We have followed the Standard Protocol Items: Recommendations for Interventional Trials guidelines for reporting on protocols [40]. This is a single-arm, prospective, explanatory, mixed-methods, pre-post design feasibility study.

The study is set in central Ottawa (Ontario, Canada), a region with broad socioeconomic diversity, including Francophones (19%), immigrants (19%), visible minorities (18%), and seniors (16%) [41].

The ARC model was implemented in 2 practice contexts to understand the different levers and barriers to implementing the ARC model: (1) the traditional primary care practice, which consists of family physicians, nurses, and administrative staff; and (2) the interprofessional practice model in which patients also have access to various allied health professionals such as nurse practitioners, pharmacists, and social workers, as well as some in-house health programs. Participating PCPs are required to provide comprehensive primary care services to the general population.

Patients of participating PCPs were eligible to participate if one or more needs, which may be addressed by services offered in the community, are identified during an encounter with their PCP, they are able to communicate in English or French or willing to be served by a cultural interpreter or translator, and have no marked cognitive deficiencies or have a family member that can provide proxy consent and participate in the study with the patient.

This study was approved by the following ethics boards: Ottawa Health Science Network Research Ethics Board (#20160914-01H), Bruyère Continuing Care Research Ethics Board (#M16-16-055), University of Ottawa Research Ethics Board (#A05-17-04), and L’Hôpital Montfort Research Ethics Board (#SD-DP-27-02-17).

We established a Collaborative Partnership of key stakeholders to inform the development, implementation, and evaluation of the ARC navigation model. The partnership includes policy makers, members from community organizations, health care providers, health planners, and people with lived experience of health care as a consumer or caregiver. The ARC intervention intends to promote equitable access to health-enabling resources by engaging PCPs in identifying their patients’ needs that could be addressed by a CR and directing them to such services. Patients’ access to these resources is subsequently supported through navigation. A logic model (presented in Multimedia Appendix 1) was developed to establish the expected links among the intervention components, planned activities, and anticipated outcomes. The ARC navigation model is presented in Multimedia Appendix 2, and details of the ARC intervention are described in Multimedia Appendix 3 (Template for Intervention Description and Replication Checklist [42]). The study promotional materials for patients are offered in Multimedia Appendices 4 and 5 and the ARC referral form and instructional video for PCPs are presented in Multimedia Appendices 6 and 7. The two main intervention components are (1) changes to the practice environment to enhance PCP recommendations for patients to access CRs that could address their needs; and (2) navigation support for these patients to help them achieve access to resources. Briefly, we relied on Rogers’ Diffusion of Innovation theory [43] to underpin the practice changes required to promote referrals to CRs in primary care practices. These changes were informed by a rapid realist review conducted by us to understand the most likely effective approaches and their potential benefit in our context [44]. The navigator’s role was founded on the Health Action Process Approach [45] and aims to support patients in achieving their health and wellness goals by promoting access to health-enabling resources in the community.

The role of the navigator is limited to nonclinical activities that support access to community services that address a wide range of health and social needs. Multiple sources were used to inform the role of the navigator including (1) peer-reviewed literature about the scope of navigation activities provided with different populations, medical conditions, and contexts [38,46-49]; (2) best practice guidelines for implementing and evaluating community health worker programs [37]; and (3) literature on navigation training programs.

The ARC navigator is a lay person with no clinical background who is hired by the research team and trained to provide navigation services. PCPs were encouraged to see that new team member as someone lending a helping hand to patients in need of support to gain access to needed services—the type of support a well-informed family member could potentially provide. It was made clear that the navigator is a lay individual and would not be expected to address any clinical issues and that should such issues arise, they would be communicated back to the PCPs for their action. We sought to hire an individual with excellent communication skills, evident empathic qualities, and good management abilities. The navigator training program is based on competencies developed specifically for their role in primary care. The key competencies are to provide basic navigation services and identify appropriate resources within our context, demonstrate effective interpersonal communication including cultural and linguistic sensitivity, collaborate and work effectively with primary care team and CR program staff, advocate for patients and intervene with services to promote access to needed resources, demonstrate commitment to professional responsibilities and ongoing learning, and educate and empower patients about CRs for their health.

The navigator would spend a minimum of one half day at the practice during which they would meet with patients referred to them and interact with the members of the primary care team. The navigator did not chart in the patient medical records. Written communication between the navigator and primary care team was by fax. The expectation was that PCPs, sensitized to the availability of CRs that could complement the care they provide to their patients and confident that more socially complex patients can receive the assistance they require to support access to these resources (from the navigator), would be more likely to identify patient needs that could be addressed by CRs, discuss these during their encounter, and if appropriate, make the recommendation to access these services. The navigator would then meet with each patient to whom a service was recommended to understand their needs, expectations and priorities, identify anticipated barriers in accessing the resource (eg, knowledge, health literacy, transportation, completing forms, caregiving responsibilities, financial, and motivation), support patients in overcoming these barriers, and facilitate access to the most appropriate resource for patients. In addition, the navigator was trained to use communication strategies, such as motivational interviewing [50], that would encourage engagement of patients in identifying what is important to them, what they perceive as challenges in accessing CRs, and the navigation activities needed. Patient self-efficacy is fostered through providing various forms of social support including providing information, emotional support, and guidance or accompaniment to access CRs. The navigator would report back to the PCP on the plan developed after the first encounter and again at the end of services. Navigation services would be discontinued when a patient has accessed the appropriate service or no longer wants or requires navigation assistance. Navigation services are offered to patients for up to 3 months.

The patient recruitment period was 9 months, and individual patient participation will be approximately 3 months.

Sample sizes of 30 [51] or 50 [52] are commonly recommended for feasibility studies. We based our sample size on the ability to adequately assess the demand for the navigation services and to estimate the potential for patients to achieve the intended access to the resource. We estimated that the participation of 4-6 practices, each with at least 3 PCPs (minimum 12 PCPs) caring for a panel size of 1500 patients, would be required. We assumed conservatively that 30% of patients at the practice could potentially benefit from a CR. In the region where the study is being conducted, >60% of the population has insufficient consumption of fruits and vegetables and 28% individuals are obese, 19% drink heavily, 12% smoke, and 22% report feeling stressed “quite a lot” [53]. The 95% CI range for these outcome measures would be <3% for referral rate (5400 potentially eligible individuals), <15% for participation rate (if conservatively 162/5400, 3.00%, are referred), and <22% for the access measure (assuming a participation rate of 81/162, 50.0%). We also aimed to interview a minimum of 1 PCP and 2 patients per practice.

The study was promoted among stakeholders. Providers expressing interest were sent a study information and consent form, and a recruitment session was scheduled with all interested PCPs at the practice.

During encounters with patients, when participating PCPs and their patients identified a need that could be addressed by a CR, the PCP completed a standardized CR referral form, briefly introduced the study, and requested the patient’s agreement to be contacted by a member of the research team. Patients who agreed to be contacted received a study information and consent package and a copy of the completed referral form identifying their need(s). A copy of that form was faxed to the study team who then contacted the patient and provided detailed information about the study. Patients who provided verbal consent to participate in the study were asked to sign and mail in the consent form included in their study information and consent package.

A subsample of PCPs and patients participated in the qualitative component of the study, which consisted of an interview at the beginning and again at the end of their participation in the study. All PCPs were invited to participate in the interviews, with the aim to enroll at least 1 PCP per practice. Patients were purposefully selected to be invited for the interviews based on their responses to the baseline survey. The recruitment aimed to maximize variation in social complexity and ensure variability in age and gender. At least 1 patient from each practice was required.

Table 1 provides a summary of the data collection tools, the dimensions measured, the population targeted, as well as how and when the tools are to be administered.

One practice member, referred to as the practice champion, is the main contact for the practice. The practice champion completes the practice survey, and each participating PCP completes the provider survey. These surveys are completed at baseline before the introduction of the patient navigator in the practice, and again immediately prior to ending the navigation services. These surveys assess the practice’s organization and PCPs’ knowledge, attitudes, and experience with reference to vulnerable populations, as well as factors that can influence the success of the intervention and its implementation from a change management perspective, including the organizational structure [54], climate of their work environment [55], as well as readiness [56] and commitment to change [57]. Furthermore, the postintervention provider survey includes questions relating to their experience with the patient navigator.

Patients complete a preintervention survey immediately after providing consent and prior to meeting with the navigator, and a postintervention survey at 3 months. These surveys assess various dimensions of access, measures of self-efficacy, social vulnerability, and their experience with CRs and the ARC navigation services.

A subset of patients and PCPs are also invited to participate in an interview following the completion of the pre- and postintervention surveys. These interviews explore patients’ access to PHC services and providers’ experience providing care to vulnerable patient populations. In addition, patients and PCPs are asked about their experience with the patient navigator. These interviews will be used to understand and build upon the survey results [58].

Conforming to a rapid cycle evaluation (RCE) approach, the designated practice champion completes regular assessments of the study progress to inform the need for adapting the intervention to meet the needs of the practice; this evaluation also informs on levers and barriers to the changes imposed by the introduction of an additional layer of services. The first evaluation was conducted immediately following the initial implementation of the intervention and assessed the practices’ experience with the introduction of the components of the ARC intervention. Subsequent evaluations were conducted bimonthly to evaluate their impressions of study progress [59].

Throughout the study, the coordinator maintains a log of activities relating to encounters with practices and weekly debriefs with the navigator. The patient navigator maintains a log of activities relating to patient support, including encounters with patients, their PCPs, and staff from the recommended CR. Furthermore, the patient navigator completes a reflective journal describing their thoughts related to day-to-day activities to promote deeper understanding of the knowledge and skills required to carry out their role.

The feasibility study will assess 8 areas of focus: Acceptability of the ARC model; Demand for the navigation service; Implementation approach viability; Adaptation required; Integration of the navigation service within the practice; Practicality of the ARC model to the practice, providers, and patients; the potential for the intervention Efficacy; and Appropriateness of the intervention evaluation approach to study participants [39]. Table 2 provides the operational definition of the areas and their data source.

Procedures developed by the ARC team and captured in various training and “how to” guides contributed to the standardized implementation of study activities related to data collection, coding, entry, and storage. Quantitative data are inscribed directly into Qualtrics, a centralized data collection tool, and transferred to SPSS (IBM) for analyses. For qualitative data, interview notes or transcripts and open-ended answers to survey questions are entered into (NVivo; QSR International), a software that facilitates content analysis.

Table 2 provides a summary of the analytical approach to each of the outcome measures related to 8 areas of focus of a feasibility study. Consistent with principles of an explanatory sequential design of a phenomenological tradition, we will begin by analyzing the patient and provider surveys, the results of which will guide the qualitative line of questioning, which will seek to further explore survey findings. The overarching research question guiding this qualitative phase is, “What is the provider and patient experience with navigation to CRs?” Questions for the patient and provider interview guides align with elements of the study’s conceptual framework including Rogers’ Diffusion of Innovation theory [43], the Health Action Process Approach [45], and elements of the Access Framework [60]. The interview questions will be further developed on the basis of the quantitative results as they emerge, and coding and content analysis will follow a sequential iterative process [61,62]. The RCE data are analyzed qualitatively to inform the ongoing adaptation of the intervention integration within each practice. While the elements of the intervention were theory driven, how these are applied within complex systems, such as primary care practices, must be informed by that context. We anticipated that team composition, use of electronic medical records tools, and clinic layout and flow would be influential factors but expected that additional context factors would emerge through the RCE [43,63,64].