Our search and review process resulted in 10 conceptual frameworks on HRH that fit the inclusion and exclusion criteria. In this section, we 1) briefly outline these 10 frameworks and how they were incorporated into the resulting logic model, and 2) detail the resulting logic model developed based on these conceptual frameworks.
1. Search results
The research team initially identified seven conceptual frameworks from the team’s existing knowledge. Six of these were included in the final analysis, while one did not meet inclusion and exclusion criteria. Our subsequent broader search yielded 36,792 results. Of these, the Google Scholar results yielded the highest number (between 160 and 20,400 per search). To manage the number of publications in the initial extraction, we reviewed the first ~150 titles for each search (total of 1,316 articles across searches). Publications with titles that appeared to fit the inclusion and exclusion criteria were selected for further review (80 publications). We reviewed these 80 publication abstracts and found that 76 publications did not meet inclusion and exclusion criteria. Where frameworks were adaptations of other frameworks, we selected the most comprehensive framework for inclusion, or included both if they were sufficiently different in terms of components and focus. For example, [10] and [26] are considered the same framework, while [8], [12], and [23] are sufficiently different to warrant inclusion of all three. Four additional frameworks were identified, leading to a total of 10 frameworks summarized in Table 1.
The 10 selected frameworks have similar high-level elements, but differ in level of detail, layout, and depicted relationships. Overlapping elements include contextual factors, health systems building blocks, HRH policy planning and implementation, processes that influence outcomes across the HRH lifecycle, and intermediate and ultimate HRH outcomes.
The frameworks differ in perspective, focus, and the outcomes they describe. Some see the labor market and/or individuals within the labor market as drivers of health workforce outcomes, and show how policies can be implemented to influence individual choices, the labor market, and/or the education sector to achieve desired health workforce size and quality [9, 11, 14, 13, 22]. Others look at more “top-down” approaches regarding HRH financing, planning, and policymaking [8, 12, 23]. Finally, some focus on organizational and contextual factors and management strategies that improve performance and outcomes [10, 24].
[Table 1: Summary of frameworks]
In addition to the frameworks, we identified several publications that did not meet our criteria (particularly that of a visual framework), but nonetheless provided insights about HRH governance and policies that informed our logic model. They are referenced in the logic model explanation that follows.
2. Logic model
We adapted the common elements and relationships depicted in the frameworks into a detailed, interactive logic model, available at [27]. The final model has five columns with components (17), subcomponents (44), and interactions between them, which allows for high-level and detailed exploration of the model elements. The arrows in the model are directed and causality flows from left to right. We intentionally chose to create the model as a directed acyclic graph (DAG) and to minimize feedback loops. Although other models emphasize causal loops [28, 29], our goal was to show how upstream factors affect downstream outcomes. Although in reality these relationships are complex, our approach makes it easier to identify the “drivers” of an effective HRH workforce. Without this DAG constraint, the complexity of the model might overwhelm its clarifying purpose.
The high-level logic model, shown in Figure 2a, includes only the top-level components, and the broad linkages across columns. Each of the top-level components has additional subcomponents, which can be seen in Figure 2b. The detailed model dives deeper and allows the user to see granular relationships between components and subcomponents.
Each of the columns, components, and subcomponents are described below.
Column 1. Contextual factors: These include factors within the broader societal, economic, political and policy environment that affect downstream decisions and outcomes within the health and HRH system [30]. We categorized contextual factors into three larger components, following adaptation of the Political, Economic, Social, Technical, Legal, and Environmental (PESTLE) analysis framework.
- Social, legal, environmental, and epidemiological factors –the legal system (regulation and law enforcement); the basic and professional educational system; epidemiological and environmental factors that affect population distribution and health needs; levels of equity and/or marginalization of different groups (e.g. socioeconomic, gender, race, ethnicity, education, etc.); and levels of corruption [31].
- Economic factors – strength of the economy, nature of health markets—including levels of care, financing, and service provision—and larger labor market forces that affect health workforce distribution, public and private decision-making, and HRH outcomes.
- Political factors – decentralization of power and decision-making; the type and capacity of political regime in power and its level of “political entrenchment”; and the influence governmental and non-governmental stakeholders have on policy adoption and implementation [32].
Column 2. Health system factors: Effective governance institutions, organizations, practices, and capacities for management, decision-making, and data use in policymaking support sound policy development and implementation. Specific areas of health and HRH systems and processes include:
- HRH system governance – leadership, processes, and capacities for governing HRH systems. Specific aspects of governance include: individual, organizational, and systemic capacity for management and decision-making; collaboration and coordination within and across sectors and ministries for multisectoral action; transparency and accountability to government and communities; and corruption within the HRH and broader health systems [33, 34, 35, 36, 37, 38, 39]. In addition, electronic human resource management systems (e.g. managing postings, performance, attendance) along with systemic capabilities to use data for workforce planning, regulation, and decision-making [34, 35]. Governance is affected by a wider national and international context, and influences HRH policy formulation and implementation.
- HRH policy formulation and implementation – areas of strategy and planning around health workforce objectives; financing allocations and mechanisms to implement HRH interventions and policies; regulatory policy and legislation around medical practice and health workers; contract compliance of entities engaged to provide health services; and day-to-day operating procedures. Policymaking and implementation are affected by factors of HRH governance.
Column 3. Health workforce processes: HRH policies affect and are implemented through HRH processes. The Health Worker Life Cycle model [41] is another framework with similar “stages” of the lifecycle, which is based on Sousa et al [9] and therefore was not included in our results. HRH processes and strategies may be implemented across four stages:
- Production of health workers – policies or factors influencing pre-service training of health workers which are generally used to affect the size, composition, competencies, and distribution of HRH. These could include admissions policies or regulation of the quality and distribution of training institutions [40].
- Entryinto the health workforce – recruitment strategies and equitable distribution of the workforce per health system needs.
- Maintenance and performance – strategies used to retain health workers (particularly in remote underserved area) and manage their performance. These include clinical quality interventions (e.g. job aids or tools to support quality and work flow), performance management systems to measure and develop health worker performance [42, 43], in-service training to maintain and strengthen competencies, supportive supervision, and regulation of practice to ensure quality of care. In addition, it includes financial and non-financial incentives to promote health worker motivation, performance, and retention in remote underserved areas [45, 46, 47, 48, 49, 50].
- Enabling environment – HRH outcomes will be affected by the environment within which the workforce operates. This includes healthcare facilities and infrastructure, availability of supplies and equipment, and other health systems building blocks, as well as living conditions including road infrastructure, housing, and quality of local schools [40, 44, 47, 48].
- Health workerexit – exit from the workforce can be a natural progression (e.g. retirement), attrition due to pursuit of alternative careers, migration to locations with better living and working conditions, or a lack of sufficient incentives or motivation to remain in the health workforce.
Column 4. Health workforce outcomes: Workforce outcomes—specifically, health workforce availability, competency, motivation, engagement, and job satisfaction—are influenced by upstream policies, processes, and contextual factors [30, 44, 45, 46, 47, 48, 49, 50].
- Availability – availability of health workers to ensure geographic coverage according to population health requires having appropriate numbers of health workers, equitable distribution across urban and rural areas, and skill-mix across the cadres of healthcare workers.
- Will do – worker and team satisfaction, engagement, and motivation are critical for the competencies of health workers to translate into practice. “Engagement” refers to vigor and energy devoted to one’s work; involvement, dedication to, and enthusiasm in work; and absorption and identification with one’s work [51, 52]. Motivation and engagement of health workers support a drive towards quality and improving health outcomes, which supports responsiveness, efficiency, and equity of care [53, 1]. This is known as the “know-do” gap between provider skills and their application of these skills when delivering services [54]. These factors are also important for influencing retention of health workers.
- Can do – required competencies (knowledge, skills, and attitudes) that are critical for health workers to provide care with high clinical and non-clinical quality, based on designated roles and responsibilities.
Column 5. Health system outcomes. Desired health system outcomes focus on “improving health and health equity, in ways that are responsive, financially fair, and make efficient use of available resources” [1, p.2]. The ultimate goals of the health workforce are to contribute to these overall health system goals, by enhancing quality, responsiveness, efficiency, and coverage. If health worker performance is combined with a well-functioning health system, the workforce can deliver high-quality health services equitably, leading to improved population health. Specific health system outcomes the workforce contributes to include:
- Quality of service delivery in accordance with predefined standards and protocols, including clinical quality and non-clinical aspects such as safety and equity [2].
- Responsiveness in how the health system meets expectations around provider treatment [55]. This may encompass aspects of non-clinical quality including people-centeredness and patient satisfaction [2].
- Coverage of health workers across both urban and rural areas according to population health needs.
- Efficiency in utilizing financial and non-financial inputs.