National stakeholders' perceptions of the processes that inform the development of national clinical practice guidelines for primary healthcare in South Africa

RESEARCH Open Access National stakeholders’perceptions of the processes that inform the development of national clinical practice guidelines for primary healthcare in South Africa Tamara Kredo 1† , Sara Cooper 1*† , Amber Abrams 1 , Karen Daniels 2,3 , Jimmy Volmink 1,4,5 and Salla Atkins 6,7 Abstract Background: There is increased international focus on improving the rigour of clinical practice guideline (CPG) development practices. However, few empirical studies on CPG development have been conducted in lowand middle-income countries. This paper explores national stakeholders’perceptions of processes informing CPG development for primary healthcare in South Africa, focusing on both their aspirations and views of what is actually occurring. Methods: A qualitative study design was employed including individual interviews with 37 South African primary care CPG development role-players. Participants represented various disciplines, sectors and provinces. The data were analysed through thematic analysis and an interpretivist conceptual framework. Results: Strongly reflecting current international standards, participants identified six ‘aspirational’processes that they thought should inform South African CPG development, as follows: (1) evidence; (2) stakeholder consultation; (3) transparency; (4) management of interests; (5) communication/co-ordination between CPG development groups; and (6) fit-for-context. While perceptions of a transition towards more robust processes was common, CPG development was seen to face ongoing challenges with regards to all six aspirational processes. Many challenges were attributed to inadequate financial and human resources, which were perceived to hinder capacity to undertake the necessary methodological work, respond to stakeholders’feedback, and document and share decision-making processes. Challenges were also linked to a complex web of politics, power and interests. The CPG development arena was described as saturated with personal and financial interests, groups competing for authority over specific territories and unequal power dynamics which favour those with the time, resources and authority to make contributions. These were all perceived to affect efforts for transparency, collaboration and inclusivity in CPG development. Conclusion: While there is strong commitment amongst national stakeholders to advance CPG development processes, a mix of values, politics, power and capacity constraints pose significant challenges. Contrasting perspectives regarding managing interests and how best to adapt to within-country contexts requires further exploration. Dedicated resources for CPG development, standardised systems for managing conflicting interests, and the development of a political environment that fosters collaboration and more equitable inclusion within and between CPG development groups are needed. These initiatives may enhance CPG quality and acceptability, with associated positive impact on patient care. Keywords: Clinical practice guideline, Primary healthcare, Qualitative study, Guideline development * Correspondence: [email protected] † Tamara Kredo and Sara Cooper contributed equally to this work. 1 Cochrane South Africa, South African Medical Research Council, Cape Town, South Africa Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Kredo et al. Health Research Policy and Systems (2018) 16:68 https://doi.org/10.1186/s12961-018-0348-3 Background Clinical practice guidelines (CPGs) have become a familiar tool in policy and clinical practice. CPGs have a range of purposes, intended to improve the efficiency and cost-effectiveness of health system utilisation and to decrease preventable mistakes [1]. They generally include statements of expected practice, benchmarks against which individuals may audit and potentially improve their practices, and guidance regarding undertaking given tasks [1]. CPGs were historically built mostly on expert opinion, which included variable (and often selective) reference to research evidence [2–4]. However, over the last decade there has been increased focus on improving the quality of CPGs and the methodological rigour of their development [5]. Globally, health system pressures are increasingly demanding that resources are effectively allocated and based on research evidence of ‘what works’[6]. Within this context, there is growing recognition that high quality, evidence-informed CPGs can serve as practical vehicles for meeting these demands and reducing the gap between evidence, policy and best practice [1,7]. This maturing CPG development culture is evidenced by the various recent attempts made by well-credentialed international collaborations to standardise and improve the credibility of CPG development practices [8]. Between 2011 and 2013, three sets of standards were independently proposed to assist CPG developers in addressing key issues of quality, as follows: the Institute of Medicine introduced 8 standards for guideline development [9], the Guidelines International Network produced 11 relatively similar standards [10], and McMaster University compiled a checklist of 18 topics and 146 items to guide developers [1]. Concurrently, two checklists (the AGREE II instrument [11] and iCAHE guideline quality checklist [12]) were developed, both providing tools to evaluate the process of CPG development and the quality of its reporting. While there are differences between these standards, they are unified in advocating for CPG development to be guided by transparently constructed and evidence-informed approaches that have a clear and applicable scope and are integrated with stakeholder consultation [8]. Despite a growing knowledge industry centred on CPG development, little is currently known about this topic in low-resource settings generally and in sub-Saharan Africa (SSA) in particular [13–15]. While numerous studies in high-income countries (HICs) have investigated the quality of CPGs and their development methodologies in a variety of settings [16], very few of these studies have been conducted in SSA. For example, a systematic review evaluating 42 guideline appraisal studies, including 626 guidelines published between 1988 and 2007, found only 6 guidelines across the studies were from Africa [17]. Nevertheless, the situation is beginning to change, with the topic receiving more empirical attention in SSA over the last 5 years. For example, recent quantitative reviews have evaluated the quality of CPGs for priority diseases in SSA generally [13] and in South Africa specifically [18]. Similarly, certain qualitative studies have provided analyses of the development processes of guidelines for eclampsia treatment and malaria control in 3 SSA countries [19, 20], as well as for maternal health [21], lay health workers [22] and primary care [23] in South Africa. This body of research in SSA has revealed many similar shortcomings in the quality of CPGs and their development as those identified in studies in HICs, including with regards to their methodological rigour, editorial independence and applicability to local practice [13,18]. However, it has also shed light on certain unique, context-specific challenges facing CPG development in the region. For example, complex political environments and interests, bureaucratic processes and budget struggles, a lack of locally relevant evidence, and limited skills have all been shown to hinder CPG development in SSA [19–23]. Taken together, the findings from this small body of research suggests that more knowledge is needed on the specific circumstances, processes and priorities underpinning CPG development in different SSA settings, and the factors that could improve their construction. This knowledge will help pave the way for better focused, locally tailored and effective interventions to improve CPG quality and development, and associated positive impact on healthcare practices and outcomes in the region. Against this backdrop, the aim of the current study was to explore national stakeholders’perceptions of current CPG development activity for primary healthcare (PHC) in South Africa. More specifically, it sought to investigate their perceptions of the processes that should inform national CPG development, and their sense of what is actually occurring. This study is a sub-study of a broader qualitative study that provided an overview of the current landscape of PHC CPG activity in the country [23]. This sub-study explores the issue of CPG development in more depth, with a focus on the perspectives of stakeholders operating at the national level. Both the larger study and this sub-study form part of the South African Guidelines Excellence (SAGE) project, which aims to understand and improve the development, adaptation, implementation and use of PHC CPGs in South Africa [24]. Research context CPGs have been part of South African clinical practice for many decades. Formal national CPG processes were put in place in the mid-90s in a bid to address historical Kredo et al. Health Research Policy and Systems (2018) 16:68 Page 2 of 14 inequity in health service delivery in the nine provinces. South Africa’s National Department of Health (NDoH) currently spearheads several primary care guideline programmes, including condition-specific guidelines (e.g. malaria, HIV, tuberculosis) and the Essential Medicines Programme, which develops comprehensive Standard Treatment Guidelines for rational prescription at all levels of care in an equitable, cost-effective manner. Additionally, academic departments and professional societies develop CPGs, addressing gaps in what is available from NDoH. Methods This study adopted a qualitative approach to understand the phenomena under investigation as experienced and perceived by the actors involved. The methods have been described in detail elsewhere [23], and thus a brief summary is provided here, together with a more detailed description of the analysis methods used in this paper. Participants and data collection The sample included 37 participants from a range of disciplines, stakeholder groups and provinces in the country (Table 1). Participants were approached in their capacity as contributing to national South African CPG development. Most were from academic contexts and providing expert input to guideline panels, many had been involved with this for more than 15 years. Besides the government players, the other participants usually held multiple roles, such that, in addition to CPG development, some were responsible for research, clinical teaching and senior management roles in academic or private sector contexts. Only one government member had a specific role in CPG implementation. Data collection comprised in-depth individual interviews, conducted together by two interviewers per interview. The interviews were based on a semi-structured interview guide, including open-ended questions and tailoring to the experiences of each specific interviewee. Key themes explored included players involved in CPG activity, CPG nomenclature and terminology, as well as the processes, contexts and values underpinning CPG development, adaption, contextualisation, implementation and use in South Africa. Data analysis The data for this paper were analysed through thematic analysis [25] and the lens of an interpretivist conceptual framework [26]. The larger SAGE project previously developed a list of conceptual components (‘open coding’), guided by the research objectives of the SAGE project. After immersion in the transcripts, codes related to ‘CPG development’within the SAGE coding framework were used to code the transcripts through multiple line-by-line readings and with the aid of Nvivo10, a software programme that aids in the management of qualitative data. Additional or revised codes were developed iteratively as determined by the data and added to the coding framework. Initial and revised codes were then collated into potential themes and produced into an overall ‘thematic map’to guide further analyses. Using the thematic map, the themes related to the processes of developing CPGs were identified, named and extracted. For each theme, the focus was on capturing how the participants were making sense of CPG development, and the values and concerns they attached to this issue. While we focused on identifying common themes, we also paid attention to the presence of potential diversities in participants’perspectives. Further analyses were then undertaken to check if these ‘fitted’in relation to the coded extracts, to refine the specifics of each theme and their relationships with each other, and to contextualise these themes within the other emerging topics in the dataset. Throughout the analysis process, we attempted to adhere to the methodological principle of reflexivity [27]. At regular intervals SC, TK and KD jointly discussed and further unpacked the emergent themes. Along with facilitating verification, validation and refinement of ideas, these discussions also provided opportunities for the researchers’interests and taken-for-granted assumptions to surface and subsequently be examined. Such reflexivity was facilitated further by the different roles the researchers occupied in relation to the research, an awareness of which provided for an illuminating interplay of emic–etic viewpoints. TK, KD and AA had jointly conducted most interviews, both had prior knowledge of many participants and TK is involved with CPG activities within academic and government settings. KD and AA are social scientists, and had been involved from project inception with planning, interviews and initial analysis. KD has been involved with health policy Table 1 Description of stakeholders sampled (n= 37) [23] Background discipline Medicine (n= 19), pharmacy (n= 5), nursing (n= 4), allied health (n= 3), dentistry (n= 1), nutrition (n= 2), nonclinical managers (n=3) Sectors and stakeholder groups National (n= 10) and Provincial Department of Health (n= 2), Professional Societies (n= 6), Private sector (pharmaceutical n= 1 and medical schemes n= 2), academia (n= 14), non-governmental organisations (n=2) Provinces represented Eastern Cape (n= 1), Gauteng (n= 16), Kwazulu-Natal (n= 3), Western Cape (n= 17) Kredo et al. Health Research Policy and Systems (2018) 16:68 Page 3 of 14 analysis in another project. AA straddles social science and quantitative research in her role at the Cochrane Centre. SC became involved in the research only at the analysis stage and had little prior knowledge of the local CPG landscape. SC’s a priori unfamiliarity with the interview content enabled the data to be explored openly and with a ‘fresh’perspective. Moreover, the interaction between researchers with both ‘distance’and ‘closeness’ allowed for previous understandings to be opened-up and questioned, and for our own positioning and associated shaping of the research process and outcomes to be critically reflected upon. Results The analysis revealed six ‘aspirational’processes that participants perceived to be most important when developing CPGs for PHC in South Africa, namely (1) evidence, (2) stakeholder consultation, (3) transparency, (4) management of interests, (5) communication/co-ordination between different CPG development groups, and (6) fit-for-context. We describe each process separately, unpacking both participants’aspirations and their views of the extent to which each process is, in reality, underpinning the development of CPGs. To preserve anonymity, certain phrases have been removed from quotations and replaced with alternative text in square brackets. Evidence The concept of an ‘evidence-based approach’featured prominently in participants’narratives. An overwhelming majority of participants, across stakeholder groups, strongly emphasised that CPGs should be driven by scientific evidence on the effectiveness, safety and cost-efficiency of a clinical process or treatment. The notion that CPGs “should be evidence-informed”(INT33), “scientifically rigorous”(INT14) and “guided by the current evidence-base”(INT25) was widespread, or as one government participant succinctly stated: “The department stand is that whatever policies or guidelines we develop are evidence based”(INT22). A common sentiment, particularly amongst academic and government participants, was that the development processes of many national CPGs have evolved over time, increasingly being informed by a more robust evidence-informed approach. Many CPGs were described as “coming a long way”(INT06) or having “over time, become more and more evidence based”(INT4). Whilst perceptions of progress were common, the need for further improvement was also communicated. More specifically, many participants expressed reservations about certain CPGs in the country, and the extent to which they are being guided by an evidence-based approach. Various reasons were provided for the inadequacies participants saw with regards to the certain CPGs’use of evidence. Here, a lack of dedicated time and funding, as well a scarcity of skills for quality CPG development emerged as overriding themes. The interviews were saturated with accounts of how those involved with CPG development are often doing it voluntarily and afterhours, and there is thus limited capacity to undertake the necessary methodological work: “People are stretched…we all have full time jobs and we’re doing it, not for money, not for kudos…so you can’t expect the kind of rigor that you’d like to see” (INT04, Academic) Along with limited time and funding, many participants also highlighted how there is a dearth of skills in the country for synthesising and incorporating evidence in CPGs. Many spoke about the “lack of competent people who are able to do this kind of work at a national level”(INT03) and that “the distribution of people with skills in evidence-based medicine…is quite a problem in the country”(INT15). The shortages of skills in evidence-based medicine within the NDoH, in particular, emerged as a key issue amongst government participants. Many attributed this deficiency to what they perceived as inadequate, or even non-existent, in-house training. As one participant put it succinctly: “We have also not been trained on the processes that need to be followed” (INT37, NDoH) Similarly, another government official replied, when asked whether he received training in evidence-based medicine: “No training, no training! You learn on the job…when you see courses being offered…in most cases it’s out of your own pocket because it’soutsideoftheHR development planning process” (INT22, NDoH) Stakeholder consultation There was considerable agreement amongst participants and across stakeholder groups that widespread stakeholder consultation also needs to form an essential part of CPG development. Many participants spoke at length about why this is critical. Along with serving as an Kredo et al. Health Research Policy and Systems (2018) 16:68 Page 4 of 14 important peer-review mechanism, widespread consultation was seen as essential for facilitating ‘buy-in’: “I think a key for any guideline is inclusivity. Because if you want people to embrace your guideline, it’s much easier if they were part of it than if you just thrust it upon them” (IN19, Professional society member) Several participants felt that it is particularly important to include the end-users of CPGs in consultation processes to ensure that guidelines are “practical…at implementation level”(IN32), are “acceptable to those implementing them”(INT22) and “so nurses understand what you want to say”(INT05). Many participants spoke about their own CPG development processes as involving stakeholder consultation, one which they perceived to be fairly wide and extensive. Participants spoke about “our strategy of wide consultation”(IN22), that “we are fairly meticulous about circulating our guidelines to all organisations concerned” (INT17) and that, ultimately, “Anybody who is on the ground, who feel they have something to contribute here, they will”(INT16). While there was a tendency amongst participants to describe their own consultation processes in relatively positive terms, a more complex picture of stakeholder engagement in CPG development also surfaced. This emerged most prominently when participants talked about their experiences with the development processes of other CPG development groups. This also materialised in certain participants’narratives of their own consultation processes, where certain reservations were revealed. Many participants spoke about various other CPG development groups as comprising “a very non-consultative process”(INT18) or as being “an authoritative entity, who never…considers clinicians’input”(INT17). Others described the inherent unresponsiveness of certain CPG groups, and their inadequacies in responding to people’s input. As one participant said: “The sense that we’ve had with all the people we’ve engaged with is that you will send a lot of feedback to the [particular CPG development group] but you’ll get no formal response to any of your feedback” (INT03, Academic) This failure to provide adequate feedback was recognised by certain participants with regards to their own consultation processes. Many explained how there actually is a very rigorous process for considering and incorporating stakeholder feedback, and yet due to capacity constraints, they are limited in their ability to adequately respond. As one participant indicated, acknowledging that this is a problem: “People will give a comment, but when the book is published they see that their comment hasn’t been incorporated…it’s not that the committee didn’t consider the comment…but because of capacity constraints, we can’t respond to each and every comment…but we’ve got to ask ourselves: how do we make it more publicly available that we have looked at your comment…without responding to each and every person on each and every point?” (INT16, NDoH) Certain participants conveyed other reservations with regards to their own consultation processes. Some questioned the level of inclusivity of their engagement processes, highlighting the problems they have engaging with particular groups. For example, many professional society participants alluded to the struggles they encounter consulting with government: “I think the process should be more inclusive…that is certainly the weakness of our current situation. But engaging with government is an extremely difficult process…there are very, serious barriers of communication with government” (INT18, Professional society member) Many other professional society participants shared this participant’s view, providing similar accounts of how “communication with the DoH has been shocking” (INT19), and how “there should be an easier way for us to engage government”(INT14). Certain government participants alluded to the difficulties they experience around engaging with various groups. Some felt that particular provinces, other than the Western Cape and Gauteng, are hard to engage with and ultimately remain weak in their participation: “I think there are weaknesses within the consultations at provincial level…in a province like Western Cape, it is done widely, but in other provinces, not really that much” (INT22, NDoH) Other government officials felt that the end-users of CPGs are another specific group that they have found Kredo et al. Health Research Policy and Systems (2018) 16:68 Page 5 of 14 hard to reach and are unsatisfactorily consulted, as reflected by this participant’s comment: “Idon’t think we have found a better…mechanism of really engaging the people who are…atthefrontlineof implementation…I think a lot of programmes struggle with what will be the best way of engaging the end users” (INT36, NDoH) Those participants who expressed uncertainties about their own stakeholder consultations suggested various aspects of the engagement process that might limit its inclusivity. For example, some indicated that the process tends to favour those who have the capacity to provide written feedback and the ability to use the internet, as depicted in the following two statements: “So [a particular CPG development group] have… collated a whole lot of emails and they send it out and also make it available on the website… butitjustdependssometimesthough,likenurses in rural areas do not have the capacity to connect online” (INT33, Academic) “The same usual suspects will give their comments because that’s their comfortable way of engaging…but there are other people who engage differently…maybe they don’t want to write something so they may need a different strategy” (INT16, NDoH) Relatedly, other participants suggested that the time given to stakeholders to provide feedback is insufficient, and may be an additional barrier to more widespread and inclusive consultation: “The consultation process doesn’t always look valid because there wasn’t given time to comment” (INT14, Professional society member) Transparency It was widely suggested that CPG development also needs to be guided by a clear, transparent process so people can understand decision-making processes. Many participants spoke about the fact that “people need to see the validity of the process”(INT01) and that “fair transparency is a critical component of guideline development”(INT23), or as one participant stated: “I think the most important thing is that…we have to have a very transparent, clear process…a level of transparency that…someone can understand why decision were made” (INT08, Professional society member) The dominant view expressed by participants in all the stakeholder groups was that CPG development in the country tends to lack sufficient transparency. Descriptions were commonplace about how CPG construction processes are “very untransparent…completely opaque to everybody”(INT14), “a complete mystery”(INT3), or as described by one participant: “I have issues with the [particular CPG development group] not being transparent…what we need is to make that process visible, because it’s actually, they have terms of reference, they have criteria, they go through a very evidence-based process” (INT30, Academic) Like this participant, many other participants conveyed a sense of trust in the rigour of various CPG development processes, and yet perceived there to be a significant gap in the documentation and sharing of the logic behind the decisions. Many suggested that there is a need for greater communication about exactly how the process unfolded, so that people can better appreciate the credibility of CPGs: “I think the communication…there’s a lot of misconceptions…but just talking to people and telling them, okay, this is how we do it, then they get the insights…that there’s a rigorous process” (INT16, NDoH) Not many participants provided reasons for why the process is not as transparent as it should be, despite probing by interviewers. The few that did reflect on this issue suggested, once again, that limited time and funding was the cause, with stakeholders lacking the capacity to adequately record and elucidate decision-making processes: “So there is an awful lot of work going on…but not a lot of capacity to engage in a very clearly documented and open process…but we do need to consider what the best means is of documenting evidence and then sharing that evidence in order to get more buy-in” (INT21, Academic) Kredo et al. Health Research Policy and Systems (2018) 16:68 Page 6 of 14 Given that few participants provided details on why transparency is an issue, it is unclear whether this view is shared by others. Management of interests A few participants stated explicitly that the declaration and management of interests need to form a key aspect of CPG development, so that decision-making processes are not influenced by inappropriate forces: “Your governance has to be clear…people have to know…what interest they need to declare, what information they need to keep confidential…and the reason…is because we don’t want undue influence on our decision-making processes” (INT16, NDoH) Although not overtly stated in most cases, many participants clearly conveyed the view that vested interests need to be considered and managed when producing CPGs. The interviews were replete with descriptions of people’s conflicts of interests and the role these are playing in CPG development. Participants spoke about these interests as being both personal/intellectual and financial. In terms of the former, many described in detail the “personal agendas”(INT5), “little hobby horses”(INT6) and “vested interests”(INT14) people involved with CPG development across the board have: “There’s a whole lot of politics…we have all these competing interests…even us sitting at the university… we have these different groups that have their own agenda…to defend their turf” (INT25, Academic) More specifically, many participants described how individuals developing CPGs may be involved with specific programmes or research projects, and often push for guidelines to incorporate these. As articulated by this participant: “There’s a lot of individuals or research institutions pushing their own agendas…like those on drug development, clinical trials…the expectation is that you would change your policy based on that…and it creates problems for us in terms of determining what should be in the guidelines…we’re put under pressure” (INT22, NDoH) Numerous participants provided analogous accounts, similarly highlighting the problems they have around managing personal interests. Certain participants also expressed uncertainty about how these agendas can and should be managed, as communicated by this participant: “We talked about this…how we probably need to also disclose grant conflicts of interest because if you’re sitting on a study that is, you know, if you change [the policy], your study is not going to continue, right?… But, then everybody has some sort of bias, so I don’t know what the ultimate answer is, like how do you make this so completely transparent” (INT08, Professional society member) In addition to interests of a personal or intellectual nature, many participants were also particularly concerned about financial interests and the fact that a “massive amount”of guidelines are “driven by industry”(INT20). As aptly revealed by this remark: “My colleagues…they don’t see the harm if industry comes and does this. You know, it’s so insidious… they’re doing subliminal advertising, and people don’t get how that can influence how you make a decision” (INT16, NDoH) When talking about the pervasiveness of financial interests, many participants were particularly worried about professional society groups in this regard. Comments about such groups having “lots of apparent influence of industry”(INT21) and “being influenced tremendously by industry”(INT18) were ubiquitous, along with descriptions of how professional society CPGs are “essentially drug company driven”(INT19). Many participants also questioned the sufficiency of the extent to which the financial interests operating amongst professional societies are being managed: “In many instances, if not most instances, there is no process to deal with potential conflicts of interest” (INT15, NDoH) Certain professional society participants themselves expressed analogous concerns, with some conveying similar apprehensions about the presence and inadequate management of financial interests within their own societies. As one member acknowledged: “It depends on the integrity of the individual…you know, I think it’s very glib now, the declaration of conflict of interest. It goes up in the first slide and you Kredo et al. Health Research Policy and Systems (2018) 16:68 Page 7 of 14 don’t even see it. Here’s my title and here’s my conflict of interest. It’s very glib” (INT18, Professional society member) In a similar manner, when asked directly about his experiences of conflict of interests within his own society, another participant explained: “I’m trying to remember…whether we actually had to declare a conflict of interest. I don’t think we did” (INT19, Professional society member) This participant went on to articulate why conflicts were not declared within his society, outlining some of the difficulties in this regard. According to him, it would be “too numerous”as “everyone is going to have to have received funding from someone for something”. He explained that “all those would all have to be declared …which would take the first four pages of the guideline”.He indicated further that “if you try and put too many rules in place”, you will ultimately createahindrancetoCPGdevelopment,orasheputit: “It then trumps people’s desire to actually do the guidelines”. Thus, while most participants felt that management of interests is a key activity for CPG development, certain stakeholders had a different perspective. As with the participant above, some questioned the value as well as the practicality of declaring and managing conflicts of interests. Communication and co-ordination between CPG development groups Various participants stressed the importance of communication and co-ordination between different CPG development groups, suggesting that “those writing guidelines must speak to each other”(INT37) and that “there’s got to be absolute linkage between programme guidelines” (INT27). This was identified as essential for ensuring “harmony between guidelines”(INT03) and that “we don’t give confusing messages to practitioners”(INT15). Many participants felt that it is particularly important for CPG developers to communicate with the Essential Drug List (EDL) committee, as exemplified by this comment: “If I was redesigning the system, I would have a… process for guideline development that has a clearing house effect that goes through the EDL, who then issues it” (INT14, Professional society member) There was much consensus amongst the participants that, in reality, communication and coordination between different CPG development groups is noticeably absent. The general picture that emerged was one of fragmentation, whereby a diverse range of groups are developing CPGs relatively independently of each other. Many described this disconnect as occurring between the private and public sectors: “Private, they do their own thing, only…where they don’t have a choice…or where they absolutely don’t know, only then do they then refer to the other guidelines” (INT06, NDoH) Other participants spoke about divisions within the NDoH and the lack of communication across government departments. CPG development processes within the NDoH were referred to as “siloed in a way that there’s not really good communication”(INT08), “disjointed pockets of activities”(INT15), or as one participant proposed: “There is a two-parallel process from the department of health, and the one side is the formal process and on the other side you’ve got stroke management, malaria management, HIV management…so all are little silos inside other silos” (INT20, NDoH) At the same time, many participants felt that some, but not all, CPG development groups are communicating with the EDL, or as two participants put it “communication is stronger with some programmes than with others”(INT37) and “often programmes haven’t checked the EDL”(INT20). This view was shared by a member of the EDL committee who, when asked whether CPGs are being circulated through the EDL, responded: “It doesn’t happen with all national departments. So [particular government department], yes, but I’m still struggling to get [particular government department] to send their guidelines to us for peer review” (INT17, Academic) This lack of communication between CPG developers was perceived to result in the replication of guidelines and a duplication of work, or in the words of one participant: “discrete pockets of people reviewing the same data”(INT11). It was thought to also give rise to contradictions between CPGs, with “a whole host of conflicting Kredo et al. Health Research Policy and Systems (2018) 16:68 Page 8 of 14 recommendations across the different guidelines” (INT03) and guidelines that “don’t fit together…to make a coherent whole”(INT02). Some participants attributed this situation to a matter of timing, suggesting that CPG inconsistencies are related to the fact that different CPGs are developed and updated “according to different schedules”(INT37) that are “not always in sync”(INT15). However, other participants conceived the problem to be of a more political nature. That is, it was suggested that the difficulties stem from the complex relations of power and control that exist within the CPG development arena in the country, as one participant proposed: “There are lots of interest groups competing for control of this, and that’s why partly it hasn’t been cohesive… The department of health, because they are in the HIV field, I think wants to keep some of that to themselves. The TB people wanted the TB to themselves…there’s just lots of competing interests around these things.” (INT14, Professional society member) Fit-for-context A final issue that featured prominently in the interviews was the need for CPGs to be contextually relevant, and thus the necessity that CPG developers think about “what is suitable for our context”(INT01) and “is this relevant to our situation”(INT37), or as specifically stated: “We feel guidelines need to be relevant to South Africa, I mean, our situations are different and our cost constraints are different, and it has to be relevant” (INT18, Professional society member) When describing CPG development processes, there was considerable agreement amongst participant groups that most CPGs in the country draw heavily on international guidelines and what is being recommended globally, particularly by WHO: “In most cases guidelines are guided by the WHO recommendations” (INT22, NDoH) Although there was widespread consensus that CPGs in South Africa are usually based on what is being done and advocated for internationally, there were divergent views amongst the participants about the use of this approach. Some were critical of this tendency, suggesting that we should not be relying on other sources to do the methodological work of CPG development: “I would say there’s an over reliance on other guidelines without looking at the primary evidence” (INT37, NDoH) Other participants expressed frustration with the dominance of global discourse, and the pressures they feel to conform to these. This was aptly conveyed by one participant who lamented: “WHO is a really sore point with a lot of our experts because they’re writing policy for Africa…Imeanwe do not lack technical expertise…and we have the evidence…I think our experts should be the ones driving the decisions” (INT14, Professional society member) In contrast to these perspectives, other participants supported the widespread use of international guidelines when developing local CPGs. Here, the common view was that it is unwarranted to repeat the work already done by other, well-respected organisations: “It’s not necessarily developing all the guidelines from scratch…because if it’s there, why reinvent the wheel” (INT25, Academic) Many participants shared this view, yet suggested further that international CPGs should be used but a process of contextualisation should ensure the “critical appraisal of international guideline”and “local adaptation”of these so they suit out local circumstances (INT37). As two professional society participants explained: “What I have been pushing for a lot is to say, take the WHO guidelines…and then adapt them to South Africa” (INT14, Professional society member) “What we want to do, [which] we haven’t always historically done…is we need to take WHO as the starting point, the baseline and then adapt from there” (INT08, Professional society member) As suggested by these two participants, and sharing the views of other participants, the process of adapting Kredo et al. Health Research Policy and Systems (2018) 16:68 Page 9 of 14