An official website of the United States government
NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.
Louw Q, editor. Collaborative capacity development to complement stroke rehabilitation in Africa [Internet]. Cape Town (ZA): AOSIS; 2020. doi: 10.4102/aosis.2020.BK85.10
Background: The identification, contextualisation and uptake of evidence-based stroke rehabilitation strategies are important to optimise the patient and healthcare system outcomes.
Aim: This chapter aims to report on the innovative methods used to produce the SA-cSRG 2017–2018 (2019) and summarises the evidence-based recommendations for implementation in the African context.
Methods: The methodology consisted of eight steps: ask clinically relevant questions; acquire evidence; appraise the evidence; extract the evidence; extracting data; grade the evidence level; endorsement and stakeholder feedback. The quality of the included component guidelines, which answered these questions, was scored using the Appraisal of Guidelines Research and Evaluation, Version II (AGREE II). The extracted recommendations were graded according to a standardised strength of the body of evidence (SoBE) method. The recommendations were then contextualised to the local context based on stakeholder input.
Findings: Seventy-eight composite recommendations were developed. Twenty-three (29%) recommendations were supported by strong evidence. Thirty-six recommendations were supported by moderate evidence (46%) and 19 recommendations were supported by low levels of evidence, no evidence or contradictory evidence.
Conclusion: This chapter outlines steps undertaken to contextualise recommendations to the African setting. These steps provide a blueprint for future guideline writers in LMIC settings to efficiently produce evidence-based guidance for other conditions that can be implemented despite local barriers to evidence uptake.
Evidence-based rehabilitation can reduce variability in care, improve accountability and promote quality care to people with stroke to optimise patient and health service outcomes (Jesus et al. 2017; Urimubenshi et al. 2017). In Africa, there is no published information (Chapter 1) to indicate that rehabilitation offered is informed by evidence or that any form of evidence-based guidance is available for stroke rehabilitation. In most African countries, access to acute and/or long-term stroke rehabilitation remains limited because of an insufficient number of rehabilitation workers, scarcity of MDTs, cost of healthcare, lack of access to care, travelling distances, cost of transport, health literacy and cultural beliefs about stroke amongst patients and caregivers (Basri, Naseen & Naz 2017; Dizon et al. 2016a; Ekeh 2017; Ned, Cloete & Mji 2017). The evidence-to-practice gap may be lengthier in LMICs because of contextual and policy barriers to implementation of evidence-based care. The implementation of CPGs has been advocated to narrow this gap; however, it is important that contextual factors are considered during CPG development and implementation to ensure the applicability and acceptability of CPG recommendations for the specific African healthcare setting (Dizon et al. 2017; Kastner et al. 2013). It is therefore imperative that any form of guidance developed in Africa is done in collaboration with all key stakeholders, including people with stroke, to strengthen the health system.
The healthcare system of South Africa is in a process of transitioning to implement the National Health Insurance (NHI) plan. The NHI white paper (2017) guides for cost-efficient, evidence-informed CPGs to improve health services in South Africa. This guidance should, therefore, be developed for prevalent, long-term conditions such as stroke, and should include the key element of rehabilitation as a cornerstone of stroke care. To our knowledge, no rehabilitation guidelines are included in any national evidence-based guidelines in South Africa (Wilkinson et al. 2018); and this may be the status in other African countries as well. This lack of local guidance for post-stroke rehabilitation may indicate that stroke care varies within the country; that rehabilitation may not be equitable across different contexts and that access to stroke rehabilitation is arguably worse in the poorest communities. This notion is confirmed in Chapter 1, where one of the main findings was that there is generally a lack of commonly recommended rehabilitation services in Africa. Congruently, the WHO (2015) estimated that in Southern African countries, only 26%–55% of people receive the rehabilitation they need; and that people with disabilities were two times more likely to access healthcare, or equipment; and that they are about three times more likely to be denied care (WHO 2015). The deficiencies in rehabilitation policy and guidelines should urgently be redressed to improve the situation and meet the needs of the growing number of people in need of post-stroke rehabilitation.
Over the last 20 years, some evidence-based stroke rehabilitation guidelines have been developed for HICs (Grimmer et al. 2018). Guidelines developed in HIC reflect HIC circumstances (and research conducted in HIC service delivery settings) and they may therefore not be readily implementable in LMIC settings because of differences in the healthcare systems, socio-cultural, societal and policy contexts (Ernstzen, Hillier & Louw 2019; Gonzales-Suarez et al. 2012). Additionally, the labour-intensive process of developing CPGs from ‘scratch’ (de novo development) may be challenging for LMICs because of limited capacity, time and resources. However, given that the existing international evidence-base is being regularly synthesised in HIC CPGs, this may undoubtedly lead to resource-wastage by ‘reinventing the wheel’ (McCaul et al. 2018). The international literature may also have limited applicability in local African contexts. Adapting or simply adopting CPGs based largely on HIC evidence may not be feasible for the African context (Dizon et al. 2016b). In the current milieu of the African healthcare context, it may be more efficient to use the existing good quality evidence-based guidelines from HICs and contextualise the guideline recommendations to ensure that they are applicable and acceptable in local contexts. During contextualisation, recommendations are sourced from existing CPGs but may be tailored to address the specific needs of the context. This process considers how each recommendation might be relevant to local circumstances, and if so, how it might be implemented effectively.
In this chapter, we report on contextualised recommendations, using innovative methodological approaches (Grimmer et al. 2019) from existing CPGs about stroke rehabilitation. This chapter outlines the innovative methods and focus taken to produce the SA-cSRG 2017–2018 (SA-cSRG 2019:8) and summarises the evidence-based recommendations that could be implemented in the African context.
The methodology that was developed by the guideline team consisted of eight steps. These steps are outlined below, and detailed methodology is published by Grimmer et al. (2018).
A stakeholder meeting was conducted to create a list of clinical questions applicable to stroke rehabilitation in the South African context. Therapists from different provinces in South Africa were contacted via email and asked to provide questions which they deemed necessary for their specific settings. A set of 38 questions regarding best practice of stroke rehabilitation in South Africa was then organised (or formulated) by the project team, based on the information obtained from the stakeholders. The questions are listed in the study by Grimmer et al. (2018).
The guideline team developed a search strategy (Grimmer et al. 2018) to identify relevant CPGs published from January 2010 to April 2017. The 7-year timeframe was set to identify the most recent CPGs that were available in full-text.
Systematic searches were conducted through www.google.com to identify potential CPGs. Specific searches were also conducted through international CPG clearing houses and CPG developers’ websites, including, but not limited to: National Guidelines Clearing House; Scottish Intercollegiate Guidelines Network (SIGN) (UK) (www.sign.ac.uk/); National Institute of Health and Care Excellence (NICE) (UK) (https://www.nice.org.uk/); NHMRC (Australia) (www.NHMRC.gov.au/); and New Zealand Guidelines Group (www.nzgg.org.nz/). These websites are listed in the SA-cSRG (2019:16).
The quality of the included component CPGs was scored using the AGREE II (2013; Dizon et al. 2016b). Potentially relevant components of CPGs were not excluded for poor quality, on the basis that any included CPG might provide answers to at least one SA-cSRG question. However, the quality of component CPGs was taken into consideration when determining the overall strength of the body of evidence (SoBE) for each composite recommendation. To classify CPG quality for the purpose of determining the overall SoBE for each composite recommendation, arbitrary total overall CPG quality score classifications were established by the methodology team as:
What constituted a “recommendation” for data extraction purposes was initially determined by the methodology team and subsequently confirmed with Brian Alper and his colleagues (who are working in the same topic area of evidence synthesis) (Alper et al. 2018; Hillier et al. 2011; Schünemann et al. 2014). This confirmation occurred at the Global Evidence Summit in Cape Town in September 2017 (Haeseler 2017). The importance of determining consensus on what constituted a “recommendation” was to reduce variability and improve efficiency in data extraction.
The inclusion criteria for recommendations were:
The exclusion criteria were:
The guideline team developed purpose-built data extraction sheets for each SA-cSRG question and recorded component CPG details, extracted recommendations and associated SoBE grading from each relevant CPG. These were CPGs that provided an answer (in part, or total) to each SA-cSRG question. Recommendations which met the inclusion criteria were extracted verbatim from the relevant included component CPGs, along with the associated SoBE grade (in whichever way it was reported).
The ways in which each included CPG reported gradings for its recommendations were collated. As has been reported by others when synthesising multiple CPG recommendations (Ernstzen et al. 2019; Gonzalez-Suarez et al. 2012; Hillier et al. 2011; Schünemann et al. 2014), CPGs often report SoBE gradings in different ways. Moreover, there is no one agreed approach to standardise SoBE grading descriptions. Thus, because multiple CPGs on stroke rehabilitation were included in the SA-cSRG, the guideline team developed an approach to standardise the component CPG SoBE gradings to assist in determining the overall SoBE discussions for composite recommendations. Thus, the different approaches in which SoBE gradings were reported in the component CPGs were extracted and aligned according to a standardised grading system for each of the extracted recommendation (Grimmer et al. 2018).
The guideline team developed a grading system for determining the SoBE for the composite recommendation, using an adapted version of the approach published by Alper et al. (2017) and Shiffman et al. (2012). However, additional criteria were used, consisting of the number of CPGs which provided recommendations for each SA-cSRG question, their methodological quality (high, moderate or poor) as determined from the overall AGREE II score (AGREE 2013; Dizon et al. 2006b) and where required, their currency (recency of publication). The grading decision-making system for the composite recommendation is reported in Grimmer et al. (2019). The detailed decision-making framework developed for each composite recommendation is available from the SA-cSRG (2019:50.
We used two concurrent processes to contextualise the recommendations. The stakeholder groups assigned an endorsement level (A, B or C) according to a decision-making framework (Grimmer et al. 2019). ‘A’ denoted that a recommendation can be adopted and implemented without change; ‘B’ denoted that the recommendation could be implemented with contextualisation processes, whilst ‘C’ meant that the recommendation was implementable with the addition of local evidence. Depending on the level of endorsement, the group nominated practice and context factors (including barriers and facilitators) that would influence the implementation of that specific recommendation within the South African healthcare system. We specifically focused on identification and reporting of service implications, as well as performance indicators, which can be used to monitor implementation of the recommendations supported by strong evidence which are relevant to all levels of care.
The guideline team produced 78 composite recommendations in total. The results are organised in three sections: the first section focuses on the recommendations that are supported by a high SoBE, the second section presents the service implications and performance indicators for the strong recommendations, and the third section presents contextual factors that were nominated by the stakeholder reference groups according to contextual factor categories. These were used to obtain information about local factors that could influence the effective implementation of recommendations.
Twenty-three (29%) of the 78 recommendations were supported by strong evidence. The strength of these recommendations is not likely to change with the subsequent addition of new research evidence. Therefore, these recommendations are deemed priority recommendations for immediate implementation in the rehabilitation journey of people living with stroke.
Table 10.1 illustrates the recommendations supporting strong evidence relevant to the health system, action for rehabilitation, information management, discharge planning and community integration. The levels of care for which the recommendation applies are also indicated in Table 10.1. Notably, many of the recommendations applied to more than one level of care.
Recommendations supported by a strong body of evidence (South African Contextualised Stroke Rehabilitation Guideline).
The South African reference group endorsement is inserted as superscripts in Table 10.1. It is positive that many of these recommendations were deemed feasible for local implementation (endorsed at A or B level) within the South African context. However, the two recommendations related to falls assessment and prevention were deemed not to be immediately implementable in local contexts. Similarly, the implementation of the recommendation related to community integration may be challenging and further local research or resources may be needed.
Figure 10.1 illustrates the recommendations supported by strong evidence (Table 10.1) which apply to all levels of care. This figure presents a simplified rehabilitation care strategy which clinicians, managers, policymakers, etc. can use to facilitate the seamless implementation of the minimum standards for the rehabilitation of people with stroke. The recommendations incorporate the organisation of the team, screening, referral to rehabilitation and rehabilitation and planning for ongoing care.
Recommendations supported by strong evidence that are relevant to all or multiple levels of care.
Table 10.2 illustrates potential service delivery implications for implementation when teams consider the implementation of the 10 recommendations supported by a strong level of evidence (Table 10.1; Figure 10.1). As part of the endorsement and contextualisation process, stakeholder reference groups provided performance measures for the implementation of recommendations. Many service implications should be considered when planning the implementation of the recommendations contained in the guideline. Also, rehabilitation teams and managers should select measures that will be feasible to monitor progress made with the uptake of the recommendations in practice (Table 10.2).
Service implications and performance measure of 10 selected recommendations supported by strong recommendations (South African Contextualised Stroke Rehabilitation Guideline) and Hebert et al. (2016).
Recommendations supported by limited, contradictory or no evidence.
Table 10.3 reports the 36 recommendations supported by moderate evidence (46% of all recommendations contained in the guideline). Although these recommendations are supported by a moderate level of evidence, some of these recommendations that are supported by a moderate level of evidence could lead to significant and costly adverse effects or death. In some instances, some of these recommendations could be considered as important as the recommendations for which there is already a strong body of evidence (Figure 10.1). For example, screening of swallowing capacity is critical to implement at all levels of care to prevent serious or fatal complications such as pneumonia. Therefore, rehabilitation teams should still consider these recommendations based on their local or institutional service needs, patient profile, capacity, resources, training, level of care, etc.
Barriers to implementation of selected recommendations and proposed strategies to overcome barriers (South African Contextualised Stroke Rehabilitation Guideline and Grimmer et al. 2019).
Table 10.4 illustrates the recommendations supported by low levels of evidence, no evidence or contradictory evidence (for which no clear judgement can be made regarding a recommendation). Although rehabilitation teams could consider the relevance of these recommendations, local applicability and importance of a specific context may be a primary reason for implementing these recommendations. It is important to note that in many instances, it is the lack of evidence (limited research) to support these recommendations, not evidence of no effectiveness. Thus, further research is crucial. In instances where there is a real need for guidance (but an absence of evidence), rehabilitation teams could consider constituting an expert advisory panel to provide interim guidance. However, regular evidence assessment will be necessary to ascertain if evidence has become available for the specific clinical question.
Recommendations supported by a moderate level of evidence and implications (South African Contextualised Stroke Rehabilitation Guideline).
Table 10.5 illustrates the framework used to identify barriers within the local context to implement a specific recommendation. This framework included elements related to the organisation, service delivery, communication and clinical care barriers into consideration. During expert reference group meetings, participating rehabilitation professionals nominated context-specific barriers and suggested strategies which could be applied to overcome those barriers within the South African context. Table 10.5 outlines example barriers and suggestions relevant to the recommendation for MDTs, stroke units, education of family members or community workers and involvement of the family in goal setting.
There is little guidance in the implementation literature about taking recommendations from HIC settings and making them locally relevant in an LMIC. This chapter provides new information on the processes of sourcing, evaluating and contextualising clinically based recommendations derived from multiple CPGs to produce a composite, evidence-based stroke rehabilitation guideline that can be implemented in South African contexts. The SA-cSRG is unique, in that it considered international recommendations through the lens of local contextual factors that may influence optimum stroke care and rehabilitation around South Africa.
To date, there is very little information about the implementation, monitoring and assessment of the implementation of evidence-based stroke rehabilitation services in Africa (Chapter 1). The SA-cCRG for stroke rehabilitation (2019) provides evidence-based guidance not only for clinicians but also for managers and policymakers, enabling them to understand, plan, evaluate and monitor rehabilitation services. Working within an environment of scarce resources and a transitioning healthcare system, it is crucial to optimise patient outcomes as well as healthcare system outcomes. It is envisaged that the SA-cSRG will play an important role in optimising rehabilitation services in South Africa, heighten the visibility of rehabilitation services for policymakers and healthcare funders, and contribute towards a better QoL in people with stroke, irrespective of geographical location, socio-economic background and ethnicity. This will assist in securing more funding for rehabilitation, as well as retaining and strengthening existing rehabilitation services.
The SA-cSRG articulates with the goals of the NHI scheme in South Africa, as it can enhance the efficient use of resources within the South African health system and to enhance patient outcomes. It also provides a way in which the South African rehabilitation workforce can demonstrate its accountability. Chapter 1 highlights that greater accountability of rehabilitation professionals is needed to promote health professions, enhance the integration of rehabilitation into health systems, and ensure better use of scarce healthcare funds (Timmermans 2005). The process of contextualising evidence-based recommendations for stroke rehabilitation, as outlined in this chapter, thus has the capacity to empower rehabilitation professionals to become more accountable for the services they provide (not just for stroke, but for other conditions requiring rehabilitation). The issues that need to be considered when providing contextually relevant care based on current best evidence will also provide South African policymakers, managers and healthcare professionals with a framework to evaluate healthcare and patient outcomes. Developing and implementing key rehabilitation standards could facilitate a better understanding of what rehabilitation services entail, and their potential to impact on person-centred and health system outcomes. As indicated in Table 10.3, it is essential to monitor how well recommendations are translated into practice, and how they impact on healthcare outcomes and resource utilisation. This could be an incentive for developing and collecting a national evidence-based minimum dataset on stroke rehabilitation.
This is the first research that we know of, in an LMIC, that has developed a step-by-step approach to progress evidence statements to practical implementation actions. The innovative methodological approach can now serve as a blueprint for other LMICs seeking to develop national rehabilitation guidelines. Possibly the most important learning from this work was the opportunity to focus on implementation, which was largely because the task of developing a de novo evidence base, had been removed. Other learnings were that:
The ongoing challenge in guidelines work internationally is the effective uptake of recommendations into everyday practice (Nilsen 2015). Evidence uptake is compounded in countries such as Africa, which have complex, multifaceted challenges to the consistent delivery of best care. Challenges include rehabilitation workforce scarcity, little accurate information on the prevalence of conditions requiring rehabilitation, unknown epidemiology of these conditions, variable patient literacy, complex healthcare beliefs and traditions, costs of accessing care, an overburdened rehabilitation workforce and dire socio-economic factors (such as poverty) that generally excludes people with a stroke from accessing services, even if rehabilitation services are available. The attention given to contextual factors, that may influence the effective implementation of CPG recommendations, such as the transitioning healthcare sector and features of LMIC healthcare system factors may play a key role in narrowing the evidence-to-implementation gap in African settings.
Having to navigate a many-paged guideline may be daunting for busy clinicians. This chapter therefore also presented a subset of key recommendations, focusing on those supported by a strong body of evidence. This approach was needed as the large number of recommendations contained in the SA-cSRG (2019:28–36) may also be perceived as a barrier to evidence implementation. Prioritising recommendations based on the strength of their evidence, and their ease of implementation, may initiate a stepwise implementation process which may be time and cost-effective, and hence acceptable in a transitional healthcare system.
Figure 10.1 in this chapter is thus useful as it provides a simplified infographic of selected key recommendations which can serve as a starting point when implementing recommendations. This infographic can be printed on a single page and/or poster and provides an easy-to-read summary of the evidence for healthcare professionals to implement into practice. It also offers a succinct summary for rehabilitation teams, managers and policymakers in other African countries who need to upscale stroke rehabilitation services. Whilst providing recommendations in an easy-to-read format may assist clinicians to get started with implementation, there must be global support for the consistent provision of evidence-based rehabilitation from national and provincial politicians, policy makers and bureaucrats, health facility managers, health insurers and patient advocacy groups. Engagement from the highest levels of government to healthcare workers should be supported by efficient communication strategies, sufficient resources and shared values. There is also an urgent need for the development of a minimum national dataset and performance indicators to assist in monitoring the quality of care, and the embedding quality assurance principles into rehabilitation.
There was a surprising number of recommendations that were able to be endorsed for immediate implementation. However, many others were not immediately implementable because of a variety of context-specific factors. One of the main factors was access to (or availability of) a rehabilitation team (human resources). To redress this requires intervention from NDoH and tertiary training institutions. Without a skilled rehabilitation workforce to implement the evidence-based recommendation contained in the SA-cCRG, ensuring equitable access to high-quality post-stroke rehabilitation in South Africa will remain a challenge. Another important contextual factor was the lack of efficient communication and information systems involving the different clinicians involved in stroke care, as well as family and/or caregivers. A national minimum dataset, and an innovative information system that captures and communicates data relevant to rehabilitation services, quality care and outcomes are urgently needed.
A side-benefit of the process of amalgamating and contextualising evidence-based recommendations from existing CPGs was the identification of gaps in knowledge. A better understanding of matters such as traditional medicine, self-management strategies that articulate with health literacy and health beliefs, and the use of telecommunication strategies is particularly important given the contextual realities of rehabilitation in African settings. A prioritised national staged research agenda to better understand effective rehabilitation in African settings is thus required.
The key limitation of this project was that there were unequal representation and involvement of the nine South African provinces. In some of the rural and remote areas of different provinces, access to email or even a computer was a challenge, and therefore it was challenging to contact clinicians and patients in these areas. Another limitation was that not all reference groups included facility managers and rehabilitation managers, which is key to the effective implementation of the evidence-based recommendations in local clinics. Future projects should consciously recruit a more comprehensive and representative group of stakeholders into the guideline reference group. Another important limitation was the lack of inclusion of an economic analysis of the cost of implementing the SA-cCRG. Although a cost estimate is not typically included in standard guideline development processes, it is a barrier for endorsement of guidelines by governing bodies in the African context because of constrained financial resources.
The recommendations presented in this chapter were contextualised for the South African context. The African continent presents many diverse settings and health contexts. Therefore, the recommendations presented in this chapter may not be generalisable to all African countries or region. We propose that the detailed contextualisation method described in this chapter is followed by African stroke rehabilitation teams to ensure that the recommendations are feasible for implementation in their local context. The methodological approaches can also be adopted and developed further by African guideline teams to develop evidence-based guidance for other conditions.
This chapter outlined the steps taken to consider the effective implementation of international evidence-based recommendations for stroke rehabilitation into African healthcare settings. It used existing evidence from international guidelines and distilled this into a composite set of recommendations that addressed clinical questions important to South African stakeholders. New processes were developed to determine a universal approach to describe the SoBE and to write one composite recommendation from multiple recommendations provided in multiple CPGs. Most importantly, however, this chapter outlines the steps undertaken to contextualise recommendations to African settings. These steps provide a blueprint for future guideline writers in LMIC settings to efficiently produce evidence-based guidance for other conditions that can be implemented despite local barriers to evidence uptake.
How to cite: Louw, Q., Dizon, J., Van Niekerk, S.-M., Ernstzen, D. & Grimmer, K., 2020, ‘Contextualised evidence-based rehabilitation recommendations to optimise function in African people with stroke’, in Q. Louw (ed.), Collaborative capacity development to complement stroke rehabilitation in Africa (Human Functioning, Technology and Health Series Volume 1), pp. 389–420, AOSIS, Cape Town. https://doi
This is an open access publication. Except where otherwise noted, this work is distributed under the terms of a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International license (CC BY-NC-SA 4.0), a copy of which is available at https://creativecommons.org/licenses/by-nc-sa/4.0/.
Your browsing activity is empty.
Activity recording is turned off.
See more...
