Exploring factors influencing physiotherapists’ perceptions of measuring reactive balance following a theory-based multi-component intervention: a qualitative descriptive study

Abstract

Purpose: Reactive balance is a critical consideration for mobility and fall avoidance, but is under-assessed among physiotherapists. The objective of this study was to explore factors influencing physiotherapist perceptions about measuring reactive balance upon completion of a 12-month theory-based, multi-component intervention to increase use of a measure of reactive balance.

Methods: A qualitative descriptive approach was used. Semi-structured interviews were conducted with 28 physiotherapists treating adults with balance impairment in three urban Canadian rehabilitation hospitals that participated in the intervention. Interviews explored perceptions of reactive balance measurement and perceived changes in clinical behavior. Thematic analysis involved multiple rounds of coding, review and discussion, theme generation, and interpretation of findings through individual analysis and team meetings.

Findings: Participants expressed contrasting views about integrating reactive balance measurement in their practice, despite consistent acknowledgement of the importance of reactive balance for function. Three themes were identified highlighting factors that mediated perceptions about measuring reactive balance: patient characteristics; trust between physiotherapist and patient; and the role of physiotherapist fear.

Conclusions: The findings highlight that decision making for measuring reactive balance in rehabilitation settings is complex. There is a need for additional work to facilitate long-term implementation of clinical reactive balance measurement, such as refining patient criteria for administration, ensuring sufficient time to establish a trusting relationship, and developing and testing strategies to address physiotherapist fear.

IMPLICATIONS FOR REHABILITATION

• Reactive balance is important for falls prevention and mobility, but is under-assessed among physiotherapists.

• This study identified three factors that influenced uptake of reactive balance measurement among physiotherapists in rehabilitation settings: patient characteristics; trust between physiotherapist and patient; and the role of physiotherapist fear.

• Knowledge of the identified factors may assist with design and use of reactive and other balance measurements.

• Strategies aimed at developing trusting relationships between physiotherapist and patient along with addressing physiotherapist fear could facilitate the uptake of clinical reactive balance measurement.

Keywords:

• Aging
• balance
• implementation research
• postural control
• rehabilitation

Background

Improving mobility and reducing falls are important components of physiotherapy practice with many clinical populations. In this regard, reactive balance is recognized as a critical functional skill for recovering from a loss of balance and avoiding a fall [1] through a rapid corrective postural muscle response, step or grasp [2]. Although reactive balance is impaired in numerous clinical populations [3–6], meta-analyses have determined that when it is improved through targeted exercise, it is associated with reduced number of falls and fallers [7,8]. And yet, in spite of this strong evidence, multiple studies have demonstrated that reactive balance is under-assessed by Canadian physiotherapists relative to other balance constructs (for example: anticipatory control, static, and dynamic stability) [9–11].

Recognition of the need to support clinicians in engaging in evidence-informed practices, such as addressing reactive balance in physiotherapy has sparked interest in advancing implementation methods focused on the process of adopting new behaviors in the delivery of health care [12]. However, the complex nature of evidence-informed physiotherapy practice applied in rehabilitation settings with diverse clinical populations and program models makes it a unique and challenging context for implementation [13]. Studying methods to increase evidence-informed physiotherapy practice can contribute to the advancement of implementation methods while enriching understanding of complex factors influencing clinical practice and the targeted clinical issue. The overarching goals of this research program are to increase reactive balance assessment and treatment among physiotherapists, and in turn reduce falls and associated injuries among adults with balance impairment. Over the past decade, we have used implementation process frameworks to identify gaps in Canadian physiotherapists assessment of reactive balance [9–11], determine barriers and facilitators to reactive balance assessment [14], determine the extent to which reactive balance is included in validated measures of balance [15], and most recently, to develop and evaluate a theory-based multi-component intervention to increase reactive balance measurement among physiotherapists working in rehabilitation settings [16].

Full details of intervention development and delivery have been previously published [13,16]. In brief, the intervention was developed using a behavior change theory for health professionals (Theoretical Domains Framework [17]). The intervention development process involved mapping six established facilitators of and barriers to reactive balance measurement onto eight domains of the framework (knowledge; skills; social/professional role and identity; beliefs about capabilities; beliefs about consequences; intentions; memory, attention and decision processes; environmental context), and then to eight established behavior change techniques for modifying each domain (information regarding behavior and outcome; rehearsal of relevant skills; feedback; social processes of encouragement, pressure and support; problem-solving, decision-making and goal setting; modeling/demonstration of behavior by others; environmental changes; prompts, triggers, cues). These techniques were organized into three intervention components delivered over a 12-month period: onsite physiotherapist local champions; group meetings involving didactic education, hands-on practice, and ongoing discussion; and health record modifications that included a reactive balance measure administration form. A summary of intervention components is described in Table 1.

Table 1. Intervention components and features.

Component	Feature
Onsite physiotherapist local champion	Physiotherapists with advanced practice or management role at each site, who were also part of research team Worked to actively engage participants by: Coordinating meetings communicating study objectives co-facilitating participant meetings modeling use of reactive balance measure in their practice mentoring participants, responding to questions and concerns brokering interactions between participants and off-site research team co-developing reactive balance measure administration forms
Group meetings	1 60-min didactic education meeting delivered by principal investigator 1 60-min hands on practice meeting facilitated by two physiotherapists 5 60-min bimonthly check-in meetings with the research team to discuss successes, challenges, questions, concerns, practice, etc.
Health record modification	Custom reactive balance measure administration form Developed by research team in consultation with participants Included administration instructions, scoring guidance, and met institutional specifications

The reactive balance measure implemented was the postural responses section of the Balance Evaluation Systems Test [18], a 6-item measure requiring no equipment and scored on a four-point, observation-based scale. The measure had established excellent interrater reliability [18], normative data [19], and documented use in multiple clinical populations [20–22]. Administration of the reactive balance measure is illustrated in Figure 1.

Figure 1. Reactive balance measure administration positions. Administering the measure required patients to lean on the hands of the physiotherapist and be physically supported by them. This position was designed to move the center of mass outside the base of support to trigger a reactive postural response when the physiotherapist unexpectedly removed physical support. The leaning was administered so that the patient moved towards the physiotherapist as a safety mechanism to catch the patient in the event the reaction was not successful (Sibley et al. 2018 [13]).

The intervention was administered to 28 physiotherapists working with adult populations having balance impairment with a range of clinical diagnoses at three urban rehabilitation hospitals in Ontario, Canada in 2015–2016. Quantitative findings from health record abstraction indicated that the reactive balance measure was used in 31% of eligible records, compared to the pre-intervention baseline period that did not include use of any standardized reactive balance measure. Questionnaires completed by physiotherapist participants showed a significant increase in knowledge about and confidence for measuring reactive balance following the intervention, and most participants positively rated satisfaction with intervention content and delivery [13].

While these quantitative findings provide evidence of some success of the intervention, they offer limited insight into the underlying factors or influences affecting the interventions’ primary outcome (use of the reactive balance measure). Multiple research methodologies, including both quantitative and qualitative approaches, are recommended in implementation research to fully elucidate intervention effects [23] and better understand the realities and nuances of clinical practice. Exploring the construct of reactive balance measurement from the perspectives of the physiotherapists who participated in the intervention presents the greatest opportunity for understanding reactive balance measurement considerations in rehabilitation settings and how the intervention worked. Together, these insights can help to refine future iterations of the intervention. Our overall intervention evaluation plan involved a sequential mixed methods approach following a QUAN → qual structure [24]. To fulfill this plan, the purpose of this study was to elaborate on the quantitative findings and explore factors influencing physiotherapists’ perceptions about reactive balance measurement after completing the intervention.

Methods

Study design

A qualitative descriptive approach was adopted for this study. This approach is naturalistic [25] and attempts to “discover and understand a phenomenon, a process, or the perspectives and worldviews of the people involved” [25]. Qualitative descriptive research has been used in pragmatic health and implementation research [26], and found to be useful for producing practical answers to real-world questions [27]. The practical nature of our research question focused on exploring physiotherapist perceptions of reactive balance to better understand intervention effects and inform future studies. These were the driving factors behind the selection of the qualitative descriptive approach. Qualitative descriptive methods are recognized in mixed methods research [28], and have been noted for incorporating the strengths of other established methodologies while maintaining flexibility where a study’s objectives do not fully align with long standing methodologies (e.g., phenomenology or ethnography) [29]. The qualitative descriptive approach shaped the development of the interview questions, emphasizing physiotherapist experiences and views on reactive balance, and shaped the analysis by focusing on the identification of trends in participants’ descriptions. Published guidance for conducting qualitative descriptive research were applied [25,30]. The Standards for Reporting Qualitative Research checklist was used to guide reporting [31]. Research ethics approval was obtained at all sites.

Participants

Physiotherapists treating adults with balance impairment working at the study sites were eligible to participate in the intervention as long as they had no involvement in other clinical balance research. All physiotherapists who participated in the intervention were eligible to participate in this study (n = 28). Participants consented to complete an in-depth interview following the intervention as part of their overall participation.

Research team

The research team consisted of registered physiotherapists, individuals who had previously studied and/or incorporated reactive balance in clinical practice, and individuals with qualitative research methods expertise. Data collection was conducted by one female postdoctoral research trainee (DCB) with expertise in kinesiology, who was known to participants prior to data collection through coordination of intervention activities, not a physiotherapist, and not involved in intervention design. The trainee received training from the principal investigator and qualitative research methods expert lead. Participants were aware that the research team was interested in learning about their intervention experiences and perceptions of reactive balance measurement as part of the study evaluation.

Data collection and analysis

Physiotherapists were contacted individually by email to participate in interviews and schedule a time convenient to them. One-on-one, in-person, semi-structured interviews were conducted at the study sites over a three-month period following the intervention. An interview guide was drafted and reviewed for clarity and appropriateness with physiotherapist research team members (Appendix 1). Interview questions explored physiotherapist experiences participating in the study, perceptions on the evaluation of reactive balance in clinical practice, and perceived changes in clinical behavior on completion of the study. Probes were used as needed throughout the interviews to expand on experiences and perceptions in detail and to clarify meaning. Interviews were digitally recorded and transcribed verbatim using a professional service.

Interview transcripts were reviewed for accuracy and de-identified prior to thematic analysis. Analysis was guided by Braun and Clark’s 5-step approach [32], and was closely aligned with the method Braun et al describe as a “codebook” approach [33] whereby we took a semi-structured approach with multiple coders and a general thematic structure that included identifying and describing domains. The specific steps were employed as follows:

1. All members of the research team independently read four transcripts (one randomly selected from each site and a fourth from the site with largest enrollment) to become familiar with the data, made notes, then met to discuss initial reflections on participant experiences and perceptions of reactive balance measurement [34].

2. The principal investigator (KMS) and a research coordinator co-investigator (MK) then read and open-coded five additional transcripts and met to compare codes and develop a coding scheme broadly informed by the research teams’ initial reflections.

3. The research coordinator coded the remaining transcripts, and organized codes into draft preliminary themes. NVivo 10 software was used to organize data and themes.

4. The principal investigator reviewed codes and draft themes, then re-organized and revised themes with a focus on physiotherapist perceptions about measuring reactive balance, and relationships between perceptions of reactive balance and intervention experiences, and then identified emergent findings and supporting participant quotes.

5. All members of the research team met to review, discuss, and refine themes. Preliminary findings were presented at an academic conference in June 2018. Themes continued to be discussed and refined with all team members until agreement was reached [35].

Credibility was established through data triangulation and an audit trail. Analyst triangulation was employed, involving multiple analysts discussing and generating key themes [36]. An audit trail was maintained by documenting discussions and decisions made throughout data collection and analysis. Trustworthiness of the data was enhanced by staying close to the data and participants’ own words throughout the analysis.

Findings

Participants

All 28 physiotherapists who completed the intervention participated in the interviews (Table 2). The majority of participants (86%) identified as female. Participants had on average 12 (SD 6.2) years of clinical experience, and 50% held an entry-to-practice degree at the Masters’ level. Most participants (68%) estimated that in a typical week, the majority of their caseload included adults with balance impairment at risk of falls. Participants worked with neurological (39%), orthopedic (29%), and geriatric populations (14%), and those with multiple or complex conditions (18%). Most (71%) worked in inpatient settings.

Table 2. Physiotherapist characteristics.

Characteristic	Mean (SD) or n (%)
Female	24 (86%)
Years since graduation	12 (6.2)
Highest degree attained
Bachelor’s	14 (50%)
Master’s	14 (50%)
Primary area of practice
Neurological	11 (39%)
Orthopedic	8 (29%)
Geriatric	4 (14%)
Multiple/ complex conditions	5 (18%)
Clinical program type
Inpatient	20 (71%)
Outpatient	4 (14%)
Undesignated	4 (14%)
Estimated proportion of caseload with balance impairment at risk of falls in a typical week
20–39%	2 (7%)
40–59%	3 (11%)
60–79%	4 (14%)
80+%	19 (68%)

Key findings

Participants expressed contrasting perceptions about measuring reactive balance in their practice, although the majority recognized the importance of reactive balance for function. One participant specifically noted the importance of considering reactive balance because of its relevance to the “real world” (P01-12, multiple clinical populations, 3 years in clinical practice). Participants could be categorized as having a generally positive perception of reactive balance measurement, or a generally negative perception that was largely related to relevance and feasibility. Some participants were enthusiastic about the measure, felt it was feasible, and targeted a skill that wasn’t addressed by other clinical tools. Such participants identified as routinely incorporating the reactive balance measure on completion of the intervention: “It’s a part of my repertoire for assessments now” (P02-001, neurological populations, 18 years in clinical practice). Meanwhile, a number of participants shared significant concerns about measuring reactive balance, questioning its relevance and reported that they did not integrate the measure in their practice. This was transparently acknowledged by P01-013 (multiple/complex populations, 8 years in clinical practice): “I’d be lying if I said I changed my practice to include that test”.

Three themes were identified, illustrating factors that were seen to mediate the contrasting perceptions about measuring reactive balance: patient characteristics, the need to establish trust between the physiotherapist and patient, and the role of physiotherapist fear. Collectively, the findings illustrated the complexity of decision-making among physiotherapists and how this process and perceptions influenced the use of the reactive balance measure.

Factor 1: patient characteristics

Patient characteristics, in many cases related to clinical assessments of overall functional ability, were strongly related to physiotherapists’ perceptions about the relevance and feasibility of measuring reactive balance. Such patient-related issues dominated physiotherapists’ perceptions of the challenges associated with measuring reactive balance, noted in some form by all 28 participants. Comments indicated that physiotherapists considered reactive balance to be a higher-order skill and some believed it was inappropriate to assess on all rehabilitation clients. As P03-001 (geriatric populations caseload, 10 years in clinical practice) discussed, “I don’t see myself using the test for the future. Maybe with very, very few selective patients which are mostly independent and highly functional”. To this end, a number of physiotherapists, often those working in inpatient settings, did not perceive reactive balance as a priority and identified different treatment goals. In multiple instances this led to triaging of the measure second to other established balance measures and constructs (such as static stability). As described by P03-005 (orthopedic populations caseload, 3 years in clinical practice): “I was only tempted to test somebody’s reactive balance if their other parts of their balance were already good cause then it would sort of be that last piece right?”

Factor 2: trust between physiotherapist and patient

The second factor that mediated physiotherapists’ perceptions about the feasibility of measuring reactive balance was the level of trust needed between themselves and their patients which they felt was required in order to administer the measure. Several physiotherapists noted that patients expressed fear in performing the reactive balance measure due to the leaning position they were required to adopt in order to administer the test. As one physiotherapist noted, “It’s too scary for them, whatever it is. I tried it [the reactive balance measure] with a couple of people and they were like please don’t do that again” (P03-006, orthopedic populations caseload, 9 years in clinical practice). To mitigate this fear, physiotherapists described the need to establish trust with their patients in order to administer the reactive balance measure, a process which often took time. As described by another physiotherapist: “I think there is a very important part of kind of building the relationship between the therapist and the patient and I do think that that [the reactive balance measure] could strain the relationship especially at the beginning, especially if they’re kind of borderline or obvious that you’re going to need to catch them” (P01-005, neurological populations caseload, 9 years in clinical practice). Establishing a trusting relationship with patients was seen as challenging as time is restricted, particularly in inpatient rehabilitation settings.

Factor 3: the role of physiotherapist fear

Lastly, fear on the part of the physiotherapist was critical to one’s perception of feasibility of reactive balance measurement, particularly as it related to fears associated with the risk of administering the measure. These fears were primarily related to causing distress, pain, or injury to patients, as one participant commented: “extra little kind of force through the leg is going to cause them a lot of pain. And that’s not something I’m willing to risk” (P01-011, orthopedic populations caseload, 15 years in clinical practice). Some participants also expressed a fear of personal injury to themselves. Fear also surfaced in comments related to the safety of the reactive balance measure itself. For some participants these fears persisted throughout the intervention and remained firmly intact on completion of the study. Others described overcoming these fears as they practiced and gained confidence using the measure during the intervention. This was articulated by P03-003 (geriatric populations caseload, 9 years in clinical practice) as, “When I was a younger PT, of course I would be way more protective, but as, you know, the years go by, you gain more experience working. You’re more willing, you’re more confident in your skills, in your patient skills, to let them take more risks, and I think that’s – that’s important, because we don’t want to hold them back from being mobile. We don’t want to limit their abilities, or their potential to - to move, and to exercise, and to be active.”

Discussion

This study provided new knowledge about perceptions of reactive balance among physiotherapists working in rehabilitation settings and identified critical factors influencing decision making related to using the reactive balance measure. Reactive balance is critical for fall avoidance, but in light of established gaps in physiotherapy practice related to balance assessment, little is known about how clinicians conceptualize this important skill. The opportunity to study physiotherapist perceptions about reactive balance from a group of therapists who actively worked to integrate reactive balance measurement into practice over a 12-month period is novel and provides rich insight from the forefront of practice. An important strength of the study was that all 28 physiotherapists who participated in the intervention agreed to be interviewed for this study, thereby achieving sampling saturation in the conduct of the study. The inclusion of this qualitative study as a component of the overall evaluation of the intervention is also a strength and recommended component in implementation research.

Understanding differences in participants’ views about the relevance and feasibility of measuring reactive is important as it helps to explain the quantitative findings that showed use of the reactive balance measure in 31% of eligible health records, despite significant increases in reactive balance knowledge and skills and a positive intervention experience. The identification of three factors (patient characteristics; trust between physiotherapist and patient; and the role of physiotherapist fear) that were seen to mediate physiotherapist perceptions and therefore use of the reactive balance measure are particularly relevant for informing future efforts to support integration of reactive balance measurement by physiotherapists, offering insight as to how to further refine and tailor reactive balance measurement practices and recommendations. For example, refined criteria for administering the reactive balance measure based on patient characteristics could address concerns expressed by these participants and facilitate use in the most appropriate rehabilitation clients. The critical role of physiotherapist fear in mediating perceptions about the feasibility of reactive balance measurement is a noteworthy finding, novel both in rehabilitation and implementation sciences, having not been previously identified as a concern in studies of reactive balance [14], nor as a prominent factor in existing implementation frameworks [17]. This finding can also contribute to refined reactive balance measurement recommendations, suggesting assessment of reactive balance somewhat later in the course of treatment and/or continuum of recovery to facilitate building of trust and thus uptake of reactive balance measure.

Few studies have explored how physiotherapists conceptualize and consider balance as a general construct in physiotherapy [37] and virtually none have specifically examined in-depth practices and perceptions related to the unique nature of reactive balance. The varied perceptions of reactive balance described by physiotherapists in this study highlight challenges influencing implementation of reactive balance measurement in clinical settings even after a dedicated 12-month intervention. This is noteworthy because some of the concerns described by participants, such as fears about the implications of administering the reactive balance measure or the role of trust, were not identified as potential barriers in previous research that informed the design of the intervention [14]. Such differences may have arisen due in part to the fact that issues identified pre-intervention were somewhat theoretical in regard to what factors “would” influence reactive balance measurement, whereas the issues identified by participants in this study post-intervention reflected actual experience attempting to adopt the target behavior. For example, the variations in patient characteristics and associated functional status treated across the scope of rehabilitation dominated physiotherapists’ perceptions of their ability to measure reactive balance. Only one previous study has explored administration of the reactive balance measure used in this intervention in a cross-section of people living with balance deficits [18]. This early work during development of the Balance Evaluation Systems Test demonstrated the “proof-of-principle” of the measure across groups, but did not attempt to administer the measure across a broad spectrum of clinical populations. Subsequent publications of the measure have focused on very distinct clinical populations and functional levels, only some of which intersected with the many individuals undergoing rehabilitation and receiving physiotherapy services in the present study [16,22,23]. The triaging of reactive balance assessment in light of more pressing therapy priorities, patient goals and/or more fundamental balance and mobility issues described by participants in this study supports previous hypotheses about potential factors influencing low rates of reactive balance assessment among physiotherapists [9], and reinforces calls for critical evaluation of existing frameworks of postural control to reflect such critical elements of balance “in action” [15].

The expression of fear is a critical finding that has the potential to supersede adoption of reactive balance measurement in spite of the recognized importance of this functional skill. It is again noteworthy that such concerns and emotional responses have not been previously identified in the literature related to reactive balance measurement [14]. The related finding about the importance of trust could also be a lynchpin in future efforts to implement reactive balance assessment and treatment in practice settings. The role of trust in clinical relationships has been recognized throughout multiple domains of health research, exemplified in a 2013 systematic mapping review exploring the evidence base on trust in healthcare provider and patient relationships [38]. Most of the identified studies addressed relationships with doctors, and it is noteworthy that the majority of studies identified trust as a secondary emergent outcome, not a pre-planned explicit examination of the role of trust. The role of trust is only recently being identified as a key issue in physiotherapy literature. In 2017, Bernhardsson and colleagues explored patient preferences for physiotherapy treatment and participation in decision-making with 20 individuals who had sought physiotherapy services in Sweden [39]. They identified the importance of patient trust in the physiotherapist as an overwhelming factor for enhancing active engagement in physiotherapy. Given the growing recognition from both patients and physiotherapists of the critical role of trust, continued work is needed to understand how to navigate the complex relationship with conjunction with recommended principles of evidence-informed practice. Indeed, this intervention did not explicitly address fear or trust, the absence of which may have reduced overall intervention effects and outcomes. Future implementation studies will need to more fully explore the role of fear in implementation and expand consideration of organizational influences related to patient safety which might play a role. Interventions will need to directly address and navigate these concerns to maximize effectiveness.

Limitations

Qualitative descriptive approaches have been criticized for providing less insight than other methodologies [29], but were selected for the present study in alignment with our goals of gaining more open-ended and in-depth understanding beyond closed quantitative scales. Additionally, these methods were considered more feasible in light of participants’ busy clinical schedules and that they had already made a substantial time commitment to the study over a 12-month period. Future iterations of this work could explore experiences in more depth, using such approaches as phenomenology through modifications to the study design, such as including additional interviews and techniques such as the “go-along” approach.

Conclusions and future directions

The present study highlights the complexities of clinical reactive balance measurement in real-life rehabilitation settings. In addition to the findings that can be used to improve reactive balance measurement practice recommendations (refined patient criteria, timing of administration during treatment), the issues discussed by participants suggest some persisting knowledge gaps which warrant additional research. In particular, the concerns about the appropriateness of the measure in specific populations warrants additional psychometric analysis to confirm suitability of the measure and facilitate interpretation of test scores. In light of the recognized role of trust in the clinician-patient relationship, future studies should explore patients’ perceptions of reactive balance measurement and the factors that may positively or negatively impact those perceptions. Finally, there is also a need to examine factors influencing physiotherapist fear for measuring reactive balance to better understand why some participants responded positively to the intervention and others did not. Future interventions may more explicitly target strategies for building confidence among physiotherapists.

Ethics approval and consent to participate

Institutional and research ethics approval was obtained at all sites: University Health Network Research Ethics Board (1408434-DE); Sunnybrook Health Sciences Centre Research Ethics Board (480–2014); Joint West Park-Toronto Central Community Care Access Centre- Toronto Grace Research Ethics Board (15–005-BP). Written informed consent was obtained from all participants.

Author contributions

All authors analyzed and interpreted data, critically revised and approved the final manuscript. In addition, KMS conceptualized and designed the study, obtained funding, and wrote the manuscript. DCB conducted interviews, MK reviewed interview transcripts and entered data into NVivo.

Acknowledgements

We acknowledge the support of the participating sites in facilitating this project and contributions of the participating physiotherapists.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This study was funded by the Canadian Institutes of Health Research (CIHR) Knowledge-to-Action Operating Grant Program [#KAL-135775] to KMS. The funding body had no role in in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript. KMS holds a Canada Research Chair in Integrated Knowledge Translation in Rehabilitation Sciences. NMS hold a Heart and Stroke Foundation Mid-Career Investigator award.

Appendix 1.

Open-ended interview questions

1. I would like to start by asking you to describe your patient caseload over the last year. Has there been a change to your caseload over the past year?

2. Can you tell me about how you were recruited to participate in the REACT study?

   • What made you want to participate?

3. Can you tell me a little bit about your experience with the study?

   • Was it what you expected?

4. The REACT study was centered around reactive balance. What are your thoughts about evaluating reactive balance among your patients?

   • How important is reactive balance? Can you give me a specific example?

   • How necessary is reactive balance? Can you give me a specific example?

5. Did anything about the REACT study and the things we did over the last year stand out for you as very helpful?

   • Was there anything that was very unhelpful?

   • Is there anything you would suggest changing about the REACT study activities if we were do this again?

6. Now that your active participation in the REACT study is complete, is there anything you do differently as a clinician?

   • Can you give me a specific example? Or perhaps a story to illustrate your thoughts?

   • How has your clinical decision making process changed in regards to balance assessment or treatment?

7. What advice would you give to other PTs about reactive balance?

   • Any advice about the BESTest?

8. Is there anything else you want to add or think we should know in relation to the REACT study?