In Canada and elsewhere, COVID-19 has highlighted the human resource gaps already present in health care systems. In New Brunswick, health care services are weakened by a shortage of registered nurses and physicians, mainly affecting rural areas, where Francophone and Acadian communities are concentrated and considered as Official Language Minority Communities (OLMCs) [1]. These OLMCs have greater social and health disparities (eg, smoking and obesity) [2] and are older than their Anglophone counterparts, which are the majority population in New Brunswick [3,4]. In other words, being part of a linguistic minority group is a potential barrier to health care and services access and makes people vulnerable to health inequities [3-5].

Since 2008, a total of 2 health networks have been providing health services and care to the population of New Brunswick: Vitalité Health Network (the “Network”) and Horizon Health Network. They are required by the Regional Health Authorities Act to provide services to users in the official language of their choice (English or French). It is worth mentioning that the same law stipulates that the language of operation of Horizon Health Network is English, whereas it is French for the Vitalité Health Network. The Network, our partner, covers 4 geographic regions, 3 of which are in rural areas where many Acadian and Francophone communities reside. The fourth zone includes a more urban area, where both health networks provide health services. The New Brunswick population currently totals 775,610 citizens, 30% of whom report French as their first language [6]. The Network manages and delivers health care and services to Francophone citizens in the province, representing approximately 241,100 people living in the 4 geographic territories [6]. It has approximately 7700 part-time and full-time employees, including 560 physicians [7].

The latest data show that there are 100 vacant physician positions [8] and 200 vacant registered nurse positions [9]. For the past year, a shortage of staff has caused the closure of the obstetric services in the Restigouche region [10]. Also, the departure of oncologists has recently affected the Dr Georges-L.-Dumont University Hospital Centre in Moncton. Several media papers have addressed these issues [11]. Provincial projections predict a critical shortage of more than 1300 registered nurses in the next 5 to 7 years, and approximately 35% of family physicians will retire [12]. Given these staff shortages, combined with the global COVID-19 pandemic, it is imperative to strengthen the Network’s retention strategies to ensure its viability and reduce the health disparities in OLMCs.

The conceptual framework is based on the “Making it Work” framework [13,14] and the “Magnet Hospital” model [15], in addition to the delivery of health care services in OLMCs. The inclusion of the linguistic minority variable will lead us back to our objectives or research questions and will provide theory-based results, increase credibility, and foster transferability to practice settings [16].

The “Making it Work” framework, which focuses on rural and remote recruitment and retention, is based on 9 key strategic elements that primarily address determinants of the environment that are grouped under 3 main activities (ie, planning, recruiting, and retaining) and are briefly described [13,17] (see Figure 1). The Network’s Board of Directors decided in 2020 to take a closer look at this framework, which has been the focus of a framework in Northern Ontario [14,18]. The framework was launched in 2011 by an international collaborative, addressing the challenges of recruiting and retaining health professionals in rural and remote northern settings and has been the focus of the Northern Ontario School of Medicine, which has sites throughout the north of Ontario, 1 being in the OLMC of Sudbury [19].

Planning thus involves assessing the population’s health needs and aligning service delivery models with them [17]. This task also includes developing job profiles for the professionals the organization wishes to recruit, including skills specific to working in rural and remote settings [17]. Recruitment focuses on sharing information about the work environment, practice, population served, and available housing in the area [17]. Recruiting also involves consideration of community engagement and support for the professional’s family, such as employment opportunities for the spouse. Finally, elements included in the retention component of the model are related to team cohesion, professional development, and training [17].

In total, 5 conditions are contingent on the success of the framework [17,22]: acknowledgment of the unique challenges of rural and remote areas (eg, the unique aspects of living and working in these areas compared to urban settings), engagement of communities and consideration of their perspectives (in workforce planning or policy-making), adequate investments (targeting professional development and training internships, in particular), an annual cycle of activities (as opposed to actions driven by urgency) that promote recruitment and retention (like attention-drawing job descriptions or performance standards), and finally, continuous monitoring and evaluation (to modify interventions based on experiences and improve quality to meet the needs of the population). Conjointly, these conditions promote the establishment of a stable and appropriately qualified professional workforce in rural and remote communities, in this case, the OLMCs.

The Magnet Hospital model emerged in the early 1980s when the American Academy of Nursing appointed a task force to study the facilitating factors and barriers to professional nursing practice in hospitals [15]. McClure et al [23] revisited the work done by these forerunners, focusing on organizational culture and quality of care:

Thus, the Magnet Hospital model, or the “magnet” hospital, means “literally, the ability to retain and attract staff, and figuratively, the ability to foster well-being to their patients” [24]. The evolution of this model over the past few decades led to the model presented in Figure 2.

The model [25] in Figure 2 builds on areas that are primarily part of the work environment: (1) transformational leadership (ie, “magnet” managers at all organizational levels), (2) exemplary professional practices (eg, the ethic of caring), (3) structural empowerment (eg, decentralized decision-making), (4) new knowledge, innovation, and improvements (eg, research- and evidence-based decision-making), and (5) empirical outcomes (eg, quality of care and organizational goals) [25,26]. These components cannot be separated from the overall context, relating to, for example, the global pandemic, and all have implications for employee retention, optimal service delivery, and quality of clinical care [27,28].

The shortage of registered nurses and physicians is particularly acute in rural areas, where are concentrated OLMCs [1,3]. These regions have difficulty attracting and retaining professionals who generally do their training and start working in urban settings [29]. Also, a preference for acute care clinical settings in urban areas and the mobility of professionals make recruitment and retention challenges difficult to overcome by rural and remote OLMCs [30].

A study [30] has shown that one of the factors affecting the retention of professionals in OLMCs is the quality of the work environment. For example, the quality of social ties, the sense of belonging, the diversity of the environment, both ethnic and linguistic, and the working conditions influenced their desire to remain in their position [30].

For OLMCs, a shortage of registered nurses and physicians can lead to inadequate access to health services and negatively impact initiatives to meet their health needs, such as those in prevention and promotion [31,32]. The provision of quality and safe health care services to OLMCs is, in other words, linked to the availability of competent professional resources [33]. Furthermore, studies have shown that the challenges to offering services to the linguistic minority group are due to the lack of professionals who are able to provide these services within OLMCs [34,35]. The smaller pool of bilingual professionals can thus have consequences on the health of these communities [36-38].

The conceptual framework presented in Figure 3 shows that varied factors in the internal and external environment influence the retention of registered nurses and physicians. The “Magnet Hospital” model provides key work environment components for understanding the retention capacities of these health professionals that are also essential to quality patient care, whereas the “Making it Work” concept helps us analyze the retention factors through the lens of strategic elements crucial to accessing high-quality health services in rural and remote environments.

The internal environment of the health organization (in yellow in Figure 3) has three components: (1) structural empowerment, (2) professional practices and innovations, and (3) transformational leadership. Structural empowerment refers to a structure that is decentralized and debureaucratized [39]. This component includes recognition measures such as awards or scholarships for innovative projects [40] and employee satisfaction with salary, workload, and work-life balance. It also considers other vital elements such as competitive pay, adequate staffing, a sense of control over decision-making and practice, and perceived stress levels [15,24].

Professional practices and innovations are associated with a philosophy of excellence in care that emphasizes evidence-based skills and best practices, a focus on quality improvement through innovative approaches, and the value placed on professional development [15,39]. This component indicates the importance of ensuring that resources are available to new employees like mentorship, access to advice, and other resources to help with their integration. A collaborative relationship between registered nurses and physicians and autonomy based on professional standards are all part of this component [15,27]. Ethical values in caring and transparency are linked to better care for people needing health care and services while helping with their self-fulfillment [24].

Transformational leadership refers to close supervision that is characterized by a manager who is visionary, respectful, attentive to their collaborators, stimulating, and mobilizing [24,41]. To counteract burnout, the caring approach to management prioritizes social ties and team cohesion as well as dialogue and trust [41]. In addition, the practice of supportive management must take place in a stimulating work environment to be effective.

The external environment of the health organization (in blue in Figure 3) is made up of three main dimensions: (1) the characteristics and needs of OLMCs; (2) the relationship established between the health organization and the OLMCs, and their involvement in attracting and retaining professionals; and lastly (3) interactions with the Department of Health and other relevant actors (such as training institutions).

The characteristics and needs of OLMCs focus on the particularities of the populations served by the health organization. In New Brunswick, the territory served by the Network is composed primarily of Francophone and Acadian populations, many of which are in rural areas. These citizens are generally more disadvantaged in terms of socioeconomic status and are older than elsewhere in the province or the country [42]. The health needs of the OLMCs served by the Network differ from one health zone to another and even from one city to another and must be reevaluated on an ongoing basis. A service model that does not meet the needs of rural and remote communities can lead to burnout and job dissatisfaction for its employees [14].

The involvement of OLMCs in efforts to counter the departure of health professionals is reported in the literature on the "Making it Work" concept [13,14]. While the involvement of rural and remote communities is crucial when recruiting health care professionals [14], policies that encourage their active and ongoing participation are also essential to ensure the retention and stability of medical personnel [43].

The interactions with the Department of Health and other relevant actors go hand in hand with the involvement of OLMCs, as it is linked to the relational dynamics in the planning stage between the health organization and other key players, particularly the Department of Health, training institutions, and nongovernmental organizations. Decisions made by actors of the external environment, sometimes outside the control of the Network, can affect the intention to stay of professionals. Furthermore, without adequate investments, such as in professional development, efforts remain focused on solving urgent problems (vacancies) at the expense of an ongoing strategy to ensure the stability of the skilled workforce [17].

Finally, the empirical results (in green in Figure 3) are influenced by “what’s going on” in the environment (internal and external) of the health care organization. Factors emanating from these 2 spheres affect the commitment and retention of registered nurses and physicians responsible for improving the quality of services offered to OLMCs.

The overall objective of this study is to support one of New Brunswick health networks in identifying and implementing concrete strategies that take into consideration its environment to promote physician and registered nurse retention.

Two specific objectives are derived from the general research objective: (1) to identify current staffing trends and examine the turnover rate in each geographical zone of the Network to determine where the most significant challenges are with recruitment and retention; (2) to identify and understand the factors that affect the retention of physicians and registered nurses (a) at the operational level, to determine the factors that make the Network’s environment (internal and external) attractive, such as the working and living conditions of professionals, and (b) at the strategic level, to make concrete and valuable recommendations to our main partner, the Network.

This study will use a sequential mixed methods design that combines quantitative and qualitative approaches [44]. This design offers a more comprehensive technique to better understand phenomena and nuance data interpretation [45].

Initially, secondary data will be compiled within existing sources to identify current organizational trends and inform our qualitative data collection strategy. In other words, the quantitative analysis of secondary data will guide the qualitative approaches’ recruitment process. Specifically, we will assess job vacancy and turnover rates for nurses and physicians to target the more problematic geographical zones and departments concerning retention. Turnover rates are commonly evaluated as an indicator of retention of hospital staff, and it is especially used for nurses [46-48]. When turnover rates are chronically high, it may indicate that staff working in that hospital unit are not satisfied at work [46] or that the work environment is not favorable to psychological well-being [48]. Research on turnover rates indicates average annual turnover rates for nurses are between 14% and 27% in Canada [49]. This finding is supported by data worldwide that found the average turnover rate for nurses to be between 15% and 36% [50,51]. For this study, we will use job turnover rates as a guide to target the areas that are more or less successful with retention.

As a secondary indicator for retention, vacant positions will also be analyzed for nurses and physicians within the Network. Using a similar strategy, the job vacancy rate will be assessed as an indicator for retention [52]. The 2 indicators, the turnover and job vacancy rates, will be analyzed together to target areas within the Network that are doing well with retention and areas with retention challenges.

We will then target these areas to recruit participants for individual and group discussions as described in section “Qualitative Approach: Data Collection and Analysis.” For physicians, position turnover rates are less reported, and researchers generally use the intention to leave as a proxy for turnover rates [53]. Of the little data available, the average turnover rate for physicians is between 4% and 25% [54].

Component 1 (quantitative secondary data analysis): to achieve specific objective 1, secondary data from the Network will be used to illustrate the proportions and the tendencies in physician and registered nurse vacancies by zone in the Network. It will also measure the extent of the current shortage and determine if there are significant differences between regions and sectors (eg, psychiatry and obstetrics) to identify those that are more successful than those that are less successful. The data from the past 5 years related to age and gender, the number of years of service in the Network at the moment of departure, family status, type of professional (specialist vs generalist), hospital unit or service area, and geographic area will be analyzed. Information on turnover rates for physicians and registered nurses for the past 5 years will be compiled and analyzed from the Regional Plan of Physician Workforce descriptive data collected by the Network’s Medical Services. To supplement this information, data from the Network’s Human Resources Department regarding medical and nursing staff who have left their positions will be used. From these data, a profile of professionals who have left the Network can be created to tailor retention strategies and improve their effectiveness.

Since we plan to use secondary data on job vacancies and turnover rates, we will be limited to the data already collected and may face an information gap. Some team members have used quantitative approaches in their research for several years. Faced with sometimes incomplete databases, the team may have to develop data collection tools to collect more quantitative data to supplement the secondary data provided by the Network. At this time, the approach considered is a web-based survey to target employees (nurses and doctors) working at the Network and people (nurses and doctors) who have left. If a survey is used, validated questions will be used to measure factors that impact job satisfaction and intention to leave the current workplace [55-57]. The survey design will be multiple choice questions with a 7-point Likert scale to maintain validation from previous studies.

The statistical analysis for specific objective 1 will be discussed first. Next, we will conduct bivariate and multivariate correlational analyses to assess the influence of the factors comprised in our conceptual framework over health professionals’ intention to stay. Considering that participants’ sociodemographic characteristics (ie, sectors, gender, age, education, and years of service) may influence the relationships between the variables under study (ie, factors comprise in our conceptual framework), analyses will be conducted to determine the magnitude of their effect [58].

Component 2 (qualitative data): to achieve specific objective 2, qualitative data will be collected through semistructured interviews, focus groups, and a document review, including gray literature such as media reports.

The project was approved by both ethical research boards: the Vitalité Health Network (101574) and the Université de Moncton (2122-105). A detailed consent form will be explained orally and in writing to all participants. Informed consent will be obtained for all participants before taking part in the individual interview and focus groups. The importance of privacy and confidentiality protection for the research team will also be expressed to all participants, as one of our research partners is their current or past employer. Thus, each participant will be given a numerical code to preserve their anonymity. The code will be kept in a locked cabinet that only the 2 primary researchers can access. As for result reporting, such as in publication or recommendation reports, it will be done so the participants cannot be identified.

Participants in the current workforce will be identified using a “snowball” technique [61]. Initial meetings with physicians and registered nurses associated with the Network should allow us to identify additional participants from these 2 disciplines who can inform us about the shortage of professional resources within the organization. A research team member will contact potential participants by phone or email to explain the research project and solicit their participation.

For respondents who have left the Network, we will seek the Network’s Human Resources support to identify potential candidates. Human Resources Department will then send an email to them, including the invitation to participate and specifying that the research is independent of the employer and that the latter will not have access to the data. The study details and the information on how to join the research team will be included in this invitation message. This procedure will ensure that no confidential employee information will be transferred to the research team. A CAD $20 (US $14.68) gift card will be offered to interview and focus group participants.