Approach

Theory of change

The ToC methodology was developed by Weiss in 1995 as a theory-based evaluation framework to describe how a prespecified long-term outcome can be achieved through a logical sequence of preconditions.14 15 Explicit attention is given to links between inputs, mechanisms of change, outputs and outcomes, and regarding external factors. The ToC framework has not previously been used in the context of global surgery but has been shown to be an effective method of identifying priorities to inform the implementation of health policies in LMICs.16–18 The framework is constructed using a backward-mapping approach, starting with the long-term outcome (the vision statement) and subsequently plotting the proximate needs or ‘elements’ required to achieve these and their more distal dependencies (see online supplemental appendix 2 for definitions of the ToC terminology).19

Identifying a vision statement

During the workshop, the participants were orientated to the ToC methodology and randomly divided into four groups of 8–10 people. The groups worked through their tasks and then summarised their findings to all participants for broader discussion and consensus. The first task was to develop an AfroSurg vision statement to reflect the group’s focus on improving equitable access to safe and timely surgical care across the respective countries, thereby defining the network’s future aims. The agreed vision statement of Afrosurg was ‘An African-led, regional network to enable evidence-based, context-specific, safe surgical care, which is accessible, timely, and affordable for all, capturing the spirit of Ubuntu.’

Identifying knowledge needs

The next task was to identify service delivery and policy elements required to achieve the vision, where service delivery is concerned with elements needed to provide surgical healthcare, and policy refers to rules and mandates required to enable better provision of surgical healthcare. Starting with service delivery, each of the four groups discussed elements that embodied the AfroSurg vision statement before consensus was achieved on the four priority service elements. Each group was then assigned one of the main elements to identify dependencies required in order to achieve the element; these were discussed in plenary and modified through consensus. Finally, groups were reconvened to draw from these dependencies and to identify the knowledge needed to inform them and thus achieve each element. The same process was repeated for policy.

The ToC process was depicted by mapping all elements, their dependencies and knowledge needed to ensure equitable access to quality surgical healthcare in the SADC region. As customary for a ToC framework, the schematic gave direction for the research and critical analysis needed to achieve the vision and, in doing so, demonstrated how providing knowledge would be essential to accomplish this. After the workshop, the writing group (DSB, MLO, KC and JD) further analysed and refined the elements, dependencies and knowledge needs, to strengthen and enhance clarity. On completion of the primary diagram, the schematic representation of the ToC was shared and subsequently reviewed by all participants. Participant feedback and comments were considered, and adjustments were made before creating the final diagram.

The final ToC schematic is presented in figure 1. In order to achieve the vision, we selected the highest priority service delivery elements of information, research and sharing, quality of care, communication and access to care. We identified 37 distal dependencies and 23 knowledge needs related to these four service delivery elements. The highest priority policy elements were mandatory data sets and surgical workforce followed by defined packages of care and stakeholder input. Twenty-four distal dependencies and 22 knowledge needs were identified for the policy elements. These are shown in tables 1 and 2.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Depicts the long-term vision statement (pink). Policy and service delivery are positioned on either side of the vision statement along with their respective elements (orange), dependencies (grey) and knowledge needs (yellow). Notably, although the order in which the workshop was conducted was from the central to the peripheral aspect, the arrows on the figure direct one from the peripheral to the central aspect according to the order in which the steps would be carried out in practice. Hence, the figure depicts the backward mapping of the ToC methodology. The policy and service delivery side of the ToC are shown separately in the online supplemental appendices and tables 1 and 2 contain expanded information on each knowledge need. ToC, theory of change.

Prioritising knowledge needs

In order to prioritise the knowledge needs identified during the ToC exercise, an electronic Delphi exercise was conducted after the workshop with all AfroSurg members. The writing group and organising/facilitation team did not participate in the Delphi exercise. All elements and their knowledge needs—for both policy and service delivery—elicited during the ToC exercise were presented to the group in the Delphi survey. The participants were asked to anonymously rank the knowledge needs from one to five on a Likert scale. A percentage score was calculated by dividing the total achieved score for each knowledge need by the maximum score that could be achieved, allowing us to rank the top 10 knowledge needs.

Of the 30 ToC participants invited to complete the Delphi exercise, 27 (90%) responded. The top 10 knowledge needs identified by Afrosurg collaborators are shown in online supplemental appendix 3; the top 5 (table 3) were (1) mapping of available surgical services, resources and providers; (2) quantifying the burden of surgical disease; (3) identifying the appropriate number of trainees; (4) identifying the type of information that should be collected to inform service planning; and (5) identifying effective strategies to encourage geographical retention of practitioners. Four of the top five prioritised knowledge needs were policy-related, and one was related to service delivery (see table 3).

Reflections

Using a ToC approach is novel in global surgery. The lack of access to quality surgical care in the SADC region is a multifactorial issue, and the ToC approach is ideal to explore this topic because it allowed the group to acknowledge and then to focus on the service delivery and political processes that underlie health system change. Although others have worked to develop consensus to ultimately benefit surgical care, previous stakeholder meetings have a priori aimed to identify research priorities.3 20 The ToC approach starts with a participatory process to agree on a common goal which allowed us to collectively construct a vision statement prior to defining priorities and their dependencies in a more structured manner. The main drawback with the use of the ToC was the lack of participant familiarity with the process. However, the initial orientation session to the process and facilitation provided by the senior author (JID) mitigated this. Our aim was not to identify empirical research questions per se, but rather to show where knowledge is lacking which could be filled by conducting empirical research studies or, for example, audits. We are not aware of previous efforts to explicitly link knowledge needs to policy or service delivery. Increasing knowledge through empirical research or practically oriented audit facilitates improved policy and service delivery, which are essential to achieving UHC for surgery.

The AfroSurg workshop was done with, and led by, surgical providers and stakeholders from six different SADC countries who used their extensive knowledge of the setting to come up with multiple knowledge needs to improve policy and service delivery in the region. This differs from the approach of much global surgery research in the region, which has been performed or directed by HICs actors with providers or researchers from LMICs participating, but rarely truly leading.21 22 To ensure that time spent in acquiring knowledge is invested well and in a way that will inform strategy, reform approaches must be based on and driven by context-sensitive understandings of surgical systems and embedded in local culture.11 23

To fulfil the vision of the AfroSurg network, the identified knowledge needs must now be addressed. The top two knowledge needs were mapping of available services, resources, and providers and quantifying the burden of surgical disease. Knowledge of supply (here referring to as available services, resources and providers) and demand (here referring to burden of disease) are essential for planning healthcare systems, but these are not well delineated enough in the region to allow this. While some investigators have aimed to ascertain surgical capacity at facilities24 and others have aimed to estimate disease burden,25 26 attendees at the meeting acknowledged that these high-level research findings are not granular enough to improve surgical care at the local level. Wide-scale use of tools at facilities can help to provide knowledge on readiness or capacity to provide surgical care. Examples include the WHO Service Availability and Readiness Assessment survey, Service Provision Assessment and the WHO Surgical Situational Analysis Tool.27–29 These should ideally be used at the national level to urgently assess supply. In addition, the pooling of data at the regional level could help develop a large-scale data warehouse to facilitate use of advanced data-science tools, enabling forecasting, benchmarking and actionable insights.

Ascertaining information on the burden of surgical disease is more challenging. In health systems where financial and access barriers are minimal, the prevalence of conditions treated at health facilities closely matches the prevalence in the population.30 However, when care is unaffordable or are when patients are unaware of available services,1 such as in many African countries,31 32 the health facility disease burden may underestimate population prevalence. Disease burden modelling can be inaccurate,33 and data collection from household surveys using validated tools may be needed to obtain accurate estimates of the burden of surgical conditions in SADC countries.

Estimating the number of trainees needed to deliver surgical services to match the burden of disease was the third highest priority. Without accurate data on the true disease burden, this will be a challenging knowledge need to fill. Knowledge of effective incentives to encourage practitioners to remain in their geographical areas and prevent ‘brain drain’ was our fifth priority and an issue that affects many African countries.34 Appraisal of initiatives that have been implemented elsewhere35 (ie, use of monetary incentives, binding contracts or training programmes that are delivered in key areas of need) can inform consensus on the best solutions to minimise ‘push’ and ‘pull’ factors for physician/surgeon movement in SADC countries.

Of the top five knowledge needs, four were policy-related, suggesting a dearth of much-needed information to develop evidenced-based effective surgical policies in the region. The only service delivery knowledge need prioritised among the top five at our meeting was to define service planning information. These results can also inform health system planning at the local and national levels. Importantly, all indicators that will be chosen cannot be difficult to collect. There was an acknowledgement during the workshop that although robust data collection is needed, indicators should be easy to obtain (eg, those collected already as part of routine clinical care) so as not to overwhelm existing scarce human resources.

Even though the need for improving access to safe, affordable surgical care has been previously articulated, there has been a lack of political priority to improve this.6 Shiffman and Smith has described four key components required to achieve political priority in global health: actor power (the strength of individuals and organisations concerned with the issue), ideas (the ways in which those involved with the issue understand and portray it), political contexts (the environments in which actors operate) and issue characteristics (features of the problem).7 AfroSurg has made progress to ensure that these four key components are met in our region. In our ToC, the need to collect more data to inform issue characteristics was clearly articulated, both for health systems and policy. Our group stopped short of defining indicators to understand the size of the problem and to monitor progress, given there is an ongoing global initiative to do this.36 37 Moreover, AfroSurg, which brings together actors from six different African countries, with support rather than leadership from HICs, aims to improve surgical care in the region with local actor power. Political contexts in the SADC region are already favourable with many countries having embarked on the NSOAP process. The ideas element of Shiffman and Smith’s framework states that a clear message on the problem is needed. AfroSurg and the ToC workshop were initial steps towards achieving strong internal framing around the needs to improve access to quality surgical care and practicalities needed to enable this change, emanating from regional key actors.

Next steps

AfroSurg is a small network, but the participants at the first meeting had a broad range of backgrounds related to surgical care from providers, policy makers and researchers. The network fulfils several of Shiffman and Smith’s requirements to improve political priority, but not all. In future meetings, we plan to identify where some knowledge needs have been filled, how we can prioritise data collection to fill remaining knowledge needs and discuss at what intervals to conduct audits to ascertain whether remaining knowledge needs have been filled. Many of the AfroSurg stakeholders are leaders in service delivery and policymaking in their respective countries and will communicate the findings from this workshop to their country’s surgical stakeholders. In addition, we plan to secure funding to meet annually and update the ToC after 5 years, after which many of the knowledge gaps should be filled and new evidence generated. We also recognise the need to invite more local interested stakeholders such as political actors, civil society organisations and service users to our meetings in order create a seamless line connecting the community to services and policy to ensure that we are working towards a shared agenda.