Journal of Physiotherapy 68 (2022) 75 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: Physical Frailty: ICFSR International Clinical Practice Guidelines for Identification and Management Date of latest update: 2 August 2019. Date of next update: 2024. Patient A comprehensive care plan for frailty should address polypharmacy, the group: Older adult people with (or at risk of) physical frailty. Intended management of sarcopenia, the treatable causes of weight loss, and the audience: All health professionals who care for older people with frailty. causes of exhaustion (strong recommendation). All persons with frailty Secondary audience: National, state and local policymakers. Additional should receive social support as needed to address unmet needs and versions: None. Expert working group: Members (multidisciplinary) of encourage adherence to a comprehensive care plan (strong recommen- the International Conference of Frailty and Sarcopenia Research. Funded dation). First-line therapy for the management of frailty should include a by: Not stated. Consultation with: External reviewer groups (healthcare multi-component physical activity program with a resistance-based providers and healthcare consumer groups). Approved by: Not stated. training component (strong recommendation). Protein/caloric supple- Location: Journal article: https://link.springer.com/article/10.1007%2Fs126 mentation is recommended when weight loss or undernutrition are 03-019-1273-z present (conditional recommendation). No recommendation was given for systematic additional therapies such as cognitive therapy, problem- Description: The task force of the International Conference of Frailty and solving therapy, vitamin D supplementation, and hormone-based treat- Sarcopenia Research (ICFSR) developed clinical practice recommenda- ment. Pharmacological treatment as presently available is not recom- tions for the identification and management of frailty in older adults. The mended therapy for the treatment of frailty. task force recommends that health practitioners case identify/screen all older adults for frailty using a validated instrument suitable for the Provenance: Invited. Not peer reviewed. specific setting or context (strong recommendation). Ideally, the Rik Gosselink screening instrument should exclude disability as part of the screening process. For individuals screened as positive for frailty, a more compre- University of Leuven, Belgium hensive clinical assessment should be performed to identify signs and underlying mechanisms of frailty (strong recommendation). https://doi.org/10.1016/j.jphys.2021.11.004 Appraisal of Clinical Practice Guideline: International clinical practice recommendations on the definition, diagnosis, assessment, intervention and psychosocial aspects of developmental coordination disorder Date of latest update: January 2019. Date of next update: December environment (n = 2) or drugs (n = 1) and monitoring (n = 2); and 2022. Patient group: Individuals with developmental coordination dis- Management of adolescents and adults, encompassing terminology/ order (DCD) across their lifespan. Intended audience: Healthcare pro- diagnosis and assessment (n = 2). Recommendations suggest that fessionals. Additional versions: Initial 2012 version https://doi.org/ physiotherapists are important in: inter-professional, multi-dimensional 10.1111/j.1469-8749.2011.04171.x, 2019 Pocket Version: https://doi.org/ assessment and diagnosis using subjective and standardised objective 10.1111/dmcn.14132 (Appendix S1). Expert working group: European tools; providing goal-oriented, strengths-based, outcomes-measured Academy of Childhood Disability (EACD) members, nominating for an intervention tailored to an individual’s functional environment and expert panel (26 members, 14 countries) and/or five writing groups (11 psychosocial status; identifying and referring for co-occurring and/or members, 12 countries). Funded by: Nil. Consultation with: Parent psychosocial issues; and caring for individuals throughout childhood organisation for individuals with learning disorders. Approved by: EACD. and adulthood. Future research highlighted for children includes: val- Location: https://doi.org/10.1111/dmcn.14132. idity and reliability of clinical reference standards, comparisons be- tween interventions, and effectiveness of interventions performed by Description: Recommendations were sought across five topics schools/parents; and for adolescents/adults includes: extended longi- (mechanisms, assessment, intervention, psychosocial issues and ado- tudinal studies of the developmental course of DCD, age-appropriate lescents/adults) using expert panel meetings at two international assessments for diagnosis and intervention evaluation, and age- conferences (DCD: 2015; EACD: 2016), a Delphi process and systematic appropriate interventions. literature reviews by five writing groups. The Grading of Recommen- dations Assessment, Development and Evaluation (GRADE) system was Provenance: Invited. Not peer reviewed. used to determine recommendations. Thirty-five recommendations were proposed across four areas: Definition/terminology (n = 1); Leanne M Johnston Assessment, encompassing diagnostic criteria (n = 4), assessment of The University of Queensland, Australia DCD features (n = 8, including motor proficiency n = 3) and comorbid- ities (n = 1); Intervention, encompassing principles (n = 7), https://doi.org/10.1016/j.jphys.2021.12.005 therapeutic approaches (n = 5), delivery mode (n = 2), impact of 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 68 (2022) 61–68 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 Clinical placements in private practice for physiotherapy students are perceived as safe and beneficial for students, private practices and universities: a national mixed-methods study Casey L Peiris a, Alan Reubenson b, Ruth Dunwoodie c, Vidya Lawton d, Alison Francis-Cracknell e, Cherie Wells f a School of Allied Health, Human Services and Sport, La Trobe University, Melbourne, Australia; b School of Allied Health, Curtin University, Perth, Australia; c School of Health and Rehabilitation Sciences, The University of Queensland, Brisbane, Australia; d Faculty of Medicine, Health and Human Sciences, Macquarie University, Sydney, Australia; e Faculty of Medicine, Nursing and Heath Sciences, Monash University, Melbourne, Australia; f School of Allied Health, Exercise and Sports Science, Charles Sturt University, Port Macquarie, Australia KEY WORDS ABSTRACT Physical therapy Question: What are the extent and characteristics of clinical placements in private practice for physiotherapy Clinical education students? What do university clinical education managers perceive to be the benefits, risks, barriers and Private practice enablers of clinical placements in private practice for physiotherapy students? What training and support are Mixed methods available for private practitioners? Design: Mixed methods study combining a national survey and in-depth, semi-structured focus group interviews. Participants: Twenty clinical education managers from Australian universities who had graduating students in entry-level physiotherapy programs in 2017 (95% response rate) responded to the survey with data on 2,000 students. Twelve clinical education managers participated in the focus groups. Results: It was found that 44% of physiotherapy graduates in Australia in 2017 completed a 5- week private practice placement. Private practice placement experiences were perceived to be safe and beneficial for students, private practices and universities. The main risks identified by clinical education managers were related to the quality and consistency of the student’s experience on placement and not risks to service or clients. The main perceived barriers were time costs (both practitioner and university clinical education managers) and perceived lost earning capacity. Clinical education managers emphasised that more time and resources to establish and support private practitioners would enable them to reduce risk and overcome barriers to increasing private practice placement capacity and quality. Engaging private practi- tioners and working collaboratively appear vital for establishing, monitoring and supporting private practice placements. Conclusion: By working collaboratively, universities and private practice physiotherapists can enhance private practice placement capacity and quality. [Peiris CL, Reubenson A, Dunwoodie R, Lawton V, Francis-Cracknell A, Wells C (2022) Clinical placements in private practice for physiotherapy students are perceived as safe and beneficial for students, private practices and universities: a national mixed- methods study. Journal of Physiotherapy 68:61–68] Crown Copyright © 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/). Introduction physiotherapy students across Australia had undertaken a placement in private practice.7 This contrasts significantly with the proportion of Clinical education in physiotherapy programs plays a critical role in ensuring that students develop and demonstrate clinical compe- physiotherapists working in private practice, which is reported to be tence upon graduation.1–4 Clinical education typically involves clin- between 43 and 60%.8,9 In addition, new graduates are increasingly ical placements where students perform physiotherapy duties under the supervision of a qualified physiotherapist (clinical educator).1,4 being employed in the private sector, which includes both private Clinical placements in Australia occur in a range of clinical settings practice and private hospitals.10,11 and areas so that students develop the competencies required to practise.5 A key aim of clinical placements is to prepare students for Recent research indicates that new graduate physiotherapists in future work practice; therefore, clinical placement settings and Australia may be underprepared for private practice employment.12,13 learning opportunities should be guided by future workforce needs.6 This has resulted in pressure from employers, graduates and profes- Traditionally, most physiotherapy student placements are completed in hospitals. In 2012, it was estimated that only 5 to 10% of sional bodies for universities to provide students with private practice placement experiences to increase preparedness.12–14 Placement provision in health professional courses is a challenge internationally15–17 and the private sector appears to be underutilised in supporting student experiential learning.18 Previous placement https://doi.org/10.1016/j.jphys.2021.12.007 1836-9553/Crown Copyright © 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/).
62 Peiris et al: Private practice placements capacity building strategies have been proposed mostly in hospital from all states/territories with physiotherapy programs, as well as a settings. One popular approach is the student-led placement model, mix of metropolitan and rural universities. Focus group interviews which Nicole and colleagues19 demonstrated increased student were conducted until data saturation was achieved (defined by the placement capacity by 213% in a previously underserviced aged-care acquisition of no new themes or data being identified with further setting. However, the student-led model may not be feasible in an interviews).24 Australian private practice context due to restrictions on billing practices for student-provided services.20 In another model, univer- Data collection sity academics provided onsite support to effectively increase place- Data were collected through an online, self-administered survey ment capacity when six or more students were at one site;17 this may not be feasible in private practice, where space has been reported as a using commercial softwarea. The survey question used a combination barrier to hosting multiple students.21,22 Therefore, strategies to in- of Likert scale, multiple-choice and open-ended questions. The survey crease the capacity of private practices to provide student placements was pilot tested by two CEMs and took approximately 20 minutes to are largely unknown. complete. These CEMs provided feedback to the researchers on readability, structure and appropriateness of survey questions, and University clinical education managers (CEMs) have a wealth of amendments were made prior to distribution. knowledge related to student placements in the context of the Australian and New Zealand Physiotherapy Practice Thresholds,23 The final survey sourced information on: student and placement which describe the threshold competence required for initial and numbers; placement setting and length; placement models; training continuing registration as a physiotherapist in both countries. and support for private practitioners; and benefits, risks, barriers and Although the specific duties may vary between universities, CEMs enablers of private practice placements. In relation to placement provide support for students and clinical educators to ensure that models, student to clinical educator ratios and three broad categories of students’ clinical education requirements are met. A CEM’s role may placement models were considered: students providing consultation include sourcing and allocating placements, providing training and under supervision (student-led); student sharing consultations with support for clinical educators, supporting students, analysing and physiotherapist (shared-care); or student observing consultations pro- responding to feedback, and initiating and maintaining partnership vided by physiotherapist (observational). For the purpose of this relationships. CEMs are therefore well positioned to provide valuable research, adverse events were defined as events reported to the uni- insight into private practice placements in physiotherapy and potential versity that occurred on placement resulting in harm to the student, benefits, risks, barriers and enablers to increasing placement capacity. patient or educator (eg, patient fall, patient complaint or student injury). One aim of this study was to establish the current status of private Semi-structured focus group interviews were conducted via practice placements in Australia (ie, placement numbers). Another teleconference calls to enable CEMs from diverse geographical loca- aim was to explore the perspectives and experiences of university tions to participate. Focus groups were chosen to explore CEM ex- CEMs in Australia regarding physiotherapy private practice placement periences, as they enable participants to explore and clarify their models, benefits/risks of providing clinical placements within private views and can help generate new ideas or remind participants of practice, training/support opportunities available to private practi- similar or different experiences.25 All focus groups were facilitated by tioners, and barriers and enablers that impact the sourcing and the same researcher (CW) who followed a flexible interview schedule maintaining of private practice placements. (Table 1). Interviews were audio recorded and transcribed verbatim. Therefore, the specific research questions for this mixed-methods Data analysis study were: For survey response rates, demographics and closed survey 1. What are the extent and characteristics of clinical placements in questions, descriptive statistics were calculated using commercial private practice for physiotherapy students? statistical softwareb. For focus group interviews, data collection and data analysis occurred simultaneously to check for data saturation.24 2. What do university clinical education managers perceive to be the Thematic analysis26 was undertaken using commercial softwarec to benefits, risks, barriers and enablers of clinical placements in pri- manage qualitative data. Two researchers (CW, CP) independently vate practice for physiotherapy students? read transcripts for open and selective coding of the data, using an approach derived from grounded theory.25 Researchers individually 3. What training and support are available for private practitioners? examined data and assigned codes (eg, risks, benefits, enablers) to portions of text and then looked for interactions and links between Method codes. Researchers CW and CP met to discuss and resolve differences. Next, they collaborated to decide on main themes and then selec- Design tively searched for further data on those themes (selective coding). Themes were sent to participants for member checking to ensure that This mixed methods study combined a national survey and in- the researchers’ interpretations were accurate.25 After feedback, re- depth, semi-structured focus group interviews. This design was searchers met to decide on final themes and relationships between chosen to enable data triangulation, as the survey provided raw data themes. Survey and focus group data were analysed separately, and on private practice placements, whilst the interviews explored the the identified themes and results were compared and contrasted experiences of CEMs in relation to these placements. Each university’s (triangulated) for overall synthesis.27 human research ethics committee approval was obtained from leading and partner institutions of all authors prior to commence- As researchers bring their own experiences into qualitative ment of the study. research and data interpretation, the researchers’ backgrounds are briefly described here to enhance reflexivity. All researchers have Participants qualifications in physiotherapy and were either current or former university CEMs. Four researchers had previously worked in private CEMs from all Australian universities who had graduating students practice (CP, VL, AR, CW) and two had a PhD qualification (CP, CW). in entry-level physiotherapy programs (Bachelor, Masters or Doctoral) in 2017 (n = 21) were identified from the Clinical Educator Managers Results Australia and New Zealand (CEMANZ) committee contact list. CEMs were contacted via email, with information and a hyperlink to Participants participate in an online survey. Participants provided informed consent to participate within the survey. Purposive sampling was used to re- Survey cruit participants for focus group interviews to ensure representation CEMs from all 21 of the eligible physiotherapy courses in Australia were invited to participate. Of these, 20 out of 21 physiotherapy
Research 63 Table 1 Theme 1: Private practice placements are increasing, but more are Flexible interview schedule. needed Aim Sample questions and prompts There’s a huge demand (P12) Explore private practice Can you describe the private practice clinical Placement data for the 2017 graduating cohort were provided by clinical placement experiences of the students at your university? 18 universities (n = 1,471 students). All universities sourced private experiences across the Prompts: practice placements, which predominantly occurred in the final year country What type of placement (core/elective/ of study. It was found that 44% of students (n = 643 of 1,471) completed a 5-week private practice placement (Table 3). Consid- introductory) and when do students ering that students complete more than one placement each, 9% of experience it? all clinical placements undertaken were in private practice Is it same/different to other settings? (Figure 1). (hours, expectations) Qualitative data confirms that CEMs have increased their efforts to Explore the perceived What do you think are the benefits of private source clinical placements in private practice, but a mismatch re- benefits of private practice practice placements for physiotherapy mains between demand and supply of placements in this setting. placements students? Prompts: We don’t have the numbers of placement offers to allocate to every Benefits for student, clinical educator, student (P12) practice, university Theme 2: Training and support are offered, but engagement is low Explore any risks associated Can you share any risks associated with We rarely get private practitioners coming (to training) (P11) with students undertaking students undertaking placements in private placements in private practice? All universities provided training on how to assess students using the practice Prompts: Assessment of Physiotherapy Practice (APP) tool, as well as individual Risks for client, student, practice phone, email and/or face-to-face support. Most universities also prov- How these compare to other settings ide other training opportunities (Table 5). However, CEMs conceded that private practitioner attendance is low (online or face-to-face). Explore barriers to sourcing What are the major barriers of sourcing and They can’t free up the time to access the training and that’s a bit of a and securing private practice securing private practice placements for barrier (P9) placements physiotherapy students? CEMs reported that lack of engagement in training was a barrier to increasing capacity and adversely affected quality and consistency. Prompts: There are wide variations in the quality of education provided and Differences to other settings, practice the assessments of students. This can be due to lack of engagement with unis (P5) needs, resources Where educators were engaged with universities and there was Explore enablers to sourcing What are the major factors that can assist you regular, two-way communication and collaboration, quality was enhanced. and securing private practice in sourcing and securing private We have a host of about ten private practices that are really great placements practice placements for physiotherapy students? supporters of our program and that we know run a good model (P2) Prompts: Theme 3: A combination of placement models is used Differences to other settings, practice I don’t think there is a model that works for everyone (P10) needs, resources Explore private practice Describe the private practice clinical placement clinical placement models models that private practices provide to your students. Prompts: Consultations: number and type of clients, shared/separate, fees Clinical and non-clinical skills Supervisions: ratios, direct/indirect Explore training and support Describe the training and support that offered to clinical universities currently provide. educators/private practices Prompts: Online versus face-to-face; content and duration of training; contact with clinical education coordinator/manager; payment Strengths and limitations of current training and support provided and how can this be improved? Check if anything has Do you have any final comments regarding been missed private practice experiences for physiotherapy students in entry-level programs within Australia and suggestions for improvement? Prompts: Priorities, next steps, stakeholder roles programs participated, giving a 95% response rate to the survey. CEMs reported that private practices often opted to use a combi- Table 2 describes the characteristics of the survey participants. These nation of models (student-led, shared-care and observational) respondents from 20 universities were able to provide data on 2,000 depending on student stage of learning, as well as patient and busi- students (Table 3), including detailed data on the 2017 cohort of ness needs. In the combination model, CEMs reported that students student placements. The survey also obtained data on the nature of typically commenced by observing their educator (estimated at 29% the clinical placements (Table 4) and the training and support given of the placement, SD 16), then progressed to shared-care consults by the universities to the private practitioners (Table 5). (34%, SD 14) and finally to student-led care (41%, SD 29). A variety of supervision ratios were employed by private practices, with no Focus groups particular preference noted (Table 4). Observation of educator con- Twelve CEMs (including representation from all states and terri- sults was perceived to be a valuable part of a student placement, but not appropriate for an entire placement if the student was assessed tories) participated in focus group interviews. Table 6 describes the with the APP. CEMs perceived there was no one placement model characteristics of the participants in the focus groups. Three focus that would suit all practices and reported working with practices to group interviews were conducted before data saturation was ach- individually tailor models. ieved, with no new themes emerging from the final focus group (n = 4 participants). In the context of meeting learning objectives. observation only is not something that we can usually entertain. There has to be some
64 Peiris et al: Private practice placements Table 2 Participants Table 3 Students Characteristics of survey participants. (n = 20) Characteristics of students at participating universities. (n = 2,000) Characteristic 18 (90) Characteristic 30 to 270 2 (10) Sex, n (%) Students (n per university), range 1,186 (59) female 6 (30) Enrolled degrees, n (%) 130 (7) male 2 (10) 466 (23) 6 (30) Bachelor with/without honours 108 (5) State, n (%) 4 (20) combined Bachelor and Masters 110 (6) New South Wales/Australian Capital Territory 2 (10) graduate entry Masters South Australia 22 (8) extended Masters (n = 1,471) Queensland 6 (5) Doctorate Victoria 17 to 173 Western Australia Detailed placement data from 2017 cohort of 18 universities 452 (31) Experience in physiotherapy (yr), mean (SD) Students (n per university), range 379 (26) Experience in clinical education management (yr), mean (SD) Enrolled degrees, n (%) 219 (15) 221 (15) kind of commitment to a shared kind of interaction and hands-on Bachelor 100 (7) opportunity. But we do have a range of options (P12) Bachelor with honours 100 (7) combined Bachelor and Masters 6 (4 to 8) Theme 4: Private practice placements are beneficial and low risk graduate entry Masters 999 (124) extended Masters The benefits are great; risk is very, very low (P3) Doctorate 8,792 5-week clinical placements per student (n), mean (range) 643 (44) Most survey participants agreed with several benefits of private Placement hours per student, mean (SD) practice placements (Figure 2), which were elaborated on con- Placements for this cohort (n), total vergently in focus groups. CEMs perceived that private practice Students who had a private practice placement, n (%) placements helped increase student employability for this sector through orientation to private practice, managing patient expecta- Some percentages do not sum to 100 due to the effects of rounding. tions and running a business. had insufficient time to support students, the quality of student It’s very beneficial for the students, in terms of their employability experience may also have been affected. (P2) Time is probably the biggest thing for a practice. [they] will often CEMs also perceived benefits for private practices (eg, screening end up just saying just watch me because I can’t afford to take the student suitability for future employment, extending patient services time away from my patient (P1) with student assistance, and enhancing practice credibility by su- pervising students and partnering with universities) and clinical ed- The time needed to supervise students was also perceived to ucators (eg, improved clinical reasoning, critical reflection, research potentially reduce a practitioner’s caseload (and therefore earning informed decision-making, and access to library resources and pro- capacity), especially when students were underperforming. fessional development). It often takes a bit more time from the practitioner [to supervise] and There’s another buy in for our practitioners, they’re looking to potentially a risk for the practice because of lost earning capacity diversify their experience and their skills. (P2) [especially] when that supervisor’s spending more time with the weaker student (P10) The risk of adverse events occurring during a private practice placement was perceived to be low, as students are supervised Perceived lost earning capacity for educators was also reported to during client consults and are academically competent before result from not being able to bill for some third-party funded services commencing placement. Quantitative data supported this assump- if a student was involved in patient care, charging less for student tion, with no major adverse events being reported. One minor consults or providing longer consults at the usual rate. However, it adverse event was reported by a CEM, where a patient experienced was proposed that cost savings would be made later in placements temporary discomfort due to incorrect use of electrotherapy by a student. Table 4 Characteristics of universities and their private practice placements. The main risk perceived by CEMs was inconsistent quality of private practice experiences. CEMs perceived that some students may Characteristic Universities not get enough ‘hands-on’ practice and that private practitioners (n = 20) were limited for time, impairing their ability to provide adequate supervision, facilitate clinical reasoning and provide feedback (which Degrees offered, n (%)a 10 (50) were identified as barriers by survey participants) (Figure 2). Bachelor 12 (60) Bachelor with honours 2 (10) It’s the quality side of things; you just don’t know what’s happening combined Bachelor and Masters 7 (35) out there as well as we do know in the big hospitals (P9) graduate entry Masters 2 (10) extended Masters Theme 5: Private practice placements cost time and money Doctorate 1 (5) It goes both ways (P9) Universities collaborating with specific types of private practice, n (%) 10 (50) majority single private practices 2 (10) Lack of time to supervise students was perceived to be a barrier majority group/multi-site private practices 8 (40) for private practices providing placements (Figure 2). If an educator mixture of single and group private practices 20 (100) Universities’ expected clientele at collaborating private practices, n (%)a 10 (50) majority musculoskeletal 7 (35) significant component of sports includes paediatrics, aged care and/or gender health 10 (50) 6 (30) Universities’ typical ratio used at collaborating private 4 (20) practices, n (%) 1 clinical educator : 1 student 2 clinical educators : 1 student 1 clinical educator : 2 to 5 students a More than one response could be selected.
Research 65 Table 5 Table 6 Universities offering training and support to private practitioners. Characteristics of the focus group participants. Type of training or support, n (%) Universities Characteristic Participants (n = 20) (n = 12) Assessment of Physiotherapy Practice (APP) tool Individual phone, email and/or face-to-face support 20 (100) Sex, n (%) 11 (92) Face-to face introductory clinical educator training 20 (100) female 1 (8) Effective supervision and support training 19 (95) male Professional development opportunities not related to clinical 18 (90) 4 (33) education 17 (85) State, n (%) 2 (17) Supporting students at risk of failing New South Wales/Australian Capital Territory 4 (33) Financial contributions 13 (65) South Australia 1 (8) University affiliation including library access 13 (65) Queensland 1 (8) 13 (65) Victoria 24 (10) Western Australia 8 (6) Experience in physiotherapy (yr), mean (SD) Experience in clinical education management (yr), mean (SD) (when students reached entry-level competency) and by attracting Some percentages do not sum to 100 due to the effects of rounding. graduates, thus reducing recruitment costs. CEMs felt that engagement and timely communication prevented CEMs felt that where it occurred, remuneration from universities this situation. provided some contribution to offset reported costs incurred by pri- vate practices. Other non-monetary benefits (eg, library resources, I hear at the end of the placement, oh we had all these troubles. like university training) were proposed by CEMs as welcomed value- why didn’t you call me? And they didn’t know who to call. Whereas adding incentives. other practices as soon as there’s a problem, they’re on the phone, it’s all sorted within a day (P9) The remuneration is generally pretty small and won’t necessarily recoup the cost of taking the student in the first place. So the benefits Discussion are much grander and much more sort of forward thinking than just saying oh well we get money for it (P11) This mixed methods study determined that 44% of final year physiotherapy students in Australia in 2017 completed a 5-week CEMs also noted significant time and costs incurred by univer- private practice placement. Private practice placement experiences sities in providing training and support for private practices to host were perceived to be safe and beneficial for students, private practices small numbers of students. and universities. The main risks identified by CEMs were related to the quality and consistency of student experience on placement and not Establishing, building, maintaining relationships and quality assur- risks to clients. CEMs perceived that the main barrier to hosting stu- ance just takes a whole lot longer because you’ve just got more dents for private practitioners was time costs incurred by the private partners. And so you have to do the same thing multiple times. practice and/or clinical educator. CEMs emphasised that more time Whereas, you know, major public hospitals where you’ve got one and resources to establish and support private practitioners would phone call or one email and you deal with everything in one go for a enable them to enhance engagement and increase private practice hundred students (P10) placement capacity. A variety of successful placement models were identified as currently being used in private practice. The adopted Theme 6: Enhancing engagement to increase capacity and quality models were influenced by contextual factors relating to the practice, stage of learner and university requirements, highlighting the need for It is built on coffee and trust (P2) flexibility to optimise and increase placement capacity. Although CEMs reported that universities offer a variety of supports to private Positive relationships and engagement were commonly reported practices, it was conceded that training was not well attended by by CEMs as being essential to developing quality private practice private practitioners, and this may impact ongoing supply and quality experiences and were identified as integral to establishing, moni- of placements. Working collaboratively to better understand the needs toring and supporting private practice placements. of both the universities and private practices related to training and support appears vital for enhancing engagement and establishing It’s our relationships that build these clinical placements and I often quality, sustainable experiences for students. say, like clinical education is kind of built on coffee and trust (P2) Universities and private practices have worked together to in- CEMs suggested that an increase in placement capacity may be crease the number of private practice placement experiences na- assisted by having additional time and resources to recruit, train and tionally more than fourfold since 2012.7 CEMs identified that private support private practice providers. CEMs reported that when prac- practice placements were safe learning environments that enhance tices are positively engaged with the university by regularly taking student experience and new graduate readiness, which has also been students, participating in training and communicating with univer- sity staff, placement quality is enhanced. Clinical placement settingPublic sector (Some) placements are stand out excellent you know. The investment Private sector that the team in private practices are prepared to put into clinical education really does stand out (P3) Private practice There were also perceptions that private practices who weren’t University clinics engaged were less equipped to manage poor-performing students, 0 25 50 75 100 and negative experiences may result in fewer placements being Percentage of students provided. Figure 1. Percentage of physiotherapy students attending clinical placements in spe- They can’t cope with anyone deemed to be difficult. they find it cific sectors of the profession. really hard to know what to do and they will go ‘oh we’re not having another one’ (P4)
66 Peiris et al: Private practice placements Percentage of participants 0% 20% 40% 60% 80% 100% Potential benefits of private practice placements Students bring up-to-date and research-informed knowledge/skills Clinical educators develop additional knowledge/skills in supervision/teaching Private practitioners/clinical educators can gauge student suitability for employment Graduates are orientated to private practice and may be more work ready Potential barriers to private practice owners providing clinical placements Limited caseload available for students Some third-party funders will not reimburse clients for student consultations Financial costs associated with supervision Extra time and effort associated with supervision Inadequate financial contribution from university Insufficient clinic space or equipment Inadequate clinical educator training and support Barriers to clinical education coordinators sourcing private practice placements Inadequate time/resources to recruit/train/support multiple private practice providers Private practices provide few clinical placements and require more work to set up Quality student experiences or supervision may not be available in private practice Students may not get their own caseload and freedom to use any hands-on skills Factors that may enable or assist private practice owners in providing clinical placements Clinical educators do not lose income by supervising students Clinical educators receive training and support from the university Flexible placement scheduling options are available University contributes financially towards costs of supervision Factors that may enable clinical education coordinators to source private practice placements Additional time/resources to recruit/train/support multiple private practice providers Committed and engaged private practice owners Strongly disagree Disagree Neutral Agree Strongly agree Figure 2. Percentage of participants reporting each category of agreement with statements in the survey. recognised by private practitioners.12,22 However, benefits for of Australian universities found that the cost of running university placement providers are traditionally more difficult to define, health courses with clinical placements exceeded the funding especially considering that service provision benefits are limited by received for these courses.31 This suggests that financial incentives funding arrangements in private practice. CEMs in this research alone may not be a sustainable option and that other non-financial identified a broad range of benefits that may be under-recognised, incentives should be explored to add value to practices so that such as future workforce recruitment and selection, continuing hosting students is more attractive. professional development opportunities, clinical educator training and university affiliation (eg, access to libraries, on-line resources, The results indicate that financial and non-financial contributions research mentors and academic expertise; potential adjunct status). by universities along with goodwill and engagement are necessary Some of these benefits have recently been recognised by private for private practice placement viability. Goodwill, together with an practitioners21 and have been identified as being vital in attracting effective model and culture of education have been long present in the provision of placements.28,29 Many parties benefit from private the public sector, reflected in the fact that 81% of all health place- practice placements for physiotherapy students and the positives for ments are in the public sector.29 For physiotherapists in public placement providers need to be optimised and emphasised to health, clinical education is a role expectation.30 To provide place- enhance placement capacity. ment opportunities, private practitioners may require more moti- vation than public health physiotherapists, in order to balance the Despite the benefits, perceived costs of clinical education may be time needed to educate students with business priorities.18 CEMs an ongoing issue in private practice. CEMs indicated that costs re- report working with a number of excellent private practices who ported by private practices were predominantly time costs and po- provide effective clinical education experiences, but this research tential lost productivity (particularly for underperforming students), did not identify any one model of education that was deemed most which is supported by Maloney et al.30 Potential for lost income may effective in private practice. This is not unexpected, as there are be a barrier to hosting students in private practice; however, it is many considerations and unique attributes of individual private unclear which model of placement provision would best address this practices. However, underpinning the success of any model is uni- barrier, and this may vary in different practice contexts. Anecdotally versities and private practitioners working collaboratively through there is a perception that funds received by universities should be engagement and communication, including appropriate training and passed on to placement providers to cover costs. However, an analysis support, leading to a perceived enhanced quality experience for all
Research 67 parties. Indeed, private practitioners reported that support from University. All participants gave written informed consent before data universities is key to resolving challenges when hosting students, and those who reported close and supportive relationships with collection began. universities perceived that challenges were easily overcome.32 Collaboration and engagement are essential for developing place- Competing interests: The authors have no conflicts of interest to ment models that support learning objectives for students while declare. maintaining business objectives. Source(s) of support: Nil. Clinical educator preparation is key to successful clinical place- ments and clinical expertise alone does not necessarily translate into Acknowledgements: We acknowledge the La Trobe University clinical educator expertise.16 A need to better prepare private prac- titioners to be clinical educators has previously been identified;14 Social Research Assistance Platform for providing research support however, despite training being available, CEMs reported low attendance by private practitioners. Similarly, Maloney et al30 re- for interview transcription. We would also like to acknowledge ported that hospital clinical educators were more likely to want to attend training than clinical educators in private practice. Engage- Cambridge McCormick and Alexis Nicholson for their assistance in ment in training and support offered by universities can strengthen relationships and communication, support the management of organising focus groups and transcribing interviews. under-performing students, enhance clinical educator access to other non-financial benefits and university resources, provide clinical ed- Provenance: Not invited. Peer reviewed. ucators with professional development, and improve placement consistency and quality. Therefore, it is important that universities Correspondence: Casey L Peiris, Discipline of Physiotherapy, and private practices collaborate to provide training that is accessible and supports participation without detracting from other business School of Allied Health, Human Services and Sport, La Trobe Univer- priorities. National collaboration among universities when devel- oping training resources would distribute the load, reduce cost and sity, Melbourne, Australia. Email: [email protected] improve consistency of support provided. References This research was focused on Australian entry-level physiotherapy student placements, so the results may not be representative of 1. Baldry Currens J. The 2:1 clinical placement model. Physiotherapy. 2003;78:540–554. student placement experiences internationally. The research only 2. Stiller K, Lynch E, Phillips AC, Lambert P. Clinical education of physiotherapy students presents the views of university CEMs, and further research is required to balance this research with views of private practitioners. in Australia: Perceptions of current models. Aust J Physiother. 2004;50:243–247. However, this research did include CEMs from almost all physio- 3. Dalton M, Davidson M, Keating J. The Assessment of Physiotherapy Practice (APP) is therapy programs in Australia, thus providing rich and robust data. Future research should also explore successful models of private a valid measure of professional competency of physiotherapy students: A cross practice placements to help guide CEMs and private practitioners to sectional study with Rasch analysis. J Physiother. 2011;57:239–246. set up new placement experiences. 4. Patton N, Higgs J, Smith M. Clinical learning spaces: Crucibles for practice devel- opment in physiotherapy clinical education. Physiother Theory Pract. 2018;34:589– In conclusion, this research found that more Australian physio- 599. therapy students are being exposed to private practice than previously 5. Physiotherapy Board of Australia. Physiotherapy Board of New Zealand. Physio- reported. Private practice placements were perceived to be safe and therapy practice thresholds in Australia and Aotearoa New Zealand. Australian beneficial, but had time and financial implications for both practices Physiotherapy Council; 2015. https://physiocouncil.com.au/media/1020/physio and universities. Modifiable risks and barriers related to the variable therapy-board-physiotherapy-practice-thresholds-in-australia-and-aotearoa-new- quality of educational experiences were identified. Universities and zealand-6.pdf. Accessed July 22, 2018. private practice physiotherapists have a shared responsibility to shape 6. Bacopanos E, Edgar S. Employment patterns of Notre Dame graduate physiothera- the future workforce, and by working collaboratively they can enhance pists 2006-12: targeting areas of workforce need. Aust Health Rev. 2016;40:188–193. private practice placement capacity and quality. 7. Health Workforce Australia. Clinical training profile: Physiotherapy. Adelaide: Health Workforce Australia; 2014. https://www.google.com/url?sa=t&rct= What was already known on this topic: Clinical placements j&q=&esrc=s&source=web&cd=&ved=2ahUKEwiR0eHFueXsAhX8wjgGHTQSD5kQF prepare physiotherapy students for future work practice, so jAAegQIAxAC&url=https%3A%2F%2Facdhs.edu.au%2Fwp-content%2Fuploads%2F20 clinical placement settings and learning opportunities should 18%2F05%2FClinical-Training-Profiles-Project-Physiotherapy.pdf&usg=AOvVaw1v match future work contexts. Despite this, most physiotherapy EZSfADvXdUuocDhRWrXY. Accessed November 3, 2020. student placements are completed in public hospitals. 8. National Health Workforce. Data Set (NHWDS) Factsheet: Physiotherapists; 2017. What this study adds: More Australian physiotherapy stu- https://hwd.health.gov.au/publications.html#alliedh17. Accessed 3 November, 2020. dents are undertaking clinical placements in private practice than 9. Pretorius A, Karunaratne N, Fehring S. Australian physiotherapy workforce at a previously reported. Private practice placements were perceived glance: a narrative review. Aust Health Rev. 2016;40:438–442. to be safe and beneficial but had time and financial implications 10. Health Workforce Australia. Australia’s health workforce series – Physiotherapists in for both practices and universities. The main risks perceived focus. Adelaide: Health Workforce Australia; 2014. https://www.google.com/url? related to the variable quality of educational experiences. sa=t&rct=j&q=&esrc=s&source=web&cd=&cad=rja&uact=8&ved=2ahUKEwj9tdeHveX Engaging private practitioners and working collaboratively foster sAhXXxzgGHWY6CXMQFjAAegQIAxAC&url=http%3A%2F%2Fiaha.com.au%2Fwp-cont private practice placements. ent%2Fuploads%2F2014%2F03%2FHWA_Australia-Health-Workforce-Series_Physioth erapists-in-focus_vF_LR.pdf&usg=AOvVaw2TIpvHkgK3-8gkpIzo0jts. Accessed July Footnotes: a SurveyMonkey software, SurveyMonkey, Palo Alto, 22, 2018. USA. 11. Mulcahy A, Jones S, Strauss G, Cooper I. The impact of recent physical therapy graduates in the workforce: A study of Curtin University entry-level physiothera- b SPSS Windows Version 24, SPSS, Chicago, USA. pists 2000-2004. Aust Health Rev. 2010;34:252–259. c NVivo 12 software, QSR International Pty Ltd. 12. Atkinson R, McElroy T. Preparedness for physical therapy in private practice: Ethics approval: The study was approved by ethics committees at Novices identify key factors in an interpretive description study. Man Ther. The University of Queensland, La Trobe University, Curtin University, 2016;22:116–121. Macquarie University, Monash University and Charles Sturt 13. Davies JM, Edgar S, Debenham JA. A qualitative exploration of the factors influ- encing the job satisfaction and career development of physiotherapists in private practice. Man Ther. 2016;25:56–61. 14. Kent FM, Richards KL, Haines TP, Morgan PE, Maloney SR, Keating JL. Patient and practitioner perceptions of student participation in private practice consultations: A mixed-methods study. FoHPE. 2015;16:42–54. 15. Smith PM, Corso LM, Cobb N. The perennial struggle to find clinical placement opportunities: A Canadian national survey. Nurs Educ Today. 2010;30:798–803. 16. Rodger S, Webb G, Devitt L, Gilbert J, Wrightson P, McMeeken J. A clinical education and practice placements in the allied health professions: an international perspective. J Allied Health. 2008;37:53–62. 17. Fairbrother M, Nicole M, Blackford J, Nagarajan SV, McAllister L. A new model of clinical education to increase student placement availability: the capacity devel- opment facilitator model. Asia-Pac J Coop Educ. 2016;17:45–59. 18. Doubt L, Paterson M, O’Riordan A. Clinical education in private practice: An interdisciplinary project. J Allied Health. 2004;33:47–50. 19. Nicole M, Fairbrother M, Nagarajan SV, Blackford J, Sheepway L, Penman M, et al. Student-led services in a hospital aged care temporary stay unit: Sustaining stu- dent placement capacity and physiotherapy service provisions. Asia-Pac J Coop Educ. 2015;16:327–342. 20. Australian Physiotherapy Association. FAQs: physiotherapy student placements in private practice; 2018. https://australian.physio/student-placements-faqs. Accessed June 18, 2020. 21. Forbes R, Dinsdale A, Dunwoodie R, Birch S, Brauer S. Weighing up the benefits and challenges of hosting physiotherapy student placements in private practice; a qualitative exploration. Physiother Theory Pract. 2020;27:1–11. 22. Wells C, Olson R, Kent F, Bialocerkowski A, Chipchase L, Scarvell J, et al. Are new graduate physiotherapists ready for private practice? Paper presented at the APA Momentum 2017 Physiotherapy Conference, Sydney. 2017.
68 Peiris et al: Private practice placements 23. Physiotherapy Board of Australia & Physiotherapy Board of New Zealand 2015. service organisation: An Evidence Check rapid review brokered by the Sax Institute Physiotherapy practice thresholds in Australia and Aotearoa New Zealand. https:// for the Hunter and Coast Interdisciplinary Training Network through the Health www.physiotherapyboard.gov.au. September 16, 2020. Education Training Institute (HETI); December 2014. http://www.heti.nsw.gov.au. Accessed June 18, 2020. 24. Saunders B, Sim J, Kingstone T, Baker S, Waterfield J, Bartlam B, et al. Saturation in 29. Buchanan J, Jenkins S, Scott L. Student clinical education in Australia: an University qualitative research: exploring its conceptualization and operationalization. Qual of Sydney scoping study. Sydney: The University of Sydney; 2014. Quant. 2018;52:1893–1907. 30. Maloney P Stagnitti K, Schoo A. Barriers and enablers to clinical fieldwork edu- cation in rural public and private allied health practice. Higher Edu Res Dev. 25. Liamputtong P. Qualitative research methods. 3rd ed. Hong Kong: Oxford University 2013;32:420–436. Press; 2009. 31. Deloitte Access Economics. Higher education teaching and learning costs. 2011. Canberra. 26. Braun V, Clarke V. Using thematic analysis in psychology. Qual Res Psychol. 32. Forbes R, Dinsdale A, Dunwoodie R, Birch S, Brauer S. Exploring strategies used by 2006;3:77–101. physiotherapy private practices in hosting student clinical placements. Aust J Clin Educ. 2020;7. 27. Thurmond VA. The point of triangulation. J Nurs Scholarsh. 2001;33:253– 258. 28. Bowles K, Haines T, Molloy E, Maloney S, Kent F, Sevenhuysen S, et al. The costs and benefits of providing undergraduate student clinical placements for a health
Journal of Physiotherapy 68 (2022) 74 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: Core Outcome Set for trials with Coronavirus disease 2019 (COVID-19-COS) Summary The Core Outcome Set (COS) for trials in Coronavirus Disease 2019 progression scale, with scores between 4 (hospitalised with no oxygen (COVID-19-COS) aimed to establish a consistent and standardised list of therapy) and 9 (extracorporeal membrane oxygenation) reflecting the increasing need for respiratory support.2,3 Validation of this scale is not outcomes to be measured and reported, as a minimum, in trials treating yet available; however, it has already been used in COVID-19 trials.4,5 patients with COVID-19.1 The COVID-19-COS was developed according Multiple organ failure is measured using the Sequential Organ to the Core Outcome Measures in Effectiveness Trial (COMET) frame- Failure Assessment (SOFA) score, which is commonly used in research work, using a series of online workshops involving adults who had and clinical practice. It has been validated for use in hospital and ICU experienced suspected or confirmed COVID-19, their family members, settings and also provides prognostic information regarding survival.6 the general public and health professionals from 111 countries.1 The Shortness of breath is measured using the Modified Medical Research COVID-19-COS specifies mortality, respiratory failure, multiple organ Council (MMRC) dyspnoea scale, with minor adaptations to the failure, shortness of breath, and recovery as the most critically impor- tant outcomes for trials involving participants with COVID-19.2 original wording and the addition of a recall period of 24 hours to Core outcome measures for the COVID-19-COS outcome domains capture daily fluctuations. The MMRC is a simple 5-point scale that were proposed,2 using the COnsensus-based Standards for the selec- has been extensively validated in patients with chronic respiratory disease.7,8 No existing outcome measure for recovery was identified, tion of health Measurement INstruments (COSMIN) framework to so a new COVID-19-COS recovery measure was proposed, consisting identify appropriate measures and their clinimetric properties. Mor- of a 5-point Likert scale with anchors ‘not recovered at all’ and ‘completely recovered’. No clinimetric data are available for this new tality is measured, according to World Health Organization (WHO) recovery measure.2 recommendations, at hospital discharge or at 60 days.2 The outcome measure for respiratory failure is a modified version of the WHO clinical Commentary Researchers across the world have rapidly responded to the ur- measure of recovery following COVID-19, given the absence of gent need for clinical trials of treatments for COVID-19. Maximising existing measurement tools, and this should be validated in future the health impacts of this research effort will need trial outcomes that studies. The greatest value of COVID-19-COS is in trials located or can be interpreted, compared and applied across different pop- commenced in acute care settings, where it provides opportunities to ulations and countries. The decision regarding which outcomes generate high-quality evidence and results that can be readily should be measured in clinical trials needs to take into consideration compared and combined with other studies.9 the relevance to clinicians and researchers but also to patients, policy makers and funders.9 A strength of COVID-19-COS is the identifica- Provenance: Invited. Not peer reviewed. tion of a small number of outcome domains relevant to all stake- holders, which can be measured using simple outcome tools, some of Mariana Hoffmana and Anne E Hollanda,b,c which are familiar to clinical trialists and already in common use. aDepartment of Allergy, Immunology and Respiratory Medicine, Monash Another important strength is the use of robust frameworks (COMET and COSMIN) to identify outcomes and understand the clinimetric University, Melbourne, Australia properties of the proposed measurement tools. bDepartment of Physiotherapy, Alfred Health, Melbourne, Australia Because COVID-19-COS was developed early in the pandemic cInstitute for Breathing and Sleep, Melbourne, Australia (March and April 2020), the top 10 priorities identified by partici- pants comprised only acute and severe outcomes. As a result, the References most robust elements of COVID-19-COS reflect the outcomes of acute care (mortality, respiratory failure, multiple organ failure). The 1. Tong A, et al. Crit Care Med. 2020;48:1622–1635. steering group elected to add a patient-reported outcome (shortness 2. Tong A, et al. Crit Care Med. 2021;49:503–516. of breath) and a longer term outcome (recovery). Since the devel- 3. WHO Working Group. Lancet Infect Dis. 2020;20:e192–e197. opment of COVID-19-COS, knowledge of the long-term sequelae of 4. Cao B, et al. N Engl J Med. 2020;382:1787–1799. COVID-19 infection has dramatically increased, including the impor- 5. Rodionov RN, et al. Lancet Microbe. 2021;2:e138. tance of other debilitating symptoms such as persistent fatigue.10 The 6. Seymour CW, et al. JAMA. 2016;315:762–774. absence of dyspnoea in acutely unwell patients has also been rec- 7. Tsiligianni IG, et al. Int J Chron Obstruct Pulmon Dis. 2016;11:1045–1052. ognised,11 such that this measure may vary in its utility across pa- 8. Sandberg J, et al. J Pain Symptom Manage. 2018;56:430–435 e2. tients. The authors are to be commended for developing a simple 9. Williamson P, et al. J Health Serv Res Policy. 2012;17:1–2. 10. Logue JK, et al. JAMA Netw Open. 2021;4:e210830. 11. Tobin MJ, et al. Am J Respir Crit Care Med. 2020;202:356–360. https://doi.org/10.1016/j.jphys.2021.06.019 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 68 (2022) 73 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 Physical Function in ICU test-scored Summary Description: The Physical Function in ICU (intensive care unit) test 80 steps/minute) if a patient can step for 3 minutes.13 Shoulder (PFIT) is a composite measure of muscle strength and physical function that was developed by Australian physiotherapists from a flexion and knee extension strength scores are derived from grading sample of medical-surgical ICU and respiratory weaning unit with the Oxford scale, ideally tested in sitting with the higher grade patients.1 In 2013, the measure was refined to improve clinimetric properties by item reduction to four tasks (sit-to-stand, marching on across left and right used. If a patient is unable to be positioned in the spot cadence, shoulder flexion and knee extension strength) and testing of a matrix for converting ordinal to interval scores, resulting sitting, strength testing can occur in recumbent positions such that in the present PFIT-scored (PFIT-s) version.2 The PFIT-s is accessible, takes approximately 10 to 15 minutes to administer,3 requires mini- patients with lower functioning can still be assessed with the PFIT-s. mal equipment (a chair and stopwatch), is available in English and Finally, interval conversion scores (range 0 to 10) are determined2 Brazilian Portuguese,4 and is extensively validated.5–15 and interpreted with a minimum detectable difference of 2.42 Methodology and scoring: Each PFIT-s item is scored from 0 to 3 points5 and minimum clinically important difference of 1.5 points.2 and summed to a maximum ordinal score of 12, with higher scores representing better physical function.2 Level of assistance for the Clinimetric properties: Clinimetric testing has occurred in several patient to safely transfer from sit-to-stand is first rated as unable, assistance of two people, assistance of one person or no assistance independent and international samples where the PFIT-s has been from therapist (scored 0 to 3, respectively).2 If standing is achieved, found to have good inter-rater reliability5 and be responsive to stepping cadence (steps/minute) is assessed by marching in place. change over time.2,3,5,6 At awakening and ICU discharge there is Assessment conventions for step cadence have included stopping the timer if a patient ceases to step for . 2 seconds or has poor foot moderate-excellent validity between the PFIT-s and other physical clearance,8 and awarding the maximum item score of 3 (equivalent to measures (namely the Functional Status Score in the ICU, ICU Mobility Commentary Scale and the Short Physical Performance Battery),2,3,5,6 and excellent construct validity with muscle strength at these same time points.6,7 Higher PFIT-s scores have predictive validity for discharge to home, post-ICU hospital length of stay and higher health-related quality of life after hospital discharge.2,3,6 Within the ICU, the PFIT-s has minimal floor and ceiling effects ( 1 to 12%)6–8 but ceiling effects are significant by hospital discharge (w25 to 27%).5,6 The PFIT-s appears to be most useful earlier in an ICU survivor’s support exercise prescription within the ICU, which is a unique recovery, as there are ceiling effects and an absence of items that feature of the PFIT-s.15 test higher levels of physical functioning. To address ceiling effects Provenance: Invited. Not peer reviewed. of the PFIT-s and floor effects of other measures, the new ‘PACIFIC’ Selina M Parrya and Claire E Baldwinb tool was recently developed, amalgamating the PFIT-s and De aDepartment of Physiotherapy, The University of Melbourne, Melbourne Morton Mobility Index through Rasch analyses to build a 10-item bCaring Futures Institute and College of Nursing and Health Sciences, Flinders University, Adelaide, Australia interval measure with improved task coverage and potential for use across acute and community settings.8 Recently, the PFIT-s has References also been used to test the validity of a new patient-rated (rather 1. Skinner EH, et al. Critical Care Resusc. 2009;11:110–115. 2. Denehy L, et al. Phys Ther. 2013;93:1636–1645. than therapist-rated) scale of physical function in the ICU and acute 3. Parry SM, et al. Crit Care. 2017;21:249. care setting.9 4. Silva V, et al. J Bras Pneumol. 2020;46:e20180366. 5. Costigan FA, et al. Can J Anaesth. 2019;66:1173–1183. The decision of when to choose the PFIT-s over other robust 6. Parry SM, et al. Crit Care. 2015;19:127. ICU physical measures such as the ICU Mobility Scale10 will 7. Nordon-Craft A, et al. Phys Ther. 2014;94:1499–1507. 8. Parry SM, et al. Crit Care Med. 2020;48:1427–1435. depend on what domains of impairment and activity limitation 9. Reid JC, et al. J Intensive Care Med. 2020;35:1396–1404. 10. Tipping CJ, et al. J Physiother. 2020;66:271. the therapist is aiming to evaluate. For example, the PFIT-s 11. Tadyanemhandu C, et al. Arch Physiother. 2016;6:12. evaluates strength, mobility and endurance,3 whereas the ICU 12. Denehy L, et al. Crit Care. 2013;17:R156. 13. Kho ME, et al. BMJ Open Resp Res. 2019;6:e000383. Mobility Scale focuses on mobility subdomains based on the 14. Mehrholz J, et al. BMJ Open. 2015;5:e008828. International Classification of Functioning.10 The key benefits of 15. Berney S, et al. Phys Ther. 2012;92:1524–1535. the PFIT-s are: minimum equipment and training requirements (including resource-limited settings);11 tasks that can be easily and regularly incorporated into physiotherapy assessment (at a minimum recommended at awakening and ICU discharge);3 utility for monitoring change in physical recovery over time and in response to rehabilitation interventions (as demonstrated in observational and randomised trials);12–14 and utility to https://doi.org/10.1016/j.jphys.2021.05.003 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 68 (2022) 80–81 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: Patient education in knee osteoarthritis Knee osteoarthritis (OA) is the most common joint disease and effects of patient education in isolation, which will help to determine affects millions of people worldwide.1 Patient education, which is whether it could be delivered alone or should be supplemented with defined as ‘any set of planned educational activities designed to other therapy (eg, exercise therapy, weight loss intervention). improve patients’ health behaviours and/or health status’, has been recommended in many clinical practice guidelines as a first- Second, the authors included all kinds of educational in- line intervention for patients with knee OA;2 however, there is a terventions regardless of the intervention development process (eg, lack of clarity about the effectiveness of patient education co-design, based on learning theory) or whether they were used as a interventions on pain and function outcomes in people with knee control, which reduced selection bias and increased the general- OA. With great interest, we read the article by Goff et al, which isability of the results. However, it should be noted that some of the found that: patient education may reduce pain and improve education programs were not disease-specific and that adherence to function compared with usual care; these effects may not be patient education varies with the delivery method, which further large enough to be clinically important; and combining patient raises doubts over the achievement of meaningful benefits in patients education with exercise therapy should be encouraged, given the who received the educational intervention. Therefore, due to these statistically superior and clinically important improvements in potential biases and confounding, the results of this review should be function compared with patient education alone.3 We applaud interpreted with caution. the authors for addressing this important issue of identifying benefits for people with knee OA; however, there are two minor The review also provides directions for future research. More well- points that are worth further discussion. designed robust trials are still warranted to determine best-practice patient education for reducing pain and improving function in peo- First, the findings of this study are different from a newly pub- ple with knee OA. Aspects that require further clarification include lished, evidence-based, clinical practice guideline.4 In this recently the delivery method, which educational modules to include, the developed guideline by the American Academy of Orthopaedic particular components of those modules and the amount of educa- Surgeons for the management of knee OA (non-arthroplasty), tion needed to improve patient outcomes. patient education programs are strongly recommended to improve pain, while the systematic review by Goff et al identified effects that Mian Tian and Shiwei Yuan may not be clinically important and questioned the value of patient Department of Orthopaedic Surgery, Dianjiang People’s Hospital of education in isolation. Although focused on the same issue, they included different studies, which may have been the main cause of Chongqing, Chongqing, China the inconsistent conclusions. The rationale of the guideline mentioned that patient education programs in studies overlap with https://doi.org/10.1016/j.jphys.2021.12.002 self-management programs such as medication compliance, pain management, pain coping strategies and stress management tech- References niques. Thus, it is possible that the clinical benefits of patient educa- tion manifest when combined with standard medical care and patient 1. Katz JN, et al. JAMA. 2021;325:568–578. education is supplementary to standard medical care, which is con- 2. McAlindon TE, et al. Osteoarthritis Cartilage. 2014;22:363–388. sisted with Goff et al’s questioning of the value of patient education in 3. Goff AJ, et al. J Physiother. 2021;67:177–189. isolation.5 Therefore, further studies are needed to evaluate the actual 4. American Academy of Orthopaedic Surgeons Management of Osteoarthritis of the Knee (Non-Arthroplasty) Evidence-Based Clinical Practice Guideline. https://www. aaos.org/oak3cpg. Accessed 31 August, 2021. 5. Riemsma RP, et al. Cochrane Database Syst Rev. 2003;2:CD003688. Correspondence: Author response to Tian et al We thank Tian and Yuan for their interest and positive comments Other methodological differences also require acknowledgement. about our systematic review.1 They correctly identify that our We clearly defined and applied the Grading of Recommendations, conclusions differ from the recently developed American Academy Assessment, Development and Evaluations (GRADE) criteria, outlined of Orthopaedic Surgeons (AAOS) guideline,2 which strongly in our paper.1 This led us to conclude ‘very-low certainty evidence’ recommended patient education programs for knee osteoarthritis. supporting patient education as a stand-alone intervention. It is un- As suggested by Tian and Yuan, different recommendations may be clear how GRADE was applied in the AAOS guidelines to conclude due to the inclusion of different trials in the two pieces of work. ‘strong evidence’ for patient education. It is also unclear if the AAOS Notably, we included control patient education interventions in our guideline patient education recommendations are based on meta- analysis and excluded trials when we were unable to retrieve analyses, or if a minimal clinically important difference (MCID) outcome data for people with knee osteoarthritis only. The AAOS threshold was applied. Mean differences from our meta-analyses for guidelines included trials for people with knee and hip patient education compared to control interventions were considered osteoarthritis, but it is unclear if they based their recommendations in the context of their MCIDs.3,4 This led to our conclusion that on outcomes for people with knee osteoarthritis only. ‘although patient education produced statistically superior short- 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 81 term pain and function outcomes compared with usual care, differ- guidelines related to the specific content, methods of delivery and ences were small and may not be clinically important.’ However, development of patient education interventions for people with knee when education was combined with exercise therapy, findings were osteoarthritis. Central to this ‘call to action’ should be consideration both statistically significant and greater than MCIDs.3,4 This led us to of the lived experience of people with knee osteoarthritis and their conclude that patient education ‘should be combined with exercise perceived educational needs. therapy to provide statistically superior and clinically important short-term improvements in function compared with education Anthony J Goffa,b, Danilo De Oliveira Silvaa and Christian J Bartona alone.’ We agree with Tian and Yuan, that the benefit of combining aLa Trobe Sport and Exercise Medicine Research Centre, School of Allied patient education with other interventions such as diet and weight- loss remains unclear and warrants further investigation. Health, Human Services and Sport, La Trobe University, Melbourne, Australia; The second point raised by Tian and Yuan was that education con- tent and development varied in our included studies. Notably, clinical bHealth and Social Sciences Cluster, Singapore Institute of Technology, practice guidelines, including the AAOS,2 provide no recommendations Singapore on how to develop education interventions, and provide limited guidance on what content to provide or how to deliver it. Our https://doi.org/10.1016/j.jphys.2021.12.008 ancillary analysis of this review,5 summarised the development, content and delivery of included education interventions. This References indicated that education interventions lack robust development and descriptions in published trials are typically broad and unclear. We 1. Goff AJ, et al. J Physiother. 2021;67:177–189. did not investigate adherence to patient education interventions, and 2. American Academy of Orthopaedic Surgeons Management of Osteoarthritis of the this should be a focus of future research. Knee (Non-Arthroplasty) Evidence-Based Clinical Practice Guideline. https://www. Building on the points raised in this correspondence, we call for aaos.org/oak3cpg. Accessed August 31, 2021. greater transparency in the content of patient education in- 3. Angst F, et al. Arthritis Care Res. 2001;45:384–391. terventions in trials, and greater guidance from clinical practice 4. Tubach F, et al. Ann Rheum Dis. 2005;64:29–33. 5. Goff AJ, et al. J Orthop Sports Phys Ther. 2021. https://www.jospt.org/doi/10.2519/ jospt.2022.10771. Accessed 15 December, 2021.
Journal of Physiotherapy 68 (2022) 80–81 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: Patient education in knee osteoarthritis Knee osteoarthritis (OA) is the most common joint disease and effects of patient education in isolation, which will help to determine affects millions of people worldwide.1 Patient education, which is whether it could be delivered alone or should be supplemented with defined as ‘any set of planned educational activities designed to other therapy (eg, exercise therapy, weight loss intervention). improve patients’ health behaviours and/or health status’, has been recommended in many clinical practice guidelines as a first- Second, the authors included all kinds of educational in- line intervention for patients with knee OA;2 however, there is a terventions regardless of the intervention development process (eg, lack of clarity about the effectiveness of patient education co-design, based on learning theory) or whether they were used as a interventions on pain and function outcomes in people with knee control, which reduced selection bias and increased the general- OA. With great interest, we read the article by Goff et al, which isability of the results. However, it should be noted that some of the found that: patient education may reduce pain and improve education programs were not disease-specific and that adherence to function compared with usual care; these effects may not be patient education varies with the delivery method, which further large enough to be clinically important; and combining patient raises doubts over the achievement of meaningful benefits in patients education with exercise therapy should be encouraged, given the who received the educational intervention. Therefore, due to these statistically superior and clinically important improvements in potential biases and confounding, the results of this review should be function compared with patient education alone.3 We applaud interpreted with caution. the authors for addressing this important issue of identifying benefits for people with knee OA; however, there are two minor The review also provides directions for future research. More well- points that are worth further discussion. designed robust trials are still warranted to determine best-practice patient education for reducing pain and improving function in peo- First, the findings of this study are different from a newly pub- ple with knee OA. Aspects that require further clarification include lished, evidence-based, clinical practice guideline.4 In this recently the delivery method, which educational modules to include, the developed guideline by the American Academy of Orthopaedic particular components of those modules and the amount of educa- Surgeons for the management of knee OA (non-arthroplasty), tion needed to improve patient outcomes. patient education programs are strongly recommended to improve pain, while the systematic review by Goff et al identified effects that Mian Tian and Shiwei Yuan may not be clinically important and questioned the value of patient Department of Orthopaedic Surgery, Dianjiang People’s Hospital of education in isolation. Although focused on the same issue, they included different studies, which may have been the main cause of Chongqing, Chongqing, China the inconsistent conclusions. The rationale of the guideline mentioned that patient education programs in studies overlap with https://doi.org/10.1016/j.jphys.2021.12.002 self-management programs such as medication compliance, pain management, pain coping strategies and stress management tech- References niques. Thus, it is possible that the clinical benefits of patient educa- tion manifest when combined with standard medical care and patient 1. Katz JN, et al. JAMA. 2021;325:568–578. education is supplementary to standard medical care, which is con- 2. McAlindon TE, et al. Osteoarthritis Cartilage. 2014;22:363–388. sisted with Goff et al’s questioning of the value of patient education in 3. Goff AJ, et al. J Physiother. 2021;67:177–189. isolation.5 Therefore, further studies are needed to evaluate the actual 4. American Academy of Orthopaedic Surgeons Management of Osteoarthritis of the Knee (Non-Arthroplasty) Evidence-Based Clinical Practice Guideline. https://www. aaos.org/oak3cpg. Accessed 31 August, 2021. 5. Riemsma RP, et al. Cochrane Database Syst Rev. 2003;2:CD003688. Correspondence: Author response to Tian et al We thank Tian and Yuan for their interest and positive comments Other methodological differences also require acknowledgement. about our systematic review.1 They correctly identify that our We clearly defined and applied the Grading of Recommendations, conclusions differ from the recently developed American Academy Assessment, Development and Evaluations (GRADE) criteria, outlined of Orthopaedic Surgeons (AAOS) guideline,2 which strongly in our paper.1 This led us to conclude ‘very-low certainty evidence’ recommended patient education programs for knee osteoarthritis. supporting patient education as a stand-alone intervention. It is un- As suggested by Tian and Yuan, different recommendations may be clear how GRADE was applied in the AAOS guidelines to conclude due to the inclusion of different trials in the two pieces of work. ‘strong evidence’ for patient education. It is also unclear if the AAOS Notably, we included control patient education interventions in our guideline patient education recommendations are based on meta- analysis and excluded trials when we were unable to retrieve analyses, or if a minimal clinically important difference (MCID) outcome data for people with knee osteoarthritis only. The AAOS threshold was applied. Mean differences from our meta-analyses for guidelines included trials for people with knee and hip patient education compared to control interventions were considered osteoarthritis, but it is unclear if they based their recommendations in the context of their MCIDs.3,4 This led to our conclusion that on outcomes for people with knee osteoarthritis only. ‘although patient education produced statistically superior short- 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 68 (2022) 69 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: A task-specific sit-to-stand training program for children with cerebral palsy improves mobility and self-care function Synopsis Summary of: Chaovalit S, Dodd KJ, Taylor NF. Sit-to-stand training for mobility (five items) domains of the Functional Independence Mea- self-care and mobility in children with cerebral palsy: a randomized sure for Children (WeeFIM), where each item is scored from 1 (total controlled trial. 2021; epub ahead of print. assistance) to 7 (complete independence). Secondary outcome mea- sures were sit-to-stand performance (Five Times Sit-to-Stand Test) Question: Does a task-specific sit-to-stand training exercise program and caregiver strain (Modified Caregiver Strain Index). Results: All for children with cerebral palsy with moderate to severe mobility participants completed the study. At the end of the follow-up mea- limitations improve self-care, independent upright mobility, and surement period (7 weeks), the WeeFIM self-care domain was sit-to-stand performance, and reduce caregiver strain? Design: improved more in the treatment group, by 2.2 units (95% CI 1.3 to 3.1), Randomised controlled trial with concealed allocation. Setting: along with the WeeFIM mobility domain by 2.2 units (95% CI 1.4 to 3.0). Rehabilitation centre, Thailand. Participants: Children with cerebral The total WeeFIM score and sit-to-stand performance were also palsy aged 4 to 12 years, gross motor function classification level III to improved, whilst caregiver burden reduced for the intervention group IV (have moderate to severe difficulty with sit-to-stand), who could compared with standard care. Conclusion: A 6-week task-specific sit- follow simple verbal instructions and had not received botulinum to-stand program for children with cerebral palsy and moderate to neurotoxin in the previous 6 months. Randomisation of 38 partici- severe mobility limitations led to small improvements in self-care, pants allocated 19 to the task-specific sit-to-stand exercise program mobility and sit-to-stand performance, and reduced caregiver strain. and 19 to the standard care group. Interventions: Both groups received routine physiotherapy for 30 minutes, five times per week Provenance: Invited. Not peer reviewed. (two sessions with a physiotherapist in an outpatient setting, three sessions supervised by caregiver at home) for 6 weeks consisting of Alicia Spittle balance and gait training. In addition, the intervention group received Department of Physiotherapy, University of Melbourne, Australia 30 minutes of sit-to-stand training using principles of task-specific training, whilst the control group had a hot-pack applied to the https://doi.org/10.1016/j.jphys.2021.10.002 lower limbs following the physiotherapy session. Outcome measures: The primary outcome was the change in the self-care (six items) and Commentary Sit-to-stand movement is a precursor skill for mobility and self- caregiver strain following the program compared with a control care. Children with cerebral palsy with moderate to severe mobility group. However, the mean between-group differences were less than limitations can have difficulty performing functional activities that the respective minimum clinically important differences for each involve sit-to-stand, including transferring to a wheelchair and outcome measure. bathing, thus making these activities a target of therapy. Prior to this trial, little evidence existed regarding the effectiveness of sit-to-stand While some components of the control intervention (eg, hot packs) training for improving functional activities in this population. Two may not commonly be considered ‘routine’, and the absence of previous randomised controlled trials tested sit-to-stand training in goal setting/measurement is notable, given the evidence supporting children with cerebral palsy; however, participants primarily had its role in improving function,3 this study provides guidance to mild to moderate mobility limitations and the dosage of training was physiotherapists that high-dose sit-to-stand training may lead to relatively low.1,2 small improvements in functional activities in this population. To date, research is limited on children with cerebral palsy who Provenance: Invited. Not peer reviewed. experience moderate to severe mobility limitations.3 This is the first randomised controlled trial to assess whether sit-to-stand training Rachel Toovey improved self-care and mobility function, along with sit-to-stand Department of Physiotherapy, University of Melbourne, Parkville, performance and caregiver strain, in children with cerebral palsy who had moderate to severe mobility limitations. The training pro- Australia gram involved a minimum of 2,250 sit-to-stand repetitions across 30 sessions over 6 weeks, including 18 home sessions. Assessment of https://doi.org/10.1016/j.jphys.2021.10.001 caregiver strain is important, given the expectations on caregivers to deliver home sessions versus the potential longer-term benefit of References improved functional activities. This study found small improvements in self-care and mobility function, sit-to-stand performance, and 1. Liao HF, et al. Arch Phys Med Rehabil. 2007;99:25–31. 2. Kumban W, et al. Dev Neurorehabil. 2013;16:410–417. 3. Novak I, et al. Current Neurol Neurosci Rep. 2020;20:3. 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 68 (2022) 70 j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j p hy s Appraisal Critically appraised paper: In people reporting dyspnoea following COVID-19 hospitalisation, additional telerehabilitation is more effective at improving exercise capacity, muscle strength and the physical component of quality of life than education alone Synopsis Summary of: Li J, Xia W, Zhan C, et al. A telerehabilitation programme in post- dyspnoea-free. Assessments occurred at baseline, 6 and 28 weeks. Results: 105 discharge COVID-19 patients (TERECO): a randomised controlled trial. Thorax participants completed the study (50 in the experimental group and 55 in the Published Online First: 26 July 2021. https://doi.org/10.1136/thoraxjnl- control group). On completion of the 6-week intervention period, the change in 2021-217382 6-minute walk distance was greater in the experimental group (MD 65 m, 95% CI 44 to 87). The experimental group also had better performance on the static Question: In people reporting dyspnoea following COVID-19 hospitalisation, is squat test (MD 20 seconds, 95% CI 12 to 28), a higher score on the physical an additional telerehabilitation program more effective at improving exercise component of the Short Form Health Survey-12 (MD 3.8, 95% CI 1.2 to 6.4) and capacity, muscle strength, health-related quality of life and dyspnoea than ed- more dyspnoea-free participants (adjusted RR 1.46, 95% CI 1.17 to 1.82). All ucation alone? Design: Randomised controlled trial with concealed allocation between-group differences, except the percentage of participants who were and blinded assessor. Setting: Three hospitals in China. Participants: Adults dyspnoea-free, were maintained at 28 weeks. Conclusion: In people reporting reporting a modified Medical Research Council dyspnoea scale score of 2 or 3 dyspnoea following COVID-19 hospitalisation, a 6-week additional tele- after COVID-19 hospitalisation. Exclusion criteria: resting heart rate . 100 bpm, rehabilitation program was more effective at producing sustained improve- uncontrolled chronic disease, history of severe cognitive disorder, and inability ments in exercise capacity, muscle strength and the physical component of to walk independently. Randomisation of 120 participants allocated 59 to an health-related quality of life than education alone. experimental group and 61 to a control group. Interventions: Both groups received 10 minutes of education and written instruction on topics such as Provenance: Invited. Not peer reviewed. physical activity, diet and sleep. In addition, the intervention group underwent a 6-week home exercise program delivered via a telerehabilitation smartphone Vinicius Cavalheri application. The program sessons were 40 to 60 minutes in duration, with three Curtin School of Allied Health, Curtin University, Australia to four sessions per week of breathing exercises, walking or running and lower limb resistance exercises. Outcome measures: The primary outcome was change https://doi.org/10.1016/j.jphys.2021.11.001 in 6-minute walk distance from baseline to 6 weeks. Secondary outcome measures included the static squat test, health-related quality of life (Short Form Health Survey-12) and the percentage of participants who were 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 Despite COVID-19 leaving millions of people with rehabilitation needs, ser- intervention. Clinicians should vigilantly screen for oxygen desaturation, post- vices have been reliant on general rehabilitation principles, clinical expertise and exertional symptom exacerbation, cardiac impairment and autonomic dysfunc- consensus guidance based on evidence from other conditions.1 This trial provides tion during and after rehabilitation interventions. Given the multi-dimensional, evidence of the effects of rehabilitation for patients admitted to hospital with episodic and unpredictable nature of Long COVID, pacing (activity management) COVID-19 who had moderate breathlessness at discharge. and symptom-titrated activities are advocated, whereas graded or fixed incre- mental exercise prescriptions are not.3 The group-level differences in the primary outcome of functional exercise capacity at 6 and 28 weeks were striking. They were twice the minimum clinically Provenance: Invited. Not peer reviewed. important difference, meaning that most participants perceived the change, re- flected in their quality of life scores. They also far exceeded changes after a similar Matthew Maddocksa and Darren Brownb intervention following admission for exacerbation of chronic respiratory disease.2 aCicely Saunders Institute of Palliative Care, Policy & Rehabilitation, This may be related to differences in pre-admission function and the potential recovery trajectory. King’s College London, UK bTherapies Department, Chelsea & Westminster Hospital The primary finding is also impressive, given the delivery model with minimal therapist contact and supervision. The COVID-19 pandemic has led to rapid growth NHS Foundation Trust, London, UK in digital healthcare interventions, making efficient use of therapist time and pro- moting patient self-management and empowerment. However, such interventions https://doi.org/10.1016/j.jphys.2021.11.003 can introduce inequality due to costs and digital capability (hardware ownership, literacy, etc). Patient choice and preference should guide use, and communication to References understand each patient’s goals, motivations and challenges remains important. 1. Goodwin VA, et al. Physiotherapy. 2021;111:4–22. Although not widely recognised early in the pandemic, ‘Long COVID’ is now 2. Greening N, et al. BMJ. 2014;349:g4315. understood as a key consideration in post-viral disease. It may explain some of the 3. World Physiotherapy. 2021. ISBN: 978-1-914952-00-5. adverse events, discontinuations and re-hospitalisations observed in this trial, which were higher in the intervention group but were not attributed to the 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/).
Journal of Physiotherapy 68 (2022) 71 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: Progressive exercise is not superior to best practice advice, and steroid injection is not superior to no injection, for rotator cuff disorders Synopsis Summary of: Hopewell S, Keene DJ, Marian IR, Dritsaki M, Heine P, single session with a physiotherapist, involving simple self-guided Cureton L, et al on behalf of the GRASP Trial Group. Progressive home exercises (5 days/week). Outcome measures: The primary exercise compared with best practice advice, with or without corti- outcome was the Shoulder Pain and Disability Index (SPADI) score costeroid injection, for the treatment of patients with rotator cuff over 12 months. Secondary outcomes included SPADI subscales, disorders (GRASP): a multicentre, pragmatic, 232 factorial, rando- quality of life, fear avoidance, pain self-efficacy, insomnia severity, mised controlled trial. Lancet. 2021;398:416-428. participant global impression of change, serious adverse events, return to desired activities, exercise adherence, health resource use, Question: Is a progressive exercise program superior to best prac- out-of-pocket expenses and work absence. Results: At 12 months, tice advice, with or without corticosteroid injection, in adults with 682 (97%) participants provided primary outcomes. Over 12 months, rotator cuff disorders? Design: Multicentre, pragmatic, superiority, there was no difference in SPADI scores between progressive exer- 232 factorial randomised controlled trial with concealed allocation. cise and best practice advice (adjusted mean difference 20.66, 99% Setting: Twenty United Kingdom National Health Service trusts. CI 24.52 to 3.20). There was also no evidence of a difference be- Participants: Adults aged 18 years with shoulder pain attribut- tween corticosteroid injection compared with no injection able to a rotator cuff disorder developed in the past 6 months. The (adjusted mean difference 21.11, 95% CI 24.47 to 2.26). Conclusion: main exclusion criteria were: significant shoulder trauma, neuro- Progressive exercise was not superior to best practice advice by a logical disease or other condition affecting the shoulder; cortico- physiotherapist. Subacromial corticosteroid injection provided no steroid injection or physiotherapy for shoulder pain in the past 6 long-term benefit. months; or being considered for surgery. Randomisation of 708 participants allocated 174 to best practice advice only, 178 to in- Provenance: Invited. Not peer reviewed. jection and best practice advice, 174 to exercise only, and 182 to injection and exercise. Interventions: Subacromial corticosteroid Nina Østerås injection was delivered by extended-scope physiotherapists before Division of Rheumatology and Research, Diakonhjemmet Hospital, exercise or advice sessions, with an optional re-injection after 6 weeks. The progressive exercise program (focused on resisted Norway shoulder external rotation, flexion and abduction) included up to six individual face-to-face physiotherapy sessions over 16 weeks, and https://doi.org/10.1016/j.jphys.2021.11.005 home exercises 5 days/week. Best practice advice was delivered in a Commentary This commentary focuses on the main effect of progressive exer- difference between the treatment groups because the confidence cise and best practice advice. Although the progressive exercise intervals do not include the clinically important difference in the treatment protocol allowed for up to six sessions of exercise and primary outcome of 8 points. This is a strong finding given that the advice, participants in this group attended an average of three ses- progressive exercise participants knew they were receiving additional sions. Hence, the progressive exercise group had only a modest in- care, which would have introduced directional bias toward this crease in supervision compared with the single session in the best group. One session of best practice advice can thus be recommended practice advice group, and this may explain why there was no in environments similar to the UK National Health Service. As this between-group difference observed for the primary outcome. trial recruited only people with an episode of shoulder pain within Further, the authors highlight that exercise to improve strength is the last 6 months, findings cannot necessarily be generalised to considered integral to management. Although more participants re- people with a longer history who may be likely to have tried (and ported doing the prescribed exercise volume at 8 weeks (ie, five ex- failed) multiple previous treatments. ercise sessions/week) in the progressive exercise compared with the best practice advice group (60% vs 43%), it is unclear if this translated Provenance: Invited. Not peer reviewed. into greater strength gains with progressive exercise. Changes in strength also depend on exercise intensity. Both groups were advised Peter Malliaras to progress exercise based on rating of perceived exertion, but exer- Physiotherapy Department, Monash University, Melbourne, Australia cise intensity was not reported. This introduces some uncertainty about how different the exercise interventions were. Despite these https://doi.org/10.1016/j.jphys.2021.11.006 limitations, readers can be quite confident that there is no meaningful 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 68 (2022) 72 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: Task-oriented exercise improved disability, pain and quality of life compared with general physiotherapy for surgically treated proximal humeral fractures Synopsis Summary of: Monticone M, Portoghese I, Cazzaniga D, Liquori V, Mar- control group underwent a program of general physiotherapy, including ongiu G, Capone A, et al. Task-oriented exercises improve disability of mobility, strengthening and stretching exercises. At the conclusion of both working patients with surgically-treated proximal humeral fractures. A programs, participants were asked to perform their exercises at home. randomised controlled trial with one-year follow-up. BMC Musculoskelet Outcome measures: The primary outcome was disability, measured by the Disord. 2021;22:293. Disabilities of the Arm, Shoulder and Hand questionnaire (0 to 100), measured immediately post-intervention (primary endpoint) and at 12 Question: Does a 12-week program of task-oriented exercises improve months (secondary endpoint). Secondary outcome measures were pain and disability, pain and quality of life in patients with surgically treated prox- quality of life. Results: A total of 63 (90%) participants completed the study. imal humeral fractures compared with general physiotherapy? Design: After intervention, reduction in disability was greater in the experimental Randomised controlled trial with concealed allocation and blinded group by 13 points (95% CI 5 to 22). At 12 months, reduction in disability was outcome assessment. Setting: One rehabilitation hospital in Italy. Partici- also greater in the experimental group by 16 points (95% CI 7 to 25). pants: Employed adults aged 20 to 65 years who were undergoing surgical Between-group differences in pain and quality of life favoured the experi- treatment of displaced and unstable proximal humeral fracture. Key mental group at both endpoints. Conclusion: A program of specific, task- exclusion criteria were: cognitive impairment; unstable cardiovascular and oriented exercise improved disability, pain and quality of life compared pulmonary diseases; systemic or neuromuscular diseases; isolated with general physiotherapy care in patients with surgically treated prox- tuberculum majus fractures; fractures involving the glenoid cavity; double imal humeral fractures. fractures; injury of the plexus and axillary nerve; and those with workers’ compensation. Randomisation of 70 participants allocated 35 to the Provenance: Invited. Not peer reviewed. experimental group and 35 to the control group. Interventions: Both groups received early mobilisation exercises for the first week. Both programs were Jane Chalmers performed as three individual 1-hour sessions per week supervised by a IIMPACT in Health, University of South Australia, Australia physiotherapist (36 sessions total). The experimental group underwent a program of task-oriented exercises specific to the individual’s job activities https://doi.org/10.1016/j.jphys.2021.11.007 and activities of daily living, and included occupational therapy (sling care information and ergonomic principles once per week for 60 minutes). The Commentary Proximal humeral fractures are one of the most common upper limb living, including occupational activities, may have contributed to superior fractures, with growing incidence among all age groups. After fracture, results.1,3 people are often referred to physiotherapy to reduce pain, improve movement and return to pre-injury activities. Movement through arm use Moving forward, I agree with the authors that evaluating program ef- and/or exercise is regularly prescribed, despite insufficient evidence for fects on return to work rates and participation is valuable. Investigating effective exercise programs.1 whether this program can be delivered with fewer consultations would make this program generalisable. To truly understand the effect of activity- This trial found superior results favouring the experimental group, focused exercise, we need to evaluate it as a standalone intervention receiving task-oriented exercises, on disability, pain and quality of life compared with usual care. This is critical to improving our understanding of measures compared with ‘general physiotherapy’. These findings are effective fracture rehabilitation interventions. clinically meaningful and an important step forward in understanding effective rehabilitation interventions following proximal humeral frac- Provenance: Invited. Not peer reviewed. ture. However, current upper limb fracture rehabilitation practices may not support 3 hours/week of physiotherapy for 12 weeks,2 so the Andrea Bruder experimental program in its entirety may be challenging to translate Department of Physiotherapy, Podiatry and Prosthetics and Orthotics, into practice. Furthermore, the ‘general physiotherapy’ (control) protocol may not reflect usual physiotherapy practice, which is more La Trobe University, Australia closely aligned to the experimental group’s person-centred, goal-ori- ented program. https://doi.org/10.1016/j.jphys.2021.11.008 The inclusion of additional weekly occupational therapy supports the References benefits of multidisciplinary care on patient outcomes. However, because of co-interventions, the superior findings cannot be attributed to any one 1. Bruder AM, et al. J Physiother. 2017;63:205–220. intervention. It is also unclear how much advice on arm use in everyday 2. Bruder AM, et al. Physiotherapy. 2013;99:233–240. 3. Bruder AM, et al. J Physiother. 2016;62:145–152. 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 68 (2022) 26–36 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 Degree and pattern of dual-task interference during walking vary with component tasks in people after stroke: a systematic review Charlotte Sau-Lan Tsang a, Shuting Wang b, Tiev Miller a, Marco Yiu-Chung Pang a a Department of Rehabilitation Sciences, The Hong Kong Polytechnic University, Hong Kong; b School of Biomedical Sciences, The Chinese University of Hong Kong, Hong Kong KEY WORDS ABSTRACT Dual-task interference Questions: What are the degree and pattern of dual-task interference during walking in people after stroke? Cognitive-motor interference How do these vary with disease chronicity and different component tasks in people after stroke? How does Stroke dual-task interference differ between people after stroke and people without stroke? Design: Systematic Systematic review review with meta-analysis of studies reporting gait-related dual-task interference. Participants: People after Meta-analysis stroke and people without stroke. Outcome measures: Measures of walking and secondary (cognitive or manual) task performance under dual-task conditions relative to those under single-task conditions. Results: Seventy-six studies (2,425 people after stroke and 492 people without stroke) were included. Manual and mental tracking tasks imposed the greatest dual-task interference on gait speed, although there was sub- stantial uncertainty in these estimates. Among mental tracking tasks, the apparently least-complex task (serial 1 subtractions) induced the greatest dual-task interference (20.17 m/s, 95% CI 20.24 to 20.10) on gait speed, although there was substantial uncertainty in these estimates. Mutual interference (decrement in both walking and secondary component task performances during dual-tasking) was the most common dual-task interference pattern. The results of the sensitivity analyses for studies involving people with chronic stroke were similar to the results of the primary analyses. The amount of dual-task interference from a mental tracking or manual task during walking was similar between people with or without stroke. Conclusions: The degree and pattern of dual-task interference vary with the choice of component tasks. When evaluating limitations to functional mobility during dual-tasking conditions and in planning in- terventions accordingly, clinicians should select dual-task assessments that correspond to the daily habits and physical demands of people after stroke. Registration: CRD42017059004. [Tsang CS-L, Wang S, Miller T, Pang MY-C (2022) Degree and pattern of dual-task interference during walking vary with component tasks in people after stroke: a systematic review. Journal of Physiotherapy 68:26–36] © 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 same time, only one is perceived, whereas the other is completely ignored. If both are perceived, the responses are often elicited in Scenarios requiring an ability to perform dual-task walking, such as succession rather than concurrently, causing a bottleneck delay, attending to traffic and memorising a shopping list while walking, are thereby impairing the perception of or response to either or both frequently encountered in daily life. In comparison to conditions where tasks.1,2 tasks are performed separately (ie, single-task conditions), a degradation in the performance of one or both component tasks may occur when Multiple resource capacity theory: Multiple resource capacity sim- performed concurrently;1,2 this is commonly referred to as dual-task ulates a hybrid of the capacity sharing and bottleneck models. Instead interference (DTI). A number of theories, which are not all mutually of one overall flexible pool of attentional resources or one single exclusive, have been proposed to explain the DTI phenomenon. operational channel, parallel or relatively independent processing resources are assumed. Each of these resources has its own capacity Capacity sharing theory: The capacity sharing theory assumes a limitations, which are allocated between tasks. Performance deteri- finite processing and attention capacity in people (resources-based).1 oration happens when tasks compete for the same resource. The This capacity and the total amount of attention employed at any given resultant degree of DTI depends on the extent to which tasks share time is flexibly shared among all involved tasks, resulting in less ca- independent resources.3,4 pacity and attention for each component task.1 When the capacity and demand of component tasks do not match, task performance Cross talk theory: Cross talk refers to a transference of energy from either deteriorates or fails completely.1 one communication channel to another.5 The cross talk model focuses on the content of the information being processed (content-oriented). Bottleneck or task switching theory: This theory assumes a stage of This may include the sensory input being presented, responses being internal processing that can only operate on one stimulus or response produced or momentary thoughts.2 When similar inputs are involved, at a time (operation-based).1 When two stimuli are presented at the the same set of processing machinery may be shared for both, making https://doi.org/10.1016/j.jphys.2021.12.009 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/).
Research 27 it easier to perform two tasks simultaneously.2 This results in a dual- (Appendix 1 on the eAddenda). The reference lists of eligible studies task facilitation (an improvement in the performance of either or and review articles were manually screened to identify additional both component tasks relative to single-task performance).6 How- articles for inclusion. ever, DTI may occur when there is an outcome conflict, in which the production of outputs, throughputs or side effects by one task hinders Study selection processing of another.2 All results were exported to reference management softwarea for DTI not only has an impact on community ambulation,7,8 but also screening and duplicate removal. Two independent reviewers on other activities of daily living,9,10 functional independence,9,11 fall assessed each title, abstract and full-text against the inclusion criteria proclivity,12–15 and community participation;16 all of these factors (Box 1) and exclusion criteria: reports published as conference pro- ultimately affect quality of life.17,18 Over the past decade, the volume ceedings, book chapters and theses. In the event of discrepancies of research devoted to investigating the effects of DTI on walking between reviewers, the principal investigator was consulted. among people with brain damage, including stroke,6,9,10,12,19–27 has grown. However, previous systematic reviews have focused on either Data extraction and quality assessment healthy participants28–30 or participants with a mix of different neurological diseases.31,32 Populations vary in how DTI influences Two reviewers extracted information regarding study design, in- cognitive function,33–35 attentional demands during walking36–40 and clusion and exclusion criteria, sample size, subject characteristics, dual-task walking ability.41–44 The degree and pattern of DTI have walking and component task characteristics, task prioritisation, also been shown to differ with component task type9,10,23,28–30,45 and outcome measures, testing procedures and reported results. Graphi- complexity21,45 among people after stroke, but not in people with cally presented data were extracted using a digital plotting toolb. Parkinson’s disease.42 A recent systematic review of 20 studies assessed dual-task ability during concurrent locomotor and cognitive The Quality Assessment Tool for Observational Cohort and Cross- tasks in adults after stroke and reported that a mutual interference Sectional Studies developed by the National Institute of Health was pattern was consistently observed among studies involving more used to assess the methodological quality of included studies.48 Two challenging walking tasks (eg, obstacle-crossing).46 Mutual interfer- reviewers independently evaluated each eligible article. If no ence was also more frequent among people after stroke than healthy consensus was reached, the opinion of the principal investigator was controls. However, no meta-analyses quantifying the effects of DTI on sought. To provide a more objective appraisal of overall quality, per- different gait parameters or secondary task performance were un- centages for the total number of items rated ‘yes’ were calculated. dertaken in that review. Their findings were also limited to dual-task Overall ratings of ‘good’, ‘moderate’ and ‘poor’ were determined conditions involving the performance of a walking task in conjunc- based on resultant percentages 75%, 50 to 74% and , 50%, tion with a cognitive task. The effect of DTI on gait during attention- respectively.49 demanding manual tasks was not examined. According to the aforementioned multiple resource capacity theory, adding a manual Data synthesis and analysis task (which involves recruitment of both motor and cognitive sys- tems) may induce greater DTI than adding a cognitive task (which To facilitate comparisons across studies and delineate the effect of mainly involves the cognitive system) during walking. This in turn different component task domains on dual-task gait performance, may have a direct influence on risk aversion and walking safety. The included studies were categorised according to the secondary tasks way in which the degree and pattern of DTI during walking varies involved.28 DTI patterns were classified according to the catego- according to task domain and stage of recovery (ie, chronicity) in risation model proposed by Plummer et al.6 The nine possible DTI individuals after stroke remains unknown. A systematic review and patterns are depicted in Table 1. meta-analysis consolidating the mounting, but fragmented, evidence on the phenomenon of DTI during walking among people after stroke To estimate the overall effect of DTI on gait and secondary task at different stages of recovery is currently lacking in the available performance, primary meta-analyses and subsequent sensitivity an- literature. alyses based on the stage of stroke recovery were conducted for groups of three or more studies reporting the same outcome. For Therefore, the research questions for this systematic review were: groups of three or more studies that reported the same outcome and included both stroke and age-matched control groups, additional 1. What are the degree and pattern of dual-task interference during meta-analyses were conducted to compare the effect of DTI on gait walking in people after stroke? and secondary task performance between people with and without a history of stroke. For studies that reported more than one dataset per 2. How do these vary with disease chronicity and different compo- participant group for the same task category, one task was randomly nent tasks in people after stroke? selected for inclusion in the meta-analysis in order to avoid a unit-of- analysis error.50 When these studies were included in more than one 3. How does dual-task interference differ between people after stroke and people without stroke? Methods Box 1. Inclusion criteria. This systematic review and meta-analysis is reported in accor- Design dance with Preferred Reporting Items for Systematic Reviews and Experimental, observational or exploratory study Meta-analyses (PRISMA) guidelines.47 Published in English Data sources and searches Participants Adults with stroke A literature search of electronic databases – including CINAHL, Cochrane Library, EMBASE, MEDLINE (19461), PsycINFO (18061) and Intervention PubMed – and a forward search for additional articles on electronic Not applicable databases, including Scopus and Web of Science, were performed with the last search conducted on 23 April 2021. No restrictions on Outcome measures study design were imposed. We extracted only baseline results for Walking performance intervention or cohort studies. Search keywords were based on population (stroke) and constructs of interests (DTI during walking) Comparisons Walking performance under dual-task condition versus without the dual-task condition Dual-task interference in people with stroke versus people without stroke
28 Tsang et al: Dual-task interference during walking in stroke Table 1 equally well or given no explicit prioritisation instructions (see Nine dual-task interference patterns. Table 3 on the eAddenda). Cognitive/manual performance Quality assessment of included studies No change No change Improved Worsened All included studies stated their objectives and inclusion/exclusion criteria. The diagnosis of stroke and outcome measures were also Mobility Improved No Cognitive/manual Cognitive/manual clearly defined. Other criteria were achieved by fewer studies: 45% interference facilitation interference provided sample size justification, power description or variance and performance effect estimates; 33% adjusted for potential confounding variables; Mobility Mutual Mobility priority 17% reported who, where and when their study population was Worsened facilitation facilitation trade-off recruited; and 4% categorised stroke type. Twelve (16%) of the 76 studies attained a good overall rating and 42 (55%) had a moderate Mobility Cognitive/manual Mutual overall rating. The remaining studies (29%) had a poor overall rating interference priority trade-off interference (see Table 4 on the eAddenda). meta-analysis (ie, for comparing the effect of DTI among different Meta-analyses task categories, between different tasks within the same task cate- gory, and/or between the stroke and age-matched control groups), The meta-analyses are shown as streamlined forest plots in the same randomly selected tasks were carried forward in subse- Figures 2 to 11. Detailed forest plots, are shown in Figures 12 to 21 on quent meta-analyses. For each study, the raw mean difference the eAddenda. (MD)51–53 was calculated for each outcome of interest. A random- effects model of the generic inverse variance method was used. Effect of secondary task domain on mobility parameters in people Heterogeneity between studies was analysed with the I2 statistic.54 after stroke Forest plots indicating effect sizes with 95% CI, heterogeneity and corresponding p-values were generated using RevMan softwarec.53 Meta-analyses were conducted for several gait parameters, For analyses with significant overall effects, publication bias was including speed, cadence, stride length, stride time and Timed Up and assessed using Egger’s regression tests.55 Further analyses with Duval Go test (Figures 2 to 6, and Tables 3 and 4 on the eAddenda). Meta- and Tweedie’s trim and fill and the classic fail-safe N methods were analyses were not performed for any cognitive measures due to the also conducted to assess the effect of potential publication bias on the limited number of studies for a given task domain (, 3). Gait speed certainty of evidence.56 These analyses were performed using com- was the most commonly assessed dual-task walking measure mercial softwared.57 (Figure 2, and Table 3 on the eAddenda). Results Overall, there was a significant effect on gait speed when a sec- ondary task was imposed during walking. A decrement in gait speed Flow of studies through the review (m/s) was found when walking was performed simultaneously with mental tracking (MD 20.11, 95% CI 20.14 to 20.08), language A PRISMA flow chart summarising the screening process is pro- (MD 20.10, 95% CI 20.15 to 20.04), manual tasks (MD 20.13, 95% vided in Figure 1. Electronic searches identified 15,824 unique re- CI 20.18 to 20.08) or discrimination and decision-making (MD 20.05, cords. After screening and assessment, 76 of 133 shortlisted articles 95% CI 20.09 to 20.01). Manual tasks induced slightly greater DTI on involving a total of 492 healthy controls and 2,425 participants with gait speed compared to cognitive tasks. Among cognitive tasks, the stroke met the criteria for inclusion (see Table 2 on the eAddenda). effects of DTI on gait speed during mental tracking and language Among the 76 studies included in the systematic review, 50 provided tasks were more pronounced than during other cognitive task do- sufficient data for conducting meta-analyses (see Table 3 on the mains (Figure 2). eAddenda). Twenty-seven of the 50 studies involved only people with chronic stroke and were included in subsequent sensitivity analyses Cadence decreased with the addition of a mental tracking task (see Table 2 on the eAddenda). (MD 210, 95% CI 215 to 24), with similar results on the sensitivity analysis (Figure 3). Adding a language task had a similar effect Characteristics of the participants (MD 29, 95% CI 215 to 23), although no sensitivity analysis was performed. A sensitivity analysis of adding a manual task showed The mean age of participants in the included studies ranged from reduced cadence among participants with chronic stroke, although 49 to 77 years. Forty-seven of 76 (62%) studies included only people not on the primary analysis (Figure 3). 6 months after stroke onset, one study included only people within 72 hours of stroke onset,58 and six studies recruited people within 6 Stride length (m) decreased with the addition of a mental tracking months of stroke onset.14,59–63 The remaining studies included a mix task (MD 20.07, 95% CI 20.13 to 20.01), language task (MD 20.09, of people in different stages of stroke recovery. 95% CI 20.16 to 20.01) or manual task (MD 20.11, 95% CI 20.18 to 20.04). The sensitivity analyses (mental tracking and manual tasks Characteristics of the dual-task testing protocols only) had similar results (Figure 4). Twenty-two studies (29%) involved more than one secondary Adding a mental tracking task induced an increase in stride time task in their dual-task assessment protocols. Among these studies, (MD 0.05 seconds, 95% CI 0.01 to 0.09), with a similar result on the five reported more than one dataset (ie, dual-task assessment) for sensitivity analysis (Figure 5). Adding a manual task had a similar the same group of participants within a single task cate- effect, although it was not evident on the sensitivity analysis. Adding gory.10,27,63–66 One study reported dual-task walking performance a mental tracking task slowed performance on the Timed Up and Go for a manual task involving the more-affected and the less-affected test (MD 23 seconds, 95% CI 1 to 6), as shown in Figure 6. hands of the same group of participants.64 Hence, 24 datasets (14%) were removed from the meta-analyses to avoid unit-of-analysis Effect of different tasks within the same cognitive or manual task errors. Mental tracking was the most adopted secondary task cate- domain on gait speed in people after stroke gory. Gait speed was the most used measure of walking perfor- mance, while secondary task measures were diverse among the We also examined how DTI was affected by different tasks within studies. Participants were either instructed to perform both tasks the same domain. Based on the available data, analyses were con- ducted for mental tracking (Figure 7), discrimination and decision- making (Figure 8), language (Figure 9) and manual task domains (Figure 10). Gait speed was the primary outcome measure across all
Research 29 Records identified through Records identified database searching (n = 18,206) through citation • CINAHL (n = 1,439) tracking (n = 6) • Cochrane Library (n = 698) • Embase (n = 5,901) • Medline (n = 3,598) • PsycInfo (n = 1,748) • PubMed (n = 637) • Scopus (n = 2,972) • Web of Science (n = 1,303) Records identified (n = 18,212) Records excluded (n = 2,388) • duplicates (n = 2,085) Records screened by title • theses (n = 119) and abstract (n = 15,824) • book sections (n = 112) • conference proceedings (n = 72) Records excluded (n = 15,691) Meta-analysis exclusions (n = 26) Records assessed as full Records excluded (n = 57) • insufficient data for effect size text (n = 133) • conference proceedings (n = 25) • no dual-task walking outcome calculation (n = 25) Records included in the • single case study design (n = 1) review (n = 76) measures (n = 15) • mixed population with results Studies included in the meta- analysis (n = 50) unavailable for stoke only (n = 7) • not English language (n = 1) • no participants with stroke (n = 3) • no dual-task assessed (n = 1) • non-research article (n = 5) Figure 1. Flow of studies through the review. domains. Within the mental tracking task domain, serial 1 sub- Comparison of DTI between people with and without stroke tractions (20.17 m/s, 95% CI 20.25 to 20.10) tended to induce a greater DTI on gait speed than serial 3 subtractions (20.10 m/s, 95% Additional meta-analyses were conducted to compare the DTI CI 20.15 to 20.05), although there was substantial uncertainty between individuals with stroke and age-matched controls. Studies inherent in these estimates. The effect of the auditory 1-back task was included in the analyses assessed walking speed during manual or uncertain (20.08 m/s, 95% CI 20.22 to 0.06) (Figure 7). For the mental tracking tasks (Figure 11). Adding a manual task increased DTI discrimination and decision-making domain, classification tasks in gait speed (MD 20.07 m/s), although the estimate was imprecise induced a small DTI on gait speed (20.07 m/s, 95% CI 20.12 to 20.02), (95% CI 20.18 to 0.03) in people after stroke (at all stages) when while the auditory Stroop tests had no effect (20.02 m/s, 95% compared with their aged-matched peers without stroke (MD 20.07, CI 20.05 to 0.01) (Figure 8). For the language domain, spontaneous 95% CI 20.10 to 20.03). This trend was slightly less evident in speech (20.11 m/s, 95% CI 20.19 to 20.02) and category naming tasks sensitivity analysis for individuals with chronic stroke (Figure 11), (20.09 m/s, 95% CI 20.16 to 20.02) induced similar DTI (Figure 9). For suggesting that both analyses may have been underpowered. The the manual task domain, the degree of DTI on gait speed was also effect of adding a mental tracking task on DTI clearly did not differ similar for cup holding (20.14 m/s, 95% CI 20.21 to 20.08) and tray between participants with and without stroke (Figure 11). holding tasks (20.14 m/s, 95% CI 20.18 to 20.09) (Figure 10). Publication bias of meta-analyses Dual-task interference pattern in people after stroke Publication bias was present in the primary meta-analyses: for Twenty-four datasets from 15 studies10,12,20,21,23,41,60,63,67–73 re- gait speed during mental tracking, language and manual tasks; for ported whether there were significant differences between the cadence during mental tracking tasks; for stride length during walking and secondary task performance during single-task and manual tasks; for stride time during mental tracking and manual dual-task conditions (see Table 3 on the eAddenda). Corresponding tasks; and for Timed Up and Go during mental tracking tasks. For the DTI patterns for dual-task conditions were identified. sensitivity analyses, publication bias was observed: for gait speed during manual tasks; for cadence during mental tracking tasks; for Among these datasets, more than half were classified as showing a stride length during manual tasks; and for stride time during mental mutual interference pattern (ie, deterioration of both the mobility tracking (see Table 5 on the eAddenda). and the secondary task performances under a dual-task condition).10,12,21,23,60,67,69,71,72 The secondary tasks included mental Discussion tracking (n = 5),10,21,60,67,69 language (n = 2),23,71 visuospatial (n = 2),10,21 short-term memory (n = 1),12 and discrimination and decision- This systematic review and meta-analysis examined the effect of making tasks (n = 2).10,72 Six of the 24 datasets23,60,63,72 were classi- different secondary tasks on gait parameters in people after stroke. fied as showing a ‘mobility interference’ pattern. There was a The results showed that the degree and pattern of DTI during dual- degradation in mobility performance, but no changes in visuospa- task walking varied with the component tasks among individuals tial,23 mental tracking (n = 4),23,60,63 and discrimination and decision- with stroke. The degree of DTI on walking speed in people with making72 task performance. Five of the 24 datasets from the two chronic stroke was comparable to their able-bodied peers. studies41,73 showed a ‘cognitive interference’ pattern. There was a deterioration in three language tasks and two mental tracking tasks, In the analyses of the effect of secondary task domain on mobility but no significant changes in walking performance. One dataset parameters in people after stroke, gait speed was the most used showed ‘no interference’ pattern.21 Neither walking nor cognitive measure of dual-task walking (Figure 2, and Table 3 on the eAd- performances demonstrated any changes during the auditory clock denda). It also appeared to be the most sensitive parameter for test (see Table 3 on the eAddenda). detecting DTI in mobility, regardless of the secondary task domain
30 Tsang et al: Dual-task interference during walking in stroke Subgroup MD (95% CI) Subgroup MD (95% CI) Study Random Study Random Mental tracking Visuospatial perception Amatachaya 2016 CA Dennis 2009 Amatachaya 2016 NCA Plummer-D’Amato 2008ᵃ Batchelor 2016 Plummer-D’Amato 2012ᵃ Chatterjee 2019 Plummer-D’Amato 2014ᵃ Cho 2015a Pooled Cho 2015b DTT Cho 2015b STT Short-term memory Curuk 2019 Hyndman 2006 Dennis 2009 Kizony 2010 5 itemsᵃ Goh 2017 Maciel 2014ᵃ Liu 2017 CDTT Pooled Liu 2017 CPT Liu 2017 MDTT Language Liu 2018 Chatterjee 2018 Manaf 2015 Meester 2018 CG Shafizadeh 2017 Meester 2018 DTT Timmermans 2018 Angusamy 2010ᵃ Walshe 2019 Baetens 2013ᵃ Baetens 2013ᵃ Plummer-D’Amato 2008ᵃ Chisholm 2014ᵃ Plummer-D’Amato 2012ᵃ Hermand 2019 1-backᵃ Plummer-D’Amato 2014ᵃ Hollands 2014ᵃ Pooled Patel 2014ᵃ Plummer-D’Amato 2008ᵃ Manual Plummer-D’Amato 2012ᵃ Canning 2006 Robinson 2011 MT Curuk 2019 Pooled Goh 2017 Chronic only Lee 2015a CA Lee 2015a LCA Discrimination and decision making Liu 2017 CDTT Canning 2006 Liu 2017 CPT Lee 2015a CA Liu 2017 MDTT Lee 2015a LCA Liu 2018 Lord 2006 Lord 2006 Walshe 2019 Manaf 2015 Bowen 2001ᵃ Walshe 2019 Eckhardt 2011ᵃ Yang 2007b CG Feld 2018ᵃ Yang 2007b EG Patel 2014 Stroopᵃ Maciel 2014ᵃ Plummer-D’Amato 2014ᵃ Robinson 2011 Slipper pickingᵃ Zukowski 2019 ᵃ Yang 2007a Tray carrying FCᵃ Pooled Yang 2007a Tray carrying LLCᵃ Chronic only Pooled Chronic only –0.6 –0.4 –0.2 0.0 0.2 0.4 –0.6 –0.4 –0.2 0.0 0.2 0.4 Dual-task interference Dual-task facilitation Dual-task interference Dual-task facilitation Figure 2. Forest plot of effect of secondary tasks on gait speed (m/s) in people with stroke. A total of 37 studies (914 participants) were involved in the primary analysis, and 18 studies (344 participants) in the sensitivity analysis among people with chronic stroke. CA = community ambulating group, CDTT = cognitive dual-task training group, CG = control group, CPT = conventional physiotherapy group, DTT = dual-task training group, EG = exercise group, FC = full community ambulating group, LCA = Limited community ambulating group, LLC = least limited community ambulating group, MDTT = motor dual-task training group, MT = mental tracking, NCA = non-community ambulating group, SS1 = serial subtractions by 1, STT = single-task training group, TG = target group. a Study included people within 6 months of stroke onset or a mix of people with sub-acute and chronic stroke. used. This is consistent with previous findings among older adults The observed reductions in gait speed and Timed Up and Go time and other neurological populations.28,30,46 With the exception of (Figure 6) during mental tracking and language task suggest that walking conditions involving discrimination and decision-making tasks requiring internally driven responses cause relatively larger DTI tasks, changes in gait speed (20.10 to 20.13 m/s) met or exceeded the minimal clinically important difference value (0.1 m/s),74 as than those requiring externally driven responses (eg, visuospatial perception or discrimination and decision-making tasks).28,30 shown in Figure 2. Common neural networks may be involved during walking,75
Research 31 Subgroup MD (95% CI) Subgroup MD (95% CI) Study Random Study Random Mental tracking Mental tracking Al-Yahya 2016 Al-Yahya 2016 Cho 2015a Chatterjee 2019 Cho 2015b DTT Cho 2015a Cho 2015b STT Cho 2015b DTT Curuk 2019 Cho 2015b STT Liu 2017 CDTT Liu 2017 CDTT Liu 2017 CPT Liu 2017 CPT Liu 2017 MDTT Liu 2017 MDTT Liu 2018 Liu 2018 Baetens 2013ᵃ Walshe 2019 Hermand 2019 1-backᵃ Baetens 2013ᵃ Plummer-D’Amato 2008ᵃ Hermand 2019 1-backᵃ Pooled Plummer-D’Amato 2008ᵃ Chronic only Pooled Chronic only Language Pohl 2011b Language Angusamy 2010ᵃ Angusamy 2010ᵃ Baetens 2013ᵃ Baetens 2013ᵃ Plummer-D’Amato 2008ᵃ Plummer-D’Amato 2008ᵃ Pooled Pooled Manual Manual Curuk 2019 Liu 2017 CDTT Liu 2017 CDTT Liu 2017 CPT Liu 2017 CPT Liu 2017 MDTT Liu 2017 MDTT Liu 2018 Liu 2018 Walshe 2019 Lord 2006 Yang 2007b CG Yang 2007b CG Yang 2007b EG Yang 2007b EG Yang 2007a Tray carrying FCᵃ Pohl 2011a MAᵃ Yang 2007a Tray carrying LLCᵃ Yang 2007a Tray carrying FCᵃ Pooled Yang 2007a Tray carrying LLCᵃ Chronic only Pooled Chronic only –0.6 –0.4 –0.2 0 0.2 0.4 –60 –40 –20 0 20 40 Dual-task interference Dual-task facilitation Dual-task interference Dual-task facilitation Figure 4. Forest plot of effect of secondary tasks on stride length (m) in people with stroke. A total of 13 studies (320 participants) were involved in the primary analyses, Figure 3. Forest plot of effect of secondary tasks on cadence (steps/min) in people with and eight studies (204 participants) for the sensitivity analyses involving people with stroke. A total of 15 studies (355 participants) were involved in the primary analyses, chronic stroke. and eight studies (220 participants) in the sensitivity analyses among people with CDTT = cognitive dual-task training group, CG = control group, CPT = conventional chronic stroke. physiotherapy group, EG = exercise group, FC = full community ambulating group, CDTT = cognitive dual-task training group, CG = control group, CPT = conventional LLC = least limited community ambulating group, MDTT = motor dual-task training physiotherapy group, DTT = dual-task training group, EG = exercise group, FC = full group. community ambulating group, LLC = least limited community ambulating group, MA = a Study included people within 6 months of stroke onset or a mix of people with sub- more-affected hand movement, MDTT = motor dual-task training group, STT = sin- acute and chronic stroke. gletask training group. a Study included people within 6 months of stroke onset or a mix of people with sub- decreased cadence and increased stride time were only found during acute and chronic stroke. walking involving the mental tracking task. Different brain activation patterns were also identified during dual-task walking with manual dual-tasking,76 and when giving internally driven and externally versus mental tracking tasks. Using functional near-infrared spec- driven responses.77 The bottleneck model suggests that greater troscopy, Lu et al revealed an increased motor cortex activation overlap of these networks may result in more competition for during walking with manual and cognitive tasks. A strong and sus- cognitive resources, and thus more severe DTI. Further investigation tained activation of the left prefrontal cortex was also observed into the underlying neural mechanism of DTI is warranted. during walking with serial 7 subtractions. This is in contrast with only weak activation of this brain region during the initial phases of The DTI effect on gait induced by manual and mental tracking single-task walking and walking while balancing a bottle of water on tasks also varied according to the specific gait parameter that was a tray.78 assessed (Figures 2 to 5). These findings are consistent with those of Lu et al,78 who compared the effect of a mental tracking task (serial 7 Also in line with the findings of Lu et al,78 reductions in stride subtractions) and a manual task (holding a bottle of water on a tray) length were more pronounced (ie, a stronger DTI effect) during on walking in healthy younger adults. Both tasks resulted in speed manual tasks than mental tracking tasks (Figure 4). This may be and stride length reductions compared to walking alone. However, related to reductions in arm-swing amplitude when engaging in
32 Tsang et al: Dual-task interference during walking in stroke Subgroup MD (95% CI) Subgroup MD (95% CI) Study Random Study Random Mental tracking Mental tracking Liu 2017 CDTT Lindvall 2014 CG Liu 2017 CPT Lindvall 2014 EG Liu 2017 MDTT Manaf 2014a Liu 2018 Denneman 2018ᵃ Manaf 2015 Kal 2018 Externalᵃ Walshe 2019 Kal 2018 Internalᵃ Hermand 2019 1-backᵃ Pooled Hollands 2014ᵃ Plummer-D’Amato 2008ᵃ Discrimination and decision making Pooled Denneman 2018ᵃ Chronic only Eckhardt 2011ᵃ Kal 2018 Externalᵃ Manual Kal 2018 Internalᵃ Liu 2017 CDTT Kim 2014 DTTᵃ Liu 2017 CPT Kim 2014 STTᵃ Liu 2017 MDTT Pooled Liu 2018 Manaf 2015 –20 –10 0 10 20 Walshe 2019 Yang 2007b CG Dual-task facilitation Dual-task interference Yang 2007b EG Yang 2007a Tray carrying FCᵃ Figure 6. Detailed forest plot of effect of secondary tasks on Timed Up and Go (s) in Yang 2007a Tray carrying LLCᵃ people with stroke. A total of six studies (256 participants) were involved in the pri- Pooled mary analyses. No further sensitivity analyses were conducted. Chronic only CG = control group, DTT = dual-task training group, EG = experimental group, External = external focus instruction group, Internal = internal focus instruction group, –0.6 –0.4 –0.2 0 0.2 0.4 STT = single-task training group. a Study included people within 6 months of stroke onset or a mix of people with sub- Dual-task facilitation Dual-task interference acute and chronic stroke. Figure 5. Detailed forest plot of effect of secondary tasks on stride time (s) in people reasonable to assume that serial 3 would be more complex than serial with stroke. A total of nine studies (168 participants) were involved in the primary 1 subtractions. analyses, and five studies (97 participants) for the sensitivity analyses among people with chronic stroke. Overall, our results are counterintuitive, in that serial 1 sub- CG = control group, DTT = dual-task training group, EG = experimental group, tractions induced a greater DTI than serial 3 subtractions. Classifica- STT = single-task training group. tion tasks also induced greater DTI than auditory Stroop tasks. These a Study included people within 6 months of stroke onset or a mix of people with sub- findings suggest that DTI, especially during cognitive-mobility dual- acute and chronic stroke. task conditions, is a complex phenomenon affected by multiple factors. In addition to the aforementioned competition for manual tasks. Previous research has shown that when one or both neural circuitry,1,82 our findings are also supported by the capacity arm-swings were restrained, a decrease in stride length was observed sharing model of attention theories. This model suggests that inter- in comparison to normal walking without restraints.79 During manual ference occurs when cognitive demand exceeds a finite attention tasks (eg, cup holding), reduced stride length could be the result of capacity.1 As indicated by Kahnamen,1 allocation of finite attention is restricted arm movements. A controlled study comparing dual-task malleable and highly responsive to individual arousal level, ever- walking with manual tasks and standalone walking with arm- changing momentary intentions, and affected by component task swings restrained at similar arm positions may help delineate the demands that exceed the available capacity.1 During dual-task effect of dual-tasking from that of arm-swing restriction during walking, people might prioritise either component task according to manual tasks. their ability, as well as their own self-perceived challenges associated with the task. For instance, participants may have considered serial 3 In the analyses of the effect of different tasks within the same subtractions to be too challenging, and thus prioritised walking over secondary task domain in people after stroke, it was interesting to the cognitive task, overtly or covertly, resulting in a lower decrement consider whether the degree of DTI during walking is more severe to mobility performance. Yang et al45 also showed that counting ac- with increasing complexity of the secondary tasks involved. Task curacy was reduced when a serial 3 subtraction task was imposed complexity can be indicated by the level of performance under single- during an obstacle-crossing task, but not when the more complex task conditions.1 However, this was rarely reported in the included (serial 7 subtraction) task was used. On the other hand, spontaneous studies, and examination of the effect of task complexity on DTI for speech questions incorporate personal experiences, opinions, and task categories was limited. Inferences concerning task complexity content relevant to daily life (see Table 3 on the eAddenda). These were instead based on previous literature. questions may be more meaningful and engaging than category naming tasks. As a result, participants may covertly be more aroused For the discrimination and decision-making domain, neutral or may overtly prioritise cognitive tasks over walking in such a sce- stimuli used in classification tasks are considered to be less difficult nario, thereby reducing walking speed. than tasks involving congruent or incongruent stimuli such as those associated with auditory Stroop tasks.80 For the mental tracking For manual tasks, walking while holding a tray demonstrated an domain, two previous studies compared the complexity of serial 3 effect comparable to cup holding (Figure 10). The relative similarity and serial 7 subtraction.45,81 Both studies reported a significantly and task-dependent movement may explain this result. For both greater number of correct responses for serial 3 than for serial 7 tasks, arm-swing was restricted, rendering a shorter stride length and subtractions under single-task conditions, thus indicating the latter to reduced walking speed.79 Participants may have also intentionally be relatively more complex than the former. Therefore, it is
Research 33 Subgroup MD (95% CI) Subgroup MD (95% CI) Study Random Study Random Serial subtraction by 3 Auditory Stroop test Dennis 2009 Walshe 2019 Goh 2017 Feld 2018ᵃ Liu 2017 CDTT Patel 2014ᵃ Liu 2017 CPT Plummer-D’Amato 2014ᵃ Liu 2017 MDTT Zukowski 2019ᵃ Liu 2018 Pooled Manaf 2015 Shafizadeh 2017 Classification task Timmermans 2018 Canning 2006 Baetens 2013ᵃ Lee 2015a CA Pooled Lee 2015a LCA Chronic only Bowen 2001ᵃ Eckhardt 2011ᵃ Serial subtraction by 1 Patel 2014ᵃ Amatachaya 2016 CA Pooled Amatachaya 2016 NCA Cho 2015a –0.4 –0.2 0 0.2 0.4 Cho 2015b DTT Cho 2015b STT Dual-task interference Dual-task facilitation Patel 2014ᵃ Pooled Chronic only Auditory 1-back Figure 8. Forest plot of effect of different discrimination and decision-making tasks on Hermand 2019 1-backᵃ gait speed (m/s) in people with stroke. A total of nine studies (154 participants) were Plummer-D’Amato 2008ᵃ involved in the primary analyses. No further sensitivity analyses were conducted. Plummer-D’Amato 2012ᵃ CA = community ambulating group, LCA = limited community ambulating group. Pooled a Study included people within 6 months of stroke onset or a mix of people with sub- acute and chronic stroke. –0.6 –0.4 –0.2 0.0 0.2 0.4 modest difference in cognitive function and mobility between the two groups. A large proportion of studies involved participants with Dual-task interference Dual-task facilitation chronic stroke who had relatively good cognitive and ambulatory function. The results of the sensitivity analyses also showed no clear Figure 7. Detailed forest plot of effect of different mental tracking tasks on gait speed differences between able-bodied older adults and people with (m/s) in people with stroke. A total of 15 studies (362 participants) were involved in the chronic stroke (Figure 11). People with chronic stroke may have primary analyses and 10 studies (301 participants) for the sensitivity analysis among regained gait automaticity, as indicated by increased speed and people with chronic stroke. CA = community ambulatory group, CDTT = cognitive dual-task training group, Subgroup MD (95% CI) CPT = conventional physiotherapy training group, DTT = dual-task training group, Study Random MDTT = motor dual-task training group, NCA = non-community ambulatory group, STT = single-task training group. Category naming a Study included people within 6 months of stroke onset or a mix of people with Chatterjee 2018 sub-acute and chronic stroke. Angusamy 2010ᵃ Baetens 2013ᵃ adjusted their gait (ie, shorter stride length) to avoid dropping objects Pooled from the tray or spilling water from the cup.78 Spontaneous speech In the analyses of dual-task interference patterns in people after Meester 2018 CG stroke, most datasets showed mobility-related interference. These Meester 2018 DTT findings suggest that people after stroke tend to reduce walking Plummer-D’Amato 2008ᵃ speed when instructed to perform both tasks equally well or when Plummer-D’Amato 2012ᵃ not explicitly instructed to prioritise either task (see Table 3 on the Plummer-D’Amato 2014ᵃ eAddenda). Similar findings were reported by Yogev-Seligmann Pooled et al,83 when dual-task walking performance was compared during conditions with and without prior task prioritisation instructions –0.4 –0.2 0 0.2 0.4 among healthy younger and older adults. Both younger and older adults showed reductions in walking speed when no explicit priori- Dual-task interference Dual-task facilitation tisation was given. Figure 9. Forest plot of effect of different language tasks on gait speed (m/s) in people Our analyses of DTI patterns indicate that mutual interference was with stroke. A total of seven studies (176 participants) were involved in the primary the most common and was often induced by a secondary task analyses. No further sensitivity analyses were conducted for people with chronic requiring internally driven responses. These findings are consistent stroke. with those of previous systematic reviews28,46 and suggest that CG = control group, DTT = dual-task training group. cognitive tasks that require internally driven responses may impose a Study included people within 6 months of stroke onset or a mix of people with sub- greater cognitive loading, thereby resulting in greater DTI. acute and chronic stroke. Intriguingly, there were no clear differences in DTI for gait speed during manual or mental tracking tasks between people with and without a history of stroke (Figure 11). This may be explained by the
34 Tsang et al: Dual-task interference during walking in stroke Subgroup MD (95% CI) Subgroup MD (95% CI) Study Random Study Random Cup holding Mental tracking Canning 2006 Batchelor 2015 Curuk 2019 Dennis 2009 Goh 2017 Walshe 2019 Lee 2015a CA Robinson 2011 Mental Trackingᵃ Lee 2015a LCA Pooled Liu 2018 Chronic only Manaf 2015 Pooled Manual task Canning 2006 Tray holding Lee 2015a CA Liu 2017 CDTT Lee 2015a LCA Liu 2017 CPT Walshe 2019 Liu 2017 MDTT Robinson 2011 Slipper pickingᵃ Lord 2006 Pooled Yang 2007b CG Chronic only Yang 2007b EG Yang 2007a Tray carrying FCᵃ –0.4 –0.2 0 0.2 0.4 Yang 2007a Tray carrying LLCᵃ Pooled More DTI in stroke More DTI in non-stroke Chronic only –0.6 –0.4 –0.2 0 0.2 0.4 Figure 11. Forest plot of comparison of dual-task interference on gait speed (m/s) between people with and without stroke. A total of six studies (127 people with stroke, Dual-task interference Dual-task facilitation 109 participants without stroke) were involved in the primary analyses. Five studies (97 people with stroke, 79 participants without stroke) were included in the sensitivity Figure 10. Forest plot of effect of different manual tasks on dual-task gait speed (m/s) analyses. in people with stroke. A total of 10 studies (235 participants) were involved in the CA = community ambulating group, DTI = dual-task interference (ie, dual-task per- primary analyses, and nine studies (205 participants) in the sensitivity analysis among formance – single-task performance), LCA = limited community ambulating group. people with chronic stroke. a Study included people within 6 months of stroke onset or a mix of people with sub- CA = community ambulatory group, CDTT = cognitive dual-task training group, acute and chronic stroke. CG = control group, CPT = conventional physiotherapy training group, EG = exercise group, FC = full community group, LLC = least limited community group, LCA = limited conductance) during single-task and dual-task performance. Task community ambulatory group, MDTT = motor dual-task training group. prioritisation instructions were not reported in 72% of studies (see a Study included people within 6 months of stroke onset or a mix of people with sub- Table 3 on the eAddenda). There was also a lack of cohort studies acute and chronic stroke. investigating the effect of stroke chronicity and other characteristics of DTI during walking. reduced gait variability.84,85 DTI is also considered to be a surrogate measure of gait automaticity.3 Overall, the lack of between-group The studies included in this review involved different designs, difference reported here was not consistent with the findings of a testing protocols and outcome measures, and the participants with recent systematic review by DeBlock-Bellamy et al,46 which found a stroke had clinical characteristics that varied in terms of stroke type, larger DTI among people after stroke compared to age-matched stage and severity. These factors may explain the heterogeneity across healthy participants. However, their conclusion was based on a studies in the primary analyses. Most studies did not report the narrative synthesis of results derived from seven studies involving impact of walking on secondary tasks, nor the resultant DTI pattern very different testing protocols and outcome measures. The results of for individual participants. The lack of information on cognitive task the current meta-analyses were based on groups of three or more performance during single-tasks also limited our ability to examine studies reporting the same mobility outcome (ie, walking speed on the effect of task complexity on DTI. Most studies did not include an level-ground). Additionally, each analysis was based on a group of able-bodied control group; this limited the comparability of DTI ef- studies involving the same secondary task category (manual tasks, fects between people with and without a history of stroke. mental tracking tasks). In contrast, studies using different secondary task categories in the testing protocol were mixed in that systematic Publication bias was present in several analyses involving lan- review to generate the overall conclusion. These factors may explain guage, manual or mental tracking tasks. This may be explained by the the discrepancy between our current review and theirs. relatively small number of studies (mostly , 10) that could be identified and pooled in the meta-analyses. However, a large number Among the 76 included studies, few reported who, where and of additional studies (9 to 949) with zero effect would be needed to when their study participants were recruited; this limits study nullify the observed effects of DTI during walking conditions replicability. Descriptions of stroke-related characteristics (eg, involving these tasks. These findings suggest that publication bias severity or location of stroke) were also missing in most studies. Less may have resulted in an underestimation in the magnitude of DTI than half of the included studies made statistical adjustments to ac- during dual-task walking, rather than an overestimation. Therefore, count for potential confounding variables, justified sample sizes and certainty of the observed evidence should be considered robust.56 power calculations, or reported variance and effect estimates. Only 16% of the studies included sample sizes 50, which may have In conclusion, this review showed that the degree and pattern of limited the statistical power of these studies.86 DTI depends on the choice of component task domain and subcat- egory. A cognitive task of relatively lower complexity also caused a As noted earlier, self-perceived challenge may have covertly greater degree of DTI during walking than a more complex task of influenced task prioritisation, which in turn affected component task the same domain. Counterintuitively, people after stroke did not performance. However, only one study87 addressed this issue by show a greater DTI for walking speed than their able-bodied peers measuring perceived challenge with physiological measures (skin during manual or mental tracking tasks. Clinicians may need to select dual-task combinations with standardised procedures in order to capture specific deficits in dual-task performance, index func- tional abilities under more challenging situations than routine clin- ical tests, and individually tailor corresponding interventions for
Research 35 people after stroke. An inventory of daily habits, ecological in- 7. Amatachaya S, Chuadthong J, Thaweewannaku T, Srisim K, Phonthee S. Levels of teractions, and aptitude in different cognitive domains (as deter- community ambulation ability in patients with stroke who live in a rural area. mined by corresponding cognitive assessments) may help in the Malays J Med Sci. 2016;23:56–62. initial identification of specific dual-task combinations that are most appropriate. There is also a need to study possible associations be- 8. Lee KB, Lim SH, Ko EH, Kim YS, Lee KS, Hwang BY. Factors related to community tween DTI and stroke chronicity, cognitive/physical impairment and ambulation in patients with chronic stroke. Top Stroke Rehabil. 2015;22:63–71. specific neuropsychological deficits (eg, decline in memory or mental tracking ability). This may produce a more comprehensive 9. Haggard P, Cockburn J, Cock J, Fordham C, Wade D. Interference between gait and picture of dual-task mobility function and assist in delineating spe- cognitive tasks in a rehabilitating neurological population. J Neurol Neurosurg cific dual-task walking mechanisms among people after stroke. This Psychiatr. 2000;69:479–486. review also reveals the great diversity in testing protocols across included studies. Dual-task mobility assessment standardisation is 10. Patel P, Bhatt T. Task matters: influence of different cognitive tasks on cognitive- needed for evaluating the effect of treatment on dual-task function, motor interference during dual-task walking in chronic stroke survivors. Top and for comparing function within and between groups at differing Stroke Rehabil. 2014;21:347–357. stages of stroke recovery. 11. Harley C, Boyd J, Cockburn J, Collin C, Haggard P, Wann JP, et al. Disruption of sitting What was already known on this topic: Performing a sec- balance after stroke: influence of spoken output. J Neurol Neurosurg Psychiatr. ond task while walking can impair performance of one or both 2006;77:674–676. tasks. In people with stroke, this dual-task interference can impact independence, participation and safety. Dual-task inter- 12. Hyndman D, Ashburn A, Yardley L, Stack E. Interference between balance, gait and ference in walking may differ with the nature of the concurrent cognitive task performance among people with stroke living in the community. task and the time since stroke, but the extensive evidence about Disabil Rehabil. 2006;28:849–856. this has not been adequately summarised. What this study adds: In people with stroke, manual and 13. Andersson AG, Kamwendo K, Seiger A, Appelros P. How to identify potential mental tracking tasks imposed the greatest dual-task interfer- fallers in a stroke unit: validity indexes of four test methods. J Rehabil Med. ence on gait speed. Although the nature of the concurrent task 2006;38:186–191. affected the extent of dual-task interference, the time since stroke did not. When evaluating mobility and planning rehabili- 14. Baetens T, De Kegel A, Palmans T, Oostra K, Vanderstraeten G, Cambier D. Gait tation, clinicians should select dual-task assessments that analysis with cognitive-motor dual tasks to distinguish fallers from nonfallers correspond to the daily habits and physical demands of people among rehabilitating stroke patients. Arch Phys Med Rehabil. 2013;94:680–686. after stroke. 15. Tsang CSL, Pang MYC. Association of subsequent falls with evidence of dual-task Footnotes: a Endnote X9, Thomas Reuters, New York, USA. interference while walking in community-dwelling individuals after stroke. Clin b GetData Graph digitizer 2.26, ShareIt!, Bochum, Germany. Rehabil. 2020;34:971–980. c Review Manager V.5.3, The Nordic Cochrane Centre, Copenha- gen, Denmark. 16. Lord SE, Weatherall M, Rochester L. Community ambulation in older adults: which d Comprehensive Meta-Analysis V.3, Biostat, Englewood, NJ, USA. internal characteristics are important? Arch Phys Med Rehabil. 2010;91:378–383. eAddenda: Figures 12 to 21, Tables 2 to 5 and Appendix 1 can be found online at https://doi.org/10.1016/j.jphys.2021.12.009. 17. Sturm JW, Donnan GA, Dewey HM, Macdonell RA, Gilligan AK, Srikanth V, et al. Ethics approval: Not applicable. Quality of life after stroke. Stroke. 2004;35:2340–2345. Competing interests: None. Sources of support: This study was supported by a research grant 18. Pound P, Gompertz P, Ebrahim S. A patient-centred study of the consequences of provided by the Research Grants Council (151594/16M). CSLT and TM stroke. Clin Rehabil. 1998;12:338–347. were supported by The Hong Kong Polytechnic University PhD stu- dentships (RUNV, RL27). The funding sources have no role in the 19. Bowen A, Wenman R, Mickelborough J, Foster J, Hill E, Tallis R. Dual-task effects of collection, analysis, and interpretation of data, in the writing of the talking while walking on velocity and balance following stroke. Age Ageing. report, or in the decision to submit the article for publication. 2001;30:319–323. http://onlinelibrary.wiley.com/o/cochrane/clcentral/articles/101/ Acknowledgements: We genuinely thank Dr Raymond Chung for CN-00371101/frame.html his advice and guidance regarding the data analyses, Dr Deborah Jehu who assisted with the data collection at the beginning of the project 20. Cockburn J, Haggard P, Cock J, Fordham C. Changing patterns of cognitive-motor and Ms Virginia Lau for her assistance in conducting the forward interference (CMI) over time during recovery from stroke. Clin Rehabil. search. 2003;17:167–173. Provenance: Not invited. Peer reviewed. Correspondence: Marco Yiu-Chung Pang, Department of Rehabil- 21. Dennis A, Dawes H, Elsworth C, Collett J, Howells K, Wade DT, et al. Fast walking itation Sciences, The Hong Kong Polytechnic University, Hong Kong. under cognitive-motor interference conditions in chronic stroke. Brain Res. Email: [email protected] 2009;1287:104–110. http://onlinelibrary.wiley.com/o/cochrane/clcentral/articles/632/ CN-00742632/frame.html References 22. Hyndman D, Ashburn A. “Stops walking when talking” as a predictor of falls in 1. Kahneman D. Attention and effort. Vol. 1063. CiteSeer; 1973. people with stroke living in the community. J Neurol Neurosurg Psychiatr. 2. Pashler H. Dual-task interference in simple tasks: data and theory. Psychol Bull. 2004;75:994–997. 1994;116:220. 23. Plummer-D’Amato P, Altmann LJP, Saracino D, Fox E, Behrman AL, Marsiske M. 3. Paul SS, Ada L, Canning CG. Automaticity of walking – implications for physio- Interactions between cognitive tasks and gait after stroke: A dual task study. Gait Posture. 2008;27:683–688. therapy practice. Phys Ther Rev. 2005;10:15–23. 4. Wickens CD. Multiple resources and performance prediction. Theor Issues Ergon Sci. 24. Regnaux JP, David D, Daniel O, Smail DB, Combeaud M, Bussel B. Evidence for cognitive processes involved in the control of steady state of walking in healthy 2002;3:159–177. subjects and after cerebral damage. Neurorehabil Neural Repair. 2005;19:125–132. 5. Merrian-Webster. Cross talk. In: Merrian-Webster Dictionary. 2009. 6. Plummer P, Eskes G, Wallace S, Giuffrida C, Fraas M, Campbell G, et al. Cognitive- 25. Smulders K, van Swigchem R, de Swart BJM, Geurts ACH, Weerdesteyn V. Com- munity-dwelling people with chronic stroke need disproportionate attention motor interference during functional mobility after stroke: state of the science and while walking and negotiating obstacles. Gait Posture. 2012;36:127–132. implications for future research. Arch Phys Med Rehabil. 2013;94:2565–2574.e6. 26. Takatori K, Okada Y, Shomoto K, Ikuno K, Nagino K, Tokuhisa K. Effect of a cognitive task during obstacle crossing in hemiparetic stroke patients. Physiother Theory Pract. 2012;28:292–298. 27. Yang YR, Chen YC, Lee CS, Cheng SJ, Wang RY. Dual-task-related gait changes in individuals with stroke. Gait Posture. 2007;25:185–190. 28. Al-Yahya E, Dawes H, Smith L, Dennis A, Howells K, Cockburn J. Cognitive motor interference while walking: a systematic review and meta-analysis. Neurosci Bio- behav Rev. 2011;35:715–728. 29. Smith E, Cusack T, Blake C. The effect of a dual task on gait speed in community dwelling older adults: a systematic review and meta-analysis. Gait Posture. 2016;44:250–258. 30. Smith E, Cusack T, Cunningham C, Blake C. The influence of a cognitive dual task on the gait parameters of healthy older adults: a systematic review and meta-analysis. J Aging Phys Act. 2017;25:671–686. 31. Ghai S, Ghai I, Effenberg AO. Effects of dual tasks and dual-task training on postural stability: a systematic review and meta-analysis. Clin Intervent Aging. 2017;12:557–577. 32. Fritz NE, Cheek FM, Nichols-Larsen DS. Motor-cognitive dual-task training in per- sons with neurologic disorders: a systematic review. J Neurolog Phys Ther. 2015;39:142–153. 33. Lees A, Smith EJB. Cognitive deficits in the early stages of Parkinson’s disease. Brain. 1983;106:257–270. 34. de Haan EH, Nys GM, Van Zandvoort MJ. Cognitive function following stroke and vascular cognitive impairment. Curr Opin Neurol. 2006;19:559–564. 35. Chiaravalloti ND, DeLuca J. Cognitive impairment in multiple sclerosis. Lancet Neurol. 2008;7:1139–1151. 36. Lapointe R, Lajoie Y, Serresse O, Barbeau H. Functional community ambulation requirements in incomplete spinal cord injured subjects. Spinal Cord. 2001;39:327. 37. Yogev-Seligmann G, Hausdorff JM, Giladi N. The role of executive function and attention in gait. J Mov Disord. 2008;23:329–342. 38. Chee R, Murphy A, Danoudis M, Georgiou-Karistianis N, Iansek RJB. Gait freezing in Parkinson’s disease and the stride length sequence effect interaction. Brain. 2009;132:2151–2160. 39. Goldie PA, Matyas TA, Evans OM. Gait after stroke: initial deficit and changes in temporal patterns for each gait phase. Arch Phys Med Rehail. 2001;82:1057–1065.
36 Tsang et al: Dual-task interference during walking in stroke 40. Patterson SL, Forrester LW, Rodgers MM, Ryan AS, Ivey FM, Sorkin JD, et al. De- 64. Pohl PS, Kemper S, Siengsukon CE, Boyd L, Vidoni ED, Herman RE. Dual-task de- terminants of walking function after stroke: differences by deficit severity. Arch Phys mands of hand movements for adults with stroke: A pilot study. Topics Stroke Med Rehabil. 2007;88:115–119. Rehabil. 2011;18:238–247. 41. Tisserand R, Armand S, Allali G, Schnider A, Baillieul S. Cognitive-motor dual-task 65. Kizony R, Levin MF, Hughey L, Perez C, Fung J. Cognitive load and dual-task per- interference modulates mediolateral dynamic stability during gait in post-stroke formance during locomotion poststroke: a feasibility study using a functional individuals. Hum Mov Sci. 2018;58:175–184. virtual environment. Phys Ther. 2010;90:252–260. 42. O’Shea S, Morris ME, Iansek R. Dual task interference during gait in people with 66. Robinson CA, Shumway-Cook A, Matsuda PN, Ciol MA. Understanding physical Parkinson disease: effects of motor versus cognitive secondary tasks. Phys Ther. factors associated with participation in community ambulation following stroke. 2002;82:888–897. Disabil Rehabil. 2011;33:1033–1042. 43. Amatachaya S, Srisim K, Thaweewannakij T, Arrayawichanon P, Amatachaya P, 67. Al-Yahya E, Johansen-Berg H, Kischka U, Zarei M, Cockburn J, Dawes H. Prefrontal Mato L. Failures in dual-task obstacle crossing could predict risk of future fall in cortex activation while walking under dual-task conditions in stroke: a multimodal independent ambulatory individuals with spinal cord injury. Clin Rehabil. imaging study. Neurorehabil Neural Repair. 2016;30:591–599. 2019;33:120–127. 68. Hackney ME, Hall CD, Echt KV, Wolf SL. Application of adapted tango as therapeutic 44. Allali G, Laidet M, Assal F, Armand S, Lalive PH. Walking while talking in patients intervention for patients with chronic stroke. J Geriatr Phys Ther. 2012;35:206–217. with multiple sclerosis: the impact of specific cognitive loads. Neurophysiol Clin. 2014;44:87–93. 69. Manaf H, Justine M, Goh H-T. Effects of attentional loadings on gait performance before turning in stroke survivors. PM & R. 2015;7:1159–1166. 45. Yang L, Lam FM, Huang M, He C, Pang MY. Dual-task mobility among individuals with chronic stroke: changes in cognitive-motor interference patterns and rela- 70. Yang L, He C, Pang MY. Reliability and validity of dual-task mobility assessments in tionship to difficulty level of mobility and cognitive tasks. Eur J Phys Rehabil Med. people with chronic stroke. PLoS One. 2016;11:e0147833. 2018;54:526–535. 71. Pohl PS, Kemper S, Siengsukon CF, Boyd L, Vidoni E, Herman RE. Older adults with 46. Deblock-Bellamy A, Lamontagne A, Blanchette AK. Cognitive-locomotor dual-task and without stroke reduce cadence to meet the demands of talking. J Geriatr Phys interference in stroke survivors and the influence of the tasks: a systematic re- Ther. 2011;34:35–40. view. Front Neurol. 2020;11. 72. Feld JA, Zukowski LA, Howard AG, Giuliani CA, Altmann LJ, Najafi B, et al. Rela- 47. Moher D, Liberati A, Tetzlaff J, Altman D. Preferred reporting items for systematic tionship between dual-task gait speed and walking activity poststroke. Stroke. reviews and meta-analyses: the PRISMA statement. BMJ. 2009;339:b2535. 2018;49:1296–1298. 48. Wardle J, Steel A. Systematic reviews in integrative medicine: a clinician’s guide to 73. Kemper S, McDowd J, Pohl P, Herman R, Jackson S. Revealing language deficits publication. Adv Integr Med. 2015;2:103–109. following stroke: the cost of doing two things at once. Aging Neuropsychol Cogn. 2006;13:115–139. 49. Koo TK, Li MY. A guideline of selecting and reporting intraclass correlation co- efficients for reliability research. J Chiropract Med. 2016;15:155–163. 74. Bohannon RW, Glenney SS. Minimal clinically important difference for change in comfortable gait speed of adults with pathology: a systematic review. J Eval Clin 50. Deeks JJ, Higgins JP, Altman DG, Group CSM. Analysing data and undertaking meta- Pract. 2014;20:295–300. analyses. Cochrane Handbook for Systematic Reviews of Interventions. Cochrane. 2019:241–284. 75. Takakusaki K, Tomita N, Yano M. Substrates for normal gait and pathophysiology of gait disturbances with respect to the basal ganglia dysfunction. J Neurol. 51. Bond Jr CF, Wiitala WL, Richard FD. Meta-analysis of raw mean differences. Psychol 2008;255:19–29. Methods. 2003;8:406. 76. Leone C, Feys P, Moumdjian L, D’Amico E, Zappia M, Patti F. Cognitive-motor dual- 52. Borenstein MHL, Higgins JPT, Rothstein HR. Effect sizes based on means. In: task interference: a systematic review of neural correlates. Neurosci Biobehav Rev. Introduction to Meta-Analysis. Hoboken, USA: John Wiley & Sons, Ltd.; 2009. 2017;75:348–360. 53. Deeks JJ, Altman DG, Bradburn MJ. Statistical methods for examining heterogeneity 77. Benedek M, Jauk E, Beaty RE, Fink A, Koschutnig K, Neubauer AC. Brain mecha- and combining results from several studies in meta-analysis. In: Systematic Reviews nisms associated with internally directed attention and self-generated thought. Sci in Health Care: Meta-Analysis in Context. Second Edition. Hoboken, USA: Wiley, Ltd; Reports. 2016;6:22959. 2008:285–312. 78. Lu C-F, Liu Y-C, Yang Y-R, Wu Y-T, Wang R-Y. Maintaining gait performance by 54. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta- cortical activation during dual-task interference: a functional near-infrared spec- analyses. BMJ. 2003;327:557. troscopy study. PLoS One. 2015;10:e0129390. 55. Egger M, Smith GD, Schneider M, Minder CJB. Bias in meta-analysis detected by a 79. Eke-Okoro ST, Gregoric M, Larsson L-E. Alterations in gait resulting from deliberate simple, graphical test. BMJ. 1997;315:629–634. changes of arm-swing amplitude and phase. Clin Biomech. 1997;12:516–521. 56. Murad MH, Chu H, Lin L, Wang Z. The effect of publication bias magnitude and 80. Hershman R, Levin Y, Tzelgov J, Henik A. Neutral stimuli and pupillometric task direction on the certainty in evidence. BMJ Evid Based Med. 2018;23:84–86. conflict. Psychol Res. 2021;85:1084–1092. 57. Borenstein M. Software for publication bias. In: Rothstein HR, Sutton AJ, 81. Bristow T, Jih CS, Slabich A, Gunn J. Standardization and adult norms for the Borenstein M, eds. Publication Bias in Meta-Analysis: Prevention, Assessment and sequential subtracting tasks of serial 3’s and 7’s. Appl Neuropsychol Adult. Adjustments. Chichester: Wiley; 2005. 2016;23:372–378. 58. Ben Assayag E, Shenhar-Tsarfaty S, Korczyn AD, Kliper E, Hallevi H, Shopin L, et al. 82. Holtzer R, Verghese J, Xue X, Lipton RB. Cognitive processes related to gait velocity: Gait measures as predictors of poststroke cognitive function: evidence from the results from the Einstein Aging Study. Neuropsychol. 2006;20:215. TABASCO study. Stroke. 2015;46:1077–1083. 83. Yogev-Seligmann G, Rotem-Galili Y, Mirelman A, Dickstein R, Giladi N, 59. Chisholm AE, Makepeace S, Inness EL, Perry SD, McIlroy WE, Mansfield A. Spatial- Hausdorff JM. How does explicit prioritization alter walking during dual-task temporal gait variability poststroke: variations in measurement and implications performance? Effects of age and sex on gait speed and variability. Phys Ther. for measuring change. Arch Phys Med Rehabil. 2014;95:1335–1341. 2010;90:177–186. 60. Chaikeeree N, Chinsongkram B, Saengsirisuwan V, Boonsinsukh R. Effect of 84. Clark DJ. Automaticity of walking: functional significance, mechanisms, measure- cognitive task on components of 7 meter timed up-and-go test in persons with ment and rehabilitation strategies. Front Human Neurosci. 2015;9. stroke. Sci Asia. 2018;44:247–256. 85. Lord S, Rochester L, Weatherall M, McPherson K, McNaughton H. The effect of 61. Denneman RPM, Kal EC, Houdijk H, Kamp JVD. Over-focused? The relation be- environment and task on gait parameters after stroke: a randomized comparison tween patients’ inclination for conscious control and single- and dual-task motor of measurement conditions. Arch Phys Med Rehabil. 2006;87:967–973. http://www. performance after stroke. Gait Posture. 2018;62:206–213. archives-pmr.org/article/S0003-9993(06)00281-4/pdf 62. Kal E, Houdijk H, van der Kamp J, Verhoef M, Prosée R, Groet E, et al. Are the effects 86. Terwee CB, Mokkink LB, Knol DL, Ostelo RW, Bouter LM, de Vet HC. Rating the of internal focus instructions different from external focus instructions given during methodological quality in systematic reviews of studies on measurement balance training in stroke patients? A double-blind randomized controlled trial. Clin properties: a scoring system for the COSMIN checklist. Qual Life Res. Rehabil. 2019;33:207–221. 2012;21:651–657. 63. Hermand E, Tapie B, Dupuy O, Fraser S, Compagnat M, Salle JY, et al. Prefrontal 87. Chatterjee SA, Daly JJ, Porges EC, Fox EJ, Rose DK, McGuirk TE, et al. Mobility cortex activation during dual task with increasing cognitive load in subacute stroke function and recovery after stroke: preliminary insights from sympathetic nervous patients: a pilot study. Front Aging Neurosci. 2019;11. system activity. J Neurol Phys Ther. 2018;42:224–232.
Journal of Physiotherapy 68 (2022) 37–42 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 Intravaginal electrical stimulation increases voluntarily pelvic floor muscle contractions in women who are unable to voluntarily contract their pelvic floor muscles: a randomised trial Flávia Ignácio Antônio a, Kari Bø b,c, Caroline Caetano Pena a, Sabrina M Bueno a, Elaine Cristine Lemes Mateus-Vasconcelos a, Ana Carolina Nociti Lopes Fernandes a, Cristine Homsi Jorge Ferreira a a Department of Health Sciences, Ribeirão Preto Medical School, University of São Paulo, Ribeirão Preto, Brazil; b Norwegian School of Sport Sciences Department of Sports Medicine, Oslo, Norway; c Akershus University Hospital, Department of Obstetrics and Gynecology, Lørenskog, Norway KEY WORDS ABSTRACT Electrical stimulation therapy Question: In women who are unable to contract their pelvic floor muscles voluntarily, what is the effect of an Pelvic floor intravaginal electrical stimulation regimen on their ability to contract the pelvic floor muscles and on self- Muscle contraction reported urinary incontinence? Design: Randomised controlled trial with concealed allocation, blinded as- Urinary incontinence sessors and intention-to-treat analysis. Participants: Sixty-four women with pelvic floor muscle function Physical therapy assessed by bi-digital palpation to be grade 0 or 1 on the Modified Oxford Scale. Intervention: For 8 weeks, participants randomised to the experimental group received weekly 20-minute sessions of intravaginal electrical stimulation with instructions to attempt pelvic floor muscle contractions during the bursts of electrical stimulation in the final 10 minutes of each session. The control group received no intervention. Outcome measures: The primary outcome was ability to voluntarily contract the pelvic floor muscles, evaluated through vaginal palpation using the Modified Oxford Scale. Secondary outcomes were prevalence and severity of urinary incontinence symptoms assessed by the International Consultation on Incontinence Questionnaire on Urinary Incontinence-Short Form (ICIQ-UI-SF) score from 0 to 21. Results: Sixty-one par- ticipants provided outcome data. After the intervention, the ability to contract the pelvic floor muscles was acquired by 36% of the experimental group and 12% of the control group (absolute risk difference 0.24, 95% CI 0.02 to 0.43). The experimental group also improved by a mean of 2 points more than the control group on the ICIQ-UI-SF score (95% CI 0.02 to 3.97). Conclusion: In women who are unable to contract their pelvic floor muscles voluntarily, 8 weeks of intravaginal electrical stimulation with voluntary contraction attempts improved their ability to contract their pelvic floor muscles and reduced the overall severity and impact of urinary incontinence on quality of life. Although the main estimates of these effects indicate that the effects are large enough to be worthwhile, the precision of these estimates was low, so it is not possible to confirm whether the effects are trivial or worthwhile. Trial registration: NCT03319095. [Ignácio Antônio F, Bø K, Pena CC, Bueno SM, Mateus-Vasconcelos ECL, Fernandes ACNL, Ferreira CHJ (2022) Intravaginal electrical stimulation increases voluntarily pelvic floor muscle contractions in women who are unable to voluntarily contract their pelvic floor muscles: a randomised trial. Journal of Physiotherapy 68:37–42] © 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 contract the PFM, as this method captures both the squeeze and the inward movement.4 There is level 1 evidence and grade A recommendation that pelvic floor muscle training (PFMT) should be the first therapeutic option for Despite being carefully taught about the anatomy and function of treatment of female stress urinary incontinence (UI).1,2 However, an essential requirement for initiating PFMT is the ability to contract the the pelvic floor, . 30% of women with pelvic floor dysfunction are pelvic floor muscles (PFM) correctly.3 When a patient is able to contract this musculature on verbal command, constriction and in- unable to distinguish PFM contraction from contractions of other ward (ventrocephalad) movement of the pelvic openings is demon- strated, which can be assessed with visual observation.4 However, muscles such as the rectus abdominis, gluteus maximus and hip vaginal palpation is considered essential in identifying the ability to adductors.5 Tibaek and Dehlendorff found that 70% of women with pelvic floor dysfunction were unable to contract their PFM correctly and 97% could perform only a weak contraction.6 Only half of women who are able to contract their PFM perform a contraction of sufficient force to increase the urethral pressure.7 https://doi.org/10.1016/j.jphys.2021.12.004 1836-9553/© 2021 Australian Physiotherapy Association. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/4.0/).
38 Ignácio Antônio et al: Intravaginal electrical stimulation for pelvic floor contraction In addition to the lack of knowledge of how to perform a correct on) of 5 seconds, relaxation time (Time off) of 10 seconds, current and efficient contraction of the PFM in the general population, the intensity defined by the motor threshold adjusted according to the pelvic region is susceptible to impairments resulting from pregnancy occurrence of accommodation and the participant’s tolerance, with a and childbirth, physical efforts with increases in intra-abdominal total stimulation time of 20 minutes. During the last 10 minutes of pressure and ground reaction forces, and decrease in oestrogen pro- the session, the participants were encouraged to attempt to volun- duction. All of these factors may contribute to inadequate PFM tarily contract during the bursts of electrical activity, with in- function, constituting aetiological factors for the development of structions from the physiotherapist. pelvic organ prolapse, UI, anal incontinence and sexual dysfunction.8 The control group did not receive any treatment during the Because PFM are not directly visible to the patient, teaching cor- intervention period. After cessation of the intervention and comple- rect contractions can be a challenge for physiotherapists. According to tion of the post-intervention outcome measurements, the control a systematic review, manual feedback and biofeedback improve PFM group participants were referred to physiotherapy. contractions.9 Recent evidence suggests that electrical stimulation may be more effective than no treatment for UI;10 however, to date Outcomes measures there is limited knowledge on the effect of intravaginal electrical stimulation (iES) to improve the ability to contract the PFM in women Primary outcome with pelvic floor dysfunction.11 The primary outcome was ability to contract the PFM, as assessed The few studies on iES have flaws such as a small sample size12 by bidigital palpation. Vaginal palpation is considered a valid and no comparison control group.12,13 One randomised trial investi- method to assess ability to contract, with an intra-rater ICC of 0.69 gated three interventions that could promote PFM contraction in between sessions.15 Assessment of the ability to contract involved women unable to contract and among those interventions iES was the following sequence of procedures.16 First, the participant found to be the least effective in promoting contraction;5 however, received information about the procedure and basic PFM anatomy, as iES was used without simultaneous attempts to voluntarily contract well as instructions on the correct way to perform a PFM contraction. the PFM. After consent was obtained, the participant was placed in the supine position with the hips and knees flexed, feet supported and legs Therefore, the research question for this randomised trial was: apart. The physiotherapist then asked the participant to perform a PFM contraction and visually observed the contraction. The in- In women who are unable to contract their pelvic floor muscles struction given to the participant before the observation was: voluntarily, what is the effect of an intravaginal electrical stimu- ‘Squeeze your PFM in the vagina as if you were holding urine’.17,18 lation regimen on their ability to contract the pelvic floor muscles The participant was instructed to squeeze and lift the PFM and to and on self-reported urinary incontinence? maintain the contraction for 3 seconds. During assessment, the command was: ‘Contract and maintain the strongest contraction you Method can’. The physiotherapist graded the PFM contraction during the contraction. Finally, the participant was instructed to completely Design relax the PFM. PFM function was graded according to the MOS,14 as shown in Box 1. A MOS score of 2 was used as evidence of the A randomised controlled trial was conducted with concealed ability to perform a correct PFM contraction. allocation, assessor blinding and intention-to-treat analysis. Women who were routinely referred to a tertiary care unit with pelvic floor Secondary outcomes dysfunction were approached about participating in the study. Reports of UI were evaluated using the Portuguese version of the Women who expressed interest in participating received verbal and written information, and were required to give their informed con- International Consultation on Incontinence Questionnaire on Urinary sent before being allocated to a group and undergoing their baseline Incontinence – Short Form (ICIQ-UI-SF), which was translated and assessment. Allocation was conducted using computer-generated validated by Tamanini et al.19 It evaluates the symptoms, severity of random numbers from a randomisation website, thereby conceal- UI, and impact that UI has on women’s quality of life. It is a short ing the upcoming random allocations. Participants in the experi- questionnaire that enables a consistent and unified assessment of UI mental group were allocated to receive an 8-week regimen of iES, symptoms and their impact on quality of life, and facilitates com- whereas participants allocated to the control group received no parison of data from different studies. It also has good construct intervention. At the first session, demographic data were collected validity and discriminates among different groups of UI. It has high and baseline measures were recorded for all participants. All par- internal consistency, good reliability and moderate to very good ticipants underwent these measures again after the 8-week inter- stability in test-retest analysis. The Cronbach’s alpha is 0.95.20 vention period. Participants, therapist Women were eligible for inclusion in the study if they were aged Adherence . 18 years, had pelvic floor dysfunction with PFM function grade 0 or Adherence to the intervention was recorded by the researcher 1 classified with the Modified Oxford Scale (MOS).14 Exclusion criteria were: neurological diseases; symptoms of vaginal or urinary tract who applied the intervention. Eight iES sessions were to be infection; pelvic organ prolapse . stage 2; suspected or confirmed completed within 8 weeks. pregnancy; and cognitive impairments. The intervention sessions were conducted by a single physiotherapist who had 20 years of Box 1. Modified Oxford Scale for grading the function of the experience in women’s health physiotherapy and who had no contact pelvic floor muscles. with the assessor or the participants’ results. Interventions 0 no contraction 1 very weak contraction The intervention was performed using a commercial electrical 2 weak contraction stimulatora once a week for 8 weeks in a physiotherapy clinic at the 3 moderate contraction hospital. A biphasic current was used and the stimulation parameters 4 good contraction were: 50 Hz frequency, pulse width of 200 ms, contraction time (Time 5 strong contraction
Research 39 Women assessed for eligibility (n = 172) Excluded (n = 108) declined to participate (n = 69) ineligible (n = 39) Measured ability to contract pelvic floor muscles and severity of urinary incontinence Week 0 Randomised (n = 64) (n = 31) (n = 33) Lost to follow-up (n = 3) Experimental group Control group electrical no intervention declined further stimulation participation (n = 3) 1 session per week 8 weeks 8 weeks Measured ability to contract pelvic floor muscles and severity of urinary incontinence Week 8 (n = 28) (n = 33) Figure 1. Design and flow of participants through the trial. Data analysis and 0.44 in the experimental group). Adopting a significance level of 5% and a test power of 90%, this sample size calculation indicated 14 The analyses were conducted using R softwareb. Data were first participants per group. The second sample size calculation was in tested for normality. For comparison between groups, continuous relation to a higher threshold (ie, an improvement to 3 on the MOS, variables were analysed using a t-test; the Anderson-Darling test was anticipating the proportion with a favourable outcome of 0.0 in the used for variables with normal distribution and a non-parametric test control group and 0.22 in the experimental group). Adopting a sig- (Mann-Whitney/Brunner-Munzel) was used for the remaining vari- nificance level of 5% and a test power of 80%, the sample size calcu- ables. A Fisher test was used for categorical variables. The primary lation indicated 28 participants per group. outcome was analysed using logistic regression and the secondary outcomes were analysed using a mixed regression model. The level of Results significance was set to 0.05. Compliance with the trial protocol Two sample size calculations were performed. These were based on a pilot sample of 9 and 14 women in receiving the experimental Recruitment exceeded the minimum sample size calculation. All and control conditions, respectively. Neither calculation included enrolled participants met the eligibility criteria. All of the outcome allowance for loss to follow-up. The first calculation was for the pri- measures in the registered protocol are reported. No additional out- mary outcome (ie, an improvement to 2 on the MOS, anticipating comes were measured or reported. the proportion with a favourable outcome of 0.0 in the control group Table 1 Flow of participants through the study Baseline characteristics of the study participants. Recruitment and data collection took place between December Characteristic Exp Con 2017 and June 2019. A total of 172 women were assessed for eligi- (n = 31) (n = 33) bility, of whom 64 met the inclusion criteria and agreed to participate in the study. Figure 1 shows the flow of the participants in the study. Age (yr), mean (SD) 53 (12) 54 (13) Demographic data are presented in Table 1. Sixty-one women pro- Body mass index (kg/m2), mean (SD) 29.3 (4.3) 29.8 (4.7) vided data that could be included in the regression analysis (28 in the Self-reported ethnicity, n (%) experimental group and 33 in the control group). Three participants 22 (71) 20 (61) in the experimental group declined further participation during the white 2 (6) 4 (12) 8-week intervention period. Among the experimental group partici- black 7 (23) 9 (27) pants who completed the study, 20 completed all eight iES sessions, other 17 (56) three completed seven sessions, three completed six sessions, one Married, n (%) 23 (74) Education (yr), n (%) 17 (52) Table 2 , 10 23 (74) 14 (42) Assessment of pelvic floor muscle function by vaginal palpation and graded according 11 to 12 3 (10) to the Modified Oxford Scale, after intervention. . 12 5 (16) 2 (6) Pregnancies (n), mean (SD) 3.6 (2.3) 3.7 (2.6) Modified Oxford Exp Con Pregnancies (n), range 0 to 11 0 to 11 Scale, n (%) (n = 28) (n = 33) Parity (n), mean (SD) 2.8 (2.1) 2.2 (1.6) Parity classification, n (%) 0 6 (21) 12 (36) nullipara 1 (3) 2 (6) primipara 8 (26) 14 (40) 1 12 (43) 17 (52) multipara 22 (71) 17 (54) Delivery mode, n (%) 2 6 (21) 4 (12) vaginal 22 (71) 25 (76) caesarean section 14 (45) 13 (39) 3 3 (11) 0 (0) Using hormone therapy, n (%) 3 (10) Urinary incontinence, n (%) 31 (100) 3 (9) Anal incontinence, n (%) 3 (10) 33 (100) 3 (9) Con = control group, Exp = experimental group. 4 1 (4) 0 (0) Some percentages do not sum to 100, due to the effects of rounding or because more than one category may apply. Con = control group, Exp = experimental group.
40 Ignácio Antônio et al: Intravaginal electrical stimulation for pelvic floor contraction Table 3 Questions on the ICIQ-UI-SF questionnaire also elicited informa- Number (%) of participants in each group who changed their Modified Oxford Scale tion about the presence or absence of specific types of UI (Table 5). grade to the threshold shown, and the absolute risk difference (95% CI) between the The estimates were also very uncertain, with the confidence intervals groups. around the estimated effect of the experimental intervention on each type of UI generally spanning both markedly favourable and unfav- Change in Modified Groups Absolute risk ourable effects. Therefore, the effect of the experimental intervention Oxford Scale grade difference (95% CI) on specific types of UI remains unclear. Exp Con Exp relative to Con Individual participant data are presented in Table 6 on the (n = 28) (n = 33) eAddenda. Improvement to 2 10 4 0.24 Discussion Improvement to 3 (36) (12) (0.02 to 0.43) The current study shows that use of an iES regimen caused the 4 0 0.14 experimental group to achieve greater acquisition of the ability to (14) (0) (0.01 to 0.31) contract the PFMs voluntarily compared to the control group. The mean estimates of this effect (ie, a 24% absolute increase in the Con = control group, Exp = experimental group. likelihood of being able to contract the PFMs after treatment) might well be considered worthwhile by many women in this clinical completed three sessions and one completed one session. There were population, but the confidence interval was unable to exclude the no reports of adverse effects. possibility that the effect might be negligibly small (ie, the lower limit of the 95% CI was 0.02). Effect of electrical stimulation on the ability to contract the pelvic floor muscles The ability to perform a voluntary PFM contraction represents a prerequisite to PFMT, which is considered a first-line therapeutic The MOS grades achieved by participants in each group at the end option for the treatment of non-neurogenic UI in women.3 As far as of the intervention period are summarised in Table 2. we have ascertained, this is the first randomised trial to investigate the effect of electrical stimulation with simultaneous instruction to When change of MOS from grades 0 or 1 to 2 was used as the attempt voluntary contractions of the PFM to improve women’s outcome criterion for the ability to perform a voluntary contraction, ability to perform a correct PFM contraction. this ability was acquired by 36% of participants in the experimental group versus 12% of participants in the control group (absolute risk The electrical parameters used in the present study follow rec- difference 0.24, 95% CI 0.02 to 0.43). In other words, the experimental ommendations for electrical stimulation for UI.16 However, the only intervention increased the likelihood of being able to perform a studies on electrical stimulation that were found did not combine it voluntary PFM contraction by 24% (absolute), as shown in Table 3. with attempts at voluntary contraction.5,12,13 Li et al assessed the effects of different protocols of electrical stimulation in the treatment When change of MOS from grades 0 or 1 to 3 was used as the of postpartum women with extremely weak muscle strength (ie, MOS outcome criterion for the ability to perform a voluntary contraction, 1).13 A total of 67 women were randomised to two intervention this ability was acquired by 14% of participants in the experimental groups. Both received transvaginal electrical stimulation but one group versus 0% of participants in the control group (absolute risk group also received some electromyographically triggered neuro- difference 0.14, 95% CI 0.01 to 0.31). In other words, the experimental muscular stimulation. The study found similar results to ours in the intervention increased the likelihood of being able to perform a group that received electrical stimulation only; 32% of those partici- voluntary PFM contraction by 14% (absolute), as shown in Table 3. pants learned how to perform a voluntary PFM contraction. This group also improved the mean electromyographic signal they could Effect of electrical stimulation on urinary incontinence sustain for 10 seconds and for 60 seconds. Unfortunately, there was no untreated control group for comparison.13 The effect of the experimental intervention on the overall severity and impact of UI on quality of life was estimated as a 2-point greater Mateus-Vasconcelos et al investigated three therapy interventions reduction on the 21-point ICIQ-UI-SF questionnaire (Table 4). The aimed at facilitating a voluntary PFM contraction, including iES, in confidence interval around this estimate ranged from an arguably 132 women with extremely weak muscle strength (ie, MOS 1).5 The worthwhile effect (around a 4-point greater reduction) to a negligible iES group again had a similar percentage who could voluntarily effect (a 0.02-point reduction). Therefore, although the experimental contract their PFMs after treatment (33%). This was greater than the intervention has a beneficial effect on UI-related quality of life, it control group (18%), who received only verbal instructions in PFM remains unclear whether that benefit is large enough to typically be contraction. However, the iES was less effective than vaginal palpa- clinically worthwhile for women with UI. tion with and without posterior pelvic tilt, and vaginal palpation was also the most effective in improving urinary incontinence. Whilst at baseline all participants of both groups had reported UI, there was some improvement in both groups at the end of the The electrostimulation parameters used in the study by Mateus- intervention period. Two of 28 participants (7%) in the experimental Vasconcelos et al5 were exactly the same as those used in the current group and one of 33 participants in the control group (3%) reported study; however, the current participants were instructed to attempt a resolution of their UI (RR 2.36, 95% CI 0.23 to 24.64). This was a very contraction of the PFM during the stimulus of the electric current. The uncertain estimate, so the effect of experimental intervention on resolving UI remains unclear. Table 4 Mean (SD) of groups, mean (SD) difference within groups, and mean (95% CI) difference between groups for the severity and impact of urinary incontinence on quality of life, as measured by the International Consultation on Incontinence Questionnaire-Urinary Incontinence-Short Form total score. Outcomes Groups Within-group difference Between-group differencea Week 0 Week 8 Week 8 minus Week 0 Exp minus Con Exp Con Exp Con Exp Con (n = 31) (n = 33) (n = 28) (n = 33) (n = 28) (n = 33) ICIQ-UI-SF 16.2 14.2 13.5 13.4 –2.7 –0.8 –2.0 (0 to 21) (3.1) (3.6) (5.3) (4.6) (3.6) (3.9) (–3.97 to –0.02) Con = control group, Exp = experimental group, ICIQ-SF = International Consultation on Incontinence Questionnaire on Urinary Incontinence-Short Form. a Analysis using mixed regression model.
Research 41 Table 5 Number (%) of participants in each group reporting each type of urinary incontinence on the Incontinence Consultation on Incontinence Questionnaire-Short Form at baseline and Week 8, and the relative risk (95% CI) of each type of urinary incontinence between groups at Week 8. Type of UI reported Groups Relative risk (95% CI) Week 0 Week 8 Exp Con Exp Con Exp relative to Con (n = 31) (n = 33) (n = 28) (n = 33) Leaks before you can get to the toilet 22 (79) 27 (82) 21 (75) 29 (88) 0.85 (0.67 to 1.09) Leaks when you cough or sneeze 24 (86) 26 (79) 22 (79) 24 (73) 1.08 (0.81 to 1.44) Leaks when you are asleep 8 (29) 10 (30) 8 (24) 0.29 (0.07 to 1.28) Leaks when you are physically active 12 (43) 19 (58) 2 (7) 16 (48) 0.96 (0.56 to 1.63) Leaks when you have finished urinating and 11 (40) 11 (33) 13 (46) 13 (39) 1.09 (0.60 to 1.99) are dressed 12 (43) Leaks for no obvious reason 13 (46) 13 (39) 16 (48) 0.59 (0.30 to 1.17) Leaks all the time 7 (25) 3 (9) 8 (29) 8 (24) 0.29 (0.07 to 1.28) 2 (7) Con = control group, Exp = experimental group. current study confirms the findings of Mateus-Vasconcelos et al, where 33% inadequate. The optimal dosage for electrical stimulation is still unknown women who received iES acquired the ability to perform a voluntary PFM and needs further investigation,10 especially in relation to women unable contraction after the intervention. Although the present study protocol to perform a PFM contraction. Future larger studies using different pro- added guidance for women to contract their PFM during electrical stimu- tocols should be conducted to evaluate the effect of electrical stimulation lation, this did not improve the efficacy of iES compared to the former study. on UI in such women. However, the study groups may not be directly comparable. Another dif- ference between the studies is that the control group in the study by In conclusion, among women who were unable to perform a PFM Mateus-Vasconcelos et al was not inactive; the participants were instructed contraction, iES with instruction to attempt simultaneous voluntary to perform 10 PFM contractions once a day at home. After the intervention, PFM contractions improved the ability to contract the PFMs. Although 18% could voluntarily contract their PFMs. In the present study the partic- the estimated effect appears worthwhile, the precision of the esti- ipants did not receive any instruction to perform PFMT at home. Despite mate was insufficient to exclude the possibility of a negligible effect. this, 12% of them could voluntarily contract their PFMs after the interven- Similarly, the regimen of iES with attempted contractions improved tion. This might be explained by a learning effect of their first consultation, overall UI-related quality of life, but the confidence interval spanned increased awareness of the pelvic floor, practice on their own or some both clinically worthwhile and negligible effects. combination of these mechanisms. Hence, thorough instruction and feed- back during assessment of the ability to contract may be sufficient for some What was already known on this topic: Pelvic floor muscle women to master a voluntarily contraction.21 In a prospective observational training reduces female stress urinary incontinence. A prerequi- study of 500 women within 1 week after childbirth who were unable to site for initiating the training is the ability to contract the pelvic perform a correct PFM contraction, 52% of the women were able to perform floor muscles voluntarily. Because the pelvic floor muscles are a contraction with inward displacement/lift of the perineum after verbal not directly visible to the patient, teaching correct contractions instruction alone.21 can be a challenge for physiotherapists. What this study adds: In women who are unable to contract The secondary aim of the current study was to evaluate change in their pelvic floor muscles voluntarily, 8 weeks of intravaginal UI. The mean estimates of the effect on overall UI-related quality of electrical stimulation with simultaneous attempts to voluntarily life (ie, a 2-point reduction in severity on the 21-point ICIQ-UI-SF) contract the muscles improved their ability to contract their might well be considered worthwhile by some women in this clin- pelvic floor muscles and improved urinary incontinence-related ical population; however, the confidence interval was unable to quality of life. Although these effects seem worthwhile, the exclude the possibility that the effect might be negligibly small (ie, precision of these estimates was low so it is not possible to the lower limit of the 95% CI was a reduction of 0.02 points). The confirm whether the effects are trivial or worthwhile. effects of the iES regimen on specific types of UI, however, could not be estimated with a useful degree of precision from the present study. Footnotes: a Dualpex Quark®, Quark Produtos Médicos, Piraci- caba, São Paulo, Brazil. A recent systematic review did not identify sufficient evidence to esti- mate the effects of different types of electrical stimulation on improving or b R software V3.6.1, R Core Team, Vienna, Austria. resolving stress urinary incontinence.10 The reviewers stated that electrical eAddenda: Table 6 can be found online at https://doi.org/10.1016/j. stimulation may be more effective than sham treatment, but the difference jphys.2021.12.004. found in favour of electrical stimulation compared with the sham treatment Ethics approval: This trial was approved by the Research Ethics was too small to have clinical importance. Committee of the Hospital das Clínicas de Ribeirão Preto, University of São Paulo (USP), Ribeirão Preto, SP, Brazil (HCRP Opinion No. The results of the current study suggest that iES with simultaneous 2.310.370) and by the Ethics Committee of a School Health Center. attempts to contract the PFMs may be used as an intervention to improve Data collection was initiated only after protocol registration, ethics the ability to perform a PFM contraction. The strengths of this study approval, and collection of signed informed consent from all women include randomisation, concealed allocation, intention-to-treat analysis who agreed to participate. and use of a trained, blinded assessor. It also used valid and reliable Competing interests: None. outcome measures: the intra-rater reproducibility of the scale used to Source(s) of support: The project was funded by the São Paulo assess the primary outcome was good15,22 and some studies have indi- Research Foundation (FAPESP proc numbers 2017/16262-5 and 2016/ cated a moderate-to-strong correlation between the MOS and assessment 24570-9), the Foundation Clinics Hospital of Ribeirão Preto Medical of PFM function using ultrasound.23,24 In addition, participants’ adherence School – University of São Paulo, and the “Coordination for the to the intervention was high and loss to follow-up was minor. A limitation improvement of Higher Education Personnel (CAPES)”. of the current study was that the intervention was performed once a week, Acknowledgements: We thank the São Paulo Research Founda- which is considered a very low dosage for electrical stimulation, and with tion (FAPESP proc numbers 2017/16262-5 and 2016/24570-9), the no additional home use (such as with a portable iES device). Although the Clinics Hospital foundation of Ribeirão Preto Medical School – Uni- frequency, dosage and duration of iES treatment for UI varies considerably versity of São Paulo and the “Coordination for the improvement of in different studies,3,10 most of the trials had programs ranging between 6 Higher Education Personnel (CAPES)” for funding this research. and 12 weeks (two to three times per week)25–27 so some readers may consider our intervention regimen of weekly sessions for 8 weeks to be
42 Ignácio Antônio et al: Intravaginal electrical stimulation for pelvic floor contraction Provenance: Not invited. Peer reviewed. strength. Medicine (Baltimore). 2020;99:e19863. https://doi.org/10.1097/MD. Correspondence: Cristine Homsi Jorge Ferreira, Department of 0000000000019863 Health Sciences, Ribeirão Preto Medical School – University of São 14. Laycock J. Pelvic muscle exercises: physiotherapy for the pelvic floor. Urol Nurs. Paulo, Brazil. Email: [email protected] 1994;14:136–140. http://www.ncbi.nlm.nih.gov/pubmed/7732416 15. Frawley HC, Galea MP, Phillips BA, Sherburn M, Bø K. Reliability of pelvic floor References muscle strength assessment using different test positions and tools. Neurourol Urodyn. 2006;25:236–242. https://doi.org/10.1002/nau.20201 1. Bø K. Pelvic floor muscle training in treatment of female stress urinary inconti- 16. Bo K, Berghmans B, Siv Morkved MVK. Visual observation and Palpation. In: nence, pelvic organ prolapse and sexual dysfunction. World J Urol. 2012;30:437– Evidence-Based Physical Therapy Pelvic Floor. Philadelphia: Elsevier Inc; 2007: 443. https://doi.org/10.1007/s00345-011-0779-8 50–56. 17. de Andrade RL, Bø K, Antonio FI, Driusso P, Mateus-Vasconcelos ECL, Ramos S, 2. Cacciari LP, Dumoulin C, Hay-Smith EJ. Pelvic floor muscle training versus no et al. An education program about pelvic floor muscles improved women’s treatment, or inactive control treatments, for urinary incontinence in women: a knowledge but not pelvic floor muscle function, urinary incontinence or sexual Cochrane systematic review abridged republication. Brazilian J Phys Ther. function: a randomised trial. J Physiother. 2018;64:91–96. https://doi.org/10.1016/ 2019;23:93–107. https://doi.org/10.1016/j.bjpt.2019.01.002 j.jphys.2018.02.010 18. Charlanes A, Chesnel C, Jousse M, Le Breton F, Sheikh Ismael S, Amarenco G, et al. 3. Dumoulin C, Cacciari LP, Hay-Smith EJC. Pelvic floor muscle training versus no Verbal instruction to obtain voluntary pelvic floor muscle contraction: Accept- treatment, or inactive control treatments, for urinary incontinence in women. ability, and understanding. Prog Urol. 2021;31:231–237. https://doi.org/10.1016/j. Cochrane Database Syst Rev. 2018;35:15–20. https://doi.org/10.1002/14651858. purol.2020.12.010 CD005654.pub4 19. Tamanini JTN, Dambros M, D’Ancona CAL, Palma PCR, Rodrigues Netto Jr N. Vali- dação para o português do “International Consultation on Incontinence Ques- 4. Bo K, Frawley HC, Haylen BT, Abramov Y, Almeida FG, Berghmans B, et al. An In- tionnaire - Short Form” (ICIQ-SF). Rev Saude Publica. 2004;38:438–444. https://doi. ternational Urogynecological Association (IUGA)/International Continence Society org/10.1590/S0034-89102004000300015 (ICS) joint report on the terminology for the conservative and nonpharmacological 20. Avery K, Donovan J, Peters TJ, Shaw C, Gotoh M, Abrams P. ICIQ: A brief and robust management of female pelvic floor dysfunction. Neurourol Urodyn. 2017;36:221– measure for evaluating the symptoms and impact of urinary incontinence. Neu- 244. https://doi.org/10.1002/nau.23107 rourol Urodyn. 2004;23:322–330. https://doi.org/10.1002/nau.20041 21. Vermandel A, De Wachter S, Beyltjens T, D’Hondt D, Jacquemyn Y, Wyndaele JJ. 5. Mateus-Vasconcelos ECL, Brito LGO, Driusso P, Silva TD, Antônio FI, Ferreira CHJ. Pelvic floor awareness and the positive effect of verbal instructions in 958 women Effects of three interventions in facilitating voluntary pelvic floor muscle early postdelivery. Int Urogynecol J. 2015;26:223–228. https://doi.org/10.1007/ contraction in women: a randomized controlled trial. Brazilian J Phys Ther. s00192-014-2483-x 2018;22:391–399. https://doi.org/10.1016/j.bjpt.2017.12.006 22. Frawley H. Pelvic floor muscle strength testing. Aust J Physiother. 2006;52:307. https://doi.org/10.1016/S0004-9514(06)70016-2 6. Tibaek S, Dehlendorff C. Pelvic floor muscle function in women with pelvic floor 23. Arab AM, Behbahani RB, Lorestani L, Azari A. Correlation of Digital Palpation and dysfunction. Int Urogynecol J. 2014;25:663–669. https://doi.org/10.1007/ Transabdominal Ultrasound for Assessment of Pelvic Floor Muscle Contraction. s00192–013-2277-6 J Man Manip Ther. 2009;17:75E–79E. https://doi.org/10.1179/jmt.2009.17.3.75E 24. Volløyhaug I, Mørkved S, Salvesen Ø, Salvesen KÅ. Assessment of pelvic floor 7. Bump RC, Glenn Hurt W, Andrew Fantl J, Wyman JF. Assessment of Kegel pelvic muscle contraction with palpation, perineometry and transperineal ultrasound: a muscle exercise performance after brief verbal instruction. Am J Obstet Gynecol. cross-sectional study. Ultrasound Obstet Gynecol. 2016;47:768–773. https://doi.org/ 1991;165:322–329. https://doi.org/10.1016/0002-9378(91)90085-6 10.1002/uog.15731 25. Demirtürk F, Akbayrak T, Karakaya IC, Yüksel I, Kirdi N, Demirtürk F, et al. Inter- 8. DeLancey JOL, Kane Low L, Miller JM, Patel DA, Tumbarello JA. Graphic integration ferential current versus biofeedback results in urinary stress incontinence. Swiss of causal factors of pelvic floor disorders: an integrated life span model. Am J Obstet Med Wkly. 2008;138:317–321. https://doi.org/2008/21/smw-12038 Gynecol. 2008;199:610.e1–610.e5. https://doi.org/10.1016/j.ajog.2008.04.001 26. Hofbauer J, Preisinger F, Nürnberger N. The value of physical therapy in genuine female stress incontinence. Z Urol Nephrol. 1990;83:249–254. http://www.ncbi. 9. Mateus-Vasconcelos ECL, Ribeiro AM, Antônio FI, Brito LGO, Ferreira CHJ. Physio- nlm.nih.gov/pubmed/2203214 therapy methods to facilitate pelvic floor muscle contraction: A systematic review. 27. Santos PFD, Oliveira E, Zanetti MRD, Martins Arruda R, Sartori M, Girão MJBC, et al. Physiother Theory Pract. 2018;34:420–432. https://doi.org/10.1080/09593985.2017. Eletroestimulação funcional do assoalho pélvico versus terapia com os cones 1419520 vaginais para o tratamento de incontinência urinária de esforço. Rev Bras Ginecol e Obs. 2009;31. https://doi.org/10.1590/S0100-72032009000900005 10. Stewart F, Berghmans B, Bø K, Glazener CM. Electrical stimulation with non- implanted devices for stress urinary incontinence in women. Cochrane Database Websites Syst Rev. 2017. https://doi.org/10.1002/14651858.CD012390.pub2 Randomization.com www.randomization.com 11. Brown CA, Sharples R. Does neuromuscular electrical stimulation increase pelvic floor muscle strength in women with urinary incontinence with an ineffective pelvic floor contraction? J Assoc Chart Physiother Women’s Health. 2014;114:56–62. 12. Rodrigues MP, Barbosa LJF, Paiva LL, Mallmann S, Sanches PRS, Ferreira CF, et al. Effect of intravaginal vibratory versus electric stimulation on the pelvic floor muscles: A randomized clinical trial. Eur J Obstet Gynecol Reprod Biol X. 2019;3:100022. https://doi.org/10.1016/j.eurox.2019.100022 13. Li W, Hu Q, Zhang Z, Shen F, Xie Z. Effect of different electrical stimulation protocols for pelvic floor rehabilitation of postpartum women with extremely weak muscle
Journal of Physiotherapy 68 (2022) 7 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 Paper of the Year 2021 The Editorial Board of Journal of Physiotherapy is pleased to announce the 2021 Paper of the Year Award. The winning paper is judged by a panel of members of the International Advisory Board who do not have a conflict of interest with any of the papers under consideration. They vote for the paper published in the 2021 calendar year that, in their opinion, has the best combination of scientific merit and application to the clinical practice of physiotherapy. The winning paper is ‘An international core capability framework for physiotherapists to deliver quality care via videoconferencing: a Delphi study.’1 The authors are Luke Davies, Rana Hinman, Trevor Russell, Belinda Lawford, Kim Bennell and the International Videoconferencing Steering Group. The paper addressed the core capabilities that physiotherapists need in order to deliver quality care via videoconferencing. Provision of care via telehealth is emerging as an effective and acceptable mode of delivering physiotherapy. The provision of physiotherapy via telehealth is associated with high rates of patient satisfaction, with clinical effects sometimes surpassing those of in-person care. Although the COVID-19 pandemic led to a dramatic up-take in telehealth delivery of physiotherapy, this sometimes occurred with limited training and preparation. The winning paper used a Delphi process with international experts (including researchers, clinicians, consumers and representatives of physiotherapy organisations) to develop a framework outlining the specific core capabilities that an international panel of experts recommend for physiotherapists to provide quality care via videoconferencing. The capabilities cover the domains of compliance, patient privacy/confi- dentiality, patient safety, technology skills, telehealth delivery, assessment/diagnosis, and care planning/management. The winning paper is one of several papers published by the Journal that relate to the COVID-19 pandemic: recommendations for acute hospital management;2 physiotherapists’ and patients’ experiences with telehealth;3 innovative clinical education models;4 the assessment of clinical competence;5 the virtual hospital model;6 and more in the current issue. Members of the judging panel commented that the topic of the winning paper will also be important beyond the pandemic, as many in the profession have come to experience the potential value of videoconferencing as one mode of delivering physiotherapy care. The winning paper has been endorsed by the International Society for Telemedicine and ehealth, and included in World Physiotherapy’s online digital physiotherapy resources. The framework has been developed into a clinician resource that is freely available for download;7 this online resource has been viewed over 500 times since being made publicly available in September 2021. The members of the Editorial Board congratulate Luke Davies and his co-authors on their success. References 1. Davies L, Hinman RS, Russell T, Lawford B, Bennell K, International Videoconferencing Steering Group. An international core capability framework for physiotherapists to deliver quality care via videoconferencing: a Delphi study. J Physiother. 2021;67:291–297. 2. Thomas P, Baldwin C, Bissett B, Boden I, Gosselink R, Granger CL, et al. Physiotherapy management for COVID-19 in the acute hospital setting: clinical practice recommendations. J Physiother. 2020;66:73–82. 3. Bennell KL, Lawford BJ, Metcalf B, Mackenzie D, Russell T, van den Berg M, et al. Physiotherapists and patients report positive experiences overall with telehealth during the COVID-19 pandemic: a mixed-methods study. J Physiother. 2021;67:201–209. 4. Dario A, Simic M. Innovative physiotherapy clinical education in response to the COVID-19 pandemic with a clinical research placement mode. J Physiother. 2021;67:235–237. 5. Tognon K, Grudzinskas K, Chipchase L. The assessment of clinical competence of physiotherapists during and after the COVID-19 pandemic. J Physiother. 2021;67:79–81. 6. Melman A, Maher CG, Machado GC. Virtual hospitals: why we need them, how they work and what might come next. J Physiother. 2021;67:156–157. 7. https://healthsciences.unimelb.edu.au/departments/physiotherapy/chesm/clinician-resources/international-core-capability-framework-for-physiotherapists-to-deliver-quality- care-via-videoconferencing https://doi.org/10.1016/j.jphys.2021.12.006 1836-9553/
Journal of Physiotherapy 68 (2022) 51–60 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 Pelvic floor muscle function differs between supine and standing positions in women with stress urinary incontinence: an experimental crossover study Márcia Maria Gimenez a, Fátima Faní Fitz a, Letícia de Azevedo Ferreira a, Maria Augusta Tezelli Bortolini a, Patrícia Virgínia Silva Lordêlo b, Rodrigo Aquino Castro a a Department of Gynecology, Universidade Federal de São Paulo, São Paulo; b Department of Physiotherapy, Escola Bahiana de Medicina e Saúde Pública, Salvador, Brazil KEY WORDS ABSTRACT Pelvic floor Question: In women with stress urinary incontinence, how does pelvic floor muscle (PFM) function differ Stress urinary incontinence between supine and standing when assessed using manometry, vaginal palpation, dynamometry and elec- Standing position tromyography? Design: An experimental crossover study. Participants: A total of 101 women with stress Evaluation study urinary incontinence were included. Intervention: The PFM evaluations were performed and compared in Physical therapy supine and standing positions. The participants were assigned to either Group 1 (assessments in supine followed by standing) or Group 2 (assessments in standing followed by supine). Outcome measures: The primary outcome was the PFM pressure during the maximum voluntary contraction (MVC). Secondary outcomes were the measures of PFM pressure at rest; PFM function (PERFECT scheme); active and passive forces (dynamometry); and PFM electromyography (EMG) activity. Results: The mean MVC pressure was significantly lower in standing (MD 27 cmH2O, 95% CI 210 to 24). The mean PFM resting pressure was higher in standing (7 cmH2O, 95% CI 5 to 10). Three measures of PFM function derived from vaginal palpation were better in supine than in standing. The PFM active and the passive forces measured using dynamometry were higher in standing (0.18 kgf, 95% CI 0.16 to 0.20). The resting EMG activity was higher in standing than in supine (MD 3.6 mV, 95% CI 2.6 to 4.5), whereas EMG activity during MVC was higher in supine than standing (MD 28.7 mV, 95% CI 212.5 to 24.8). Conclusion: The pressure and EMG activity during MVC, and PFM function were lower in standing. The resting pressure, the passive and active forces of the PFM and the resting EMG activity of the PFM were higher in standing. [Gimenez MM, Fitz FF, de Azevedo Ferreira L, Bortolini MAT, Lordêlo PVS, Castro RA (2022) Pelvic floor muscle function differs between supine and standing positions in women with stress urinary incontinence: an experimental crossover study. Journal of Physiotherapy 68:51–60] © 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 is usually easy to standardise and therefore recommended in clinical practice.9 Arnold Kegel suggested pelvic floor muscle (PFM) contractions for the treatment of stress urinary incontinence (SUI) and observed a To date, few studies have evaluated PFM function in the standing cure rate . 84%.1,2 Since then, PFM training has been extensively studied, and it is currently considered the first-line treatment for position in women with SUI. Among them, one study investigated SUI,3 with a success rate of around 60 to 75% when it is performed under the supervision of a physiotherapist.4,5 The training aims to PFM strength measured by manometry in standing compared to su- improve muscle impairment components such as reduced strength, altered activation time or poor muscle coordination, and to decrease pine positions in women with SUI after a 3-month intensive PFM the symptoms of PFM dysfunction such as SUI.6 Thus, assessing PFM training program.9 The authors showed that vaginal resting pressure function before and after training is important to determine whether the intervention yields significant changes.7 was significantly higher in the standing position, and neither maximal strength nor holding time were different between the two Methods commonly used by physiotherapists for PFM assessment positions.9 Three studies have assessed PFM contraction using ultra- include the clinical inspection of the movement of the perineum, sonographic parameters, with divergent results.11–13 Arab et al re- vaginal palpation, manometry, electromyography (EMG) and dyna- mometry.8 Some studies have reported that physiotherapists and ported that PFM contraction is greater in the standing position in patients prefer to perform the evaluation in the supine position;9,10 it relation to the supine position in women because the base of the bladder is displaced further in the cranial direction, irrespective of continence status.11 Dietz and Clarke observed that the effect of a pelvic floor contraction was not influenced by posture in women with urinary symptoms,12 and Peng et al observed a decrease in the acti- vation of PFM to contain the increase in intra-abdominal pressure in https://doi.org/10.1016/j.jphys.2021.12.011 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/).
52 Gimenez et al: Pelvic floor function in different positions Figure 1. Design and flow of participants through the trial. women with SUI in the standing position.13 Two studies have eval- Therefore, the research question for the observational study was: uated PFM activity using EMG.14,15 Aukee et al (2003) showed less muscular activation in incontinent women compared with asymp- In women with stress urinary incontinence, how does pelvic floor tomatic ones mainly in the standing position; however, those authors muscle function differ between supine and standing when used the prototype of a biofeedback device that was not previously assessed using manometry, vaginal palpation, dynamometry and tested as a valid instrument.14 Aukee et al (2002) reported increased electromyography? PFM strength in women with SUI in both supine and standing posi- tions after 12 weeks of PFM training with and without biofeedback,15 Method but the influence of the standing position on PFM function was not discussed. One study evaluated the PFM with digital muscle testing, Design manometry and transabdominal ultrasound in standing and supine positions. Digital muscle testing and vaginal squeeze-pressure scores This observational, cross-sectional, analytical study was per- were highest in the supine position, and vaginal resting pressure and formed between June 2018 and February 2020. The study occurred in transabdominal ultrasound scores were highest in the standing po- the Sector of Urogynecology and Reconstructive Pelvic Surgery at the sition.10 However, the study recruited pelvic health physiotherapists Universidade Federal de Sao Paulo, Brazil. The purpose and content of as study participants to have their PFM function assessed; only a the study were explained to potentially eligible women, and those minority of these physiotherapists reported any urinary incontinence who agreed to participate in the study provided written informed symptoms, which were mild. consent. The PFM were assessed by a single physiotherapist with 16 years of experience in PFM rehabilitation (MMG). The exam was The evaluation of PFM contraction in the standing position performed in a single day and lasted about 1 hour. Participants served potentially allows a more functional PFM assessment in women with as their own control by undergoing the assessments in both supine SUI because urine loss occurs in the upright position, with gravity and standing (order randomised) as shown in Figure 1. The STROBE influencing the muscles’ function. Therefore, it is important to (Strengthening the Reporting of Observational Studies in Epidemi- investigate PFM contraction in the standing position where urinary ology) guidelines were followed for reporting this study.16 incontinence usually occurs, and whether different tools produce different results when the body position changes. Participants Methodological studies with high accuracy are required to find Consecutive patients with urinary symptoms were evaluated by a optimal assessment protocols, with sensitive, reproducible and valid urogynaecologist who performed a clinical examination (patient instruments to measure PFM function and strength in the standing history and physical examination, including a stress cough test) and position, so that the relationship between body position and SUI 1-hour pad test17 as part of routine standard care. Patients with SUI or mechanism can be understood.9 mixed urinary incontinence with predominant SUI symptoms were considered candidates for conservative treatment. SUI was diagnosed This study was developed to use four different PFM measurement based on the patient’s report of leakage during physical stress and the tools, each of which evaluates different aspects of PFM contraction, in condition that bothered them the most in the case of mixed urinary order to give the scientific community and clinicians a broader un- incontinence (stress urinary or urgency urinary incontinence), and derstanding of PFM function. The primary aim was to assess and the diagnosis was confirmed by positive cough stress test results and compare the maximum voluntary contraction (MVC) pressure of the pad test findings ( 1.5 g of leakage).18 The patients were referred to a PFM in standing and supine positions by manometry in women with physiotherapist for PFM examination. Women were excluded if they: SUI. The secondary aims of the study were to assess in both positions: had chronic degenerative diseases, uncontrolled metabolic diseases, the resting pressure by manometry; the pelvic floor function by vaginal palpation (PERFECT scheme); the muscle strength (active force) and resting tone (passive force) by dynamometry; and the EMG activity of the PFM at rest and during MVC.
Research 53 or neurological or psychiatric diseases; had previously participated in normally. For assessment of the PFM during MVC, they were asked to pelvic floor re-education programs and/or undergone pelvic floor squeeze and lift their PFM as if they were preventing the escape of surgery; were undergoing another type of treatment for UI; had flatus and urine.25 Between each measurement tool and each posi- pelvic organ prolapse greater than stage I on the Pelvic Organ Pro- tion, the participants had a 5-minute rest to minimise fatigue. lapse Quantification system;19 were unable to contract their PFM; or reported discomfort with the PFM assessment. Resting and MVC pressure of the PFM by manometry The MVC pressure of the PFM was evaluated using a commercial Before the PFM evaluation, the physiotherapist (MMG) instructed all women about the location of the PFM with anatomical figures and manometera. The catheter was covered with a new sterile condom for muscle function was explained verbally. The subjects were instructed each patient and inserted into the vagina until the entire length of the how to contract their PFM properly. The correct contraction was compressible part of the device was above the level of the hymenal identified by vaginal palpation and observation of inward movement ring, and the resting pressure of the PFM was measured in cmH2O for of the perineum during contraction. Feedback on their performance 3 minutes. Then the device was reset to zero, three consecutive MVCs was provided by the physiotherapist,20 with discouragement of pelvic were performed with an interval of 10 seconds between each tilt and synergistic contractions of abdominal, hip adductors and contraction,21 and the best of the three values was recorded (primary gluteal muscles. Participants unable to perform the correct contrac- outcome).24 Only the contractions during which an inward move- tion of the PFM after receiving instructions were not included in the ment of the balloon catheter was observed and synergistic contrac- study. tions of the hip adductors and gluteal muscles or pelvic tilt were not observed were included in the analysis. Figure 2 illustrates the po- Pilot testing of the measurement procedures sitions of the assessor and the probe in the vagina. To test the intra-rater reliability, 10 women were evaluated on two PFM function by vaginal palpation different occasions separated by a 1-week interval prior to beginning The PFM function was assessed by vaginal palpation and quanti- the study. To determine the intraclass correlation coefficient (ICC) of the qualitative variable (Oxford scale), the Cohen’s Kappa criteria was fied according to the PERFECT scheme.22 The PFM strength was used. The weighted Kappa 95% for the Oxford scale was 1.000 for the assessed by the modified Oxford Grading Scale (0 to 5).17 The par- supine position and 1.000 for the standing position. To determine the ticipants were instructed to perform three contractions lasting 5 ICC of quantitative variables (manometry, endurance, fast contraction, seconds, each with the greatest strength possible (maximum PFM EMG, dynamometry), Cronbach’s alpha was used. Cronbach’s alpha contractions). The periods of contraction were separated by 1-minute for the resting pressure and MVC pressure of the PFM (manometry), rest periods. Endurance was expressed as the length of time for up to endurance and fast contraction (vaginal palpation), muscle resting 10 seconds that a maximal vaginal contraction could be sustained. activity and PFM contraction strength (dynamometry) was 1.000 in Thus, the contraction was timed until the muscle started to fatigue. the supine position and 1.000 in the standing position. Chronbach’s The fast contraction variable was recorded as the number of fast alpha for the resting PFM electrical activity was 0.982 in the supine maximal PFM contractions that could be repeated (up to 10).22 position and 0.997 in the standing position. Chronbach’s alpha for Figure 3 illustrates the position of the assessor and the fingers in MVC electrical activity was 0.997 in the supine position and 0.999 in the vagina. the standing position. PFM active and passive forces by dynamometry Measurement procedures A wireless unit with dynamometerc was used to assess the PFM. In The participants were assigned with equal probability to either brief, the dynamometer comprises a computerised central unit, a Group 1 (assessment in supine then standing) or Group 2 (assessment peripheral unit and a dynamometric speculum. The dynamometer in standing then supine). The allocation sequence was generated by a was calibrated prior to PFM assessment. After thorough disinfection, research assistant (FFF) using a computer-generated random number the probe was covered with a new condom and appropriately lubri- table in a ratio of 1:1 and it was concealed in sequentially numbered, cated with water-soluble jelly. The dynamometer probe was equip- sealed, opaque envelopes. The envelopes were kept in a closed locker, ped with a 2-cm load cell on its base, and the passive force and the which only the research assistant could access, at the centre. The maximal PFM strength (active force) were measured in kgf. The envelopes were given to the physiotherapist (MMG) immediately passive force was recorded over a period of 10 seconds. The mean prior to the PFM examination. In both the supine and standing po- value was considered an index of PFM tonicity. Then, the maximal sitions, the same sequence was used to evaluate the PFM: resting and PFM strength was recorded.23 Three 10-second contractions sepa- MVC pressures were measured in cmH2O using a commercial man- rated by a 3-minute rest period were performed. The best of three ometera;21 muscle function was assessed by vaginal palpation contractions was used for analysis.23 Figure 4 illustrates the position (strength, endurance and fast contraction);22 resting (passive force) of the assessor and the probe in the vagina. and unidirectional compression strength and antero-posterior forces (active force) were measured in kilogram-force (kgf) by dynamome- PFM electrical activity by surface EMG tryb;23 and electrical activity at rest and during the MVC was The surface EMG of the PFM was performed using a transvaginal measured in microvolts (mV) through surface EMGb.24 probe. To eliminate external interference, a reference electroded was During PFM assessment in supine, the participants adopted a used on the styloid process of the radius.26 The electrodes were dorsal recumbent position in a gynaecological chair with the body connected to a wireless unit with electromyographc, and the EMG slightly elevated, the feet positioned on stirrups, the knees flexed to signal, expressed in mV, was recorded by commercial softwaree and 30 deg and the upper limbs placed alongside the trunk. During PFM stored on a portable computer. To evaluate the mean value of elec- assessment in standing, the participants stood with the feet hip- tromyographic activity of the PFM at rest, a 10-second period was width apart, the knees extended to maintain the vertical alignment used. Then, the electromyographic activity of the PFM during an MVC of the trunk with physiological curvatures of the spine, the upper was recorded. Three 10-second contractions separated by 3 minutes limbs relaxed beside the trunk and the gaze directed towards the of rest were performed, and the surface EMG signals were recorded horizon. When movement of the perineum during the contraction according to the surface EMG for the non-invasive assessment of was to be observed, this was facilitated by a mirror placed between muscles (SENIAM) guidelines.27 The best of three trials was used for the participant’s feet. analysis.24 To minimise possible confounding in the EMG signals, the evaluations were always performed in the morning and the size of The same verbal instructions were used for all evaluations of the the electrodes was also standardised, but hydration was not stand- PFM. For assessment of the PFM resting tone, the women were ardised. Figure 5 illustrates the position of the assessor and the probe instructed to keep their body and PFM relaxed and to breathe in the vagina.
54 Gimenez et al: Pelvic floor function in different positions Figure 2. Manometry: positions of the assessor and the probe in the vagina. Figure 3. Vaginal palpation/visual observation: positions of the assessor and the fingers in the vagina.
Research 55 Figure 4. Dynamometry: positions of the evaluator and the probe in the vagina. Figure 5. Electromyography: positions of the assessor and the probe in the vagina.
56 Gimenez et al: Pelvic floor function in different positions Table 1 Demographic and clinical characteristics of the study participants. Characteristic Total Group 1 Group 2 (n = 101) (assessed in supine then standing) (assessed in standing then supine) Age (yr), mean (SD) 54 (11) (n = 52) (n = 49) Body mass index (kg/m2), mean (SD) 29.4 (5.3) Marital status, n (%) 53 (11) 55 (12) 19 (19) 28.9 (5.2) 29.9 (5.4) single 62 (61) married 12 (12) 9 (17) 10 (20) divorced 32 (62) 30 (61) widowed 8 (8) 7 (14) 5 (10) Education level, n (%) primary/secondary 86 (85) 4 (8) 4 (8) university 15 (15) Ethnicity, n (%) 44 (85) 42 (86) Asian 1 (1) 8 (15) 7 (14) Black 7 (7) Caucasian 49 (49) 1 (2) 0 (0) Mixed race 44 (44) 2 (4) 5 (10) Type of UI diagnosed, n (%) 25 (48) 24 (49) stress 51 (51) 24 (46) 20 (41) mixed 50 (50) UI symptom duration (mth), median (IQR) 36 (24 to 96) 24 (46) 27 (55) Obstetric history (n), median (IQR) 28 (54) 22 (45) pregnancy 3 (2 to 4) 36 (24 to 60) 36 (24 to 120) caesarean 0 (0 to 1) vaginal birth 1 (0 to 2) 3 (2 to 4) 3 (2 to 4) Pad test (g), median (IQR) 6 (2 to 18) 0 (0 to 1) 1 (1 to 2) Positive cough test, n (%) 101 (100) 1 (0 to 2) 1 (0 to 3) Menopause age (yr), mean (SD) 6 (2 to 18) 6 (3 to 19) 48 (6) 52 (100) 49 (100) 50 (5) 47 (6) UI = urinary incontinence. Some percentages do not sum to 100 due to the effects of rounding. Data analysis Results The sample size calculation was based on published data.9 In a Participants conservative approach, the values were always rounded so that the difference decreased and the variability increased, thereby increasing A total of 105 women were contacted; of them, 101 met the the sample size. The following statistical parameters were used: eligibility criteria, with 52 randomised to assessment in supine before normal distribution, type-I error , 0.05 (two-tailed) and statistical standing and 49 randomised to assessment in standing before supine. power of 80%. The primary outcome used to calculate the sample After inclusion, 12 women were excluded because they reported size was a difference of 2 cmH2O (SD 4) between the PFM MVC discomfort during the PFM assessment (Figure 1). The mean age of all pressures in the supine and standing positions. Thus, the sample size the included women was 54 years (SD 11). The two randomly allo- required to achieve a statistical power of 80% was at least 63 pa- cated groups had similar demographics and clinical characteristics. tients. To allow for loss to follow-up, the initial sample size was Most participants were married (61%), Caucasian (49%) or mixed race increased to 101. (44%), with primary and secondary education level (85%). About half had SUI and about half had MUI with predominant SUI symptoms. Data analyses were performed by another professional who was They had a median of three pregnancies, mean body mass index of blinded to group allocation. Commercial statistical softwaref was used 29.4 kg/m2, mean menopausal age of 48 years and a mean duration of for the statistical analyses. The categorical data are presented as the urinary symptoms of 36 months (Table 1). absolute number and the respective percentage, and the continuous data are presented as the mean and the respective standard devia- Effect of body position on PFM assessment using manometry tion. The related samples from the two different positions were compared with tests for paired data. For inferential statistical analysis The pressure recorded using manometry during an MVC (primary of the continuous data, the normality of all the data was tested with outcome) was lower in standing than in supine, with a mean difference the Kolmogorov-Smirnov test, and the Wilcoxon test was subse- of 27 cmH2O (95% CI 210 to 24), as shown in Table 2. This effect of quently used in place of the paired t-test, where applicable. Spearman body position on the manometric pressure was observed regardless of correlation was used to identify the degree of association between which body position was tested first (see Table 3 on the eAddenda). two variables (eg, rest and maximum voluntary contraction) and between the measurement tools (eg, manometry, dynamometry and The pressure recorded using manometry at rest was higher in electromyography). The degree of association between the variables standing than in supine, with a mean difference of 7 cmH2O (95% CI 5 was classified as high (0.80 to 1.00), moderate (0.60 to 0.80) or weak (, 0.59), according to the criteria reported by Richman et al.28 The Table 2 adopted level of significance was 5%.29 Manometric assessment of resting and MVC pressures of the pelvic floor muscles in the supine and standing positions, and the mean difference (95% CI) of standing relative to To determine the intraclass correlation coefficient (ICC) of the supine. ordinal variables (Oxford scale), the following Cohen’s Kappa criteria for the level of agreement were used: 0 = zero, 0.10 to 0.39 = weak, Manometry (cmH2O), Supine Standing Mean difference 0.40 to 0.69 = moderate, 0.70 to 0.99 = strong and 1 = perfect. To mean (SD) (n = 89) (n = 89) (95% CI) determine the ICC of continuous variables (endurance, fast contrac- tion, EMG, manometry and dynamometry), the following Cronbach’s Resting 33 (9) 40 (9) 7 (5 to 10) alpha criteria were used: , 0.21 = small, 0.21 to 0.40 = reasonable, MVC 32 (18) 25 (13) 27 (210 to 24) 0.41 to 0.60 = moderate, 0.61 to 0.80 = substantial and 0.81 to 1.00 = perfect. A significance level of 5% was used.30,31 Shaded row indicates the primary outcome. MVC = maximal voluntary contraction.
Research 57 Table 4 difference 22 (95% CI 23 to 22), as shown in Table 4 and Figure 6c. Three aspects of function of the pelvic floor muscles assessed by vaginal palpation in This effect of body position on the number of fast contractions was the supine and standing positions, and the median difference (95% CI) of standing observed regardless of which body position was tested first (see relative to supine. Table 5 on the eAddenda). Vaginal palpation, Supine Standing Median difference Effect of body position on PFM assessment using dynamometry median (IQR) (n = 89) (n = 89) (95% CI) The passive force of the PFM recorded using dynamometry was MOS (0 to 5) 2 (2 to 3) 2 (1 to 2) 0 (21 to 0) higher in standing than in supine, with a mean difference of 0.18 kgf Endurance (sec) 4 (3 to 4) 2 (2 to 3) 21 (21 to 21) (95% CI 0.16 to 0.20), as shown in Table 6. This effect of body position Fast contractions (n) 6 (4 to 8) 3 (2 to 5) 22 (23 to 22) on the passive force of the PFM was observed regardless of which body position was tested first (see Table 7 on the eAddenda). MOS = Modified Oxford Scale. The active force of the PFM recorded using dynamometry was also to 10), as shown in Table 2. This effect of body position on the higher in standing than in supine, with a mean difference of 0.14 kgf manometric pressure was observed regardless of which body position (95% CI 0.08 to 0.19), as shown in Table 6. This effect of body position was tested first (see Table 3 on the eAddenda). on the active force of the PFM was observed, regardless of which body position was tested first (see Table 7 on the eAddenda). In standing, there was a correlation between the resting pressure and the MVC pressure: r = 0.48 (95% CI 0.30 to 0.62), which was a In standing, there was a correlation between the active and pas- weak correlation.28 This correlation was observed regardless of which sive forces: r = 0.69 (95% CI 0.56 to 0.78), which was a moderate body position was tested first (data not shown). In supine, no clear correlation.28 In supine, a weak correlation was evident between the correlation was evident between the resting pressure and the MVC active and passive forces: r = 0.54 (95% CI 0.37 to 0.67). These cor- pressure: r = 0.08 (95% CI 20.13 to 0.28). This absence of correlation relations were not influenced by which body position was tested first observed regardless of which body position was tested first (data not (data not shown). shown). Effect of body position on PFM assessment using vaginal palpation Modified Oxford Scale Effect of body position on PFM assessment using surface EMG Grading with the Modified Oxford Scale demonstrated weak PFM The electrical activity recorded using EMG at rest was higher in contractions without elevation of the vaginal wall in both groups and standing than in supine, with a mean difference of 3.6 mV (95% CI 2.6 positions. The median score in each group was 2, although the dis- to 4.5), as shown in Table 8. This effect of body position on EMG tribution favoured the supine group, as shown in Table 4 and activity was observed, regardless of which body position was tested Figure 6a. This effect of body position on the Modified Oxford Scale first (see Table 9 on the eAddenda). score was observed regardless of which body position was tested first (see Table 5 on the eAddenda). The electrical activity recorded using EMG during a MVC was lower in standing than in supine, with a mean difference of 28.7 mV Endurance (95% CI 212.5 to 24.8), as shown in Table 8. This effect of body po- The duration for which a maximal PFM contraction could be sition on EMG activity was observed regardless of which body posi- tion was tested first (see Table 9 on the eAddenda). sustained was shorter in standing than in supine: median difference 21 second (95% CI 21 to 21), as shown in Table 4 and In standing, there was a correlation between the electrical activity Figure 6b. This effect of body position on the duration of a sustained at rest and during an MVC: r = 0.58 (95% CI 0.42 to 0.70), which was a contraction was observed regardless of which body position was weak correlation.28 In supine, a weaker correlation was evident be- tested first (see Table 5 on the eAddenda). tween the electrical activity at rest and during an MVC: r = 0.32 (95% CI 0.12 to 0.49). These correlations were not influenced by which Fast contractions body position was tested first (data not shown). The number of fast maximal PFM contractions that could be Individual participant data are available in Table 10 on the repeated (up to 10) was lower in standing than in supine: median eAddenda. a b c 5 6 10 5 48 4 36 3 24 2 1 12 Modified Oxford Scale (0 to 5) Endurance (s) Fast contractions (n) 0 0 0 Supine Standing Supine Standing Supine Standing Figure 6. Distributions of scores derived from vaginal palpation in supine and standing: (a) Modified Oxford Scale, (b) duration of sustained maximal contraction and (c) number of fast maximal contractions.
58 Gimenez et al: Pelvic floor function in different positions Table 6 controlled in this study: the sample size was calculated to answer the Dynamometric assessment of passive and active forces of the pelvic floor muscles in primary objective, the SUI was detected by the pad test and the the supine and standing positions, and the mean difference (95% CI) of standing participants had not previously performed PFM training. relative to supine. Our evaluation of performance, endurance and muscle coordina- Dynamometry (kgf), Supine Standing Mean difference tion by vaginal palpation, and using the PERFECT scheme classifica- mean (SD) (n = 89) (n = 89) (95% CI) tion revealed that PFM function score was lower in standing than in supine in women with SUI. Physiotherapists use vaginal palpation to Passive 0.34 (0.11) 0.52 (0.12) 0.18 (0.16 to 0.20) assess PFM function because both the squeeze pressure and lift can be Active 0.74 (0.36) 0.88 (0.29) 0.14 (0.08 to 0.19) assessed, and this method is a low-cost and relatively easy method to conduct.6 Frawley et al had few subjects grouped in each PFM grade Discussion and did not consider it appropriate to draw conclusions about their results.10 The sensitivity of digital palpation using the modified Ox- This study demonstrated that assessment of the PFM of women ford scale in detecting the component of elevation of PFM activity has with SUI in supine versus standing led to differences in the scores been questioned when compared to a more objective measure of obtained using manometry, vaginal palpation, dynamometry and elevation of PFM such as ultrasound. A separate qualitative scale of EMG. Some of these assessment methods produced better results in the elevation component of the PFM has been suggested for digital standing, while others produced better results in supine. For each muscle testing.10 The Oxford scale also evaluates the squeeze pressure outcome measure, the difference in scores arising from testing in of the PFM during contraction, and we believe that squeeze pressure supine versus in standing was very consistent, regardless of which produced by PFM in the standing position and assessed by vaginal position was tested first. palpation can reproduce the actual capacity of the PFM to contract when standing still in a manner to prevent loss of urine. During the When testing the pressure recorded using manometry during an vaginal palpation, the therapist can instruct the patient in how to MVC (primary outcome), the MVC pressure was significantly lower in perform a contraction correctly and tell her about coordination skills the standing position. The same was demonstrated by Frawley et al, and strength. Testing in other positions and in dynamic tasks would who investigated pelvic health physiotherapists who reported no or be needed to assess the full picture of the function of the PFM. mild urinary incontinence symptoms.10 In contrast, Bø and Finck- enhagen did not find a difference in vaginal squeeze pressure be- We found that the EMG activity of the PFM during the MVC was tween the two positions in women with SUI assessed after a 3-month lower in the standing position in women with SUI. This result is in intensive PFM training program;9 the previous treatment may explain accordance with Laycock,33 who compared surface vaginal EMG signals their better-than-expected scores in the standing position.9 Vaginal for PFM and observed a significant lower PFM activity in standing squeeze pressure measurements can be invalid if there is involve- compared with supine positions.33 Rett et al34 and Aukee et al14 also ment of other muscle groups that would increase the measured reported similar results to ours, even though they evaluated PFM EMG values. The simultaneous observation of the inward movement of the activity after a pelvic floor training program.14,34 Rett et al failed to perineum indicates that the contraction is being performed show that the PFM contraction capacity would be similar in supine and correctly.6 Anatomic placement of the device is another factor to re- standing positions after a PFM training program, and explained that a cord to achieve reproducible measurements.32 In the present study, large force of gravity on the PFM in standing would have caused the physiotherapist held the catheter in position while the participant overload, and the capacity of the PFM to perform satisfactory muscle contracted her PFM, and ensured that the catheter did not move contractions would have decreased, thereby exhibiting smaller am- between the contractions. A mirror was placed between the partici- plitudes of EMG signals.34 Aukee et al did not discuss the influence of pant’s feet in the standing position to monitor the contraction. the standing position on PFM function.14 Our study showed that the mean EMG amplitudes at rest were higher in the standing position, This study detected a high resting pressure in the standing posi- which was similar to that reported by Rett et al.34 According to the tion when compared with the supine position, which is in agreement authors, higher amplitudes in the standing position have also occurred with previous reports by Bø and Finckenhagen9 and Frawley et al.10 in response to the action of gravity and activation of tonic fibres rather There is still limited knowledge of the mechanisms underlying this, than phasic fibres, which are needed more during tasks requiring great which need to be addressed in future research. Bø and Finckenhagen effort.34 EMG can be used to indicate denervation/reinnervation in discussed that this higher resting pressure may be directly influenced striated PFM. In cases of mild-to-moderate partial denervation, EMG is by gravity, and speculated whether the increase in this pressure is a very limited in providing data on muscle strength (which is logically direct consequence of strong PFM and whether there is a positive impaired due to denervation).35 Therefore, other methods in addition association between high resting vaginal pressure, resting urethral to EMG need to be used to evaluate PFM. Surface EMG is considered to pressure and continence.9 Frawley et al did not consider the vaginal be a reproducible measure; however, the quality of the electrical sig- resting pressure to be a reliable method in upright positions and nals can be affected by many factors.35 Factors that can influence the pointed out that this can limit the usefulness of the information amplitude of surface EMG signals include the location and size of the obtained from this measure in vertical positions.10 The results of our electrodes, the hydration level of the individual and the diurnal vari- study suggest that in women with SUI with no previous treatment, ation in muscle bioelectric activity;27 therefore, evaluation of PFM EMG gravity and intra-abdominal pressure impact the PFM (due to the activity may be more valid for the elaboration of PFM training higher resting pressure), which interferes with the voluntary PFM considering the daily muscle condition, and not for evaluation of contraction (ie, lower MVC pressure). Factors that could interfere in muscle function before and after treatment. the assessment of PFM in the standing position reported in the studies by Bø and Finckenhagen9 (3-month intensive PFM training The passive and active forces of the PFM by dynamometry were program, a small number of subjects) and Frawley et al10 (a small, higher in the standing position in our observation of women with SUI. convenience sample of pelvic health physiotherapists) were Dynamometry is a reliable tool for assessing the PFM forces in the supine position in women with pelvic floor dysfunction.36,37 The Table 8 literature revealed a 92% increase in the resting vaginal closure force Electromyographic assessment of activity of the pelvic floor muscles at rest and during when a woman moves from the supine to the standing position, and MVC in the supine and standing positions, and the mean difference (95% CI) of the increased intra-abdominal pressure or greater resistance from the standing relative to supine. PFM would be two potential contributors for this result.38 In relation to vaginal closure force, Morgan et al showed that the pelvic muscle Electromyography (mV), Supine Standing Mean difference force elevating the speculum is independent of body orientation.38 mean (SD) (n = 89) (n = 89) (95% CI) Morgan et al also revealed that the subjects were able to isolate the PFM and avoid increasing the intra-abdominal pressure during the Resting 6.2 (2.7) 9.8 (4.3) 3.6 (2.6 to 4.5) MVC in supine, but they had difficulty contracting the PFM without MVC 37.5 (19.9) 28.8 (13.7) 28.7 (212.5 to 24.8) MVC = maximal voluntary contraction.
Research 59 increasing intra-abdominal pressure while standing.38 Therefore, the grouping of knowledge, collaborating for a greater discussion in the results need to be interpreted with caution because higher measures scientific community. do not always indicate better PFM function. What was already known on this topic: Pelvic floor mus- According to the literature, the supine position is one of the cles are commonly assessed in supine. The few studies that have easiest positions in which pelvic floor evaluations can be per- compared pelvic floor muscle function in supine and standing in formed.9,10,14 The standing assessment has the following disadvan- women with stress urinary incontinence have some conflicting tages compared with supine assessment: patients experience results. discomfort, therapists require more time to complete the examina- What this study adds: Measurements of pelvic floor muscle tion, there is no consensus about the best equipment to use,9,14,34 and function vary between standing and supine positions in women leakage of urine during the examination is more likely, which can with stress urinary incontinence with no previous treatment. The embarrass the patient. The discomfort caused by the examinations in upright position influences the ability of women to perform different positions was not evaluated and graded in this study; pa- satisfactory muscle contractions, where gravity and abdominal tients were only excluded if they reported discomfort with the de- pressure may overload the pelvic floor muscles. Although vices when the examinations started (supine or standing). The examining these muscles in supine is convenient, data obtained position of the therapist was standardised during the evaluations to in standing capture the natural action of the muscles in many minimise possible discomfort. More time is usually required for the activities of daily life. So, the standing position may be more valid examinations in both supine and standing, and the costs associated for analysing the function and mechanism of action of the pelvic with the longer examination time need to be considered. floor in relation to stress urinary incontinence. Some factors need to be considered in the interpretation of the re- Footnotes: a PeritronTM, Cardio Design Pty Ltd, Oakleigh, Victoria, sults of the present study. The literature has investigated the intra- Australia. abdominal pressures during physical activity and showed that the progression of the activity, such as walking to running, increases intra- b Miotec®, Porto Alegre, Rio Grande do Sul, Brazil. abdominal pressure in a similar way. The increased intra-abdominal c New Miotool Wireless, Miotec, Porto Alegre, Rio Grande do Sul, pressure in response to the increased demand for tasks may be due Brazil. to the role that intra-abdominal pressure plays in the stability of the d Medi-TraceTM, Kendall, Mansfield, MA, USA. spine and is aggravated by the high acceleration of impact forces.39 e Miotec SuiteTM software, version 1.0, Miotec, Porto Alegre, Rio Identifying the values of intra-abdominal pressure (using specific and Grande do Sul, Brazil. validated equipment) in the standing position can contribute to the f SPSS Version 17.0, SPSS Inc, Chicago, USA. understanding of its impact on PFM function and its capacity for muscle eAddenda: Tables 3, 5, 7, 9 and 10 can be found online at https:// contraction. We did not evaluate intra-abdominal pressure because doi.org/10.1016/j.jphys.2021.12.011. there is no specific instrument in our practice for measuring intra- Ethics approval: The Research Ethics Committee of the Uni- abdominal pressure during assessment of PFM in the standing position. versidade Federal de Sao Paulo (UNIFESP) in Brazil, approved this study (2.351.777). All participants gave written informed consent In addition, our findings were obtained from evaluation of a before data collection began. specific low socioeconomic population of women who attended the Competing interests: The authors declare they do not have any public health system in Brazil. Brazil suffers from great social conflicts of interest. inequality, where people with a high socioeconomic and educational Source(s) of support: This study was financed in part by the status usually are of low parity, with high rates of caesarean section Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - and adequate access to private healthcare and nutrition, whereas Brasil (CAPES) – Finance Code 001. those in the low-income population usually have higher parity and Acknowledgements: Coordenação de Aperfeiçoamento de Pessoal vaginal deliveries, less access to the public healthcare system and de Nível Superior - Brasil (CAPES) – Finance Code 001. may have deficient nutrition, among other differences. Given that, the Provenance: Not invited. Peer reviewed. findings should be confirmed in other populations nationally and Correspondence: Marcia Maria Gimenez, Federal University of São worldwide before being generalised. Paulo, São Paulo, Brazil. Email: [email protected] This observational, cross-sectional, analytical study followed the References standard guidelines for performance and report. Its strengths included randomisation of the order to perform the exams to mini- 1. Kegel AH. Progressive resistance exercise in the functional restoration of the mise interference of the starting position on the assessments and the perineal muscles. Am J Obstet Gynecol. 1948;56:238–248. https://doi.org/10.1016/ performance of the intra-examiner reproducibility test for all tools. 0002-9378(48)90266-x The sample size calculation was performed properly and the intended sample of 101 participants with SUI was enrolled. 2. Kegel AH. Stress incontinence and genital relaxation; a nonsurgical method of increasing the tone of sphincters and their supporting structures. Ciba Clin Symp. In conclusion, this study consistently showed that measurements 1952;4:35–51. of PFM function vary between standing and supine positions in women with SUI with no previous treatment. It demonstrated that 3. Dumoulin C, Cacciari LP, Hay-Smith EJC. Pelvic floor muscle training versus no the upright position influences the ability of women to perform treatment, or inactive control treatments, for urinary incontinence in women. satisfactory muscle contractions, as evidenced by manometry, EMG Cochrane Database Syst Rev. 2018;10:CD005654. and vaginal palpation. Those findings suggest that the action of gravity and abdominal pressure in the standing position causes 4. Castro RA, Arruda RM, Zanetti MRD, Santos PD, Sartori MGF, Girão MJBC. Single- overload on PFM. Physiological differences in the function of PFM can blind, randomized, controlled trial of pelvic floor muscle training, electrical stim- have wide implications for patient care, which may explain the dif- ulation, vaginal cones, and no active treatment in the management of stress ferences in symptom control and in the efficacy of the PFM training in urinary incontinence. Clinics. 2008;63:465–472. https://doi.org/10.1590/s1807- the treatment of SUI in the literature. Although it is always more 59322008000400009 convenient to study the action of PFM in the supine position, the data obtained in the upright posture capture the natural action of the PFM 5. Bø K, Hagen RH, Kvarstein B, Jørgensen J, Larsen S. Pelvic floor muscle exercise for in the position in which they work in daily life. So, the standing po- the treatment of female urinary stress incontinence: III. Effects of two different sition may be more valid for analysing the function and mechanism of degrees of pelvic floor muscle exercises. Neurourol Urodyn. 1990;9:489–502. action of PFM in relation to SUI. The precise description of the PFM https://doi.org/10.1002/nau.1930090505 assessment in the standing position, as performed and illustrated in the present study, will help to standardise this assessment, so that 6. Bø K. Vaginal squeeze pressure measurement. In: Bø K, Berghmans B, Mørkved S, future studies can be carried out, and thus allow comparison of Van Kampen M, eds. Evidence-based Physical Therapy for the pelvic floor. Edinburgh: findings between studies, consistency in publications and the Elsevier; 2015:61–66. 7. Bouchard C, Shephard RJ, Stephens T. Physical Activity, Fitness, and Health: Interna- tional Proceedings and Consensus Statement. Champaign: Human Kinetics Inc; 1994. 8. Bø K, Frawley H, Haylen B, Abramov Y, Almeida F, Berghmans B, et al. International Continence Society (ICS) joint report on the terminology for the conservative and
60 Gimenez et al: Pelvic floor function in different positions nonpharmacological management of female pelvic floor dysfunction. Int Urogy- 23. Dumoulin C, Morin M. Pelvic floor dynamometry. In: Bø K, Berghmans B, necol J. 2017;28:191–213. https://doi.org/10.1007/s00192-016-3123-4 Mørkved S, Van Kampen M, eds. Evidence-based Physical Therapy for the pelvic floor. 9. Bø K, Finckenhagen HB. Is there any difference in measurement of pelvic floor Edinburgh: Elsevier; 2015:61–66. muscle strength in supine and standing position? Acta Obstet Gynecol Scand. 2003;82:1120–1124. https://doi.org/10.1046/j.1600-0412.2003.00240.x 24. Grape HH, Dedering A, Jonasson AF. Retest reliability of surface electromyography 10. Frawley HC, Galea MP, Phillips BA, Sherburn M, Bø K. Reliability of pelvic floor on the pelvic floor muscles. Neurourol Urodyn. 2009;28:395–399. https://doi.org/ muscle strength assessment using different test positions and tools. Neurourol 10.1002/nau.20648 Urodyn. 2006;25:236–242. https://doi.org/10.1002/nau.20201 11. Arab AM, Chehrehraz M, Parhampour B. Pelvic floor muscle assessment in 25. Laycock J. Clinical evaluation of the pelvic floor. In: Schussler B, Laycock J, Norton P, standing and lying position using transabdominal ultrasound: comparison be- Stanton SL, eds. Pelvic Floor Re-education, Principles and Practice. London: Springer tween women with and without stress urinary incontinence. Aust NZ Continence Science; 1994:39–51. J. 2011;17:19. 12. Dietz HP, Clarke B. The influence of posture on perineal ultrasound imaging pa- 26. Madill SJ, McLean L. Quantification of abdominal and pelvic floor muscle synergies rameters. Int Urogynecol J Pelvic Floor Dysfunct. 2001;12:104–106. in response to voluntary pelvic floor muscle contractions. J Electromyogr Kinesiol. 13. Peng Q, Jones R, Shishido K, Constantinou CE. Ultrasound evaluation of 2008;18:955–964. https://doi.org/10.1016/j.jelekin.2007.05.001 dynamic responses of female pelvic floor muscles. Ultrasound Med Biol. 2007;33:342–352. 27. Hermens HJ, Freriks B. The SENIAM CD-ROM: European recommendations for surface 14. Aukee P, Penttinen J, Airaksinen O. The effect of aging on the electromyographic electromyography. Netherlands: Roessingh Research and Development. [S.l.: s.n.]; activity of pelvic floor muscles. A comparative study among stress incontinent 1999. 1 CD-ROM. patients and asymptomatic women. Maturitas. 2003;44:253–257. https://doi.org/ 10.1016/s0378-5122(03)00044-6 28. Richman J, Mackrides L, Prince B. Research methodology and statistics. Part 3: 15. Aukee P, Immonen P, Penttinen J, Laippala P, Airaksinen O. Increase in pelvic floor measurement procedures in research. Physiother Can. 1980;32:253–257. muscle activity after 12 weeks’ training: a randomized prospective pilot study. Urology. 2002;60:1020–1023. 29. Brennan P, Silman A. Statistical methods for assessing observer variability in clinical 16. Von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP, measures. BMJ. 1992;304:1491–1494. https://doi.org/10.1136/bmj.304.6840.1491 STROBE Initiative. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: guidelines for reporting observational studies. 30. Fonseca JS, Martins GA. In: Atlas, ed. Curso de Estatística. 3a. ed. São Paulo; 1982. Int J Surg. 2014;12:1495–1499. https://doi.org/10.1016/j.ijsu.2014.07.013 31. Pagano M, Gauvreau K. Princípios de Bioestatística. São Paulo: Pioneira Thomson 17. Ferreira CH, Bø K. The Pad Test for urinary incontinence in women. J Physiother. 2015;61:98. https://doi.org/10.1016/j.jphys.2014.12.001 Learning; 2004. 18. Krhut J, Zachoval R, Smith PP, Rosier PFWM, Valansky L, Martan A, et al. Pad weight 32. Bø K. Pressure measurements during pelvic floor muscle contractions: the effect of testing in the evaluation of urinary incontinence. Neurourol Urodyn. 2014;33:507– 510. different positions of the vaginal measuring device. Neurourol Urodyn. 1992;11:107–113. 19. Bump RC, Mattiasson A, Bø K, Brubaker LP, DeLancey JOL, Klarskov P, et al. The 33. Laycock J. Comparison of vaginal electromyography (EMG) in lying, sitting and standardisation of terminology of female pelvic organ prolapse and pelvic floor dysfunction. Am J Obstet Gynecol. 1996;175:10–17. standing. In: ICS Annual Meeting. Denver, Colorado, USA; 1999:325. 20. Bø K, Kvarstein B, Hagen R, Larsen S. Pelvic floor muscle exercise for the treat- 34. Rett MT, Simões JA, Herrmann V, Marques AA, Moraes SS. Existe diferença na contra- ment of female stress urinary incontinence: II. Validity of vaginal pressure measurements of pelvic floor muscle strength and the necessity of supple- tilidade da musculatura do assoalho pélvico feminino em diversas posições? Rev Bras mentary methods for control of correct contraction. Neurourol Urodyn. Ginecol Obstet. 2005;27:20–23. https://doi.org/10.1590/S0100-72032005000100005 1990;9:479–487. 35. Vodusek DB. Electromyography. In: Bø K, Berghmans B, Mørkved S, Van Kampen M, 21. Bø K, Finckenhagen HB. Vaginal palpation of pelvic floor muscle strength: inter- eds. Evidence-based Physical Therapy for the pelvic floor. Edinburgh: Elsevier; test reproducibility and comparison between palpation and vaginal squeeze 2015:53–60. pressure. Acta Obstet Gynecol Scand. 2001;80:883–887. https://doi.org/10.1034/j. 36. Brazález BN, Lacomba MT, La Villa P, Sánchez BS, Gómez PG, Del Barco AA, et al. The 1600-0412.2001.801003.x evaluation of pelvic floor muscle strength in women with pelvic floor dysfunction: 22. Laycock J, Jerwood D. Pelvic Floor Muscle Assessment: The PERFECT Scheme. a reliability and correlation study. Neurourol Urodyn. 2018;37:269–277. https://doi. Physiotherapy. 2001;87:631–642. org/10.1002/nau.2328 37. Romero-Cullerés G, Peña-Pitarch E, Jané-Feixas C, Vilaseca-Grané A, Montesinos J, Arnau A. Reliability and diagnostic accuracy of a new vaginal dynamometer to measure pelvic floor muscle strength. Female Pelvic Med Reconstr Surg. 2020;26:514–519. https://doi.org/10.1097/SPV.0000000000000796 38. Morgan DM, Kaur G, Hsu Y, et al. Does vaginal closure force differ in the supine and standing positions? Am J Obstet Gynecol. 2005;192:1722–1728. https://doi.org/10. 1016/j.ajog.2004.11.050 39. Dietze-Hermosa M, Hitchcock R, Nygaard IE, Shaw JM. Intra-abdominal Pressure and Pelvic Floor Health: Should We Be Thinking About This Relationship Differ- ently? Female Pelvic Med Reconstr Surg. 2020;26:409–414. https://doi.org/10.1097/ SPV.0000000000000799 ̌
Journal of Physiotherapy 68 (2022) 5–6 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 Pelvic health K Jane Chalmers a, Mark R Elkins b,c a IIMPACT in Health, Allied Health and Human Performance, University of South Australia, Adelaide, Australia; b Editor, Journal of Physiotherapy; c Faculty of Medicine and Health, University of Sydney, Sydney, Australia This Editorial introduces a new addition to Journal of Physiotherapy’s and in women.6 In the only male-specific article in the collection, article collections. These are collections of papers in a specific field of Nahon’s review5 provides a comprehensive overview of how conti- research, published in the Journal of Physiotherapy within the past nence is maintained in men, and the unique mechanisms and decade and curated to: facilitate access to recent important findings in management strategies for urinary, faecal and post-prostatectomy the field; highlight trends in the study designs, populations and in- incontinence. This review concludes with directions for future terventions addressed by the research; and provide a scoping overview research, with a focus on mirroring the work undertaken in female of avenues for further research. The studies in this article collection incontinence in defining exact protocols for management and the use relate to the role of physiotherapy in pelvic health. of adjunct therapies such as electrotherapy and biofeedback. ‘Pelvic health’ is an umbrella term used to describe the optimal The Invited Topical Review by Bø focuses on urinary incontinence functioning of the pelvic region, including the bladder, bowel, in females.6 The functional anatomy of continence in females is reproductive organs and pelvic floor muscles (PFMs). Examples of introduced, followed by the mechanisms of action of the PFMs in urge pelvic health conditions include urinary and faecal incontinence, and stress incontinence. Prevention and treatment strategies are pelvic organ prolapse (POP) and pelvic pain. While pelvic health explored, with particular detail on the role of group PFMT. Bø de- conditions are commonly associated with women, men can also scribes the evidence about the effects of group PFMT, concluding that develop problems. Several pelvic health conditions overlap between it is not inferior to individual PFMT, although there are concerns over the sexes: both men and women can develop incontinence and pelvic the correctness of the PFM contraction achieved with group PFMT. pain, while other conditions are unique, such as period pain in This review concludes with a specific, evidence-based protocol for women and prostatitis in men. In order to reflect that physiothera- PFMT and directions for future research focused on further rando- pists manage conditions beyond continence in both women and men, mised trials in unique populations, such as postpartum women and the Australian Physiotherapy Association changed the name of the those with POP. ‘Continence and Women’s Health group’ to the ‘Women’s, Men’s and Pelvic Health group’ in 2016. Pelvic floor muscle training and education This article collection focuses on several aspects of pelvic health Systematic reviews have demonstrated that PFMT is beneficial in physiotherapy: the management of incontinence, the role of pelvic the management of urinary incontinence in women.7 The third article floor muscle training (PFMT) and education in pelvic health and in this collection is a randomised controlled trial investigating the the effects of physiotherapy interventions for labour pain and effectiveness of PFMT for urinary incontinence in postmenopausal period pain. While the collection presents important research women who were and were not using hormone therapy.8,9 PFMT was supporting the role of physiotherapy in pelvic health, it is effective at increasing PFM strength and reducing prevalence of uri- acknowledged that not all pelvic health conditions are covered, and nary incontinence, although subgroup analyses showed that it was there is a particular emphasis on female pelvic health in research, more effective for women who were not on hormone therapy than which is reflected by the high proportion of female-focused articles women who were. Adherence rates to the PFMT program were in the collection. similar across both groups of women, but decreased when women transitioned from supervised sessions to a home program. Incontinence Adherence to home-based PFMT programs is often poor,6 and the Incontinence describes the involuntary loss of urine or faeces. fourth article in this collection investigated new methods of Prevalence rates of different kinds of incontinence across women and improving adherence to home-based PFMT programs for women men are under-researched, with the primary focus of prevalence studies with urinary incontinence.10 This randomised trial enrolled 86 post- being on urinary incontinence. The prevalence of urinary incontinence menopausal women with urinary incontinence and tested the effects is variable.1 While it can affect both men and women, urinary inconti- of additional video testimonials, prompting reminders and structured nence prevalence is increased two-fold to three-fold in women goal setting on adherence to a PFMT program. The trial clearly compared to men, regardless of age bracket.2 The prevalence of urinary showed that these interventions had no worthwhile effect on the incontinence increases with age, increased body mass index, pregnancy level of self-efficacy or adherence to the program. Both the experi- (in women), and surgery for prostate disease or cancer (in men).1 Sys- mental and control groups had moderate adherence to the PFMT tematic reviews support the role of physiotherapists in managing uri- program. These moderate adherence rates were higher than reported nary incontinence in both women3 and men.4 in many other studies; this may reflect strategies common to both groups such as the use of daily self-efficacy and exercise completion The first two articles in this collection are Invited Topical Reviews diaries, which may have motivated adherence behaviour. The authors that detail the physiotherapy management of incontinence in men5 https://doi.org/10.1016/j.jphys.2021.12.003 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/).
6 Editorial conclude that more research is needed to examine other strategies to experimental intervention consisted of additional exercise on a Swiss enhance home-based PFMT program adherence, but physiotherapists ball at 4 to 5 cm of dilation; lumbosacral massage provided by a can be confident that the strategies tested here were not clinically physiotherapist at 5 to 6 cm dilation; and a warm shower at . 7 cm useful. dilation. The experimental group had significantly lower pain severity immediately after: exercises (MD 24 mm, 95% CI 15 to 34), massage Education has also been investigated as a strategy to improve (14 mm, 95% CI 4 to 25) and showering (17 mm, 95% CI 5 to 29), which adherence to, and thereby outcomes from, PFMT programs. A rand- both reduced and delayed the use of pharmacological analgesia. omised trial by de Andrade et al11 investigated the effects of an ed- Participants in the experimental group also had a shorter expulsion ucation program (including instructions to perform PFM contractions time during active labour (by 18 minutes, 95% CI 5 to 30). No adverse during activities that increase intra-abdominal pressure) on PFM effects on mother or baby were found with the additional function, urinary incontinence, sexual function and women’s interventions. knowledge of PFMs. The control group received no intervention and no education. Participants were community-dwelling adults with Period pain varying degrees of urinary incontinence, including some without. At a 4-week follow-up, the group that received education were around The final article in this collection synthesises the evidence two or three times more likely to correctly answer questions about regarding physiotherapists’ roles in relieving menstrual pain in a PFM function, dysfunction and treatment options. Although the mean systematic review.18 Based on their review, Kannan and Claydon estimates of the effect on other outcomes all favoured the experi- recommend the use of heat, TENS and yoga, although these recom- mental group, these estimates were very imprecise, making effects on mendations were made on single studies for each intervention. Sig- these outcomes uncertain. These results suggest that education alone nificant reductions in pain were also found for acupressure and may not replace a rigorous PFMT program, but confirm that it pro- acupuncture; however, the authors noted that these were likely due vides additional benefits such as increasing women’s knowledge of to placebo effects because in trials where acupressure/acupuncture urinary incontinence and options for treatment. was compared with a sham rather than no intervention, the effect on pain severity was much less promising. New research has highlighted The importance of education is further supported in a longitudinal the role of exercise, including yoga, in reducing period pain,19 study by Fernandes et al,12 who interviewed community-dwelling although physiotherapy and related interventions are still not rec- women about their experience in attending a group-based pelvic ommended in recent clinical guidelines.20 floor education program. Participants reported varying degrees of PFM dysfunction, including urinary incontinence, faecal incontinence, In summary, the role of physiotherapy interventions, especially POP and sexual dysfunction. The education program was viewed PFMT, is well established for people with urinary incontinence, positively and increased participants’ knowledge of PFM function and although much more so for women than for men. PFMT is also dysfunction. Many participants also assimilated their new knowledge effective for some other types of pelvic dysfunction (eg, POP and by adopting a PFMT program at home and sharing their knowledge sexual dysfunction), although there is still need for further research with friends and family. It is possible that education not only pro- about optimal dosage for different conditions and ways to get people motes PFMT uptake and adherence, but could also result in wider to adhere to the programs. The role of physiotherapy interventions in awareness in the community, helping to prevent issues such as PFM managing labour pain is an emerging area of research with promising weakness and urinary incontinence in the first place. results. Future research directions should examine combining in- terventions to maximise pain relief. The effects of physiotherapy in- Although the bulk of the literature focuses on urinary inconti- terventions on period pain are heavily under-researched; this reflects nence, the benefits of PFMT extend beyond maintaining continence. the wider pain literature, where not much has been undertaken on In particular, PFMT is considered the first-line treatment for POP.13 the condition (especially primary dysmenorrhoea). Given that phys- The final article pertaining to PFMT in the collection is a rando- iotherapists are often first contact for people with persistent pain, it is mised controlled trial that investigated whether perioperative PFMT an important area for us to expand in our clinical practice and in patients undergoing POP surgery could improve outcomes research. compared with surgery alone.14 Interestingly, the perioperative PFMT did not make a clinically worthwhile difference to POP symptoms at Competing interests: Nil. either 40 or 90 days after surgery. While PFMT protocols for urinary Source(s) of support: Nil. incontinence and conservative POP management are well estab- Acknowledgements: Nil. lished, the same cannot be said for PFMT as an adjunct to POP sur- Provenance: Invited. Not peer reviewed. gery. The authors postulate that a different dosage of PFMT may Correspondence: Mark R Elkins, Centre for Education & Workforce result in different outcomes. In any case, given the evidence that Development, Sydney Local Health District, Sydney, Australia. Email: PFMT reduces POP symptoms and severity stage,13,15 PFMT should [email protected] still be considered first-line treatment for POP before surgery is considered. Labour pain References Two of the randomised trials in this article collection relate to the 1. Buckley BS, et al. Urology. 2010;76:265–270. role of physiotherapy in reducing labour pain. Santana et al16 assessed 2. Milsom I, et al. Climacteric. 2019;22:217–222. the effect of transcutaneous electrical nerve stimulation (TENS) on 3. López Liria R, et al. J Womens Health. 2019;28:490–501. pain and use of pharmacological analgesia during labour. In their 4. Hall LM, et al. Neurourol Urodyn. 2020;39:533–546. sample of 46 low-risk primigravida women, TENS reduced labour 5. Nahon I. J Physiother. 2021;67:87–94. pain severity by a mean of 15 mm on a 100-mm visual analogue scale 6. Bø K. J Physiother. 2020;66:147–154. (95% CI 2 to 27). Also, the mean time to pharmacological analgesia 7. Dumoulin C, et al. Cochrane Database Syst Rev. 2018;10:CD005654. was 5 hours longer in the TENS group (95% CI 4 to 6). These results 8. Antônio FI, et al. J Physiother. 2018;64:166–171. highlight that the application of TENS in addition to routine obstetric 9. Antônio FI, et al. J Physiother. 2020;66:7–8. care reduces the severity of pain during labour and postpones the 10. Sacomori C, et al. J Physiother. 2015;61:190–198. need for pharmacological pain relief. 11. de Andrade RL, et al. J Physiother. 2018;64:91–96. 12. Fernandes ACNL, et al. J Physiother. 2021;67:210–216. In the other trial, Gallo et al17 investigated the sequential appli- 13. Li C, et al. Int Urogynecol J. 2016;27:981–992. cation of physiotherapy interventions on labour pain, duration of 14. Duarte TB, et al. J Physiother. 2020;66:27–32. labour and time until pharmacological analgesia use in labour. Both 15. Brækken IH, et al. Neurourol Urodyn. 2014;33:115–120. the control and experimental groups received routine care. The 16. Santana LS, et al. J Physiother. 2016;62:29–34. 17. Gallo RBS, et al. J Physiother. 2018;64:33–40. 18. Kannan P, et al. J Physiother. 2014;60:13–21. 19. Armour M, et al. Cochrane Database Syst Rev. 2019;9:CD004142. 20. Burnett M, et al. J Obstet Gynaecol Can. 2017;39:585–595.
Journal of Physiotherapy 68 (2022) 8–25 j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j p hy s Invited Topical Review Physiotherapy management for COVID-19 in the acute hospital setting and beyond: an update to clinical practice recommendations Peter Thomas a, Claire Baldwin b, Lisa Beach c, Bernie Bissett d,e, Ianthe Boden f,g, Sherene Magana Cruz h, Rik Gosselink i,j, Catherine L Granger c,k, Carol Hodgson h,l,m,n, Anne E Holland o,p, Alice YM Jones q, Michelle E Kho r,s,t, Lisa van der Lee u, Rachael Moses v, George Ntoumenopoulos w, Selina M Parry k, Shane Patman x a Department of Physiotherapy, Royal Brisbane and Women’s Hospital, Brisbane, Australia; b Caring Futures Institute, College of Nursing and Health Sciences, Flinders University, Adelaide, Australia; c Department of Physiotherapy, The Royal Melbourne Hospital, Melbourne, Australia; d Discipline of Physiotherapy, University of Canberra, Canberra, Australia; e Physiotherapy Department, Canberra Hospital, Canberra, Australia; f Physiotherapy Department, Launceston General Hospital, Launceston, Australia; g School of Medicine, University of Tasmania, Launceston, Australia; h Australian and New Zealand Intensive Care Research Centre, Monash University, Melbourne, Australia; i Department of Rehabilitation Sciences, KU Leuven, Leuven, Belgium; j Department of Critical Care, University Hospitals Leuven, Leuven, Belgium; k Department of Physiotherapy, The University of Melbourne, Melbourne, Australia; l Alfred Health, Melbourne, Australia; m Department of Critical Care, School of Medicine, University of Melbourne, Melbourne, Australia; n The George Institute for Global Health, Sydney, Australia; o Central Clinical School, Monash University, Melbourne, Australia; p Departments of Physiotherapy and Respiratory Medicine, Alfred Health, Melbourne, Australia; q School of Health and Rehabilitation Sciences, The University of Queensland, Brisbane, Australia; r School of Rehabilitation Science, McMaster University, Hamilton, Canada; s St Joseph’s Healthcare, Hamilton, Canada; t The Research Institute of St Joe’s, Hamilton, Canada; u Physiotherapy Department, Fiona Stanley Hospital, Perth, Australia; v NHS Leadership Academy, Leadership and Lifelong Learning, People Directorate, NHS England and Improvement, London, UK; w Department of Physiotherapy, St Vincent’s Hospital, Sydney, Australia; x Faculty of Medicine, Nursing and Midwifery, Health Sciences & Physiotherapy, The University of Notre Dame Australia, Perth, Australia KEY WORDS ABSTRACT Physical therapy This document provides an update to the recommendations for physiotherapy management for adults with Coronavirus coronavirus disease 2019 (COVID-19) in the acute hospital setting. It includes: physiotherapy workforce COVID-19 planning and preparation; a screening tool for determining requirement for physiotherapy; and recom- mendations for the use of physiotherapy treatments and personal protective equipment. New advice and recommendations are provided on: workload management; staff health, including vaccination; providing clinical education; personal protective equipment; interventions, including awake proning, mobilisation and rehabilitation in patients with hypoxaemia. Additionally, recommendations for recovery after COVID-19 have been added, including roles that physiotherapy can offer in the management of post-COVID syndrome. The updated guidelines are intended for use by physiotherapists and other relevant stakeholders caring for adult patients with confirmed or suspected COVID-19 in the acute care setting and beyond. [Thomas P, Baldwin C, Beach L, Bissett B, Boden I, Cruz SM, Gosselink R, Granger CL, Hodgson C, Holland AE, Jones AYM, Kho ME, van der Lee L, Moses R, Ntoumenopoulos G, Parry SM, Patman S (2022) Physiotherapy management for COVID-19 in the acute hospital setting and beyond: an update to clinical practice recommendations. Journal of Physiotherapy 68:8–25] © 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 update recommendations for physiotherapy practice and service deliverya,b. The recommendations remain focused on adult patients Recommendations for physiotherapy management for corona- in acute hospital settings and are structured around: physiotherapy virus disease 2019 (COVID-19) in the acute hospital setting1 were workforce planning and preparation; delivery of physiotherapy produced in March 2020 in response to the emerging pandemic and interventions, including both respiratory and mobilisation/reha- urgent need for guidance for physiotherapists world-wide. Since bilitation; and PPE requirements for physiotherapy service delivery. then, COVID-19 cases have exceeded 258 million2 and deaths have They have also been expanded to address the long-term impacts of exceeded 5.1 million.2 The experience of healthcare providers and COVID-19 and the implications they have for acute hospital phys- policy-makers in dealing with the pandemic and research specific iotherapy services. These recommendations will continue to be to the COVID-19 population has evolved rapidly. The aim of this updated, as required, in response to future development of evi- second document is to inform physiotherapists and key stake- dence necessitating a change in physiotherapy practice for hospi- holders of relevant changes in the management of COVID-19 and to talised adults with COVID-19. https://doi.org/10.1016/j.jphys.2021.12.012 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/).
Invited Topical Review 9 Methods Table 1 World Health Organization categories of COVID-19 disease severity in adultsa. Consensus approach Category Definition All previous authors were invited to contribute to this update. The skills and experience of the authors were reviewed and an invitation Non-severe Symptomatic patients without evidence of viral pneumonia extended to two additional cardiorespiratory physiotherapy experts (ie, no fever, cough, dyspnoea or hyperpnoea) and without hypoxia (LB, AEH) who brought additional expertise in pandemic leadership (ie, SpO2 90% on room air) and models of care (LB) and pulmonary rehabilitation (AEH). A con- sumer representative with lived experience of COVID-19 (SMC) was Severe Clinical signs of pneumonia (fever, cough, dyspnoea or also invited to review the recommendations. hyperpnoea)b with at least one of the following: The AGREE II framework3 was used to guide reporting. To guide - respiratory rate . 30 breaths/min the revision of original or the development of new recommendations - severe respiratory distress and decision-making, all members of the authorship group assisted in - SpO2 , 90% on room air conducting literature searches and reviewing international guide- lines. Given the rapid evolution of evidence and wide scope of our Critical Requires the provision of life-sustaining therapies such as guidance, systematic reviews or guidelines were sought for each section wherever possible. However, at times, the most relevant mechanical ventilation (invasive or non-invasive) or vasopressors primary studies were chosen using best clinical and methodological judgement. with presentations including: - Acute respiratory distress syndrome179 All authors reviewed the previous recommendations and nomi- - Sepsis180 nated recommendations that should be revised or revoked. The lead - Septic shock180 author (PT) circulated a draft document that included previous rec- ommendations and items that were nominated to be revoked, revised COVID-19 = coronavirus disease 2019, CT = computerised tomography, SpO2 = or added. All authors had the opportunity to vote to revoke items, or oxyhaemoglobin saturation. approve new or revised recommendations, with 70% agreement for approval. Votes were conducted independently via return to the lead a Adapted from the Clinical management of COVID-19 patients: living guidance.181 author. Votes were tallied and any feedback collated and de- b While the diagnosis can be made on clinical grounds, chest imaging (radiograph, identified, then presented back to all authors. All new and revised recommendations were discussed in a follow-up videoconference, CT scan, ultrasound) may assist in diagnosis. where minor alterations in recommendations were made if required. health;12 this includes the Alpha, Beta and Gamma variants. The Delta After the guidelines were developed, a consumer (SMC) was variant, which was first detected in India in October 2020, is currently invited to review all recommendations and provide feedback. the ‘variant of concern’.12 Variants of concern appear to be signifi- Endorsement of the revised recommendations was again sought from cantly more transmissible and are associated with a higher viral load, physiotherapy societies, physiotherapy professional groups and World Physiotherapy. longer infectious periods, increased risk of severe illness requiring hospitalisation, and mortality.12,13 The emergence of variants is Epidemiology and key public health measures for COVID-19 anticipated to continue and will require ongoing research to under- While the global number of COVID-19 cases now exceeds 258 stand the consequences of the different variants on initial acuity of million,2 the weekly incidence of COVID-19 cases and deaths has been gradually declining in all regions, except Europe, since late August presentation, long-term sequalae and trajectories for recovery. 2021.4 Classifications for disease severity have now been defined by The cornerstone of disease prevention remains a combination of the World Health Organization (WHO)5 (Table 1). Similar classifica- tions are incorporated within Australian guidelines, which include public health measures for infection control and vaccination. Guid- additional clinical descriptors.6 In Australia and the United States, the ance on public health measures and exposure risk controls have majority of people with COVID-19 have non-severe disease; however, approximately 13% are admitted to hospital and 2% require intensive changed since the start of the pandemic, as evidence about the spread care unit (ICU) admission.7,8 Similar rates of severe (14%) and critical (5%) disease have been reported in China.9 Mortality associated with of COVID-19 has developed. Early in the pandemic the WHO advised COVID-19 appears higher in the United States (5%)8 compared to that transmission of the virus between people was primarily through China (2.3%)9 and Australia (1%).7 This may be explained by many droplet and contact routes;14 this advice has since changed.15 There is factors, including regional differences in population demographics, now substantial evidence supporting the airborne transmission of local healthcare responses and the robustness of data reporting. COVID-19.15–21 Subsequently, public health recommendations for While at the beginning of the pandemic the incidence of COVID-19 preventative measures have shifted to include the use of three-layer was highest in elderly people aged at least 60 years, a shift has been seen in this second pandemic year with highest case numbers face masks and ensuring natural ventilation of enclosed spaces, in now in individuals aged , 40 years.10 In 2021, the highest rate of addition to the standard messaging of physical distancing by at least infection in Australia has been in the 20 to 29-year age group and a 1 m and avoiding crowded places.15,17,22 slightly higher rate of infection is seen among males than females.7 While higher case numbers are occurring in younger people, hospi- The development and testing of safety and efficacy of vaccines for tal admissions remain predominantly in older age groups.11 Ethnicity COVID-19 has been instrumental in managing COVID-19. As of 25 may also impact on the severity of COVID-19: for example, patients of November 2021, more than 7.4 billion vaccine doses have been Indian and Pakistani origin have been identified as a higher-risk delivered world-wide, with 3.1 billion people being fully vaccinated,2 group in the UK.11 which reflects approximately 39% of the world’s population.23 How- ever, there have been and continue to be large differences in vaccine Genetic lineages of COVID-19 have been emerging and circulating access and rollout across countries.24 For example, African regions around the world. Several variants currently classified as ‘variants being monitored’ have had significant and sustained reduction in have on average approximately 12.7% of their population fully vacci- regional proportions over time or now pose lower risk to public nated versus European regions, which average approximately 53.7%.23 The inequitable access to vaccines increases the risk of emergence of new lineages of COVID-19 that may be even more threatening and require ongoing development of vaccines to ensure their effectiveness. Of critical relevance for healthcare is that COVID-19 in the hospital setting is now becoming a disease predominantly of the unvacci- nated. The probability of severe or critical disease from COVID-19 is ameliorated through vaccination,25,26 with substantially lower rates of emergency department utilisation, hospitalisation and admission to ICU in vaccinated populations.11,27 However, even after vaccination, there is an elevated risk for hospital admission and death due to COVID-19 for some groups. High-risk groups appear to include: people with Down’s syndrome; immunosuppression due to chemo- therapy, previous solid organ transplantation (particularly kidney transplantation) or recent bone marrow transplant; HIV and AIDS; liver cirrhosis; neurological disorders, including dementia and Par- kinson’s disease; and residents in aged-care facilities.11 Increased susceptibility may also be seen with conditions such as chronic
Search
