-workforce/health- workforce-classifications/modified-monash-model. Accessed 19 May, 2021. 26. Hsieh HF, Shannon SE. Three approaches to qualitative content analysis. Qual Health Res. 2005;15:1277–1288. 27. Assenza C, Catania H, Antenore C, Gobbetti T, Gentili P, Paolucci S, et al. Continuity of care during COVID-19 lockdown: a survey on stakeholders’ experience with telerehabilitation. Front Neurol. 2020;11:617276.
Research 209 28. Malliaras P, Merolli M, Williams CM, Caneiro JP, Haines T, Barton C. ‘It’s not hands- physiotherapists’ perceptions of such a service for knee osteoarthritis. A qualitative on therapy, so it’s very limited’: telehealth use and views among allied health study. Arthritis Care Res. 2018;71:545–557. clinicians during the coronavirus pandemic. Musculoskelet Sci Pract. 33. Hwang R, Mandrusiak A, Morris NR, Peters R, Korczyk D, Bruning J, et al. Exploring 2021;52:102340. patient experiences and perspectives of a heart failure telerehabilitation program: a mixed methods approach. Heart Lung. 2017;46:320–327. 29. Tenforde AS, Borgstrom H, Polich G, Steere H, Davis IS, Cotton K, et al. Outpatient 34. Tsai LLY, McNamara RJ, Dennis SM, Moddel C, Alison JA, McKenzie DK, et al. physical, occupational, and speech therapy synchronous telemedicine: a survey Satisfaction and experience with a supervised home-based real-time videocon- study of patient satisfaction with virtual visits during the COVID-19 pandemic. Am J ferencing telerehabilitation exercise program in people with chronic obstructive Phys Med Rehabil. 2020;99:977–981. pulmonary disease (COPD). Int J Telerehabil. 2016;8:27–38. 35. Turolla A, Rossettini G, Viceconti A, Palese A, Geri T. Musculoskeletal physical 30. Negrini S, Donzelli S, Negrini A, Negrini A, Romano M, Zaina F. Feasibility and therapy during the COVID-19 pandemic: is telerehabilitation the answer? Phys acceptability of telemedicine to substitute outpatient rehabilitation services in the Ther. 2020;100:1260–1264. COVID-19 emergency in Italy: an observational everyday clinical-life study. Arch 36. Lin CC, Dievler A, Robbins C, Sripipatana A, Quinn M, Nair S. Telehealth in health cen- Phys Med Rehabil. 2020;101:2027–2032. ters: key adoption factors, barriers, and opportunities. Health Aff. 2018;37:1967–1974. 37. Edirippulige S, Armfield NR. Education and training to support the use of clinical 31. Miller MJ, Pak SS, Keller DR, Barnes DE. Evaluation of pragmatic telehealth physical telehealth: a review of the literature. J Telemed Telecare. 2017;23:273–282. therapy implementation during the COVID-19 pandemic. Phys Ther. 2021;101:193. 32. Lawford BJ, Delany C, Bennell KL, Hinman RS. “I was really pleasantly surprised”: first-hand experience with telephone-delivered exercise therapy shifts
Journal of Physiotherapy 67 (2021) 151–155 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 Physiotherapists can reduce overuse by Choosing Wisely Priti Kharel a,b, Joshua R Zadro a,b, Chris G Maher a,b a Institute for Musculoskeletal Health, Sydney Local Health District; b School of Public Health, Faculty of Medicine and Health, The University of Sydney, Sydney, Australia Choosing Wisely is a major public awareness campaign that aims physiotherapy3 and the positive effects of Choosing Wisely in other to facilitate conversations between patients and health professionals fields, it is vital that physiotherapy associations join the campaign. about overuse.1 Overuse is the provision of a health service (test or treatment) that has no net benefit or causes harm.2 The term is often Missed opportunities from physiotherapy associations interchangeably used with overlapping concepts of overdiagnosis, overtreatment and low-value care.2 There is increasing awareness of Choosing Wisely has made a great start towards increasing overuse in physiotherapy and the need for strategies to help phys- awareness of overuse of musculoskeletal healthcare, with over 150 iotherapists provide evidence-based care.2 This Editorial explains recommendations targeting inappropriate care for common muscu- why physiotherapy associations need to join the Choosing Wisely loskeletal conditions (eg, low back pain, knee osteoarthritis, shoulder campaign, outlines the missed opportunities from physiotherapy as- pain, rheumatoid arthritis).7 However, as we pass the 8-year mark sociations around the world and proposes solutions to maximise the since the campaign began, it appears that physiotherapy associations impact of Choosing Wisely on future physiotherapy practice. have missed several key opportunities to leverage Choosing Wisely to improve physiotherapy practice. There are over 120 physiotherapy Why physiotherapy associations need to join the Choosing Wisely associations worldwide: only four have joined the campaign and campaign published their own ‘do not do’ list of tests and treatments relevant to physiotherapy practice. More involvement is needed if physiotherapy Choosing Wisely began as an initiative of the American Board of is to be viewed as a profession taking the fight against overuse Internal Medicine (ABIM) Foundation in April 2012,1 inspired by seriously. earlier initiatives such as the Blue Cross Blue Shield Medical Necessity Project of the mid-1970s and ‘Medicine’s Ethical Responsibility for Physiotherapy associations currently involved with Choosing Health Care Reform – The Top Five List’. Today, Choosing Wisely is Wisely have also missed opportunities to increase the impact of the endorsed by over 250 professional associations across 20 countries, campaign because of the limitations in their lists. Choosing Wisely including the Australian Physiotherapy Association, and has inspired lists from physiotherapy associations are developed using a panel of similar initiatives to reduce overuse in physiotherapy such as the expert clinicians and researchers from various fields, and input from Journal of Orthopaedic Sports Physical Therapy’s ‘Overcoming Overuse’ society members. These lists are meant to target tests or treatments series.2 that represent overuse and are frequently provided by physiothera- pists. Inspection of current Choosing Wisely recommendations from In recent years, increased awareness of overuse in medicine has physiotherapy associations – presented in Table 1 – suggests that shifted guideline recommendations for musculoskeletal conditions many recommendations are not precisely targeted. Some recom- from medication and surgery to non-pharmacological interventions mendations target tests and treatments that are more within the such as exercise. This shift has given physiotherapists an opportunity scope of practice of other health professionals and would not affect to become key providers of evidence-based care for musculoskeletal the scope of practice or income of physiotherapists, for example: conditions and help combat overuse. However, ensuring that patients imaging for non-specific low back pain, cervical spine trauma and do not receive unnecessary surgery or harmful medications is only acute ankle sprains (ie, three out of six recommendations from the half the battle against overuse. To adequately combat overuse, Australian Physiotherapy Association), whirlpools for wound man- physiotherapists must also pay close attention to the care they pro- agement, and bed rest following the diagnosis of acute deep vein vide. A 2019 systematic review of 94 studies across 19 countries thrombosis after initiation of anticoagulation therapy. Other recom- found that two in three physiotherapists provided recommended care mendations target treatments rarely provided by physiotherapists for common musculoskeletal conditions such as back pain, knee (eg, only 14% use electrotherapy for acute ankle sprains, yet this is osteoarthritis and ankle sprains, while one in four provided care that targeted by the Italian Association of Physiotherapists) or where there guidelines recommended against.3 If nothing is done to ensure that are no data on use (eg, prescribing under-dosed strength training physiotherapists choose wisely, recent shifts in guideline recom- programs for older adults) (Table 1). Physiotherapy associations are mendations towards non-pharmacological care might reduce overuse not the only profession guilty of poorly targeted recommendations, in one area (eg, surgery) but create overuse in another (eg, electro- whether this be intentionally or unintentionally. A content analysis of physical agents for low back pain). 1,293 Choosing Wisely recommendations from eight countries found that professional associations often point to practices of other pro- Choosing Wisely appears to be reducing overuse in several areas fessions rather than their own members, particularly when the of medicine, with decreased use of autoimmune and thyroid tests,4 inappropriate practice generates income for its members.15 For blood and plasma transfusions5 and computerised tomography (CT) example, nine of the 48 recommendations from eight associations of imaging for suspected renal colic.6 Given the problem of overuse in https://doi.org/10.1016/j.jphys.2021.06.006 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/).
Table 1 Characteristics of Choosing Wisely recommendations from physiotherapy associations and physiotherapists’ use of featured t Association Recommendationa Test or treatment Target American Physical Don’t use continuous passive Treatment Therapy Association motion machines for the postoperative management of patients following uncomplicated total knee replacement Don’t use whirlpools for wound Treatment management Don’t use (superficial or deep) Treatment heat to obtain clinically important long-term outcomes in musculoskeletal conditions Don’t prescribe under-dosed Treatment strength training programs for older adults. Instead, match the frequency, intensity and duration of exercise to the individual’s abilities and goals Don’t recommend bed rest Treatment following the diagnosis of acute deep vein thrombosis after the initiation of anticoagulation therapy, unless significant medical concerns are present Australian Physiotherapy Avoid using electrotherapy Treatment Association modalities in the management of patients with low back pain Don’t routinely use incentive Treatment spirometry after upper abdominal and cardiac surgery Don’t provide ongoing manual Treatment therapy for patients with Test adhesive capsulitis of the Test shoulder Test Don’t request imaging for patients with non-specific low back pain and no indicators of a serious cause for low back pain Don’t request imaging of the cervical spine in trauma patients, unless indicated by a validated decision rule Don’t request imaging for acute ankle trauma unless indicated by the Ottawa Ankle Rules (OAR) (localised bone tenderness or inability to weight-bear, as defined in the Rules)
Search
