Barriers and facilitators associated with the upscaling of the Transmural Trauma Care Model: a qualitative study

This study was conducted alongside a multicentre trial that aims to evaluate the effectiveness and cost-effectiveness of TTCM in nine Dutch hospitals [19]. The research team at AmsterdamUMC, location VUmc, coordinated and supervised the implementation of the TTCM at each site. The implementation process involved using procedures tailored to each hospital’s respective context [24, 25]. The methods for conducting the current process evaluation were based on those described in Wiertsema et al. [18] and the guideline for evaluating implementations in healthcare [26]. The study was reported according to the COnsolidated criteria for REporting Qualitative research (COREQ) checklist [27] (supplementary file 1).

Participants were purposively selected from the nine hospitals involved in the aforementioned multicentre trial. The relevant stakeholders that were represented included trauma surgeons, NPs, HBPs, and patients. Three researchers (JR/SW/JvD) were responsible to recruiting participants. The recruitment procedure involved contacting potential participants via email or telephone, explaining the study purpose and procedures, and inviting them to participate in the study. Care was taken to include healthcare providers and patients who were positive about the TTCM as well as those who were not. If potential participants were willing to participate and gave informed consent, an in-person interview was scheduled at a time and location convenient to the participants. An interview by video conferencing was also an option.

Data were collected using semi-structed interviews. These interviews were conducted by a two- or three-person team, consisting of a (3rd-year) student enrolled in a Bachelor of Health Sciences degree program at the VU University and one or two researchers (JR/RO/JvD/SW). The professional and academic backgrounds of the researchers were as follows: clinical epidemiology (JR/RO), human movement sciences (JvD/SW), physiotherapy (JR/SW/RO), or health technology assessment (JvD). Two researchers (JvD/SW) were experienced in conducting qualitative research [19, 28, 29] and all four student interviewers had successfully completed coursework on qualitative research and interviewing methods. Before the formal interviews were conducted, all interview team members were trained on procedures.

In sum, interviews were guided by a topic list and an audio recording was made [30]. Topic lists were based on the literature, a theoretical framework (see section on data analysis), and previous experience [18]. During the interview phase of the study, the topic list was adjusted based on knowledge and experience from previous interviews and adapted to the stakeholder in question [28] (supplementary file 2). The interview procedure involved one researcher leading the interview, while the other(s) probed areas for further questioning, kept track of the topic list, and made notes. Researcher objectivity was optimised by keeping a reflective diary [29]. To enhance the data’s trustworthiness, a member check was performed after each interview by sending the participants a brief summary of the interview and its transcript [31]. The interviews were conducted between April 2022 and August 2022.

Descriptive statistics were used to analyse participant characteristics (i.e., age, gender, stakeholder, and if applicable, years of professional experience, experience with TTCM [yes/no], type of injury, and time since discharge) and the degree to which certain parts of the TTCM were implemented/upscaled (i.e., reimbursement of the HBP, joint outpatient consultations, the exchange of patient information, accreditation of the network). For this, the following variables were described and compared between university medical centres and supra-regional hospitals: reimbursement of the HBPs (i.e., completely, partially, or not), care providers acted as a team (i.e., completely, partially, or no), information exchange between primary and secondary care (i.e., yes/no), and accreditation (i.e., was arranged for the network activities, yes/no).

Data from the interviews were analysed using a framework method, a hierarchical, matrix-based method for ordering and synthesising qualitative data [32]. Our theoretical framework was based on the “constellation approach”, which assumes that a healthcare system consists a set of interrelated practices and relevant, interrelated, structuring elements that define and fulfill a function in the more extensive system as in a constellation [22]. Within a constellation, there is a continuous interaction between the “structure, culture and practice triplet” (Fig. 1). A more detailed description of the constellation approach can be found in Supplementary file 3.

The applied framework method consists of seven steps. First, we transcribed the recorded interviews verbatim (IE/RM/AG/JK). Second, we familiarised ourselves with the content in the interviews by listening to the audio recordings and rereading the transcripts (IE/RM/AG/JK/JR). Third, we labelled text fragments relevant to the research question by relevant codes (open coding)(IE/RM/AG/JK and JR). Fourth, we developed a working analytical framework by grouping codes according to structure, culture, and practice categories of the constellation approach (IE/RM/AG/JK/JR). We developed final codes by applying an iterative process of refining through discussion until the criterion of saturation (i.e., no novel codes emerged from subsequent iterations) was met (JR/RO/JvD). Our approach to identifying themes and codes was both deductive and inductive: we used themes and codes defined by Wiertsema et al. [18] as a starting point (deductive), while new themes and codes were generated from the data (inductive). Fifth, working in pairs, we systematically reread each transcript, highlighted each meaningful text passage, and selected and attached an appropriate code from the final analytical framework (IE/RM/AG/JK/JR/JvD). Sixth, we charted the data by generating a framework matrix in which data were summarised by category and stakeholder group, categorised into the matrix, followed by adding illustrative quotes from participants to the matrix (IE/RM/AG/JK/JR). Lastly, we used the framework matrix to interpret the data together with the interview/coding notes. Two researchers (JR/JvD) assessed the “value” of the participants’ statements based on the intensity, frequency, persuasiveness, and contrast with which they were made. To ensure rigour and credibility, two other researchers (SW/RO) reviewed the generated matrix and checked whether the selected quotes were relevant to the themes. Disagreements were resolved by discussion. All steps were conducted using word-processing software. Quotes were translated from Dutch to English by an English native speaker and were edited slightly to make them more readable without losing their meaning.

A total of 33 stakeholders were invited to participate; however, one trauma surgeon declined the invitation due to limited availability. In the end, 32 interviews (31 via Zoom/TEAMS; one in-person) were conducted with eight trauma surgeons, eight NPs, eight HBPs, and eight patients. Five (63%) of the trauma surgeons, seven (88%) NPs, and four (50%) HBPs worked at a university medical centre. Six patients (75%) were treated at a university medical centre. The healthcare providers’ professional experience ranged from 2 to 40 years (mean=11.78; SD=10.09) and their experience with the TTCM ranged from 1 to 54 months (mean=15.19; SD=11.13)(Table 1).

Of the participating hospitals, one had successfully arranged reimbursement for the HBP at the outpatient trauma clinic, three had partially arranged it, and five had not made any arrangements for reimbursement. Additional findings on the extent to which the TTCM was implemented are summarised in Table 2.

Stakeholders shared the belief that the TTCM held the potential to improve both the quality and efficiency of trauma rehabilitation. Nonetheless, various barriers and facilitators associated with the upscaling of the TTCM were identified for each category of the constellation approach and are discussed below. Similarities and differences between the various stakeholders also were observed. An overview of all themes, sub-themes, and illustrative quotes are presented in Table 3.

With regards to the structure category, five themes were identified: “communication structure”, “financial structure”, “physical structures and resources”, “rules and regulations”, and “organisation of the network”. Each theme was associated with its unique barriers and/or facilitators.

"Communication structure" refers to the exchange of patient information between primary and secondary care. Typically, this takes place via an encrypted email system (i.e., ZorgMail) that allows healthcare providers to send and receive messages, documents, and images securely. The use of this system was perceived as both a facilitator and a barrier. On one hand, communication was sometimes hampered by incompatibility between a given hospital’s electronic patient record system and that of a network or a primary care practice, resulting in extra work (i.e., healthcare providers had to write separate emails instead of the information being automatically transferred). One NP noted:

Some HBPs considered alternative encrypted email systems (e.g., ZorgDomein), but these systems had similar incompatibility issues. If, however, the electronic patient record system and the encrypted email system were compatible, the communication structure was perceived as a facilitator.

"Financial structure" refers to the reimbursement of HBPs. The lack thereof was deemed an critical barrier to implementing TTCM by all healthcare providers. One trauma surgeon noted:

In hospitals that were successful in securing full reimbursement for the HBP, HBPs were able to be present during all joint outpatient trauma consultations. In most hospitals, however, only partial reimbursement could be achieved (e.g., as in for a limited proportion of the consultations and/or for a limited period); thus, joint outpatient trauma consultations were performed inconsistently or offered only temporarily.

"Physical structures and resources" refers to the availability of adequate rooms and number of computers to conduct joint consultation. Trauma surgeons and HBPs noted that the implementation of the TTCM was sometimes hampered by a lack of adequate rooms and/or an insufficient number computers. In some hospitals, the problem of insufficient resources was pronounced by the wish of trauma surgeons to work separately from HBPs, and hence requiring two rooms per consultation. In other hospitals, the number and size of rooms was simply insufficient. One trauma surgeon noted, for example, that the presence of a HBP meant that there was no longer space for a medical resident.

"Rules and regulations" refer to existing rules and regulations that impacted the implementation of the TTCM. NPs frequently mentioned that the number of TTCM patients they received was relatively low, because of the freedom to choose, some patients disregarded the referral to a network practice. Additionally, regulatory issues, such as “benchmarking” and “reimbursements” limited the number of physiotherapy sessions per patient. “Benchmarking” refers to the Dutch healthcare performance index that compares the average number of sessions per patient across physiotherapy practices. While the aim of this index is to monitor efficiency, some insurance companies use this index as leverage during contract negotiations with physiotherapy practices and/or audits. For physiotherapy practices, this can translate into less money per session, which in turn negatively impacts treatment decisions [33]. One NP indicated, for example, that even if he/she wanted to treat a certain patient three times a week, he/she would not do so, because of the benchmark. Moreover, the number of physiotherapy sessions that is reimbursed through the Dutch basic insurance package is limited. That is, physiotherapy sessions following a hospital admission are only reimbursed after the 20^th session and within the one year following discharge. Even though people have the option to purchase supplemental insurance that would cover physiotherapy sessions prior to the 20^th session, only 35% of the Dutch population has this coverage [34].

"The organisation of the network" refers to the set-up, content, website, and accreditation of the network. Physiotherapists were eligible to join a TTCM network after completing an online training on how to provide care according to the TTCM. Due to the COVID-19 pandemic, training sessions were organised online, which was perceived as both a facilitator and a barrier. Most NPs appreciated the convenience of not having to travel for training; however, they also noted a reduction in opportunities for personal interaction and networking. The number and duration of the training sessions differed between networks, and depended on the participating networks’ prior experience.

Five themes were identified: “commitment”, “acting as a team”, “quality and efficiency of care”, “patients’ experience”, and “job satisfaction”.

"Commitment" was the most common theme identified from the interviews and the most contributions came from healthcare providers. This theme refers to their, as well as their colleagues’, intrinsic motivation to work according to the TTCM. A high level of “commitment” was perceived as a facilitator, while a lack thereof was perceived as a barrier. Most healthcare providers were committed and felt some responsibility for the successful implementation of the TTCM. Some HBPs, however, noted that their direct colleagues and/or colleagues from other departments (e.g., trauma surgery) were less committed. In their opinion, this was detrimental to the successful implementation of the TTCM.

"Acting as a team" refers to the “contact between trauma surgeons and hospital-based physiotherapists at the outpatient trauma clinic” and the “contact between the network and the hospital team.” In the hospitals with an inconsistent presence of a HBP during outpatient trauma consultations, both types of contact were affected negatively. In some cases, contact between trauma surgeons and HBPs was limited due to trauma surgeons, contrary to what was intended, expressing the desire to work separately from the HBPs. As one HBP noted:

Some patients also noted that their outpatient consultations were not provided jointly by a trauma surgeon and HBP, which they perceived as a barrier.

Patients and NPs also reported problems with the communication between the hospital team and NPs. As one patient noted:

If consultations were provided jointly and an effective communication channel was in place, stakeholders perceived the improved levels of communication between primary care and secondary care as a critical facilitator. When working together, trauma surgeons and HBPs indicated that they were respectful of professional boundaries and that their respective responsibilities were clear. That is, they believed that they complemented each other in terms of knowledge and expertise. Also, most trauma surgeons indicated that their communication with the NPs (via the HBPs) had improved since implementing the TTCM. NPs, on their part, indicated that their contact with the hospital had improved and they believed that they played a more significant role in the rehabilitation process of patients with traumatic injuries.

"Quality and efficiency of care" refers the belief among healthcare providers and patients that the TTCM could enhance the quality and efficiency of trauma care. A trauma surgeon noted:

Some healthcare providers indicated, however, that “the applicability of the TTCM” was not always clear. Specifically, they found it challenging to anticipate when and if the presence of a HBP would contribute value to a particular patient's treatment. Trauma surgeons believed that HBPs provided significant added value for patients with complex injuries. In addition, trauma surgeons frequently mentioned that they experienced a “lower administrative workload” since the implementation of the TTCM, because they were no longer responsible for the communicating with NPs.

"Patient experience" refers to the patients’ experience and satisfaction with the TTCM. In some hospitals, patients reported “feeling rushed” or “not feeling heard”. In most of these hospitals, however, HBPs were inconsistently and/or only temporarily present during the outpatient trauma consultations. Some patients also indicated they were unaware of what had been communicated between the hospital and their NP, and/or noticed that “care providers contradicted each other”. As one patient noted:

"Job satisfaction" refers to the anticipated or experienced effect that working according to the TTCM had on the healthcare providers’ job satisfaction after its implementation. One trauma surgeon was particularly enthusiastic about his/her increased collaboration with HBPs:

A HBP noted that he/she would prefer to spend his/her entire work week treating patients according to the TTCM.

Two themes were identified: “practical issues at the outpatient clinic” and “knowledge gained”.

"Practical issues at the outpatient clinic" refers to the fact that some HBPs and trauma surgeons experienced some practical problems/issues while working with the TTCM. An important practical issue was the lack of appropriate consultation rooms. In some cases, there was a shortage of consultation rooms at their outpatient clinic, which became pronounced when trauma surgeons wanted to work separately from HBPs. In others, there was insufficient space in the available consultation rooms to allow both a HBP and medical resident to be present with the patient, and/or to place enough computers for each healthcare provider to enter notes simultaneously.

"Knowledge gained" refers to the fact that most healthcare providers indicated that they gained expertise in treating patients with traumatic injuries since working according to the TTCM. As one trauma surgeon noted:

This study identified various barriers and facilitators associated with the upscaling of the TTCM.

Under the structure category of the “constellation approach”, the main barriers to upscaling the TTCM were “communication structure” (i.e., incompatibility of electronic patient records), “financial structure” (i.e., absence of reimbursement for the HBP), “physical structures and resources” (i.e., unavailability of rooms/computers), “rules and regulations”, and “the organisation of the network” (e.g., online training). Under culture, the presence of “commitment” and “acting as a team during the consultations” were perceived as facilitators and the lack thereof as barriers. In some hospitals, contact between trauma surgeons and HBPs and between the hospital team and NPs was suboptimal and considered a barrier. In hospitals where contact between healthcare providers was improved, the improvement appeared to coincide with two perceived facilitators: increased level of “job satisfaction” and a “lower administrative workload for the trauma surgeons”. Under the practice category, “practical issues at the outpatient clinic” (e.g., inadequate or insufficient consultation rooms) was perceived as a barrier. With regards to “knowledge gained”, most healthcare providers indicated that they appreciated the fact that their expertise in treating patients with traumatic injuries increased since working according to the TTCM. Most stakeholders, including patients, believed that if the barriers were overcome, the TTCM could significantly improve trauma rehabilitation.

In line with the pilot study, we found that most stakeholders, including patients, believed that the TTCM could significantly improve trauma rehabilitation if implemented successfully. Many of the identified barriers and facilitators were in line with those of the pilot study [18]. In both studies, the inability to refer Dutch patients to a designated healthcare provider was identified as a barrier. This interferes with patients with traumatic injuries from receiving treatment from physiotherapists specialised in trauma rehabilitation (i.e. NPs), and impedes effective collaboration between primary and secondary care. Another barrier that was identified in both studies was the challenge stakeholders faced with arranging reimbursement for HBPs. The main reason for this difficulty arises from the entrenched financial boundaries between primary and secondary care in the Netherlands, which have also impeded the reimbursement of various other transmural care models [35, 36]. Bloemen-Vrencken et al. [37], for example, found that organisational and financial constraints interfered with the implementation of a transmural care model for spinal cord injury patients. In the pilot study, efforts to secure funding for the entire TTCM were not successful either, however, full reimbursement for the HBP was arranged by adjusting the pricing of medical specialist care (i.e., the trauma surgeon). We planned on using the same funding strategy in the current multicentre trial, but this was not feasible due to the suspension of negotiations amid the COVID-19 pandemic. This unforeseen circumstance further complicated the intricate challenge of navigating financial and organisational obstacles in the implementation of transmural care. Consequently, outpatient trauma consultations were performed jointly in less than 50% of the participating hospitals, and generally only for a limited proportion of the scheduled consultations and/or a limited period. Another notable discrepancy between the current multicentre trial and the pilot study was the reluctance of certain trauma surgeons to collaboratively conduct outpatient consultations in the present study; this was not the case in the pilot study. This discrepancy is likely explained by the “not-invented-here syndrome”, that is, the tendency of people and organisations to avoid things they did not create themselves [38, 39]. Such an attitude can act as a barrier to upscaling (healthcare) interventions [40]. Indeed, findings from other studies indicate that “key actors”, “ownership”, and “leadership engagement” (i.e., commitment, involvement, and accountability of leaders with the implementation) are conditional requirements for change management, and upscaling activities in particular [35, 36, 40]. Therefore, it is crucial for trauma surgeons, who frequently hold leadership positions in hospitals [41, 42], to serve as "key actors" during the implementation and/or scaling of the TTCM. In an ideal situation, this would be established along with strong support from highly committed HBPs. This might be achieved by providing comprehensive training programs to trauma surgeons, HBPs, and NPs; fostering a culture of collaboration and shared responsibility; and establishing clear communication channels between stakeholders. Furthermore, it is crucial for the overall leadership of a hospital to champion the implementation of a new healthcare intervention, as a supportive organisational environment is a critical success factor for effective implementation and/or upscaling [43, 44]. Another barrier that impacted the upscaling of the TTCM is the fact that many of the electronic patient record systems used in Dutch hospitals are incompatible with the available encrypted email systems. This incompatibility severely complicates communication between primary and secondary care providers, which is an integral part of TTCM and many other transmural care initiatives. Indeed, the challenges of compatibility between electronic patient record systems and encrypted email systems have been identified in a systematic review [45] and emphasises the necessity for standardised communication platforms between primary and secondary care [45‐47].

This process evaluation had several strengths. First, we used a theoretical framework to construct an analytical framework that enabled a systematic exploration of the data. Second, all stakeholders groups who provided treatment according to the TTCM were represented in the study. We made deliberate efforts to include participants from diverse hospitals and networks (including those that were who were positive as well as negative about the TTCM) to enhance the transferability of the results. Third, the credibility of data was improved by performing a member-check [31] and keeping a reflective diary [29]. Finally, to optimise reliability and reproducibility, the role of the researcher, the location, the order of the questions, and the description of the coding were described as precisely as possible [29].

The study also had some limitations. Participants were purposively selected, potentially introducing a bias in the sample towards individuals were more positive about the TTCM than the average healthcare provider or patient. Given that the sample is skewed towards individuals who express higher satisfaction with the TTCM compared to the average healthcare provider or patient, the bias may lead to an overestimation of the observed facilitators. Furthermore, we did not include representatives of other healthcare professionals, such as nurses or orthopedic casting specialists, who might have also been affected by the implementation and/or upscaling of the TTCM. Future research endeavors may benefit from interviewing individuals at different departments to capture a more comprehensive perspective. Furthermore, data were obtained through interviews with researchers involved in the development and/or evaluation of the TTCM, which may have caused “social desirability bias”. Consequently, participants may have overstated their positive experiences with the implementation of or working according to the TTCM. For future research, we therefore recommend researchers to obtain additional data through other methods, such as surveys or focus groups (preferably conducted by researchers who are not involved with the TTCM). Third, it is essential to acknowledge the fact the current study only assessed the barriers and facilitators associated with the upscaling of the TTCM during a period of nine months. However, upscaling procedures in the context of healthcare transitions may unfold over more extended periods [23. As such, we may have missed some barriers and facilitators and/or the identified barriers and facilitators may have been experienced more intensely by the stakeholders due to the fact that implementation process had just started.