A monocentric, explorative, prospective concurrent mixed method study was performed to assess the feasibility of newly implemented digital treatment pathways for patients with orthopedic trauma in an urban, level-2 trauma center and teaching hospital. A concurrent mixed methods triangulation design was used, using quantitative and qualitative research methods simultaneously with equal weight [16]. One team (GJAW and JFS) collected and analyzed quantitative data, and another team (KAGJR and EGEM) separately collected and analyzed qualitative data. Using the Bowen feasibility framework, both data types were categorized within the following themes: acceptability, demand, implementation, integration, and limited efficacy [17].

Our study was conducted among adult patients who received follow-up treatment through a digital treatment pathway at our institution between October 1, 2022, and April 1, 2023. Patients were eligible for a digital treatment pathway if they (1) were aged 18 years or older, (2) had an active account for the web-based patient portal, (3) were able to read and write in Dutch, and (4) had no cognitive impairment or preexisting motor impairment. Exclusion criteria were: multiple injuries, pathological fracture, and initial presentation or follow-up treatment at another institution. Additionally, a control group was formed using previously collected anonymous data from 100 consecutive patients to compare secondary health care use–related pre- and postimplementation outcomes of the digital treatment pathways. These data were collected between October 1, 2021, and February 10, 2022, as part of a clinical audit.

This study was reported according to the GRAMMS (Good Reporting on a Mixed Methods Study) criteria (Multimedia Appendix 1) and COREQ (Consolidated Criteria for Reporting Qualitative Studies; Multimedia Appendix 2) [18,19].

This study was ethically reviewed and approved by the Medical Ethical Committee of Utrecht, the Netherlands (WO 22.049). All included participants provided informed consent before study participation and all participants had the ability to opt out at any given time during the study. Prospectively collected data were deidentified and all retrospective data were used anonymously. Furthermore, all study data were stored on a secure institutional drive, only accessible by 2 researchers (GJAW and KAGJR). Study participants did not receive any compensation for their participation in this study.

Patients with orthopedic trauma at our institution are treated according to a virtual fracture care review protocol [20,21]. To optimize this protocol, additional digital treatment pathways were introduced for the 8 most frequent types of hand, wrist, ankle, and foot fractures on October 1, 2022, including both nonoperative and operative treatment (Table 1). Multimedia Appendix 3 offers a detailed description of the virtual fracture care review workflow and the digital treatment pathway.

The mode of delivery of the digital pathways was the web-based patient portal and its accessory app: “MyChart,” an extension of the electronic patient record (EPR) system: Epic (Epic Systems Corporation). An impression of its layout is provided in Figure 1.

In selected pathways, patients could directly influence their follow-up treatment through an additional anchor questionnaire sent prior to the final routine function check appointment. Based on the response, the need for an appointment was evaluated and planned accordingly, as shown in Figure S1 in Multimedia Appendix 3. Only patients who indicated stagnation or deterioration of recovery or dissatisfaction with their treatment received a follow-up appointment. Function check appointments that included imaging could not be guided by an anchor questionnaire, as this could not account for imaging results in its current form. An example of the course of a digital treatment pathway is provided in Figure 2.

The sample size for quantitative data collection was set at 70, based on general recommendations for the design of feasibility studies [17,22]. For the quantitative part, patients were recruited via convenience sampling. A researcher informed them about our study the day after their ED visit via an information email and by phone, and sent an information letter and informed consent form by email to patients indicating willingness to participate (Multimedia Appendix 4). Study participants provided digital informed consent before participation. Upon providing consent, patients also indicated willingness to partake in an interview. For the qualitative part of this study, patients were included using a purposive maximum variation sampling method to ensure heterogeneity in terms of gender, age, type of injury, and treatment strategy. The qualitative sample size was guided by the principle of data saturation and determined a posteriori [23]. The expected number of interviews was 15 [23].

In total, 251 patients were eligible for a digital treatment pathway (Figure 3).

Of these patients, 125/251 (49.8%) were eligible for study inclusion, of whom 66/125 (52.8%) were included. Baseline characteristics are shown in Table 2.

Similar characteristics were compared between the control and digital treatment pathway groups, and no differences were identified. All included patients (66/66, 100%) completed the T0 study survey, 61/66 (92%) completed the T1 survey and 52/66 (79%) completed the T2 survey. The respondent groups at various time points did not differ in baseline characteristics (Table 2).

Thirty of 66 (45%) patients consented to interview participation of whom 15/66 (23%) were interviewed.

The reasons for not participating were not obtained. Nine out of 15 (60%) interviewees were female and 6/15 (40%) were male. Interviewee age ranged between 23 and 77 (median 48, IQR 35-62) years.

This was assessed though qualitative data provided by interviewees focusing on the ease of integration of digital treatment pathways into patients’ daily activities. Interviewees generally found the MyChart app easy to incorporate into their lives, noting that its mode of delivery was familiar to smartphone users. However, they pointed out that successful integration depended on users’ digital proficiency. Furthermore, participants mentioned that the app’s language settings and layout made it time-consuming to navigate to the desired section, which could be challenging during daily activities such as work, hindering a seamless integration (Multimedia Appendix 7, quote 8).

Limited efficacy was assessed by comparing secondary health care resource use before and post implementation of the digital treatment pathways. This was complemented by patients’ perspectives on the necessity of function check appointments and the new scheduling process through the anchor questionnaire.

Quantitative data showed a reduction of follow-up appointments following the implementation of the digital treatment pathways (P<.001; Table 4).

Primarily, this was the result of a reduction of appointments by phone. Control group patients had a median of 1 (IQR 0‐1) follow-up appointment by phone versus 0 (IQR 0‐0) in patients treated with a digital treatment pathway (P<.001). The number of face-to-face appointments remained similar with a median of 2 (IQR 1‐4) in the control group versus 2 (IQR 1‐4) in the digital treatment pathway group (P=.07). Fewer health care professionals, particularly physicians (P<.001), were involved in the follow-up of the intervention group compared to the control group. Patients in the digital treatment pathways had fewer follow-up radiographs (P=.01), while the number of computed tomography and magnetic resonance imaging scans remained unchanged. ED reattendances were rare in both groups, mostly due to cast-brace issues. No differences in complaints or complications were observed within 6 months. One patient in the intervention group developed a wound infection, treated with surgical debridement and antibiotics, leading to full recovery. This patient independently sought care, unaffected by the anchor questionnaire.

Qualitative data revealed varying outcomes regarding the need for follow-up appointments among participants. Some individuals found follow-up appointments unnecessary, while others deemed them essential (Multimedia Appendix 7, quote 9). Another participant shared an experience around persistent pain and mobility issues that were not picked up by the questionnaire (Multimedia Appendix 7, quote 10). This participant scheduled a follow-up appointment on their own initiative after receiving questionnaire feedback, emphasizing the importance of this option.

Our findings show that using a digital treatment pathway for patients with orthopedic trauma with extremity fractures is feasible, reflected in high usage, satisfaction, positive experiences, and reduced health care use. Participants saw it as a promising step toward digitizing trauma care, expressing willingness for future use. Nevertheless, clear areas for improvement were identified, including patient expectation management, user-friendliness of the patient portal and accompanying application, the implementation process, and protocol compliance by health care professionals.

The positive expectations and readiness among patients toward the digital treatment pathways, as reflected by the high questionnaire completion rates, aligns with other studies on the use of digital tools and home-monitoring within orthopedic patients [27-29]. The decline in completion rates during later stages of follow-up is a known challenge within this population, suggesting this it is not unique to the digital treatment pathways but rather associated with other characteristics of the orthopedic trauma population [30].

Although results were predominantly positive, approximately one-third of survey respondents expressed unmet expectations, a sentiment reinforced by interview findings. This was primarily due to a lack of user-friendliness, a common barrier in digital health care innovation [31]. In our study, this was attributed specifically to the web-based patient portal and the accessory MyChart app, both an extension of our EPR system. While integration with the EPR enables direct digital feedback, these systems were not designed for home monitoring, introducing usability issues [8,15,32-35]. Furthermore, these systems are generally used hospital-wide, requiring a wide array of different features and functions for different patient types. This could confuse patients, as confirmed by our participants. Specialized digital tools and pathways designed from an end user perspective could enhance usability and improve this mode of care [8,9]. Such digital tools have already demonstrated positive usability outcomes; however, these are generally costly and lack necessary EPR integration [36-38]. Applications and EPRs capable of independent integration would offer an ideal solution, leveraging the strengths of both approaches. However, this requires substantial financial resources and close collaboration among the commercial sector, hospitals, health insurance companies, and government.

A key challenge we encountered was the poor protocol compliance by health care professionals following implementation. During this study inclusion phase, 117/251 (46.6%) of eligible patients were excluded due to improper initiation of the digital pathway, and protocol deviations due to unawareness were high with 14/52 (27%). These rates were unexpectedly high compared to similar studies, highlighting flaws in our implementation process [36,39,40]. An essential area for improvement in this context is the early involvement of diverse health care professionals, particularly those whose daily tasks are altered [31,35,41,42]. This could facilitate integration into daily workflows and consequently improve protocol compliance. This principle also holds true for patients. Existing research in this field underscores the significance of a user-centered design strategy and early stakeholder engagement in the development and implementation of digital tools in daily practice, transforming patients from testers to designers [15,40,43]. Furthermore, using hands-on training and on-the-job guidance for involved health care professionals, complementary to an instructional training system, could yield better results [34,44,45].

Digital treatment pathways effectively reduced secondary health care use by replacing routine function check follow-up appointments with a more patient-centered approach. Notably, the reduction in follow-up appointments exceeded the number of anchor questionnaires sent, suggesting other aspects such as improved patient information and the ability to address concerns via the patient portal and MyChart app also played a role. Our results emphasize the importance of providing patients with an option to schedule an appointment, regardless of questionnaire outcomes as a safety net. Allowing shared decision-making is crucial, as standardized automated systems may not adequately address all concerns and needs [46]. The decline in radiographic imaging likely resulted from fewer routine follow-up appointments, which were previously often linked to routine scheduling of imaging. The added value of follow-up radiographs in extremity injuries has been strongly debated [47]. These findings align with the aims of the digital treatment pathways, in which these radiographs are only performed on clinical indication, rather than routinely. Despite reduced resource use, ED reattendances and complications remained similar, indicating treatment safety was not compromised. These findings are in accordance with current literature; however, future studies with larger sample sizes that focus on the functional outcomes for specific injuries are needed to make definitive statements regarding treatment safety [9].

While the digital pathways improved resource use, their full potential was limited by user-friendliness issues and poor protocol adherence. Addressing these could enhance their benefits. The strain on health care resources and medical personnel warrant reconsideration and innovation of the trauma care system. In this, digitization of the trauma care chain may prove an important tool. However, these new approaches must be carefully designed and (not hastily) implemented, to ensure alignment with the needs of all users.

This study has several strengths. First, a concurrent mixed methods design was used, enabling a comprehensive assessment of the digital treatment pathways. Quantitative data measured specific effects, while qualitative data enabled direct interpretation. The concurrent data collection prevented one type of data informing the other, thus ensuring independence of data sources [48,49]. Second, separate teams conducted quantitative and qualitative data collection and analysis independently. Subsequently, they collaboratively engaged in a triangulation process, enhancing the reliability and validity of the results [26]. Additionally, the use of quotes enhanced transparency in presenting qualitative findings [50]. Lastly, using a validated framework provided a structured examination of various feasibility parameters [51].

Several limitations must be addressed. The primary issue is this study’s sample, which consisted of patients with relatively high digital literacy and potentially positive attitudes toward digital tools, due to the convenience sampling method used. This introduces selection bias. Additionally, patients could choose to make an appointment regardless of questionnaire results, meaning those who opted for the digital appointment may have been early adopters. Therefore, our results might not reflect the general patient population accurately, limiting their generalizability, especially to those with lower health literacy or digital skills [44,52]. Future research should validate these findings with a more representative trauma population.

Furthermore, the high protocol noncompliance among health care professionals resulted in a substantial loss of potential study participants. Nonetheless, the final sample size was still deemed sufficient for the purposes of this feasibility study [17,22]. Again, regarding sample representativity, the number of operatively treated patients in our quantitative sample, 30/65 (45%), was relatively high compared to a general monotrauma population. It is unknown whether this influenced our results, this would require a validation study on a larger scale. Importantly, the other baseline characteristics were deemed representative for the monotrauma population at our institution.

Finally, we could not account for the perspective of health care professionals in this study. Future qualitative evaluation in this stakeholder group would be valuable to gain a complete understanding digital treatment pathway feasibility and potentially to identify other points of improvement.

This study shows that patients with orthopedic trauma consider our digital treatment pathways in follow-up care as feasible and satisfactory, with positive attitudes toward current and future use. This patient-centered approach has the potential to empower patients and reduce secondary health care use. Opportunities to further improve feasibility lie in early stakeholder involvement in design and implementation, integrating specialized tools with existing EPR systems, and providing hands-on training for health care professionals. Clinicians and policy makers can use these insights to better integrate similar digital tools into clinical practice and future guidelines.