Primary care redesign for person-centred care: delivering an international generalist revolution

Person-centred primary care is a priority for patients, practitioners and healthcare policy. Despite this, data suggests we are still not consistently achieving person-centred care – and indeed that in some areas, care may be worsening. Whole person care is the expertise of the medical generalist – an area of clinical practice that has been neglected by health policy for some time. It is internationally recognised that we need to rebalance specialist and generalist primary care. Drawing on fifteen years of scholarship within the science of medical generalism (the expertise of whole person medical care), I describe a 3-tiered approach to primary care redesign; describing changes needed at the level of the consultation, practice set up, and strategic planning. The changing needs of patients living with complex chronic illness has already started a revolution in our understanding of healthcare systems. This paper outlines work to support that paradigm shift from disease-focused to person-focused primary healthcare.

4 highlights a gap between the language of policy and the reality of practice. Perhaps most alarming in the UK was a report last year highlighting a reported decline in person-centred care in general practice (National Voices 2017) -the medical discipline which defines itself as 'specialising in the patient'.
So how do we respond?
In this paper, I draw on the UK Society for Academic Primary Care's model of blue sky thinking to propose a 'Dangerous Idea': an idea that challenges the status quo but with a commitment to action (https://sapc.ac.uk/article/sapc-dangerous-ideas-soapbox). Based on a critical reflection on 15 years of my own work in this area, I propose a 3-tiered approach to primary care redesign to reverse the trend in person-centred care, to address the changing healthcare needs of our population, and to respond to our quiet revolutionaries. I will outline the changes needed at the level of the consultation -the interaction between patient and professional; the organisation of practice teams; and the strategic overview of our health care system (Table 1). My account is largely grounded in a UK perspective, but evidence suggests that the proposals are relevant to an international audience.
My intention is to spark the wider conversations needed, 'now more than ever', to deliver truly person-centred primary care.

Tier 1: The Consultation
Patients highlight a perceived lack of person-centred care in their interactions with health care professionals (National Voices 2017, Reeve et al. 2012, Reeve andCooper 2016a). We need to take a fresh look at the consultation -the process by which patients and professionals interact to determine and address healthcare need.
Addressing the decline in person-centred care: tailoring care to the individual Living with long term conditions can be exhausting (Carel 2008). A rich literature records a biographical account of the work living with chronic illness (Lockock and Ziebland 2015). On top of the work of everyday life comes the burden of illness, and an increasing burden of healthcarewhether from medication, managing appointments, or the tasks of monitoring health status (Mair 2014). Primary care has long recognised this broader experience, as described within a biopsychosocial understanding of illness (Weiss 1980). But patients tell us that although we acknowledge this wider experience of illness, we don't necessarily take it into account when we make clinical decisions (Reeve et al. 2012 (2015) report on polypharmacy highlighted the importance of capacity for "compromise" between professional and patient priorities if we are to address the challenge of problematic polypharmacy. Denford et al.'s (2014) review of medication use provides further evidence of patients reporting a lack of such tailoring of care.
What stops health professionals from tailoring care to meet individual needs? To understand this, we must look more closely at the way that health care professionals make clinical decisions about what is wrong (diagnosis) and what needs doing (Intervention). The dominant model of clinical practice in current western healthcare is that of specialist medicine -condition-or system-specific healthcare. The specialist model currently dictates how we define 'best practice' (the timely and correct identification of disease status), and so how we train health professionals to deliver care, and design the systems that they work in. The wider literature, including my own research, highlights how and why this model of practice has become a barrier to the delivery of individually tailored care.
Person-centred care requires a clinician to tailor the use of evidence to individual circumstances (Denford et al. 2014), potentially requiring compromise between biomedical and individual perspectives (Kings Fund 2015) in order to deliver healthcare decisions that recognise personcentred goals focused on continued daily living (Health Policy Partnership 2016). Although Evidence-Based-Medicine (EBM) and clinical governance mechanisms encourage the use of clinical judgement, they also place clinical opinion at the bottom of a hierarchy of knowledge (Sackett et al. 2006). No guidance is offered to practitioners as how to distinguish between 'clinical judgement' and the form of evidence found at the bottom of EBM hierarchy of evidence -namely 'profession opinion'.
Practitioners consistently report feeling constrained in challenging the evidence, defending judgement/opinion and so tailoring care (Reeve et al. 2013a(Reeve et al. , 2018a. I have previously described how scientific practice could address this barrier by recognising that generalist practice is grounded in a different epistemological framework to specialist practice (Reeve 2010). Specialist practice is deductive -theory driven practice that assesses the likelihood that a diagnostic theory can be applied to this person. Generalist practice is inductive -data driven practice in which multiple elements (all believed to be robust) are combined to infer an explanation/conclusion. What generalists have lacked is a framework to legitimise the process and output of this action -a gap which has contributed to generalist practice being overlooked (Reeve et al. 2013) in recent primary care redesign. Scientific practice may help address that gap.
The scientific literature, especially within qualitative and applied traditions, highlights the intellectual tasks needed to deliver trustworthy inductive interpretation (Reeve 2010). I have translated these into a guiding framework for 'defendable-clinical-decision-making' (ibid), andworking with GP tutors -into a consultation model for teaching (Reeve 2015). The 'SAGE consultation model' outlines the five steps needed to deliver effective, safe clinical decision making within an inductive model of practice. (Further details can be found in this account of the model applied to a clinical case of managing multimorbidity: http://primarycarehub.org.uk/images/SAGE/SCM.pdf) . The SAGE model aims to contribute to rebalancing the hierarchy of evidence, to visibly place professional wisdom back at the top of the chain (Reeve 2018b). It contributes to shifting our understanding of generalist practice from a 'jack of all trades' view that describes the generalist in terms of what she/he does (the range of work); to a scholarship model that understands generalist practice by how the work is done -the intellectual task of effective, safe interpretive practice.
This intellectual work happens in the context of daily practice. Surveys of professional practice highlight key enablers and barriers to the work of generalist decision making and individually tailored care (Reeve et al. 2013a(Reeve et al. , 2018a: including a lack of clarity in what generalist practice is, a failure to prioritise the intellectual task (and cognitive load) of generalist practice in the wider General Practice day; a lack of training; a shortage of necessary resources for practice including continuity of care and collective action; a failure to support ongoing practice through feedback and performance assessment; and a lack of coordinated/optimised delivery of care once decisions are made and implemented.
Drawing this work together we start to recognise generalist practice as a complex intervention consisting of multiple interacting parts, defined by the distinct expertise that is whole-personcentred clinical decision making (Reeve et al. 2013a) - Figure 1. My own research now seeks to apply these principles to rethinking primary care design and delivery in the critical situations of prescribing practice, mental health care and acute care. My intention is to use this work to refine the generalist model as a tool to support a person-centred primary care 'revolution'.
There is still work to be done, but efforts to date prioritise four key issues in this first tier of redesign.
Experience shows that it is insufficient to simply describe the principles of generalist practice -of whole person medical care -and assume that professionals can or will deliver this model to care. To deliver person-centred primary care we need to reimagine generalist practice, focusing on 4 elements: • Recognising the distinct intellectual task that underpins the everyday pragmatism of Family Physicians and General Practitioners delivering person-centred care, recognising 'Every GP a Scholar' (Society for Academic Primary Care 2017) • Describing that intellectual task: the five steps to trustworthy interpretive practice (Reeve 2010) and so training people in this distinct model of care The data highlighting barriers to delivery of whole-person-centred generalist care recognises the impact of context -the organisation of practice and practice teams -in supporting or undermining care. My second tier of change looks at the organisation of practice necessary to support delivery of generalist care.
Addressing the decline in person centred care: redesigning primary care teams We currently design teams to deliver disease focused care -described in the largely linear models of care pathways. Person-centred, generalist care, is a complex intervention -a non-linear model of care. We need to redesign teams to deliver complex interventions (Reeve et al. 2018a).
We still know relatively little about how to do that. In a recent systematic review, Lau et al (2015) highlight that whilst we know more about how to change individual professional behaviour to support implementation, there is a lack of evidence on the process and effectiveness of implementation of complex interventions at organisational levels -evaluating change at a whole practice level.
I have completed two studies evaluating the implementation of complex interventions within primary care -both grounded in the ideas described within Normalisation Process Theory, a sociological theory of the implementation and embedding of organisational innovations (www.normalizationprocess.org/). One study was a prospective evaluation of introducing a new whole-person, generalist model of primary mental health care in the UK -the Bounceback project (Reeve et al. 2016b). The second was a retrospective analysis of the implementation of a new frailty initiative within the General Practice setting in England (Reeve et al. 2018a). Critical review of the findings from this work offer us useful insights into understanding the practice level changes needed to deliver a person-centred revolution in primary healthcare delivery.
A key finding from both studies was the need for models of practice that support the iterative redesign of complex interventions as an integral part of their implementation and delivery. Novel complex interventions, even when evidence-based, should not be seen as a 'bolt-on' to existing services, but rather as a stimulus for re-evaluation and evolution of existing models of care (Reeve et al. 2018a). Implementation and adaptation creates a need for extended expertise within practice teams to support the critical development of knowledge-in-practice-in-context (Gabbay and le May 2010). Our experience supported the observations described by Evans and Scarborough (2014) of the need to 'blur' rather than simply 'bridge' the gap between clinician and research skills to optimise implementation and delivery of new interventions. All of which may require review and refinement of contractual and quality assurance mechanisms that traditionally focus on delivering a described model rather than adaptation and implementation (Reeve et al. 2018).
Primary healthcare systems around the world are redesigning models of practice. The current focus is on developing integrated, extended multidisciplinary teams to address the growing complexity of patient health needs -a focus on who is in the team. My research highlights that we also need to think again about what they are doing and how they are working; providing a framework to consider when commissioning new models of practice (Reeve et al. 2018).

Tier #3: Whole system strategic redesign
Practice teams represent units of care delivery within a wider system of healthcare. Strategic priorities within that system determine the drivers that shape practice teams, the resources offered to them, and the performance management processes that influence ongoing delivery. To change the way we deliver care, we must look at the strategic context within which health care happens

Addressing the decline in Person-Centred Care: designing a system of care around a clear vision of balanced generalist-specialist care
We see growing international consensus on the need for a revival of generalist, whole person, carea strategic shift away from an excessive focus on the "command and control of disease" to delivery of person-centred care (WHO 2008); a rebalancing of healthcare systems (Heath 2011). We have a substantial knowledge base on the strategic design of specialist, condition-focused models of care; but lack a clear understanding of the design of generalist systems. We lack a 'blue print' to guide policy makers and commissioners in achieving a goal of a balanced generalist-specialist healthcare system. Lewis's (2014) editorial started to address this gap. In this work, he recognised two emerging conceptualisations of generalist care. The first is the systems-focused model of generalism described as Integrated Care -the coordinated delivery of accessible, comprehensive potentially multi-faceted healthcare. Integrated Care seeks to overcome the healthcare burden created by fragmented models of condition-specific care -enabling the smoother navigation of a whole person (patient with multiple needs) through a complicated system. Integrated Care is the main strategic focus of most current primary care redesign; although it is also a model of care for which the evidence-base is still mixed (RAND Europe 2012). Taking Lewis's editorial as a starting point, Byng and I recently proposed that balanced primary care redesign needs both approaches -integrated delivery of condition-specific care, and capacity for individually tailored (personalised) care. We described a United Model of Generalism (Reeve and Byng 2017) which recognises Lewis's (2014) accounts as two axes in a single system of healthcare design ( Figure 2). Our blueprint describes four new categories of healthcare defined by a personcentred need (for generalist or specialist care) and the health systems requirements to deliver care (based on simple-technical, or complex-integrated models). The blueprint provides us with a map to re-define a epidemiology of need, to describe an updated workforce model to deliver care, and so to re-balance resource and demand. We are now starting discussions with commissioners, patients and professionals to consider how we might take this work forward.

Concluding thoughts
In 2008, the World Health Organisation (2008) outlined why we need to revitalise person-centred primary care, "Now More Than Ever". Travelling forward to 2018, we see a continuing commitment across political, policy and practice contexts to that vision. But the main efforts of health services remain focused on improving the coordination and integration of existing models of care (Lewis 2014, RAND Europe 2012, rather than a true shift to a more "revolutionary view" (Lewis 2014) of individually tailored generalist care.
The changing needs of our patients are driving a paradigm shift in healthcare design from diseasefocused to person-focused care (Reeve 2017). So far, a vision of redesign of care around the patient has failed to deliver this. I therefore propose that to achieve person-centred healthcare, we need to redesign healthcare around the expertise of the generalist clinician in making whole-person, goaloriented clinical decisions.
Clinicians and scholars around the world are actively engaged in work that addresses individual elements of the ideas I have outlined (for example see Chambers et al. 2013, Spencer-Bonilla et. al. 2017, Sinnott et. al. 2017. My intention in this paper is to start conversations, collaboration and shared actions that translate scientific innovation into systems 'revolution'. Taking this work forward will require leadership and collaboration from across the clinical and academic primary care communities (Society for Academic Primary Care 2016). I look forward to conversations arising. But any such action will only succeed if we draw on the greatest resource of all -our quiet revolutionaries. Our patients.

For patients
Re-focused healthcare service emphasising outcomes targeted on 'a life for living' -health as a means to an end, not an end in itself • Patient engagement in co-construction of solutions to health problems • Need defined by health related impairment to daily living (including resilience) rather than disease status (alone)

For clinicians
Revisions to training and organisation of practice to recognise/support the INTELLECTUAL TASK of expert generalist practice.
• Enhanced Scholarship Training (Reeve 2018b) • Restructure of working day to recognise Cognitive Load of generalist task (survey) • Reprioritisation of resources for generalist care (Figure 1)

For systems
Reshaped around an understanding of person-centred healthcare as delivery of complex interventions, requiring us to • Redefine 'best' care -supporting an outcome of daily living, rather than 'command and control' of disease • Expand expertise within teams -including skills in implementation as well as delivery of interventions • Sustain expertise within teams (building professional capital -Gabbay and le May (2010))