“A way to think of the client holistically”: Factors inﬂuencing students’ ICF regard and uptake

Adoption of the International Classiﬁcation of Functioning, Disability and Health (ICF) may facilitate holistic delivery of health and social care and improve interprofessional practice, however there is limited uptake across the spectrum of health professions. Improved commitment will partially depend on student education, yet related educational research is scant. increased ability to apply it to interprofessional practice and, importantly, a greater sense of agency to eﬀect change.


Introduction
The International Classification of Functioning, Disability and Health (ICF) is a universal framework and classification system with multiple applications, including as a clinical tool, for managing statistics, research, social policy and in education (World Health Organisation, 2013), with the potential to enhance interprofessional practice through the use of a common language (Allan et al., 2006;Stephenson and Richardson, 2008). While there is widespread acceptance of the ICF (Jelsma, 2009), limited and inconsistent uptake across the spectrum of health professions is reported (e.g. Darrah, 2008;Peters-Brinkerhoff, 2016;Ross, Bickford and Scholten, 2018;Stewart et al., 2013).
The ICF supports person-centred practice (PCP) and promotes a biopsychosocial model of care (World Health Organisation, 2013). However, a positive cultural change towards such holistic consideration of functioning and disability will depend on "education of the next generation (or re-education of the current generation)" (Bornbaum et al., 2015, p.179). Health professions' accreditation bodies typically stipulate inclusion of the ICF in university curricula to meet professional standards. However, ICF implementation or evaluation in higher education has received limited examination (Bornbaum et al., 2015) and the extent to which this is enforced is not known.
A scoping review (2001 -2011) appraising the understanding and teaching of functioning and disability in the health professions (Bornbaum et al., 2015) included 18 largely conceptual papers. Notable amongst these is the study by Reed and colleagues (2008) who included evaluation of short programs involving students. The first compared instructor-led training (a single 2 hour lecture) and a self-directed learning module. Participants in both learning formats demonstrated significant, albeit limited, knowledge gains. However, only instructor-led training resulted in significant coding skill improvement, although insufficient for reliable clinical application. More positive attitudes about the ICF were expressed by students attending the lecture. The second program delivered ICF related content through five online modules including a discussion forum, with instructor and peer contact. Students in this program reported positive learning experiences and ICF-related views. The authors argue that interactive distance learning methods may help to overcome the weaknesses of self-directed training in comparison to face-to-face training. Recently Stallinga and colleagues (2018) conducted a study similar to Reed's instructor-led program, finding that participants' knowledge gains following 4 hours of training were retained three months later, although their initial appreciation of the ICF's usefulness was not maintained. Other papers included in the scoping review describe ICF teaching within discipline specific subjects with recursive presentation of a biopsychosocial approach to care across the semester (Dougall et al., 2014;Ghul and Marsh, 2013;Nguyen et al., 2016).
Embedding learning more broadly across a program's curriculum might be most conducive to supporting a paradigm shift from a biomedical to the biopsychosocial model of healthcare (Stallinga et al., 2018). Five of the 18 reviewed papers described such integration within health professions curricula (Cockburn, Trentham and Kirsh, 2005;Darrah et al., 2006;Gutenbrunner et al., 2010;Jones, 2011;Skarakis-Doyle and Doyle, 2008), the latter being interdisciplinary. Three further papers not included in the scoping review discuss health related programs in which the ICF infuses the curriculum: two postgraduate (de Brouwer et al., 2017;Ottenbacher, 2000), and one undergraduate program (Sandborgh et al., 2018). Only the latter includes any evaluation, with authors concluding that an integrated curriculum resulted in their graduates having improved competency. While description of ICF coverage in educational curricula is limited, there are even fewer investigations of university students' involvement with the ICF. Snyman, Von Pressentin and Clarke (2015) explored the experiences of undergraduate medical students and their practice educators (PEs) using the ICF during interprofessional rural placements, and how patients perceived their care. Findings suggested patients and their caregivers preferred the care provided by medical students educated in the ICF to that provided by other doctors. Patients reported that the students thought more broadly and considered them as a person, with the researchers concluding that ICF application assisted students to use PCP. Students requested both more direct teaching and clinical exposure to the ICF. While most students viewed application of the ICF optimistically, others indicated that the ICF was "too time-consuming, unnecessarily detailed and not always practical given the clinical workload" (ibid, p. 318).
Peters-Brinkerhoff (2016) investigated graduate physical therapy students' and PEs' experiences and perceptions of the ICF during their clinical internship program. Most PEs reported a poor understanding of the ICF model or how it relates to PCP; both PEs and students reported no to minimal learning experience related to the ICF model. No assessment of the ICF model was mentioned in student evaluations. Peters-Brinkerhoff concluded that learning experiences of all domains of the ICF model are generally not being presented to physical therapy students on placement.
In conclusion, health professional accreditation standards typically specify embedding the ICF framework in practice, in keeping with its intended universal application. However, health professionals across jurisdictions demonstrate inconsistent and restricted uptake, and ICF education appears to be variable. A cultural change from the traditional biomedical model to a more holistic way of practising is likely to depend in part on student education, yet research in this arena is sparse, with insubstantial evidence about effectiveness.
In order to inform health professions' educational practices related to the ICF this exploratory study aimed to describe how the ICF is regarded and used by one group of students, speech-language pathologists (SLPs), including what learning approaches, experiences and resources are valued and factors contributing to uptake.

Methods
Return of the online questionnaire, supported by LimeSurvey (Schmitz, 2015) implied consent to participate. All currently enrolled senior Australian student SLPs in entry-level programs (3 rd and 4 th year undergraduates and all masters' students) were eligible to contribute, with recruitment occurring primarily via invitation from a professional association newsletter and program websites.

Research
Instrument. An existing survey (Stewart et al., 2013) was adapted, with permission, to reduce redundancy, improve data retrieval processes, and add literature-supported detail regarding potential uses and benefits of the ICF. The revised questionnaire contained three main sections: I. Use/Application of the ICF, targeted familiarity, level of comfort, knowledge development and nature of use; II. Utility of the ICF, reflected perceived value, barriers and enablers related to use; and III. Demographics, provided students' study-related details. Table 1 outlines questions for the first two sections and the full survey is available on request. Survey completion required 15-20 minutes. Statistical Analysis. Statistical analysis of multiple choice, categorical and scale questions involved descriptive statistics, Chi-square goodness of fit to check for sample bias, normality testing of interval data to inform inferential testing, and exploratory factor analysis. Partial data were used where possible; data were analysed using IBM SPSS (IBM Corporation, 2015). Significance was established at a p-value of less than .05. Effect sizes were calculated for all inferential tests.
Thematic Analysis. Thematic analysis (Braun and Clarke, 2006) of open responses was conducted via the lens of familiarity, with measures taken to enhance rigour. The researchers' familiarity with the ICF and related literature guided initial coding, although we remained receptive to new ideas that emerged from the data. Two researchers were experienced with qualitative methodology. Analysis was aided by NVivo software (QSR International, 2015). Memos were written during the analytic process to further develop interpretations and understanding. All researchers read responses multiple times to enhance familiarity with the content. One researcher (IS) coded the data using a recursive style. Initially, line-by-line coding proceeded as the researcher asked "What is happening here?" A process of constant comparison of codes continued throughout analysis such that open codes were linked to develop emerging categories which were shared and discussed with the team. In the final stage of selective coding, these categories were grouped to form themes. A second researcher (JB) coded 20% of the data independently and following further team discussion all responses were re-coded using the derived coding scheme.

Results/Analysis
Demographics. Responses were received from 101 student SLPs, with 79 full responses (all female) and 22 partial responses (gender unspecified). Demographic data included gender, study program (bachelor or master's), and clinical placement factors. See Table 2. 100% a Missing data were excluded from calculations. b A few respondents specified their area of experience (e.g. speech/language or multimodal communication) but did not indicate whether this was adult, paediatric or both. This was classified as missing data Level and comfort of ICF knowledge. Typically students were confident about their ICF knowledge base and somewhat comfortable in its use, indicating "good" or better knowledge (max.=5); (n=101; x=3.30; SD=0.855) and comfort levels (n=101; x=3.47; SD=0.870).
Use of and regard for the ICF. Following exploratory factor analysis, two factors were extracted for each of the scale question sections using the Kaiser Meyer Olkin (KMO) statistic, accounting for 56%, 62% and 49% of the variance respectively: components (Survey section I. 4); adoption (Survey section I. 5); and utility (Survey section II. 2).
Use of ICF components. The two factors obtained for ICF components used were General Framework and Coding. Overall reported use was low; the most frequently used ICF components fall within the General Framework factor (the framework itself [n=98; x=1.82; max.=4; SD=1.287] and terminology [n=100; x=1.58; max.=4; SD=1.257]); fewer students used coding or core set aspects of the ICF (n=98; x=0.57 [max.=4]; SD=1.065).
Adoption. The extent to which students use the ICF is consistent with the two derived factors of Clinical Focus and Service/Management Focus, with more positive responses (max. 4) provided in the context of tasks with a clinical focus, such as: goal-setting (n=94; x=2.96; SD=1.100); describing activities and participation (n= 96; x=2.65; SD=1.098); personal factors (n=95; x=2.65; SD=1.120); environmental factors (n=96; x=2.5; SD=1.081); or as a guide to selection of assessment tools (n=95; x=2.31; SD=1.202). Conversely, the ICF was applied less frequently in management tasks: supporting service integration and management (n=94; x=2.12; SD=1.283); assessing service quality (n=94; x=2.04; SD=1.341); assisting planning at a service level (n=93; x=1.85; SD=1.132); or establishing eligibility for services (n=93; x=1.54; SD=1.274). Over 40% (n=39; 42.4%) of students indicated that it was either not applicable or they never used the ICF for evaluation of therapy outcomes, which also loaded onto the management factor. Perceived Utility of the ICF. Students indicated their regard for the ICF across multiple tasks. Factor Analysis supported two underlying dimensions, Person-centred thinking and Enhances communication. Students deemed the ICF promotes person-centred thinking (max. 5): encourages holistic thinking (n=81; x=4.62; SD=0.621); fosters clinical reasoning and decision-making (n=81; x=4.19; SD=0.754); has great benefits (n=80; x=4.3; SD=0.973); and enhances ethical service provision (n=81; x=3.92; SD=0.755). However, students generally had more neutral views regarding many of the variables associated with the Enhances communication factor, namely whether the ICF helps clarify team roles (n=80; x=3.13; SD=0.848); or fosters teamwork (n=80; x=3.18; SD=0.786).

Trends in uptake.
There were no significant differences for any factors in relation to study program or previous study, nor regarding students' knowledge confidence or level of comfort with respect to their placement settings or regarding areas of practice. Likewise, students' knowledge confidence and ICF component use were unaffected by their number of placements. However, students' placement experiences did influence a range of factors, with small to medium effect sizes: Students' level of comfort with the ICF (p = .046, d = .232), application of the ICF to person-centred thinking (p = .005, d = .323), clinical focus (p = .005; d = .327) and service and management focus (p = .009; d = .306) all increased with two or more placements.
There were also significant differences for clinical focus in relation to area of practice (paediatric, adult or both; p = .006); post hoc testing revealed that students with only a paediatric placement indicated lower levels of ICF application for tasks with a clinical focus compared with students who had both adult and paediatric experience (p = .02).
Thematic Analysis. We considered 85 students' open responses to three survey questions (most influential ICF factors; advantages and disadvantages; barriers and facilitators). Comments ranged from 1 -143 words (mean = 27.1). Thematic analysis revealed two principal and one secondary theme: ICF (framework) as a "way of thinking"; experiential learning optimises ICF uptake; and rudimentary understanding restricts ICF uptake (See Fig. 1). See Appendix for sample quotes according to theme. ICF (framework) as a "way of thinking". It is clear that with few exceptions students equate "the ICF" with the conceptual framework, particularly the activity and participation components, rather than its classification or coding features. The ICF model was used as a loose guide rather than overtly applied. Students recognised the ICF as guiding holistic practice. The biopsychosocial approach to care was perceived to lead to improved client outcomes, by guiding goal-setting and intervention. However, students also indicated that the ICF is infrequently used in specific settings or situations, in particular in acute hospitals and in paediatric practice. There may also be a lack of focus on the ICF in relation to the students' other learning needs.
Experiential learning optimises ICF uptake. When commenting on factors that best enabled them to implement the ICF, students expressed their appreciation for relevant, clear and specific learning resources. They valued active learning experiences, both in the university setting and on placement. Both authentic and hypothetical case studies were highlighted as being particularly useful. Students mentioned the benefits of applying the ICF to university assessment tasks, although not all shared this majority view. Opportunities for discussion, with peers, lecturers, and PEs, and implementing the ICF on placement were valued.
While students appreciated the benefits to learning of appropriate university experiences and placements, the absence of these was also noted. The reported top barriers to ICF implementation related to learning experiences, including a recognition of the student's inadequate understanding of the ICF, theoretical lectures without clinical application, lack of workshops and tutorials, and limited opportunities while on placement.
Few students commented on their own role in the learning process, mentioning the importance of actively seeking and working through relevant information or using specific strategies such as keeping a diagram of the ICF framework on hand. Rudimentary understanding restricts ICF uptake. Some students are unaware of misunderstandings and restrictive attitudes that potentially limit further development of their ICF knowledge-base. For example, some students believe that the ICF must be applied in its entirety, that its use must always be explained or that only certain language can be used to describe it to clients.
The most commented on barrier to learning was the ICF's complexity. Students described their frustration with the terminology, scope of the ICF and time required to "conquer" it. The distinction between activity and participation was especially problematic.
Model of ICF uptake. Our combined statistical and thematic findings contribute to a preliminary model depicting the process of integrating the ICF into practice (Figure 2). The cycle begins with learning experiences related to the ICF which generate knowledge and affect ICF regard and are in turn influenced by knowledge level, barriers and facilitators. Level of uptake, depicted on the far right of the model as using the ICF as a "way of thinking", is optimised by experiential learning. It is influenced by both regard for the ICF and facilitators and barriers, including the restrictive influences of rudimentary understanding. Practice and continued education sustain ICF use, closing the learning loop. Results that informed these interpretations are highlighted within the model.

Discussion
This study investigated how the ICF is regarded and used by student SLPs, including factors influencing uptake, favoured learning resources, experiences and approaches. Key findings are organised and discussed here in relation to our proposed Transition from Theory to Practice model which implies where changes might be made to improve ICF uptake and adoption of a biopsychological approach to care. health professionals reported earlier; typically the conceptual framework alone is applied to "document and guide clinical decision making" (Darrah, 2008, p. 150). Similarly, students use the ICF for clinical rather than management tasks.

ICF (framework) as a "way of thinking". Responses about students' ICF use and appreciation resembled those of
Findings suggest that students' education has focused on the ICF framework, emphasising activity and participation, with less attention paid to other aspects. A review of Nordic papers also found "activity" was the most mentioned ICF component (Maribo et al., 2016). Creating effective change will demand specific attention to other ICF components and its relevance at the societal level.
Students indicated that the ICF promotes holistic practice. Albeit not directly referencing the ICF, research addressing preferences for PCP and related performance is relevant. While students and health practitioners value person-centeredness (Bellon-Harn, Hartwell Azios and Dockens, 2017;Dockens, Bellon-Harn and Manchaiah, 2016), explicit person-centred interventions are uncommon (Torrence et al., 2016). Interventions typically address impairment and skill-based activity levels, focusing on tasks and target behaviours, rather than considering other ICF components, notably life participation (Kjellberg et al., 2012;Ramklass, 2015;Torrence et al., 2016). Likewise, supervisory practices may target concrete, discipline-specific knowledge and skills at the expense of psychodynamic or intrapersonal factors (DiLollo and Favreau, 2010). It is vital to identify publications that detail sound participation-focused interventions from which students might learn (Torrence et al., 2016).
In our study, students did not identify the ICF as supporting workplace communication, either with same-profession colleagues or interprofessionally. The likely reason is that they are not typically working interprofessionally on placement (Ramklass, 2015). Positive student experiences have been achieved by basing shared learning experiences around the ICF: by experienced PEs modelling interprofessional practice (Snyman, Von Pressentin and Clarke, 2015), by following a simple structured tutorial process (Cahill et al., 2013), and holding regular interprofessional team discussions, carrying out joint treatment sessions and home visits and socialising together (Kloppers et al., 2015). Learning occurs through interaction with others in a work context. It is important for students to "engage actively with the roles, beliefs, values and cultures of other professionals. Collaborative learning through tasks and discussions helps to achieve shared understanding and, in terms of clinical practice, shared goals" (Thistlethwaite, 2012, p.65).

Facilitators and barriers.
In this section, we consider influences on ICF uptake: issues pertaining to practice settings and exposure; limited opportunity to bridge theory and practice; the time-consuming nature of ICF application; rudimentary understanding of the ICF; and the absence in the data of an appreciation of a broader perspective.
Whether or not students had only a paediatric placement and the number of completed placements significantly influenced responses.

Issues in Paediatric Practice.
Students' views that ICF use is context dependent are consistent with those expressed by Australian SLPs. Practitioners working with, and in settings predominantly assisting children indicated significantly lower confidence in ICF knowledge and use for clinical tasks than those working with adults in combined acute and rehabilitation settings (Ross, Bickford and Scholten, 2018). If PEs aren't using and discussing the ICF, students will not appreciate its relevance to practice. This reduced ICF adoption in paediatric settings may be partly because the paediatric perspective was more recently assimilated into the ICF (World Health Organisation, 2012). In addition, developmental issues are underrepresented in the ICF coding structure (Darrah, 2008). Finally, a health-related framework may be considered irrelevant in disability and educational settings where different frameworks and priorities apply. Castro and Palikara's (2017) book about ICF implementation in educational and Scholten I, Ross K, Bickford J MedEdPublish https://doi.org/10.15694/mep.2019.000061.1 Page | 10 care systems may help to demonstrate the universal applicability of the ICF. Improved ICF uptake with two or more placements. The relationship between improved ICF uptake and two or more placements may pertain to SLP students' development of professional competency -a combination of knowledge, skills and personal qualities (McAllister et al., 2011). A hierarchy of difficulty exists, with earlier development of easier competencies involving readily explained behaviours, such as punctuality or maintaining confidentiality, or those practised more frequently. More challenging competencies require complex cognitive skills such as data synthesis and analysis and interpretation of what this means for the client (ibid.). Initially, students are typically self-focused, gradually attending more outwardly, including to their clients. Therefore, one explanation for the finding that students' knowledge confidence and appreciation for the ICF framework is stronger following two or more placements is that, in spite of protestations that it is complex and difficult to learn, it is actually internalised by students as involving relatively basic concepts, especially when considered alongside the multifaceted and complex areas of professional practice. Hence it develops early in the learning trajectory, but only once students have established a level of comfort with the clinical process.

Limited opportunity to bridge theory and practice.
Students experienced an unsettling disconnection between their academic learning and practice experiences, consistent with our previous report of limited ICF uptake by practitioners (Ross, Bickford and Scholten, 2018). To facilitate translation and broader implementation of the ICF, models such as the Behaviour change wheel (Michie, Van Stralen and West, 2011) or the 10-step model for inducing change in professional behaviour (Grol and Wensing, 2004) may be considered. Michie and colleagues (2011) identify that motivation, capability and opportunity are important factors underpinning behavioural change. With this in mind, university educators could offer PEs training and skill development opportunities in relation to the ICF. Such training will increase awareness of what students are learning and reinforce professional practice guidelines. It may also motivate PEs to identify opportunities for students to apply the ICF on placement. A participatory design approach (Appleby and Tempest, 2006) could also be used to further understand the barriers and facilitators to ICF use and support its implementation.
ICF is time consuming to implement. The ICF's perceived complexity presented as a barrier to students in two ways. The first is in its clinical application. Students' concerns about the time consuming nature of using the ICF may contribute to their lack of emphasis on coding and classification elements, and is consistent with findings elsewhere (Bornbaum et al., 2015;Jelsma, 2009). As practitioners become more familiar with the ICF and tools are developed to streamline documentation, the process should become less demanding, especially when applying the framework alone.
Rudimentary understanding restricts ICF uptake. Secondly, students found the ICF complex, confusing and difficult to learn. Like their professional colleagues, students expressed confusion between the concepts of activity and participation (Van de Velde and De Vriendt, 2018) which, paradoxically, were the most cited influential components. However, the ICF is unlikely to be more complex than other professional content which is grasped by all successful students. The perceived complexity and how the framework is taught appear to have resulted in students holding limiting beliefs. Some students described having "mental blocks" about using the framework as well as clearly holding some inaccurate views that obstruct further learning. A strong curricular theme of reflective practice and the development of metacognition should help address students' independent learning skills.
The ICF framework itself is deceptively simple; however, it has multiple levels and uses. When incorporating the ICF into curricula, educators must consider how best to present, integrate and revisit the model and underlying biopsychosocial approach to practice, across different subjects and years of a program's curriculum, without Scholten I, Ross K, Bickford J MedEdPublish https://doi.org/10.15694/mep.2019.000061.1 Page | 11 oversimplification. Ways to improve student learning are addressed within the context of the final theme below.

Absence of a broader perspective.
A notable absence in students' responses was mention of the potential for the ICF to be used at the societal level, including reference to barriers. While the academic focus goes beyond students' understanding of relevant underlying conditions and mastery of specific treatment techniques, more comprehensive perspectives may require additional attention. If we want our graduates to be involved in shaping health reform and environmental change at the societal level, consideration of the macro perspective must be clarified and clearly integrated into curricula.

Experiential learning optimises ICF uptake.
Students value much about their educational experiences in relation to acquiring ICF knowledge and skills and identified the absence of these features as obstacles to learning. In order to change professional behaviour, problematic issues must be addressed within student education. The health professions literature contains examples of concrete tools, conceptual models, intervention methods, and curricular approaches (subject-based and whole program) to support ICF learning experiences. Below we consider how we might implement a biopsychosocial approach to practice in health professions' curricula, underpinned by the terminology and framework of the ICF (and notwithstanding the myriad of other factors to be considered, such as relevant disciplinary content, interprofessional, evidence-based and reflective practice).

Learning through the use of clear tools and models.
Students value clear and specific resources. There are tools and conceptual models that foster ICF application skills to improve practice, including decision-making frameworks to enhance students' reflection and reasoning e.g. (Atkinson and Nixon-Cave, 2011;Dal Bello-Haas, Wojkowski and Kho, 2015;Darrah et al., 2006;Jarl and Ramstrand, 2018;Jelsma and Scott, 2011). Embedding the ICF framework into routine tasks, such as writing reports, determining appropriate intervention or conducting case presentations, should enhance acceptance of the ICF's relevance to practice (Yardley, Teunissen and Dornan, 2012).

Curricular Change.
A number of papers describe the adoption of the ICF as a framework for health professions curricula to promote a paradigm shift towards a biopsychosocial approach to care (Bornbaum et al., 2015), including interdisciplinary programs (Ottenbacher, 2000;Skarakis-Doyle and Doyle, 2008;Stephenson and Richardson, 2008). Sandborgh and colleagues (2018) present their experiences in developing and testing a renewed entry-level undergraduate physical therapy curriculum that integrates professional best practice with a biopsychosocial approach and the ICF. The authors describe the basis for the curriculum, its development and execution, challenges faced, outcomes from over a decade of graduates, and related research. Their findings suggest that graduates' practice embodies ICF principles in the context of strong professional competencies.
Assuming a biopsychosocial perspective of the individual's health condition is a key factor in adopting PCP which is often overshadowed by social, economic, and political inequalities (Fleming-Castaldy, 2015). Whereas Sandborgh and colleagues (2018) omitted interventions without a strong evidence base to accommodate the addition of a biopsychosocial approach in the curriculum, (Duchan, 2001) provocatively suggested facilitating a more social approach to practice by refocusing disproportionate attention away from impairment. Duchan (2014), Fleming-Castaldy (2015 and Ghul and Marsh (2013) describe taking a macro perspective in their teaching, sharing examples designed to develop students' emancipatory skills. Students consider sociocultural and historical views of health and social care through the use of rich narratives, presented by clients and caregivers, case studies, historical documents, film, and students' own stories. The aim is to exploit students' personal reactions and experiences in order that they learn important professional perspectives, including appreciation of each person's unique experience, leading to greater understanding of person-centeredness and, in particular, the complexity of barriers and enablers of participation. Explicit adoption of key themes for curriculum reform, including a biopsychosocial approach, the ICF and an emancipatory orientation, in the context of harnessing features identified by students in the present study as enabling their learning, would help students develop their professional identity as powerful change agents.
Limitations. Study findings are exploratory given the low number of respondents, the fact that students' ICF knowledge was not explicitly assessed, their level of interprofessional experience was not identified and the survey tool, although used previously, was not tested for psychometric properties. Also, there were no opportunities for more in-depth interaction with participants to explore their perspectives. Likewise, students from a single health profession are represented, the placement structure for which likely differs from other student groups. However, the fact that issues related to ICF education and application have been described across health professions may support the generalisability of our findings.

Conclusion
Since the ICF's launch increasing attention has been paid to the value of taking a biopsychosocial approach to practice as well as to educating future health professionals in its application. By explicitly incorporating a biopsychosocial approach to practice within and across health professions' curricula, including active learning experiences that challenge students' prior conceptions and require their engagement with an emancipatory framework, and use of authentic assessment tasks, it is expected that students will have a deeper understanding of the ICF, increased ability to apply it interprofessionally and, importantly, have a greater sense of their own agency to effect change.

Take Home Messages
Students appreciate the ICF framework for its ability to foster holistic practice, applying its components loosely in order to improve client outcomes, in some settings more than others. Students find the ICF challenging to learn and apply. Active engagement in learning experiences, with specific, relevant and clear resources, are appreciated in both university and placement settings and may be barriers to learning when absent. Students experienced a disconnection between academic and practice learning. Targeted training and skill development in relation to the ICF should support practice educators to ensure more cohesive learning and enhanced appreciation of the ICF's many benefits.