A new kind of research

The GP curriculum and a new kind of research

Professional module 2.04 : Enhancing professional knowledge states the need for GPs to be able to learn, critically appraise and teach in relations to:

Evidence-based practice and statistics

A new kind of research for a new kind of practice

Research in the ‘middle-ground’

Multiple new ways of working, and new models of general practice and primary care, are being developed and piloted, such as alternatives to face-to-face consultations, different triage systems, extending the role of the practice nurse, the training of advanced practitioners, direct access to physiotherapists for acute as well as chronic musculoskeletal problems, and direct patient care delivered by pharmacists within the practice, to name but a few. All of these initiatives are welcome, and should reduce pressure on GPs, help alleviate the workforce crisis in general practice, and allow GPs to have more time to provide generalist care to patients with complex needs. However, extending the role and complexity of the multidisciplinary team also brings challenges, including roles and responsibilities and continuity of care. These changes are also taking place within a wider agenda to better integrate health and social care. The direction of travel is exciting and much needed, but it means that the landscape (both for patients and for GPs) has never been more complex.

This potential transformation of general practice requires research evidence to guide it, and evaluation to inform it. The ‘gold-standard’ model of research - based on randomised controlled trials (RCTs) is still much needed. These provide the most definitive scientific answer as to whether an intervention has benefit or not (although only within the confines of the trial sample, which is often not representative of the broader population).

However, there is also an important research ‘middle-ground’ that sits between tightly controlled RCTs (usually led by academics) and poorly conducted service evaluations (usually done by over-stretched NHS staff) that can evaluate innovations in general practice over a shorter time scale and in real-life settings (Guthrie et al., 2017). Such a ‘realistic research’ approach can inform policy and service development much faster and more cheaply than traditional RCTSs, and be of high relevance to NHS partners. This approach to research would exploit the strengths of academics and of frontline clinicians and managers (often complementary, but seldom brought together in jointly planned programmes of work). Table 1 highlights some of these strengths and weaknesses. If academics and NHS staff could work together more effectively in this ‘middle-ground’ then the strengths and skills of both groups could be better utilised for the benefit of patients.

OPEN IN VIEWER

Table 1.

Strengths and weaknesses in innovation development and evaluation.

	Frontline clinicians and managers	Academics
Creating interventions and new models of care	Normal business for NHS innovators. Strong on feasibility, but often does not draw on strongest existing theory and evidence.	Normal business for health services researchers. Strongly based on existing theory and evidence, but often inadequate attention paid to feasibility.
Evaluating interventions and new models of care	Often not focused on from the start, and evaluations done tend to use weaker designs that have significant risks of bias.	Emphasise pre-planned, ‘as strong as possible’ evaluation design to minimise bias.
Translating new ideas into practice and ensuring spread and sustainability	The experts in real-world implementation, but often do not draw on existing theory and evidence.	Often under-estimate the complexity of real-world implementation and many perceive translation to be someone else’s responsibility.
Evaluating widespread implementation	Often not focused on from the start, and evaluations done tend to use weaker designs that have significant risk of bias.	Have relevant methodological expertise, but not commonly engaged in real-world evaluation, although now partly incentivised by Research Excellence Framework requirements to demonstrate impact.

Guthrie B, Gillies J, Calderwood C, et al., (2017) Developing middle-ground research to support primary care transformation. British Journal of General Practice, 67(664), 498–499. Reproduced with permission.

Why do we need middle-ground research?

The practical challenge

The evidence-base for how best to manage patients with multimorbidity in general practice is extremely limited. A Cochrane Review in 2016 found only 18 RCTs worldwide, and mostly these did not show any evidence of benefit (Smith et al., 2016). The National Institute for Health and Care Excellence (NICE) Guidelines on multimorbidity highlight the need for more evidence (NICE, 2016). Multimorbidity also raises challenges of polypharmacy, and the associated problems of adverse drug reactions, drug–drug interaction, and drug–disease interactions, thus making patient safety a key concern. Patients with multimorbidity may also experience ‘treatment burden’ given the considerable self-management demands from multiple complex conditions, and the fact that the health service, including general practice, is largely organised around a single-disease model. Thus, in addition to self-management, patients with multimorbidity face numerous different appointments with different specialists, and even with different nurses within the same practice for their annual chronic disease Quality Outcomes Framework reviews.

Despite this lack of evidence, and a fragmented system based on single conditions, GPs are, of course, doing their best to manage large numbers of patients with multimorbidity in a holistic way every working day. They do so by a judicious use of guidelines and evidence-based medicine, but also by using ‘practice-based evidence’ as defined and discussed by Professor Joanne Reeve at the University of Hull (Reeve et al., 2013). Practice-based evidence relates to the role of the GP as an expert generalist (see Box 1).

OPEN IN VIEWER

Box 1.

Expert generalist practice.

This is defined by Reeve et al. (2013) as being:

The distinct and defining component of the generalist, whole-person focused therapeutic approach, is defined by the principle of person-centred decision making which recognises health as a resource for living and not an end in itself. Which is in turn underpinned by the practice of interpretive medicine: integrating multiple sources of knowledge (including biomedical, biographical and professional) in a dynamic exploration and interpretation of individual illness experience. Practice leads to decisions about what is wrong, and what is needed to intervene, which support an outcome of health as a resource for living.

Expert generalism lies at the core of general practice and is especially relevant to the population challenges of multimorbidity outlined above. Without expert generalism to maintain the holistic and cost-effective balance of realistic care, there is a real risk that patients with multiple conditions and complex needs will face increasing levels of fragmented, burdensome, and, at times, potentially harmful care from numerous different, uncoordinated specialists. Without the expert generalism within general practice, the NHS would very quickly become unstable and financially unsustainable.

Evidence-based medicine meets practice-based evidence

Our proposal for middle-ground research is borne out of our conviction that in the rapidly changing policy landscape of health and social care, and especially in primary care, the transformation of general practice required to meet the changing needs of the population cannot happen without being informed by high-quality evaluation and research. The spiralling costs of the NHS, the crisis in the general practice workforce, and the rising needs of the population (from multimorbidity, ageing, and widening inequalities) all need effective solutions now, not 10 or 15 years hence, if universal coverage and care for all, free at the point of delivery, is to survive in the UK. This will require innovations and new ways of working, which will need to be rigorously evaluated in order to ensure effectiveness and cost-effectiveness. The urgency is such that evidence of delivery is needed more quickly than large RCTs can deliver such evidence. A single RCT in the health care system can take up to 20 years to complete, from conception to implementation.

Middle-ground research offers the production (or rather co-production) of useful evidence over a much shorter time-frame, not instead of, but in addition to large RCTs. Middle-ground research is not a philosophy, nor a single methodology, but rather a pragmatic approach to fill the ‘no-man’s land’ between RCT-based academic research and NHS innovations with little or no evaluation. The type of research and evaluation that is possible in the middle-ground will depend on many factors, and the methods required will depend on the nature of the questions that need to be answered. Some may be answered by data-driven innovation, utilising large data sets of routine general practice data. Others may require mixed-methods approaches that combine and integrate quantitative and qualitative approaches.

Our experience since proposing this middle-ground approach (Guthrie et al., 2017) suggests there is widespread support from clinicians and researchers. However, making it happen will require buy-in and commitment from policy-makers, commissioners, NHS staff and other academics. It will also require a real commitment to work together with patients and staff in a co-production model. It will also require funding from governments.

Such research requires a clear identification of what the research questions are, and crucially, whether they can be answered in a timely and efficient fashion. Thus, evaluability assessment is an important first step. It will also be important to differentiate the types of research required.

For example, some research projects will set out to help fill a gap in the available evidence (such as with multimorbidity, where very few definitive RCTs have been conducted that provide definitive evidence of ‘efficacy’) and to guide practice in a pragmatic way until definitive RCTs can be completed. Such ‘good enough’ research will aim to give a clear steer to policy-makers, managers and staff on whether a particular intervention appears to be effective and cost-effective.

Some research will consider the implementation of an intervention with a good evidence-base (perhaps RCTs conducted in a different health care setting or country) and seek to identify the barriers and facilitators of local implementation. Such implementation-focused research will identify whether an intervention has been fully implemented, and if not, what are the key barriers hindering implementation.

In both these types of research, it is also helpful to know what works, for whom, and under which circumstances, given the complexity and heterogeneity of primary care in different geographical settings across the four nations.

Getting realistic

Middle-ground research is part of a broader move within the UK and internationally to make health care more focused on the actual needs and goals of patients. In Scotland, the Chief Medical Officer’s vision for ‘Realistic medicine’ has been widely welcomed by GPs and other primary care staff (NHS Scotland, 2016). The key principles are:

Moving towards shared decision-making

This vision aligns closely to the future vision for general practice and primary care. Many of these principles are shared with approaches in other nations. In Wales, ‘Prudent Healthcare’ is based on four principles:

Public and professionals are equal partners through co-production

The global ‘Choosing Wisely’ movement, backed by the Academy of Royal Medical Colleges in the UK, aims to encourage shared decision-making by empowering patients to discuss with their doctor what treatments are best for them and which may not be necessary.

These developments have come about due to a growing awareness of the issues of overdiagnosis and overtreatment of disease (Treadwell and McCartney, 2016). Indeed, the RCGP established the Standing Group on Overdiagnosis (Supporting Shared Decisions in Health Care) in 2014. Middle-ground studies could help to address this trend by a generalist approach to examining the harms and benefits of proposed new changes in diagnostic criteria or therapeutic approaches with marginal gains.

Again, given the ageing population and the widening chasm of health inequalities, a focus on multimorbidity (which underpins both) would seem warranted. Middle-ground research offers a vision for ‘realistic research for realistic medicine’ which will be essential if realistic medicine is to become a practical reality with impact at a general practice and population level.

KEY POINTS

The NHS is dependent on a strong generalist primary care system that is able to deal holistically with the needs of the population served