Rethinking Barriers and Enablers in Qualitative Health Research: Limitations,Alternatives,and Enhancements

Barriers and Enablers Can Lead to Shallow Research Findings

Barriers and enablers studies tend to generate lists of ‘good things’ and ‘bad things’, often with a lack of in-depth analysis and little explanation of precisely what those things are (Checkland et al., 2007). When exploring and synthesising research data, it is necessary to ask what class of phenomenon they relate to: individual, interpersonal, cultural, organisational, or structural? These details have huge implications for change because they point to different leverage points and different intervention strategies. For example, we recall supervising a thematic analysis that identified a list of enablers to high-quality patient education which were, in fact, all adaptive strategies developed by health professionals in challenging circumstances. Research findings that recognise this and provide recommendations which build on frontline innovation and support the development of champions are likely to provide more applicable and strategic direction than a generalised conclusion that we should strengthen enablers (get better at doing the ‘good things’) and tackle barriers (reduce the ‘bad things’).

A key strength of qualitative research methods is that they are inherently flexible, exploratory, and responsive to the phenomena they seek to understand (Renjith et al., 2021). The use of inductive and abductive analysis allows researchers to learn from the data, both noting and responding to new concepts. However, the B&Es approach is often used as a static deductive structure that data must fit into. There is nothing inherently wrong with using deductive structures to guide qualitative research (as we illustrate below) providing they serve a clear purpose and, in most cases, are ‘held lightly’, allowing them to be critiqued in response to emergent concepts in the data. But using a simple B&Es framework may result in reductive analysis which does little more than code data as a ‘barrier’ or ‘enabler’ and avoids the ‘interpretive leap’ (seeking meaning within the data) that high quality qualitative research often demands (Braun & Clarke, 2023; Järvinen & Mik-Meyer, 2020).

Pope and Mays (2009) stress the need to push qualitative analysis beyond superficial descriptions towards the work of interpretation in which deeper meanings are surfaced:

Everyone thinks they can do qualitative research now, and I’m not always convinced we do justice to what we were trying to achieve when I started doing health research …. those papers that report thematic analyses. The kind that report six themes and don’t explain what the relationship is between the themes and so don’t really go anywhere in terms of trying to explain the data. Sometimes the problem is the word limits set by journals, but sometimes it’s just superficial work …. A real strength of qualitative research is induction – interpreting the data. That’s where you find the unexpected. I sometimes worry that we don’t push qualitative research far enough. I’d like it to be less descriptive and for us to try harder to explain things …. Checkland and colleagues highlighted this in their study of the implementation of clinical best practice guidelines in general practices. The GPs they interviewed listed commonsense implementation barriers such as lack of time. However their accounts could not be understood literally; instead, they related to the GPs’ underlying beliefs about how work should be allocated in a general practice and what it meant to be a GP. The things that stopped GPs implementing guidelines actually had little to do with information management and shortage of time, the main reasons they gave to the interviewer. (Pope & Mays, 2009)

Some studies treat the categories which dominate the data in B&Es studies as more important than infrequently coded concepts (Graham-Rowe et al., 2018). This implies (erroneously) that the frequency with which a concept is coded reveals something significant about its magnitude, regardless of the different classes of phenomena involved and their relations to one another. Codes in qualitative data analysis act as containers for concepts (Braun & Clarke, 2019). The higher the level or the more abstract the code, the more the data is likely to be attached to it. B&Es are very high level. This offers little guidance for priority-setting in relation to policy or practice change, or further research. As Bach-Mortensen and Verboom (2020) argue:

… simply describing the most frequently reported factors may produce overly simple representations of the underlying issue being investigated. In other words, identifying barriers and facilitators to health-related phenomena will not, in isolation, provide insights on how this outcome can be improved through intervention.

The categorisation of a barrier or enabler as products of the research process is seldom reflected on in B&E studies, despite the fact that a recent review of systematic reviews which used the B&Es approach found a tendency towards identifying more salient, common, uncontroversial, and easily communicated factors, as well as those in which primary study authors had an a priori interest (Bach-Mortensen & Verboom, 2020). This suggests that data collection and analysis approaches which frame the research topic in terms of B&Es may be likely to elicit safe, ‘top of mind’ responses that potentially neglect messier and more hidden aspects of human experience. Further, as interview and focus group studies framed using B&Es become more ubiquitous, they are more likely to produce familiar findings. Do we really need more studies that tell us that information alone is seldom a catalyst for change, that health professionals struggle to perform everything that is asked of them optimally due to lack of time, or that people are more inclined to exercise if it is enjoyable?

Barriers and Enablers Can Bypass Critical Reflexivity

There is a tendency for the B&Es approach to bypass critical thinking about different ways of seeing the topic of interest. Taking the example of process evaluation in intervention research, the B&Es approach asks ‘What is getting in the way (of this intervention or ideal behaviour)?’ and ‘What is assisting it?’ (Checkland et al., 2007). These may be valid questions, but they are far from the only questions we should be asking if we want to understand complex socially embedded phenomena. We need to think critically about target behaviours and the interventions designed to generate them, including how fit-for-purpose and desirable they are within different settings; however, as Wellstead argues, “Categorizing any factor or process as a ‘barrier’ reduces complex and highly dynamic decision-making into simplified, static, and metaphorical statements about why current outcomes are ‘incorrect’” (Wellstead et al., 2018).

Consequently, questions such as ‘Why does the problem happen this way?’, ‘What exceptions are there and what causes them?’, ‘What other perspectives are there?’, ‘What do stakeholders gain and/or lose with this intervention?’, ‘Does the intervention address this community’s priority concerns?’, ‘How socially credible is it?’ ‘Are there any unintended consequences/potential harms from this intervention?’, and ‘How can we generate broader/deeper understanding?’ may be equally valid, depending on the research topic, aims, and context. Crucially, qualitative researchers need to be responsive to the phenomena they seek to understand. This often means trying out different ways of seeing and thinking about phenomena as the research progresses, and new learning deepens our understanding and challenges previous assumptions. So perhaps a better generic starting point is to ask, ‘What is going on here? What matters to people?’ rather than ‘What are the barriers and enablers to X?’.

Barriers and Enablers Are Untheorised (But May Not Appear So)

Understanding the complex socially embedded phenomena that health research seeks to tackle often demands some engagement with theory (Lynch et al., 2018; Meyer & Ward, 2014; Willis et al., 2007). Theory may be held lightly as sensitising ideas or used as a conceptual guide that structures the research. Theory can provide a taxonomy for the description of phenomena that allows for systematic explanation, new hypotheses, or recommendations that can then be further tested (May & Finch, 2009). Alternatively, theory may be developed within the research such as with grounded theory and theory-driven methods like realist evaluation. Well-theorised qualitative research and evaluation has greater transferability, positions the study better within what is already known, and allows others to build on it more effectively than a-theoretical studies that produce a list of findings (Maxwell & Chmiel, 2014).

Barriers and enablers can act as a false friend in this regard, appearing to offer a ‘common-sense’ conceptual framework that, in reality, may be little more than a convenient heuristic or pathway metaphor comprising obstacles and aids, rather like a game of snakes and ladders. The B&Es approach provides an illusion of theory-informed qualitative analysis. If it can claim any theoretical basis, arguably, it would be that the approach is loosely underpinned by a linear-rationalist behavioural paradigm which posits that people will enact a desired behaviour (e.g., implement a clinical guideline, collaborate more effectively, attend a health service, exercise more, or use a new technology in their work) if obstacles are minimised or removed and/or facilitators are enhanced or introduced. This assumes that the target behaviour and the intervention used to encourage it are both appropriate and desirable, irrespective of individual and contextual variations. This stance tips towards a positivist, reductive world view which is largely incompatible with qualitative approaches that seek to shed light on messy intersubjective realities.

The B&Es approach is also unable to provide an objective, a-theoretical methodology. No human knowledge is developed in a theoretical vacuum: we are all products of our culture and perceive the world through the lens of social and historical contexts (Korstjens & Moser, 2017). Theories are always present in the knowledge-making of both qualitative and quantitative research, even when they are not acknowledged (Bradbury-Jones et al., 2014). Research questions, data collection instruments, and analytical methods are inevitably informed by researchers’ ideas about how the world works (ontology) and what we can know about it (epistemology). As Carter and Little (2007) point out:

A reflexive researcher actively adopts a theory of knowledge. A less reflexive researcher implicitly adopts a theory of knowledge, as it is impossible to engage in knowledge creation without at least tacit assumptions about what knowledge is and how it is constructed.

For example, when we (and our research participants) explain phenomena in terms of good/bad or what works/doesn’t work, it is important to consider whether these are ontologically ‘real’ phenomena at work in policy, practice, and everyday life settings, or if we are using ‘barriers’ and ‘enablers’ as a metaphor or heuristic that helps us cope with complex underlying realities (or something else altogether) (Bach-Mortensen & Verboom, 2020).

The Barriers and Enablers Approach Struggles to Engage With Context

A key strength of qualitative research is its ability to explore questions of how, why, and under what circumstances phenomena arise (Liamputtong, 2020). This involves considering how themes and contextual features relate to one another: are there associations, dependencies, nested relationships, and causalities? Is a barrier always a barrier – in all contexts, for all people, in all circumstances? What mechanisms are at play when factors become barriers or enablers, and how are they mediated by context? And how does a barrier or enabler manifest – what does obstruction or facilitation actually entail? (Bach-Mortensen & Verboom, 2020). The B&Es approach has a tendency to present findings without the contextual, historical, and process-focused specificity that can help to answer these questions.

This is a particular concern in implementation research which seeks to identify and enhance factors that contribute to implementation success. Understanding local interactions between implementation strategies, people, and context is central to this endeavour. However, the B&Es conceptualisation of implementation suggests a linear, unidirectional process of ‘transfer’ in which contextual features are ‘sources of obduracy and interference’ that impede smooth implementation (May et al., 2016). Yet, context is not something that happens around social phenomena such as an intervention; it is intrinsic to it. As Greenhalgh et al. (2004) note:

Context and “confounders” lie at the very heart of the diffusion, dissemination, and implementation of complex innovations. They are not extraneous to the object of study; they are an integral part of it. The multiple (and often unpredictable) interactions that arise in particular contexts and settings are precisely what determine the success or failure of a dissemination initiative.

Contexts, and the interventions that interact with them, are dynamic. Contextual factors such as ‘social influence’ or ‘organisational culture’ which constitute normal conditions of practice may appear as barriers in some places or at some times but function as facilitators in others (May et al., 2016). During COVID-19 pandemic lockdowns, for example, rapid changes to social and policy contexts meant that community-based prevention programs were forced to adapt to extraordinary circumstances, shifting goals as well as program activities. New modes of program delivery necessitated system-level adaptations such as new partnerships and funding agreements which, in turn, affected further program design and delivery (Loblay et al., 2022). Thus, just as interventions are often adapted in local contexts, interventions may reshape contexts in unplanned and unexpected ways through their implementation (Rhodes & Lancaster, 2019). In organisational research, history and unfolding change within the organisation are vitally important for understanding social phenomena and designing and evaluating intervention and implementation strategies. In these studies, B&Es are usually identified from interviews or focus groups are considered to provide insights into participants’ experienced reality. However, these perceptions may change over time, including over the duration of an implementation process (Petersen et al., 2021).

Barriers and Enablers Encourage Binary Thinking That Is Ill-Suited to Complexity

B&Es are a heuristic device: a mental shortcut that allows a simplified representation of the world (Bach-Mortensen & Verboom, 2020). The approach presents a binary framework driven by a dualistic logic. A binary framework may appear to offer a way of cutting through complexity to identify simple, tangible solutions; yet there is increasing recognition of the need to understand and respond to complexity in the challenges faced by health services and public health more broadly (Crane et al., 2022; McDaniel et al., 2003). This draws on the seminal distinction between simple, complicated, and complex problems outlined in Box 1.

Box 1: Differences Between Simple, Complicated, and Complex Problems

• Simple problems are like baking a cake. If you have the right ingredients and measurements and follow the recipe, you will probably produce a successful cake every time.

Consider a ‘Whole Package’ Research Methodology

Unlike thematic analysis or content analysis which are a-theoretical and can be used in very different ways according to the research’s theoretical framework, some research approaches using qualitative methods provide a whole methodology. That is, in addition to the methods (what you are doing), they provide a rationale and theoretical basis for the overarching research strategy (why you are doing it that way). Grounded theory, phenomenology, realist evaluation, photovoice, and soft systems methodology are just some examples. These approaches can suggest alternatives to B&Es for framing research questions and design but can also accommodate an exploration of B&Es. For example, Huntingdon et al. (2020) explored B&Es within a grounded theory study which built a conceptual model that describes the process of sexual adjustment to HIV and intervention points for providing support. Erdem et al. (2020) conducted an interpretative phenomenological study exploring B&Es associated with male mental health help-seeking to develop in-depth understating around themes which included ‘denial as preservation’. Labbé et al. (2021) explored B&Es to social inclusion of people with disabilities using photovoice’s participatory methodology to develop themes such as ‘the paradox of accessibility’ which examines conflicting needs, while King et al. (2023) “go beyond a simple listing of barriers and facilitators” to identify context–mechanism–outcome configurations in their realist evaluation of AI use in pathology. Many realist studies also incorporate an investigation of B&Es, but the inquiry is underpinned by questions of what works for whom and under what circumstances (Fletcher et al., 2016). Realist evaluation also helps researchers “avoid being overwhelmed by the enormity of potentially important contextual factors by focusing on … [those] that affect the mechanisms ‘sparked’ (or inhibited) by an intervention …. rather than asking generic questions about contextual enablers and barriers” (Punton et al., 2020).

Use Theories and Conceptual Frameworks to Guide the Research

Formal theories and conceptual frameworks can offer an alternative way of guiding research questions, data collection, and analysis which can also increase the transferability of research findings (Willis et al., 2007). As Meyer and Ward (2014) argue, “The role of theory in qualitative health research is paramount for translation into practice and policy, since it moves beyond pure description of data, allowing interpretation of the social processes underpinning and potentially ‘explaining’ findings.” Importantly, theories and conceptual frameworks help researchers explore the relationships between concepts in their data and think critically about data that might be overlooked, misinterpreted, or uninterrogated without a theoretical framework (Collins & Stockton, 2018), including paying attention to political and power dynamics (Huffman & Tracy, 2018). For example, in their study of obesity in Cuba and Samoa, Garth and Hardin (2019) reject focusing on barriers to weight management. They argue that previous studies using a barriers lens conflated different classes of phenomena and risked compounding stigma and disempowerment because of the inherent assumption that once barriers are identified they should be surmountable and when they are not overcome, the individual is to blame. Instead, they use social visibility as an analytic lens to critique the persistent failure of most obesity interventions and draw on concepts that seek to explain contextual factors such as obesogenic environments.

Bacchi offers an analytic strategy for interrogating ‘solutions’ to social problems which is particularly helpful for planning and guiding research. The ‘What’s the Problem Represented to Be?’ approach comprises six questions that provoke critical scrutiny of how policies and programs implicitly frame (represent) the problems they purport to be tackling. The questions support researchers to examine problem and solution representations as socio-political constructions with power implications rather than natural phenomena which are ‘out there’ waiting to be discovered (Bacchi, 2012, 2016). A recent study using this approach analysed the representation of problems underpinning the introduction of routine mental health outcome measures, highlighting their neglect of social determinants of mental health in meeting consumer needs and expectations (Oster et al., 2023).

A priori theories and models such as self-determination theory (Tettero et al., 2022), social-ecological approaches (Brutzman et al., 2022; Lun et al., 2022), normalisation process theory (Kousgaard et al., 2022; Simpson et al., 2018), the theoretical domains framework and associated COM-B model (Akiba et al., 2022; Handley et al., 2016; McDermott et al., 2022; Petersen et al., 2021), and social cognitive theory (Fuller et al., 2019; van Rijen & ten Hoor, 2023) can also accommodate an exploration of B&Es, situating them within a broader conceptual context and ensuring attention is paid to different levels, classes, and dimensions of the phenomena, as well as providing a way of theorising about how the B&Es relate to one another. Critical perspectives which attend to structural inequities and power-based features, such as critical disability theory, can also be incorporated and have particular value in challenging cultural assumptions and revealing systemic barriers (Aylsworth et al., 2022; Lin & Melendez-Torres, 2017; Naidu et al., 2023). For example, van Heesewijk et al. (2022) use queer theory as a theoretical lens to investigate B&Es to implementation of transgender health education in medical school curricula, identifying normative gender ideologies that classify transgender people as ‘unhealthy Others’.

There are well-described frameworks for constructing effective qualitative research questions for clinical studies and research syntheses (Bennett & Bennett, 2000; Foster & Jewell, 2017) (including a useful overview provided here: https://lib.guides.umd.edu/SR/research_question). And implementation science boasts a huge range of tools for guiding nuanced implementation planning and evaluation (e.g., Proctor et al. (2011)). These include process models, determinant frameworks, classic theories, implementation theories, and evaluation frameworks (Nilsen, 2015). Researchers can use an open access website to select the most appropriate tool: https://dissemination-implementation.org/tool, and Lynch et al. provide guidance related to the ‘top ten’ approaches (Lynch et al., 2018). At the very least, a B&Es approach should be augmented with careful attention to acceptability. Sekhon et al.’s framework for evaluating acceptability of healthcare interventions is useful here as it includes concepts that B&Es may not capture such as feasibility, affective responses to the intervention, and opportunity costs (Sekhon et al., 2017). Tools are also available for assessment and planning of public health and clinical program sustainability in relation to key success factors: a more sophisticated way to consider ‘enablers’ (Center for Public Health Systems Science, 2023).

Use Participatory Methods to Better Understand Participants’ Lived Experiences

Researchers adopting a critical stance often strive to use their work and the process of conducting it as a tool for emancipation and so may prefer participatory or action research designs where participants become co-researchers (Pope & Mays, 2020). This is an especially important approach for populations who have been disenfranchised by traditional methods of research, such as First Nations people (Baum et al., 2006). Research which is co-produced in partnership with stakeholders asks questions that are most relevant to them, provides richer insights, and facilitates research translation because stakeholders are invested in the process and outcomes. Specific participatory action research methods such as health impact assessment (den Broeder et al., 2017), photovoice (Walker et al., 2020), soft systems thinking (Checkland & Poulter, 2007), and Our Voice citizen science (Wood et al., 2023) can provide a methodological framework which could be used instead of, or to enhance, a B&Es approach.

Use Sensitising Concepts to Guide Inquiry

Those who do not wish to adopt a full flung theory or to be constrained by an existing framework can engage with sensitising concepts: ‘background ideas’ that provide a point of departure for developing the research question and design without directing the scope or shape of the work (Bradbury-Jones et al., 2014; Kraus, 2018; Timmermans & Tavory, 2012). Again, these concepts may enable researchers to identify alternative approaches to B&Es, or they can be used to enhance a B&Es inquiry. For example, a study of B&Es to social inclusion experienced by children with disability which is informed by the social model of disability – in which environments are regarded as enabling or disabling (Barnes, 2003) – is far more likely to highlight structural and socio-political barriers than one which is informed by a biomedical model.

Qualitative research that investigates or explores interventions in health services could benefit from attention to sensitising ideas in the organisational and knowledge mobilisation literature. These include sensemaking (Weick, 1995), organisational learning (Argote, 2011), communities of practice (Brown & Duguid, 2001; le May, 2009), and various models of research utilisation that expose the fallacy of linear pipeline models of knowledge transfer (for which a B&Es inquiry would be appropriate) and focus instead on social, discursive models of knowledge transformation in context (Brown & Duguid, 2001; Glasgow et al., 2012; Haynes et al., 2011; Nutley et al., 2007; Weiss & Bucuvalas, 1980). Ecological metaphors and key ideas from systems thinking can help researchers move away from these unhelpful mechanistic or linear conceptions of change processes. These approaches raise questions such as: How does change happen in this setting? Do the people targeted accept the premise of the intervention/desired behavioural change? How does it intersect with their personal or professional identity, autonomy, values, wishes, and circumstances (what’s in it for them?)? How do people make sense of it? How is knowledge shared and legitimised in this setting? These questions may be particularly suited to understanding new interventions that have not been co-designed where researchers may have overlooked some of messier and inconveniently human aspects of the phenomena.

An example of research informed by a community-of-practice lens is the ethnographic study of guidelines use in GP practices conducted by Gabbay and le May (2004, 2011, 2016). Rather than asking what the B&Es were to using guidelines, the researchers sought to explore how GPs made everyday healthcare decisions in their practices. They found GPs developed ‘mindlines’: personal ‘guidelines-in-the-head’ that melded evidence from myriad sources with tacit knowledge, contextual understanding, and continual learning – shaped by coffee-room chats with colleagues – to form an internalised guide to practise. These findings illuminate the fluidity of ‘knowledge-in-practice-in-context’ in everyday healthcare and provide new ways of thinking about how to facilitate the use of guidelines in primary care.

Investigate How an Identified Barrier or Enabler Works in Depth

A commonly cited barrier to implementing interventions introduced in organisational settings such as hospitals and clinics is a lack of time. Findings typically reflect a sense among staff that they lack capacity to take on anything new but seldom examine this critically. Yet there may be numerous structural, systemic, or individual factors underlying this experience of ‘lack of time’ which are not volunteered (or even perceived) by participants. What if, for example, the ‘rush culture’ has convinced health staff that hospitals are not appropriate sites for health promotion (Kelley & Abraham, 2007) or that staff feel the need to be seen to be working at capacity irrespective of ebbs and flows in their practice? What if the underlying problem is burnout or compassion fatigue, or deep resentment about workplace dynamics, but it’s too hard to say those things? Thus, concepts of time – as with other B&Es – might better be understood not only in terms of good/bad but also as a point of departure for further analysis. This includes questioning the assumption that more time is needed at all. Légaré et al. (2008) address this in their systematic review of shared decision-making in clinical practice in which lack of time was identified as a ‘universal perceived barrier’. However, they note that:

… there is no robust evidence that compared to usual care, more time is required to engage in shared decision-making in clinical practice …. Therefore, it remains essential that future studies investigate whether engaging in shared decision-making actually takes more time or not ...

Whelehan et al.’s (2021) study of barriers to reducing fatigue in surgeons exemplifies this approach. They sought to better understand the problem of fatigue, asking about its causes and impacts, in order to rethink possible solutions including individual, team-based, and environmental-level interventions.

Develop and Test Implementation Strategies for Addressing Barriers and/or Enablers

Descriptive studies of B&Es frequently conclude with high-level recommendations about possible intervention strategies, but these may be left dangling with no research efforts to operationalise, implement, and evaluate them. Thompson and Reeve (2022) highlight this gap in the literature in their study of deprescribing in which they state that much research in this field:

… focused on identifying barriers to and facilitators of deprescribing in clinical practice … However, with continuing research around barriers and facilitators, we need to be mindful to undertake research that builds on existing knowledge, addresses known gaps, and advances the field … [by] translating existing knowledge into strategies and tools that can impact clinical practice and lead to practical and sustained deprescribing …

Martin Ginis et al. (2016) make a similar point in relation to studies of physical activity participation by people with physical disability:

… we urge researchers to move beyond conducting simple descriptive studies. If the ultimate goal is to increase LTPA [leisure-time physical activity] participation, then it is time to focus on selecting, developing, testing, and implementing interventions, rather than simply generating lists of barriers and facilitators. For this to occur, researchers must be able to address a variety of barriers/facilitators that cut across multi-level, multi-sector structures.

Use the Existing Barriers and Enablers Literature to Develop a New Model or Theory

Given the plethora of B&Es studies, synthesis of this literature may be warranted. Herber et al. (2019) used meta-synthesis techniques to integrate identified barriers and facilitators to heart failure self-care and developed a new theoretical model expressed in a series of propositions that can be tested in further research. Noyes et al. (2014) developed a model for reconceptualising children’s complex discharge, drawing on B&Es and existing theory. Nilson (2015) points out that many frameworks that focus on implementation determinants were developed synthesising results from empirical studies of B&Es to implementation success.