Tables 1-3 present a synthesis of the models shortlisted to inform our health workforce planning process (Table 1), our service requirement and capacity projections, and our allocation of service requirements across cadres (Table 3), respectively. These synthesis tables also describe the shortlisted models’ alignment with the needs of the TC LHIN.
Based on the findings of our model assessment, we developed a hybrid HWP toolkit for primary care services. Because no single model identified through our search strategy fully accommodated the TC LHIN’s needs, we integrated key features from a number of existing approaches to develop a fit-for-purpose HWP process that aligns with the specific planning needs and objectives of the TC LHIN.
The overarching HWP process that we recommended to the TC LHIN combines promising elements from three distinct HWP frameworks. England’s Robust Workforce Planning Framework (Centre for Workforce Intelligence, 2014) informed the recommended process for health workforce planning and scenario development. Australia’s Health Workforce Planning Tool (Health Workforce Australia, 2014) informed the recommended process for stakeholder and workforce engagement. Finally, our recommended workforce planning process integrates a number of environmental scanning tools presented by New Zealand’s Workforce Intelligence and Planning Framework (National DHB General Managers Human Resources & Health Workforce New Zealand, 2014).
These promising practices in HWP nest quantitative HWP models within broader health workforce and health system planning processes that are both iterative and interactive in nature. The toolkit we proposed (depicted in Figure 2) outlines a qualitative workforce planning process that provides opportunities for primary care workforce, stakeholder, and patient engagement and facilitates the evaluation and selection of policy interventions that are robust to uncertainty across a range of possible futures.
Horizon Scanning
The cyclical workforce planning process presented in England’s Robust Workforce Planning Framework (CfWI, 2014) begins with a horizon scanning exercise to map the driving forces present within the system. Within the context of the TC LHIN, we have recommended that an internal planning group engage in a one-day horizon scanning workshop using the environmental scanning tools presented by Health Workforce New Zealand (National DHB General Managers Human Resources & Health Workforce New Zealand, 2014) to identify driving forces that could influence workforce and population health trends over the defined planning period.
Planners can use PESTLE analyses (political, economic, sociological, technological, legal, and environmental) and SWOT analyses (strengths, weaknesses, opportunities and threats) to engage in the identification of factors that can affect the ability of a system to achieve optimal or appropriate alignment between service requirements (population health needs) and service capacity (workforce supply).
First, planners can use PESTLE analysis to identify macro-level contextual factors that merit consideration in the HWP process due to their potential impact on the health workforce or on population health and demography within a particular region. As a means of enriching discussions surrounding these six categories of factors, we encourage planners to refer to a systems framework for HWP, and we employed an example specific to the Canadian context (Bourgeault, Demers, & Bray, 2015). This framework, and its applications in this exercise, are more fully described in [Bourgeault, Chamberland-Rowe & Simkin, SUBMITTED COMMENTARY]. By consulting such a framework, planners can ensure that their discussions account for the complex network of system-level inputs and policy levers that must be mobilized in order to allow for population health needs to serve as the drivers of health workforce planning and deployment.
Second, SWOT analyses allow planners to categorize external (contextual) and internal (organizational) factors as either favourable or unfavourable to the desired system outcome (e.g., a balance of population health needs and health workforce supply and capacity), and to the ability of planners to achieve this outcome through targeted planning and intervention. As an initial step for SWOT analysis, planners can categorize the contextual factors identified through the PESTLE Analysis as either opportunities or threats. Planners can then identify internal organizational factors that should be considered in the workforce planning process and categorize them as either strengths or weaknesses.
These analytical tools allow planners to account for their sphere of influence and the policy levers at their disposal to control the factors identified. Internal factors are within the planners’ sphere of influence, and so these factors are more readily reinforced or remedied, whereas planners must develop strategies to leverage external opportunities and mitigate external threats that are beyond their sphere of influence. We have recommended that planners synthesize the outputs of this horizon scanning workshop into a brief report that can serve to frame a broader consultative process.
Planners can use environmental scanning tools in the horizon scanning phase of workforce planning to explore the breadth of factors that interact within the health region as a complex adaptive system. In subsequent stages of scenario generation and policy analysis, planners can use these same tools to delve deeper into particular issues of concern in the delivery of primary care within the region. Furthermore, all of the included environmental scanning tools can be used for both internal brainstorming and external consultation and engagement throughout the HWP process.
Scenario Generation
Scenario generation allows planners to elicit, develop and focus on HWP scenarios that are relevant to their communities. The scenario generation process is also critically important to inform the ultimate data requirements for quantitative modelling. We recommended that planners conduct scenario generation workshops at the sub-region level as well as at the full-LHIN regional level, ensuring that both local and region-wide workforce issues can be addressed. These one-day workshops are designed to bring together a broad range of stakeholders to augment the list of factors generated by the horizon scanning exercise, and develop narrative scenarios shaped by the uncertainties that may influence the future state of the system (CfWI, 2014).
Stakeholder consultation promotes the modelling process and reinforces the relevance of its outputs (Kinsella & Kiersey, 2016). Furthermore, stakeholder engagement can foster buy-in and facilitate the acceptance of projections as a trusted evidence-base for policy action (Crettenden et al., 2014). To supplement the work conducted internally by the TC LHIN and infuse the scenario generation process with local workforce intelligence, we have recommended that planners invite clinical leads from each concerned primary care cadre, patient advisors, and other relevant experts to participate in scenario generation workshops.
During these workshops, participants develop narrative scenarios that describe a reference future, which is considered to be the most probable and reasonable baseline future given current trends, as well as alternative futures that reflect the potential effects of the driving forces identified during the horizon scanning workshop. In addition to the environmental scanning tools described in the previous section, planners can use causal loop diagrams during scenario generation workshops to map the complex web of interactions between factors and system components. Once the causal loop diagram has been drawn, participants are asked to elaborate on a series of narrative scenarios that describe its interactions, and their potential impact on service requirements and capacity. Causal loop diagrams can assist workshop participants in gaining a more holistic understanding of the challenge, allow them to elaborate consistent and valid narrative scenarios, and enable them to identify the quantitative variables that require manipulation to simulate this scenario using the HWP model.
The toolkit then bridges qualitative and quantitative approaches by employing the elicitation methods described by England’s Centre for Workforce Intelligence (2015) - including traditional Delphi Processes, the EFSA Delphi approach, and the Sheffield Elicitation Framework - to gain expert consensus on the estimated quantitative input parameters of narrative scenarios. These inputs reflect the potential influence of these driving forces on service requirements and capacity. We recommended that the TC LHIN host an elicitation workshop to define the parameters of the reference future using the Sheffield elicitation framework, and that the parameters for alternative scenarios be elicited remotely using the EFSA Delphi Approach. Both of these approaches allow planners to define probability distributions for each elicited parameter, including upper and lower bounds of the plausible range of values, a median value, and upper and lower quartiles.
Workforce Modelling
Embedded within the proposed HWP process is a quantitative HWP model. This model begins with the development of population health and workforce profiles that inform service requirement and service capacity projections, respectively. Planners then conduct an initial assessment of alignment between service capacity and service requirements which is supplemented by a descriptive allocation process designed to explore workforce capacity to meet population health needs under alternative models of care (optimizing the distribution of service requirements across the full spectrum of cadres contributing to integrated primary care).
Three models informed our initial assessment of alignment between service capacity and service requirements in the TC LHIN: the Canadian Institutes for Health Information Population Grouping Methodology (CIHI, 2017), the Needs-Based Health Human Resource Planning Framework (Birch et al., 2007), and Manitoba’s Needs-Based Planning for Generalist Physicians (Roos et al., 1997). The descriptive allocation process outlined in the toolkit is inspired by adjusted service target-based planning approaches (Dreesch et al., 2005; Guerra Arias et al., 2017; ten Hoope Bender et al. 2017; Jansen et al., 2014). Simkin et al. (SUBMITTED PART 2 PAPER) present the development and output of the quantitative service requirement and capacity projection models included in this toolkit.
The quantitative scenario parameters identified through the elicitation processes can be used as inputs for the modelling stage. The HWP model should be run using the reference future scenario, as well as all scenarios defined in the previous step of the workforce planning process. Planners can introduce scenarios to assess the impact of alternative population health and workforce profiles, and of alternative allocations of services across cadres with relevant scopes of practice.
Policy Analysis
Finally, planners can hold structured workshops to explore potential policy interventions that could be conducive to remedying any misalignments highlighted by the model’s gap analysis.
We have recommended that the TC LHIN invite the expert participants who were engaged in scenario generation, and a broader range of primary care workers and patients to participate in these discussions.
Planners can develop the narrative description and quantitative input parameters for identified policy scenarios using the tools prescribed for scenario generation. The influence of potential policy interventions can then be measured against all identified scenarios, which represent a number of potential futures. Policies are therefore considered “robust” to uncertainty if they produce favourable workforce outcomes against a high proportion of potential futures (CfWI, 2014).
As an additional layer of robustness, Porter’s Five Forces Framework can be used to identify key forces with the potential to influence the implementation of proposed workforce policies and interventions. Planners are encouraged to assess whether the implementation of an intervention could be influenced by the bargaining power of suppliers and buyers, or pose a threat to the existing workforce through the introduction of new entrants or substitutes. This framework is particularly amenable to the identification of dynamic interactions between actors and interests within health systems that could influence the implementation of proposed workforce policies and interventions. These considerations are salient given the social and political context within which HWP occurs. In developing scenarios, and interpreting health workforce projections, planners must take into account the whole picture, acknowledging that political and social contexts can influence the levers at their disposal and their capacity to act upon the evidence generated by these models in order to achieve desired outcomes.