Evidence for public health practice: Health Inequalities National Support Team
Evidence for public health practice: Health Inequalities National Support Team
Abstract and Keywords
From 2007 until it was closed down by the Coalition government in 2011 HINST operated on the premise that it is possible to narrow gaps in health outcomes by targeting local policy and action. This chapter provides practical examples of approaches to reduce health inequalities and effective practice at a local level. In so doing an insight into issues encountered in developing an applied approach to health inequalities is developed. A ‘Christmas Tree’ diagnostic is unpacked across a series of component principles for commissioning services to reduce health inequalities. The HINST’s experience suggests that it is possible to narrow gaps in health outcomes based on policy action and that a systematic approach can impact on the range of components that constitute the complex problem that is health inequalities.
This chapter is a case study that summarises the learning from a team that worked with local partnerships in the most deprived areas of England to reduce health inequalities – the Health Inequalities National Support Team (HINST). The team used two underpinning models that are congruent with the underlying thesis of this book – that work to address health inequalities and the complex dynamic relationships that support the status quo (Figures 7.3 and 7.4, pp 160, 163). The chapter provides practical examples of approaches to reduce health inequalities and effective practice at a local level. Appendix A links to the practical techniques developed by the Team. Although the structure of statutory agencies charged with addressing health inequalities has changed, the principles remain relevant, and the chapter provides a case study of the practical issues encountered when the underlying theses of this book are applied to practice.
From 2007 until it was closed down by the Coalition government in 2011, HINST operated on the premise that it is possible to narrow gaps in health outcomes by targeting local policy and action, which is one of the principle themes explored in this book, as outlined in the first chapter. Its aim was to promote effective practice in this area of public health, as a resource in public health practice for academics, researchers and public health practitioners. Much of what follows can be viewed in terms of issues around the boundaries, dynamics, self-organisation and co-evolution of systems involved in addressing health inequalities. As such, the chapter makes implicit connections to the complexity-based framing of this book, and these connections are reflected upon more explicitly in the concluding chapter.
The chapter begins by providing a brief introduction to the context and evolution of HINST, which includes an overview of the Team’s programme, a summary of its evolution and some reflections on the health inequality outcomes. The next section provides a more detailed account of (p.156) the principles that underpinned the HINST approach. The following section provides a series of practical illustrations of effective practice identified by HINST. The changing environment under the Coalition government is then considered, with a focus on requirements for delivering population-level change through the new policy structure. The chapter concludes with some reflections on how the HINST model can be capitalised on.
Context and evolution of HINST
The national support teams programme
From 2006, the Department of Health (DH) established ten thematic national support teams to support local areas in tackling complex public health issues using the best available evidence. They operated within a shared framework, adopting similar processes based on a common model of change.
The national support team process involved a pre-visit desk review of selected documentation and data-based intelligence, followed by interviews with key informants, often in combination with a series of workshops or focus groups. Collation and analysis of findings was immediate, and the findings – including strengths and recommendations – were fed back straight away to the key local players at each site. Recommendations were accompanied by offers of support, either at the time of reporting or as through a package of follow-up activity to address agreed priorities.
The national support teams were appointed as change agents. Iles and Sutherland’s National Health Service (NHS) manual on organisational change underpinning this approach encouraged the teams to challenge the ‘implicit assumptions that dictate the direction of seemingly disparate and unrelated decisions, (that) shape the change process by emergent rather than “planned” change’ (2001: 14).
The evolution of HINST
HINST was founded in 2007 with the explicit purpose of helping the local teams working in the most deprived fifth of local authority areas to reduce the gap in health inequalities. It worked, therefore, with the 70 Spearhead areas. These were selected on the basis that they fell into the lowest quintile of all English local authorities in at least three of the following five criteria:
(p.157) • male life expectancy at birth;
• female life expectancy at birth;
• cancer mortality rate in under-75s;
• cardiovascular disease (CVD) mortality rate in under-75s;
• Index of Multiple Deprivation 2004 (Local Authority Summary), average score.
The 70 Spearhead areas were in the North, the Midlands and London (see Figure 7.1).
HINST worked to an explicit public service agreement (PSA) target, which was refined over time to include a mixture of improvements in absolute terms (such as mortality rates), and relative terms (such as the gap in inequalities between authority areas). As the PSA target deadline of 2010 came closer there was a discernible change of focus in the recommendations made through this target, emphasising those things that would make a difference to the target population in the time that remained.
(p.158) • Interventions were recommended that had to be applied systematically and on an ‘industrial’ scale to have sufficient impact on life expectancy at a population level.
• The potential impact of interventions on the number of lives to be saved was calculated, enabling choices about local priorities and the scale of change against investment to be planned.
• There was an intensified focus on health service changes that could be made which would add years of life to those who were otherwise facing an early death.
Recommendations made by HINST to the Spearheads it visited included a balance of actions ranging from those that would have an immediate impact (within the timeframe of the PSA target) through to those which would have a more sustained impact over time, as Figure 7.2 shows.
The lessons that had been learned about delivering short-term targets
(p.159) The outcome
Following the change in government in 2010 the PSA target was dropped in favour of local targets before the end of its timeframe, and national monitoring ceased. However, the Public Health leadership in North West Region maintained their component of the national target as an interim measure, while local targets were developed, and analysis has continued there. This shows that the Spearhead life expectancy target would not have been reached within the timeframe planned, but there were some successes.
The best analysis so far is given by Tom Hennell, Senior Public Health Analyst, Department of Health North West46 in a presentation that concludes:
The good news
• The national Spearhead target for male life expectancy appears to have been met in 2009–2011.
• Although the national Spearhead target for female life expectancy appears to have been missed in 2009–2011, there are signs of narrowing gaps since 2006–2008.
• Whatever interventions Spearhead areas have been applying to reduce inequalities in public health over the last four years, they should keep them up, because they’re working.
• The three cause-specific objective targets for 2010 were achieved in Spearheads in areas in 2009.
The bad news
• Progress is locally variable; with the most successful narrowing of gaps in London and the North East.
• Improvement in the female cancer mortality gap is badly lagging that for the male cancer mortality gap.
Principles that underpin the HINST approach
Principles for intervention
HINST evolved through its interaction with local teams – drawing on and disseminating a growing body of effective practice to underpin its recommendations. The team emphasised that interventions within a (p.160) given population by a given date would require systematic programmes of action that are known to be effective and reaching as many of the target audience of people as possible who could benefit.
Programme characteristics included being:
• evidence-based – concentrating on interventions where research findings and professional consensus are strongest;
• outcomes-oriented – with measurements locally relevant and locally owned;
• systematically applied – not depending on exceptional circumstances and exceptional champions;scaled up appropriately – ‘industrial-scale’ processes require different thinking from small-scale projects or pilots (‘bench experiments’);
• appropriately resourced – refocusing on core budgets and services rather than short bursts of project funding;
• persistent – continuing for the long haul, capitalising on, but not dependent on fads, fashion and changing policy priorities.
Interventions can be delivered through three different approaches to drive change (focused on the population, individual and community levels) as illustrated by Figure 7.3.
(p.161) Three approaches to population-level change
Direct population-level interventions include developing healthy public policy, legislation, regulation, taxation and public funding strategies. These elements should support making ‘healthy choices easy choices’ for individuals and communities.
The impacts of such population-level interventions, however, will not automatically ‘trickle down’ to everyone, often missing those who are socially excluded for various reasons. Strategies for targeted communication and education, service support and even enforcement may be required to achieve full impact.
Individual approaches through services
Some interventions taken up at the individual level – such as support for behaviour change, therapies, treatments and rehabilitation – can improve individual risk of premature mortality significantly, in some cases by 30–40%. The challenge is to achieve enough of those individual successes that they add up to percentage change at the population level. This will be achieved only if services take into account issues of system and scale to enable this to happen, and work to address population-level outcomes as well as those for individual service users.
Improvements in health and wellbeing will require some reorientation of health and other services to take a more holistic view of individual circumstances. These should take account of any personal characteristics/subpopulation group status or socioeconomic status and focus on development of personal skills of staff and service users, thus promoting healthy choices and actions.
Individuals will only choose to use and benefit from certain behaviours and actions if those behaviours fit with the cultural and belief system of their own community. Communities can be based on place (neighbourhood, school, workplace), culture (ethnicity, faith) and others (disability, sexual orientation). Community development is one way of facilitating communities’ awareness of the factors and forces that affect their wellbeing, health and quality of life.
Community engagement is often patchy, favouring those communities that already have leadership, organisation and some (p.162) resources. Instead, it needs to be systematic in bringing top-down and bottom-up priorities together into effective action plans. The aim should be to strengthen community action by creating more supportive environments for health improvement and developing the knowledge and skills of community members.
Service links into communities can be superficial, of poor quality, unsystematic and based on low levels of understanding. Connectivity between services can be disorganised and confusing. Use of the voluntary, community and faith sectors as a bridge between services and community-based structures needs to be systematic and based on need rather than supply. Effective commissioning is key to realising these objectives.
A holistic approach to commissioning: the ‘Christmas tree’ diagnostic
Substantial progress can be achieved in making an impact in the short, medium and long term in relation to inequalities in mortality and life expectancy through a focus on existing services. Because of this, extra attention is given here to extracting maximum benefit from delivery of interventions for which there is strong evidence base of effectiveness. In addition, there is a deliberate emphasis wherever possible on improving access to services at a scale that will bring about a population-level improvement in mortality and life expectancy within a 2–3-year period.
The detail is illustrated in the following diagram of commissioning services to best population-level outcomes, otherwise known as the ‘Christmas tree’ diagnostic, with an accompanying description of its component principles. The framework balances two sets of factors that determine whether optimal outcome can be achieved at the population level from a given set of personal health interventions (see Figure 7.4).
The right hand side of the diagram (1 to 5) – a challenge to providers – links the factors that will influence health service outcomes: that is, how can we construct the most effective service? However, optimal outcomes at the population level will not be obtained without combining this service focus with the other side.
The left hand side of the diagram (6 to 10) – a population focus – identifies those factors that determine whether a community makes best use of the service provided; for example, whether the benefits of personalised improvements to services are having a systematic impact on reducing health inequalities at the population level.
The balance between the two sides of the diagram is critical in determining what outcomes are secured, and designing an effective balance is the commissioning challenge: aiming for equality of (p.163)
The central elements of the diagram are concerned to ensure that, when the most effective services/interventions are identified that are fully acceptable, accessible and effective in terms of take-up and compliance, there is adequate capacity to meet the need. Effective leadership and networks are needed to ensure that all of these elements are being kept under review and to encourage continuous improvement and equality of morbidity and mortality outcomes.
The next section draws on some of the effective practice identified by HINST to illustrate each of the component parts of the Christmas tree model.
There is a well-established and strong evidence base in areas such as coronary heart disease, CVD and diabetes, which makes the case for those interventions that can be effective in reducing mortality. HINST commissioned modelling of a number of these to demonstrate the potential impact of the most effective of these interventions, and the tangible impact that systematic and scaled applications could have on mortality rates at the population level in a short timescale. The model was applied to all Spearheads to show the potential impact of evidence-based interventions. These were used by HINST as the basis of recommendations for evidence-based strategies during Spearhead visits.
Blackburn with Darwen Council/PCT, for example, addressed its variance from the inequalities target and used a systematic approach to quantify the impact of different, costed interventions on mortality, using the HINST evidence base. It produced a costed, evidence-based strategy that was adopted by the primary care trust (PCT) and local authority, based on HINST calculations of the numbers needed to be treated to postpone the death of a given number of people, using a tested set of interventions (see Figure 7.5).
Once the evidence base is established, the impact of its application can still vary considerably, from place to place and even from clinician to clinician within a general practitioner (GP) practice, for example. HINST placed heavy emphasis on reducing local variation in primary care performance. It was possible to use the Quality and Outcomes Framework (QOF), collected as part of the GP contact, to identify significant variation and drive change. Figure 7.6 illustrates the difference between an area where there was unmanaged variation and, in contrast, the performance of Wakefield PCT. Here, GPs and primary care, supported by the PCT CVD team, were able to bring the performance of all practices up to the level of the best.
It is important that interventions intended to produce population-level impact can be cost-effective, so that they can be deployed at sufficient scale. HINST encountered examples where large gains could be achieved by improving the efficiency of current spending on such interventions. For example, Rotherham PCT illustrated the relationship between the average expenditure per patient with diabetes and their clinical outcomes, such as their level of blood sugar control (see Figure 7.7).
This diagram illustrates that there was no statistical relationship between the level of spend and outcome. The PCT was able to analyse the different prescribing patterns of GPs with high costs and poor outcomes, compared to the more cost-effective prescribing of those with low costs and good outcomes. The less cost-effective practices were then supported to improve their approach to prescribing to achieve a population health gain, while at the same time reducing costs.
Bringing services that deliver cost-effective interventions as close to the target population in need as possible is an important way of reducing barriers to access. In delivering its smoking cessation support programme, Knowsley PCT was providing the service through four centres across the district. The uptake was not sufficient to meet the PCT’s ‘quit target’ or to achieve their ambition to reduce significantly the prevalence of smoking, and its impact on mortality. They therefore changed their strategic approach in order to achieve a (p.166)
The Knowsley campaign generated a 154% increase in numbers coming into the service in the first six months, and almost a 200% increase in smoking quitters. (p.167)
Engagement with users and potential users of services is critical to ensure that services meet the needs of patients rather than those of the delivery systems themselves. As part of this it is important to connect with those who failed to take up an offer of service, or dropped out of a service programme, rather than those who are easier to access or more compliant with the current service design.
Four phases of rehabilitation are recommended following a heart attack or heart surgery. In most systems patients can be lost at various points along the pathway. In Sunderland they had managed to reduce this attrition substantially through the first three stages of the process, but then recorded a dramatic fall-off before stage IV. This stage was the point at which the venue for rehabilitation moved from a health care setting into leisure services. An exercise of tracking patients who failed to attend at stage IV uncovered real concerns that they were worried about leaving the perceived safety and care of professional health care staff to transfer to non-medical leisure services staff.
This was addressed by bringing the leisure services staff into the programme at stage III. This enabled patients to get to know leisure services staff and build confidence in their ability to maintain levels of (p.168) care before the transition. Attendance for stage IV of the programme increased significantly.
Population needs assessment
Population health programmes require interventions to be delivered in complex environments. Elements in these environments will influence the success or otherwise of delivery. HINST found that the public health evidence base has not paid sufficient attention to defining the context of delivery.
Public health teams in South Yorkshire worked with Sheffield Hallam University on cluster analysis of previously defined ‘natural neighbourhoods’. By modelling a range of measures available at the neighbourhood level the team was able to identify 20 cluster types, each of which was distinct in its characteristics from the others (see Figure 7.8).
The importance of this was that neighbourhoods of the same cluster type/colour had more similar characteristics to each other than they had to their geographical neighbours. It followed that interventions that were effective in a neighbourhood of one cluster type could be tried in other neighbourhoods that shared that cluster type. This could then be used as the basis of planning interventions
(p.169) Seven neighbourhood cluster types were identified:
• older large estates
• new estates
• rural and small towns
• ex-coalfields communities
• mixed young families
• established non-Caucasian ethnic
• mobile young.
It can be seen that the basis of deprivation in an ex-mining community is likely to be different from that of a neighbourhood largely comprised of ethnic minorities or an inner-city estate.
Figure 7.9 gives a stark representation of the problems that HINST came to describe as ‘the missing thousands’. The map of the City of Birmingham on the left gives the number of patients registered with their GP as having coronary heart disease. The map on the right shows the standardised mortality rate due to heart disease. The two together demonstrate the complete mismatch between the level of expressed demand registered with GPs and the actual level of need.
Even if the quality of services is good, if patients drop out and do not connect with them appropriately they will have little impact. The best example HINST saw of a systematic initiative to address this problem
As a result, within the year 85% of the eligible population was screened. This was balanced across all districts of the borough, including the most deprived areas. Dramatically, some 800 people were newly diagnosed with diabetes and some 1,200 with renal disease. This previously undiscovered reservoir of unmet need could then be registered, thereby enabling GPs to provide treatment systematically for people who were previously missing out.
The issue of the ‘missing thousands’ is one that was found to be almost universal across all the Spearhead areas visited. Hence, HINST began to represent the scale of this problem using data shown in Figure 7.10. This diagram builds on the chart showing the intermediate outcomes of health care in Figure 7.6 earlier, but in addition, the lightest grey section of each bar provides an estimate of how many patients with each condition are currently missing from the relevant practice disease register.
In addressing health inequalities it is important to tailor the resources available to meet differential needs across the population. However, the focus needs to be less on equality of inputs, according to need, and more on the resource necessary to achieve equitable outcomes. HINST (p.171) found that in most areas a proportion of the population containing the most vulnerable people living in the most chaotic environments needed a disproportionate amount of resources to achieve the same sort of outcomes that were possible for others.
An important example was found in the QOF incentive scheme, which rewards achievement of only up to 70% of target blood pressure in people with coronary heart disease. The problem from a health inequalities perspective is that many of the most vulnerable people are likely to remain in the unmanaged remaining 30%. This was addressed, again in Bolton, by recasting the incentive scheme to reward effort on an exponential scale that heavily incentivised reaching successful outcomes with people who were hard to reach (see Figure 7.11). An enhanced payment system was introduced to maximise CHD assessments by GPs. It can be seen from Figure 7.11 if a Practice achieves only 50% coverage they are paid £2 per patient and if it achieves 90% it is paid £9 for every single patient, recognising the exponential effort and resource that is needed to include hard to reach patients.
When patients do come forward and present to services it is critical that their early contacts are positive and engaging, to ensure that the opportunity is taken to diagnose their illness at a point when it is accessible to treatment. In Doncaster, where lung cancer mortality among men is a significant problem, it was found that most patients were presenting too late for effective treatment. Part of the solution (p.172) was to raise the knowledge and expectations of the population, to encourage people to come forward (increasing ‘expressed demand’). However, Suckling (2010) has suggested that patients often did not think it worthwhile to come forward as their GP, when faced with new chest symptoms, would just prescribe antibiotics without carrying out further tests. A second study showed that because a large proportion of these patients were smokers and had a chronic cough the GPs rarely referred them for X-ray (which could potentially have detected cancers at an earlier, treatable stage), believing that it would be unlikely to be helpful in many cases.
A systematic programme was launched to raise the awareness of GPs of the National Institute for Health and Care Excellence (NICE) guidance on the indicators of early lung cancer and the criteria for referral for chest X-rays, which had been recently published. This led to a significant change in practice. The hospital radiology department was included in the initiative so that it was able to manage the increased number of referrals resulting from a change in the threshold of referral.
The increase in patient self-referral, combined with more responsive frontline services, resulted in a dramatic shift in the percentage of patients potentially eligible for surgical treatment from 11% to 19% (see Figure 7.12).
Once diagnosis is made and care packages are prescribed, patients need to manage their use of the interventions themselves for much of the time. If not used appropriately, outcomes will be limited, however (p.173) potentially effective the planned treatment was. When a range of medications remains untaken ‘in the bathroom cabinet’, for example, the predicted outcomes will not be achieved. Preparing patients and supporting them to self-manage medication and other treatment should be an important part of the care plan. However, the HINST experience is that this was often not the case.
A significant example was provided by a major teaching hospital, which has received accolades for the quality of its self-management training for patients diagnosed with type 2 diabetes. Patients provided very positive feedback and the service was held up as an exemplar for others on how to run such a service in innovative ways. However, further review showed that the capacity of the programme was just over 200 patients per year, while the number of newly diagnosed diabetics was over 1,000 per year. This was reflected in the population outcomes, when the proportion of patients whose blood sugar levels were within target levels was shown to be well below 50%.
From a population perspective it would be important to provide effective self-management training at a scale to make it available to the majority of the population, rather than providing excellent training to a relatively small minority.
Adequate service volumes
HINST predicted, and subsequently observed, that for action to make an impact at the population level it needs to be systematic and resourced at the right scale. It is important that adequate service volumes are commissioned to aim for, and ensure, acceptable access times. ‘Process mapping’ can assist in the commissioning process here. For example, have all partners involved in delivering diabetes care carried out a process mapping exercise to help identify areas where there are inefficiencies, lack of resources or duplication? Have the outcomes been acted upon? A second approach to developing adequate service volumes is related to ‘walking the patient journey’. For example, have techniques been used to gain a real picture of the issues and problems for patients as they move through the care pathway (for example, patient shadowing, tracing the patient’s journey through a study of records)?47
Balanced service portfolio
Patient pathways of care and the more complex array of support required, for example in the management of long term conditions, can sometimes be baffling for patients. There may be gaps in the pathway (p.174) or barriers to parts of their care programme. It is clear that patients in the most complex and vulnerable situations will be among the first to fall at any of these hurdles. This was found to be the case by the Wakefield Diabetes Care Service.
When the capabilities of GPs were reviewed across the patch they found that there was great variability on the level of knowledge and skills relating to diabetes care. They were able to assess these and grade the capability of each practice on a multi-point scale.48 In this respect, a Level 1 practice would be limited to a smaller number of capabilities than a Level 4 (these capabilities are outlined in Figure 7.13). This meant that, if all patients were to receive their required level of service, a variable level of compensatory input was necessary from secondary care specialists (see Figure 7.13).
However, assessment suggested that services did not match exactly and that patients would ‘fall through the gaps’ (see Figure 7.14).
In an agreed programme the secondary care clinicians came into primary care to provide a period of joint management of complex patients together with the patient’s own GP. This raised skill levels across primary care demonstrably, so that after two years no practices were classified within the lower levels of capability 1–2 (see Figure 7.15).
The resulting agreements on joint working and increased clinical confidence in primary care led to significant improvements in clinical outcomes for patients and greater levels of professional satisfaction among the practitioners
Networks, leadership and coordination
Networks provide a framework for identifying and sharing effective practice. HINST worked with the Merseyside and Cheshire Cancer Network to scope the ways that cancer networks could have an impact on health inequalities. This Network was chosen because it had a well-developed strategy for reducing heath inequalities, with a range of effective, practical tools.49
The steps identified in the development of a coherent strategy to reduce mortality from cancers were as follows.
• Develop a clear vision shared across the Network about what the future could look like – taking everybody forward together.
• Develop a strategic plan which prioritises areas of greatest need where change is possible – tackling shared problems together.
• This needs to be based on best evidence and regional experience – taking into account cost-effectiveness. (p.175)
(p.176) • The strategic plan also needs to include a transparent process for reviewing progress and feeding back to the constituent members of the Network – including the emerging commissioners.
Cancer networks have the potential to establish a direct link between poor outcomes at the population level and conventional service improvement approaches by:
• identifying anomalies in outcomes (particularly when incidence is average);
• producing explanations for them;
• recommending interventions of sufficient scale to have a quantified impact on mortality;
• targeting the patient pathways that result in poor outcomes;
• engaging all stakeholders (patients and potential service users, clinical groups, primary care, commissioning and public health) involved along the pathway in a comprehensive action plan;carrying out a series of major network locality change management events to ensure translation of needs into action and sharing of good practice;
• assisting in obtaining finance for the key actions.
The variations that exist are startling – Figure 7.16 shows an example of the route that patients took to reach the specialist treatment they needed, by GP practice. Networks, working with NHS England and clinical support units, are ideally placed to explore why some practices are able to manage patient referral – and may have effective practice
• the work could link to work getting underway through National Cancer Action Team (NCAT) on the Map of Medicine – a library of evidence based practice for clinicians. For interest see: http://mapofmedicine.com/about;
• the ambition should be on a ‘spine’ for the whole pathway, from patient awareness and decision to present with symptoms/signs, through reception and management in primary care, to referral and engagement by specialist services, living with cancer, and finally end-of-life care;the emphasis would be determining fixed elements of process, but allowing room for substantial variation by locality;
• the pathway should define the ‘what’, leaving the ‘how’, where appropriate, to local ownership and innovation.
These underlying principles apply to most clinical networks, which are going to be in a critical position in having both an overview vantage point to be able to identify local effective action and those areas where there are issues at one of the other 13 points of the Christmas tree diagnostic.45
The new environment: requirements for delivering population-level change through the new policy structure
Localism is the foundation of the public health system introduced in 2013, and progress against health inequalities will require strong local partnerships, led by local government, to address the ‘wicked’ public health problems that sustain health inequalities. The scale of these problems is such that they will require industrial-scale interventions, rather than piecemeal initiatives. Many partnerships have achieved great success, but many have not and will struggle to do this without support, given the fiscal environment of the austerity current measures. The new public health system will not have a centralised performance-management system and Public Health England (PHE), in its role supporting transformational change locally, will have to build and adapt approaches that enable partnerships to choose their own priorities, and come together and review their collective endeavours against international evidence and the successes of others elsewhere.
(p.178) The approach of HINST is consistent with these principles. Ten aspects of the HINST approach are now highlighted that may be of use to PHE and others working to support local partnerships in the new environment. Applying the evidence-based learning from HINST to the emerging environment, it is possible to identify the following ways in which the new public health system could be developed to address health inequalities.
Achieving percentage change at the population level for any selected outcome means taking account, particularly, of system, scale and sustainability of evidence-based interventions. Delivering within the new policy structures may require the following issues to be addressed.
Governance: who is running the show?
If health and wellbeing boards (HWBs) are to deliver on the health and wellbeing strategy they will need to establish strong governance arrangements. As statutory bodies they will be held accountable for their own delivery by local residents/electorate, by constituent bodies such as the cabinet and clinical commissioning groups (CCGs), and through some national agreements – for example, PHE health premium plans. Style will include decisions as to whether they constitute a ‘strategic forum’ for collective action, or whether they develop a performance-management function with which to deliver integrated programmes
Joint strategic needs assessments
Joint strategic needs assessments (JSNAs) should start from a systematic balanced, non-judgemental overview of health, social care and wellbeing issues, followed by a more detailed drilling down on emerging important issues. They should combine a top-down more quantitative analysis with bottom-up more qualitative inputs from communities, service users and frontline staff. Each prioritised topic should be broken down into meaningful causative elements that might be addressed. JSNAs should not stop at describing problems, but should proceed to highlighting the ‘so-what’: possible consequences and potential actions. Ideally, needs-based assessment could be balanced with an assessment of assets. A JSNA should be a process, not a document (or website). Findings should ideally move from dry analysis to ‘marketing’ the findings to a range of audiences (elected members; local authority officers; GP commissioners; HealthWatch and the public).
(p.179) Priority setting: how does it really work?
Priority setting will usually involve a consultation process and debate, for which officers will be involved in making cases and options appraisal. This is the part of the process where corporate ownership of issues is won or lost, reaching for the hearts and minds of members. The ‘anatomy’ of a decision is said to be in three parts: evidence, ethics and politics. Case makers will need to take into account all three elements, not just rely on evidence. Debate in a local authority setting is more often emotive rather than scientific/technical. A good example of working with this is referring to ‘the number of men or women dying’ for comparative purposes, rather than ‘differences in mortality rates projected for each sex’.
Setting targets: locally relevant and meaningful
In setting targets, choosing testing but achievable benchmarks for change (for example, levels achieved by ‘best in class’ for Office for National Statistics (ONS) peer authority cluster) and setting ‘SMART’ (specific, measurable, achievable, realistic, timely) objectives, with numbers where possible, are both important issues to address. Where relevant and it makes sense, national outcomes frameworks and indicators should be taken into account. Distributional factors need to be considered, via use needs assessments and equity audits to establish the need for graded or targeted responses within an area. Comparison should be meaningful to local stakeholders. Offers of support to local partnerships need to combine those areas the local partnerships want to work on with those identified as having the biggest potential impact on life expectancy. The two may be different. In the new public health system there is no mandate for centrally defined, hierarchically managed targets. Support from regional and national bodies will have to incorporate locally defined priorities.
Whole system approach
Complex programmes for population-level change will not all be delivered through conventional services, and may well involve all three points of the population-level intervention triangle:
• population-level interventions (healthy public policy, legislation, regulation, licensing);
(p.180) • systematic and scaled intervention through primary, community and hospital based services; wraparound social care; third sector support);
• systematic community engagement.
The triangle (see Figure 7.3 earlier) could provide a useful framework and specification for the population-based working of the HWB.
Select interventions: strongly evidence-based
The evidence base should be reviewed to establish interventions that could make a substantial contribution to the target. This should capitalise on evidence sources across the sectors involved (including health, social care, housing and community engagement). The size of the potential contribution for each proposed intervention should be modelled or estimated. The scope for improvement utilising the proposed interventions should explored, and consideration given to what systems will be necessary to deliver the scale of intervention necessary.
Develop a business plan: economic case for change
With systems, scale and timescale established, cost and potential return on investment should be modelled to make a viable business case. It may often be a case for changing focus of existing programmes, rather than starting a whole new programme afresh. In standing up to competition for resources, the persuasiveness of the case can be supported through a well-managed and widely owned prioritisation process (see above). Finally, the business plan will need to take into account the different business planning models and cycles of the component sectors and organisations.
Programme planning: who is responsible?
The components of strategy contributing to a priority outcome should be brought together in a defined cohesive programme. Different components will often be delivered by different organisations. However, it would be preferable if there were a single empowered lead answerable to the local Health and Wellbeing Board for each prioritised health and wellbeing strategy outcome.
(p.181) Information governance: sharing intelligence
The health and wellbeing strategy would benefit from being accompanied by a number of support strategies, in particular:
• an information governance strategy exploring and facilitating the flows of data and information around the system, necessary for delivery of the strategy and its programmes;a communication strategy, marketing information through partner organisations and the wider public, about the JSNA findings, and the health and wellbeing strategy and its progress.
Maximise impact: minimise inequalities
In relation to delivery through services, there are many factors that influence whether an intervention, or group of interventions, has the optimal impact at the population level. HINST used a checklist of 13 factors, pulled together as the Christmas tree diagnostic (see Figure 7.4 earlier). Many of the 13 factors in this diagnostic are often neglected in designing and running programmes, resulting in patchy delivery, well below the programme potential. HWBs could provide a milieu, through connectivity between services and with community infrastructures, to develop population connectivity with services (left hand side of the diagnostic) which they have previously not been utilising appropriately.
Concluding discussion: capitalising on the hinst model
In the new policy context there has been a move away from nationally defined targets, towards localism. Desired outcomes of policy are to be locally defined and owned, with encouragement to set some of them within a set of national outcomes frameworks. Accountability will be more to resident populations and their elected representatives, rather than through performance management to national government. However, such accountability will still need to be demonstrable, with significant change at the (local) population level. It will also, pragmatically, need to demonstrate measureable progress within the short timescale between local elections. Localism offers the benefits of local ownership, relevance and accountability. However, among the possible dis-benefits of such a ‘devolved’ structure can be isolationism, patchiness and variability.
(p.182) The model of support offered by the national support teams, including HINST, could be particularly relevant in this new environment, for a number of key reasons.
• The teams worked separately from any performance-management structures, bringing an external perspective and an approach that was ‘high challenge; high support’.
• The teams were linked to ‘the Centre’ (the Department of Health), bringing empowerment and credibility, but not of the Centre and feeding back judgements.
• The local area leadership was designated as the ‘client’, with feedback and recommendations directed to them only, and onward distribution by them at their discretion.
• Once trust in these principles was established, it was possible to get more honest, and less sanitised, inputs through interviews and workshops.
• Using the structured approaches to assessment and inquiry, the teams were able to identify good practice as well as common gaps and barriers to progress. These could then be disseminated:
• horizontally, to other local areas by matching local specific needs to relevant good practice;
• vertically, assimilating frontline experience and feeding back to the Centre in aggregated and anonymised form, to influence policy.
• Beyond assessment and recommendation, teams were able to use the principles of change management to influence and ‘unblock’ local politics – for example, by introducing external perspectives and benchmarks to local leaders and opinion formers.
In the context of this book the work of HINST offers an insight into the issues encountered in developing an applied approach to health inequalities. The HINST experience suggests that it is possible to narrow gaps in health outcomes based on policy action. The Team did not have the remit to test the alternative premise, set out in chapter one above, that health inequalities are a wicked problem that require change in the structure of the economy and society. However HINST came to believe that the systematic approach they developed could impact on the range of components that constitute the complex problem that is health inequalities.
(45) This diagram and those following in this chapter have been published previously in a Department of Health resource pack of best practice guidance (DH, (2010) Redoubling efforts to achieve the 2010 National Health Inequalities Life Expectancy Target, Gateway reference 13785).
(46) Tom Hennell, Senior Public Health Analyst, Department of Health North West, Narrowing the health inequalities gap: What went right and what went wrong?, in Symposium: ‘Where next for health inequalities research?’, 5–6 December 2012.
(47) A comprehensive range of tools to facilitate improvement of pathway design and management is available on the NHS Institute for Innovation and Improvement website, available at: www.institute.nhs.uk/option,com_quality_and_service_improvement_tools/Itemid,5015.html.
(49) For more information on the HINST approach to reducing health inequalities due to cancer, see the report produced through the collaboration with the National Cancer Action Team: How to reduce excess mortality from cancer in areas with the worst health and highest levels of deprivation, available at: http://ncat.nhs.uk/sites/default/files/work-docs/HINST%20NCAT%20How%20to%20Cancer%20final%20_%2009032011.pdf.