Barriers to the implementation of point-of-care ultrasonography by physiotherapists in haemophilia treatment centres in Canada: a modified Delphi approach

Knowledge translation (KT) addresses the gap between research and knowledge synthesis and the implementation of this work into clinical practice ^[23].

The Knowledge-to-Action (KTA) Framework is a conceptual framework for KT based on planned action theories and the knowledge creation process ^[23]. The KTA process is “iterative, dynamic, and complex”, and takes into consideration the fluid boundaries between knowledge creation and the application (action cycle) of knowledge into practice ^[23]. The framework provides steps to guide the process of KT, and can be used to conceptualise the KT gap around the implementation of POC-US in haemophilia care (Figure 1). As research evidence, protocols and training programmes support the use of POC-US by physiotherapists, the ideal entry into the KTA framework to address the KT gap for implementation of physiotherapy performed POCUS is at the level of barriers to knowledge use in the action cycle.

Figure 1

The Consolidated Framework for Implementation Research (CFIR) can also help to inform the discussion of POC-US implementation and the barriers to use by physiotherapists in HTCs in Canada ^[24]. As a determinants of implementation framework, it combines multiple implementation theories, such as the Diffusion of Innovation Theory, into a single framework, providing a foundation for development and verification around what works, where and why, across multiple contexts ^[24]. The CFIR includes five domains, each built on a menu of constructs that together provide a practical or theory-based approach to offer guidance on the factors associated with successful implementation of innovation. The five domains (visualised in Figure 2) are:

Figure 2

The CFIR provides a pragmatic structure for investigating the complex, multi-level interactions around the barriers to POC-US implementation in haemophilia care ^[25]. It fits well with the national structure of haemophilia care in Canada as the framework allows for comparison between HTCs across the country. This study examined three CFIR domains: intervention characteristics, the individual and the inner setting. Constructs of these domains were investigated to identify and achieve a consensus on the barriers to the implementation of POC-US, and to provide guidance on strategies to overcome these barriers ^[26]. Focusing on these three CFIR domains allowed for the in-depth analysis of the strengths of POC-US, the features of the physiotherapist, and the position of the HTC within healthcare organisations in Canada. Given the national structure of haemophilia care in Canada, it was anticipated that commonalities exist in the implementation process and outer setting.

The Delphi method is a research technique that provides a flexible structure for group communication, with the aim of achieving a consensus of opinion on a common problem ^{[27, 28, 29]}. It uses a series of questionnaires delivered in multiple iterations to collect data from a selected group of experts, and is effective when the goal is to improve understanding of a problem, investigate barriers and/or look for solutions ^{[27, 28, 29]}. The recommended number of participants is variable and there is no set calculation for sample size, although it is generally agreed that the sample size should constitute a representative pooling of the opinions and experiences of the target audience/professional group across geographical regions ^[27]. Consensus is determined if a percentage of responses falls within a set range, historically set at ≥70% ^[27,30].

The classical Delphi technique uses an open-ended questionnaire designed to generate ideas and opinions on a topic from a panel of experts, and involves three or more rounds of consensus-building ^[30]. In a modified Delphi approach, the first round of the process is modified through replacing the first round of openended questionnaires with a face-to-face meeting, or using a validated and structured questionnaire that is supported by the literature ^[27,30]. This study used a modified Delphi approach that incorporated the BARRIERS Scale.

The BARRIERS Scale is a self-report questionnaire designed to assess clinicians’, researchers’ or administrators’ perceptions of barriers to the implementation of research findings into practice ^[31]. It is standardised and was considered a valid and reliable instrument to assess the barriers and/or facilitators to POC-US use in haemophilia care. The questionnaire consists of 29 items divided into four subscales:

Respondents rate the extent to which each item is a perceived barrier to the use of research in practice using a four-point Likert scale (1= “to no extent”; 2 = “to a little extent,” 3 = “to a moderate extent,” and 4 = “to a great extent”), and also have the option to record a “no opinion” response ^[31,32]. There is also a free text option, enabling respondents to introduce other possible barriers, to rank the three greatest barriers, and to list possible facilitators of research implementation. The BARRIERS Scale demonstrates acceptable internal consistency, with Cronbach’s alpha for the subscales of 0.80, 0.80. 0.72 and 0.65 respectively, and acceptable test retest reliability (r=0.68 – 0.83) ^[31]. To enhance the Scale’s validity, it is recommended that the barriers/facilitators under investigation are considered in the context of the implementation ^[31]. For the purposes of the current study, the BARRIERS Scale was therefore modified to specifically address physiotherapy-performed POC-US in haemophilia care (see Appendix 1). As all physiotherapists practice within the comprehensive care model of the HTC in Canada, following the same standards and guidelines, it was reasonable to assume that the context of implementation was similar across respondents.

Physiotherapists involved in the treatment of patients with haemophilia within HTCs in Canada are formally organised into the Canadian Physiotherapists in Hemophilia Care (CPHC) group. Members of the CPHC are the musculoskeletal experts within the comprehensive care team, and have been the target of education, training and research initiatives in POC-US. There are 25 HTCs in Canada, each with a minimum of one physiotherapist who is a member of the CPHC. The CPHC maintains a network of physiotherapists in haemophilia care to provide a high level of physiotherapy management, education and research for people with haemophilia and related bleeding disorders ^[33]. In the current study, CPHC members formed the expert panel for the application of the Delphi approach. All CPHC members who attended the Canadian Hemophilia Society Annual Conference in 2017 were eligible to participate. This meeting was chosen for the purposes of the study to minimise attrition, to allow the research to be completed with minimal direct cost, and to reach a representative sample of physiotherapists in HTCs in Canada.

The Modified BARRIERS Scale (MBS) was administered to physiotherapists daily over the course of the three-day meeting. Study participants were provided with the questionnaire upon arrival to the conference each day, and were asked to complete the questionnaire independently and not to discuss their responses, in order to maintain blinding and anonymity. The researchers did not collect identifying information (e.g. name, address, email, HTC) on the questionnaire in an effort to maintain anonymity and to ensure that the project could not link individual questionnaires to the participants.

Following Round 1, a summary of the anonymous findings was collated, analysed and distributed back to participants. The summary consisted of a list of the top five items on the MBS rated as barriers, and a summary of the open text questions on the barriers, facilitators and future directions regarding physiotherapy-performed POC-US in haemophilia care.

On Round 2, the participants were provided with the Round 1 summary and the MBS; the summary from Round 2 was then collated. The summary comprised a list of the items on the MBS that achieved consensus as barriers to physiotherapy-performed POC-US.

On Round 3, the MBS was administered in an effort to confirm consensus on the barriers/facilitators to POC-US use. Items on the MBS that achieved consensus in Round 2 were not removed from the questionnaire for Round 3; however, these were not considered in data analysis for the final round.

To achieve consensus on each item of the MBS, two criteria had to be met:

This level of agreement is consistent with the literature ^[30].

Comments from participants in the free text questions on the barriers, facilitators and future directions around physiotherapy-performed POCUS in haemophilia care were collated and reported back to participants in the summary of findings for Round 1. The list of free text comments were completed only for Round 1 to provide participants with a broad context for physiotherapy-performed POC-US, as both physiotherapists who had completed the POCUS in Hemophilia Training Program and those who had not completed the training were included in the study.

Permission for the modification and use of the BARRIERS Scale was provided by Dr Sandra Funk of the University of North Carolina at Chapel Hill, United States. Research ethics approval was obtained from the Hamilton Integrated Research Ethics Board.

Twenty-two physiotherapists attended the 2017 Canadian Hemophilia Society Annual Conference and agreed to participate in the study. The participants represented 18 different HTCs in Canada; 13 of the physiotherapists had successfully completed the POC-US in Hemophilia Training Program for Acute Hemarthrosis and Synovitis. Twenty participants completed the MBS in Round 1 and 21 participants completed Rounds 2 and 3. Given the anonymity of the process, it is unclear which participants declined to participate in each round.

Table 1 presents the descriptive statistics and percentage of participants rating the items as a moderate or great barrier in Round 1. As none of the items in the MBS reached a median of 3.25 for consensus, the top five barriers were identified using the percentage of participants rating the item as a moderate (score 3) or great (score 4) barrier. The top five barriers in order of ranking were:

Participants’ free text reports of the barriers to physiotherapy-performed POC-US in haemophilia care included: lack of POC-US machines, lack of time and space in clinic, lack of support from administration, unknowns about college requirements, opportunities to take training, general comfort level with POC-US, insecurity regarding interpretation of POC-US findings, and clinic visits that are generally focused on annual reviews not acute bleeding episodes.

Participants’ free text reports of the facilitators to physiotherapy-performed POC-US included: support from the HTC team, engagement of the radiologist from the initiation of the implementation process, development of medical directives, funding for POCUS education, support from physiotherapy colleagues, patient advocacy and support for POC-US, support of administration, support from industry for machine/ training/competency evaluation, observing changes in clinical practice such as changing weight-bearing recommendations following a bleed, and increased patient adherence.

Future ideas for consideration to assist with the implementation of physiotherapy-performed POC-US included the development of a discussion board or forum within the CPHC website, and the formation of POC-US “rounds” or webinars to ensure appropriate skills.

Table 2 presents the descriptive statistics and percentage of participants rating the items as a moderate (score 3) or great (score 4) barrier in Round 2. Two items on the MBS achieved consensus:

Items that did not achieve a median of 3.25, but which approached consensus on analysis of the percentage of participants rating the items as a moderate or great barrier, were communicated back to the participants in Round 3. These included:

Table 3 presents the descriptive statistics and percentage of participants rating the items as a moderate (score 3) or great (score 4) barrier. Following this final round, the following items achieved consensus with both methods of analysis:

No additional items approached the 70% rating, thus expert consensus was achieved following the three rounds of the Delphi approach. The four items of the MBS that reached consensus were then analysed to reflect the three domains of the CFIR under consideration: intervention characteristics, inner setting, and characteristics of the individual involved.

This study was a first step in investigating the perceived barriers to the implementation and use of physiotherapy-performed POC-US from the practitioner perspective. In addition to providing a basis for future studies that involve patient groups and hospital administrators, the findings can be used to provide direction for the next step in the KT cycle to select, tailor and implement interventions ^[23]. In the context of the HTCs, possible interventions could include written fact sheets and a knowledge broker ^[38,39]. As printed education material, fact sheets are one of the most commonly used forms of KT, and enable the passive dissemination of research, skills and attitudes directly to the healthcare professional, thereby helping to translate research findings into clinical practice, with resulting enhanced patient outcomes ^[38]. They are a low-cost KT intervention and a convenient medium for information transfer, especially for a comprehensive topic targeted at a specific audience. A knowledge broker can be an individual or organisation that strives to provide a link between researchers and the end user. Given that the barriers identified in the current study represent constructs from the inner setting domain of the CFIR, using a knowledge broker is a preferred KT intervention, as the knowledge broker has the skills and capacity to facilitate the transfer of knowledge into local practice and policy ^[40]. Knowledge brokers are perhaps an ideal medium for addressing the barrier related to the isolation of physiotherapists in haemophilia care, as they have been shown to be effective in creating networks of people, and can facilitate a community of practice and provide opportunities for increased information sharing among the target audience ^[39].