The organisation and delivery of health improvement in general practice and primary care: a scoping study

Article history

The research reported in this issue of the journal was funded by the HS&DR programme or one of its preceding programmes as project number 09/1001/52. The contractual start date was in February 2011. The final report began editorial review in January 2013 and was accepted for publication in July 2014. The authors have been wholly responsible for all data collection, analysis and interpretation, and for writing up their work. The HS&DR editors and production house have tried to ensure the accuracy of the authors’ report and would like to thank the reviewers for their constructive comments on the final report document. However, they do not accept liability for damages or losses arising from material published in this report.

Declared competing interests of authors

none

Copyright statement

© Queen’s Printer and Controller of HMSO 2015. This work was produced by Peckham et al. under the terms of a commissioning contract issued by the Secretary of State for Health. This issue may be freely reproduced for the purposes of private research and study and extracts (or indeed, the full report) may be included in professional journals provided that suitable acknowledgement is made and the reproduction is not associated with any form of advertising. Applications for commercial reproduction should be addressed to: NIHR Journals Library, National Institute for Health Research, Evaluation, Trials and Studies Coordinating Centre, Alpha House, University of Southampton Science Park, Southampton SO16 7NS, UK.

Structure of this report

The main elements of the report provide details of the literature review. The review was developed from a reading of all selected papers, but not all these are referred to in the report because of the substantial amount of material. Chapter 3 provides details of how we approached the literature and dealt with the complexities of conducting the review. Chapters 4–6 provide details of the literature. Chapter 4 describes the main aspects of the literature including the quality, topic areas and extent of the literature found. The chapter also includes a brief outline of some of the key themes emerging from discussions with practitioners. Chapter 5 provides details of the range of activities identified in the literature. Chapter 6 provides details of the evidence about effective ways to organise and deliver health improvement interventions in general practice or by the primary health care team. Given the changes currently being introduced in the English NHS, in Chapter 7 we examine how these changes impact on the public health role of general practice. The aim of this chapter is to set the findings of the scoping review within the context of the recent changes to the organisation of public health and health-care services in England resulting from the Health and Social Care Act 2012. 21 Finally, in Chapter 8 we identify some of the key messages arising from the review and set out the implications for practice and further research.

Defining public health and primary care

There is some ambiguity about whether the term ‘public health’ is primarily about NHS activity or something wider. This ambiguity reflects a longstanding debate about the nature of public health itself. The most widely used definition of public health is that provided in 1988 by Acheson,42 who defined public health as ‘the science and art of preventing disease, prolonging life, and promoting health through the organised efforts of society’ (p. 1). The UK Faculty of Public Health endorses this definition; it suggests that public health is population based; it suggests that there is a collective responsibility for health, protection and disease prevention; it accepts the key role of the state; and it suggests that tackling public health problems involves a concern for the underlying socioeconomic and wider determinants of health as well as disease, and involves partnerships between all those who contribute to the health of the population. The faculty identifies three key domains of public health practice:

1. health improvement

   1. inequalities

   2. education

   3. housing

   4. employment

   5. family/community

   6. lifestyles

   7. surveillance and monitoring of specific diseases and risk factors

2. improving services

   1. clinical effectiveness

   2. efficiency

   3. service planning

   4. audit and evaluation

   5. clinical governance

   6. equity

3. health protection

   1. infectious diseases

   2. chemicals and poisons

   3. radiation

   4. emergency response

   5. environmental health hazard.

In the 2010 Public Health White Paper,37 the government set out five domains of public health which were to be covered by a public health outcomes framework:

• domain 1 – health protection and resilience: protecting people from major health emergencies and serious harm to health

• domain 2 – tackling the wider determinants of ill health: addressing factors that affect health and well-being

• domain 3 – health improvement: positively promoting the adoption of ‘healthy’ lifestyles

• domain 4 – prevention of ill health: reducing the number of people living with preventable ill health

• domain 5 – healthy life expectancy and preventable mortality: preventing people from dying prematurely.

The scope of this review falls mainly in the area of health improvement and domains 3 and 4 outlined in Healthy Lives, Healthy People. 37 However, identifying clear-cut public health activities or responsibilities is complex, and many activities overlap the domains defined by the Faculty of Public Health and those in the public health outcomes framework. For this review, the definition of health improvement draws on that used by the World Health Organization, first outlined by Leavell and Clark in 1958. 43

Primary prevention: activities designed to reduce the incidence of an illness in a population and thus to reduce, as far as possible, the risk of new cases appearing and to reduce their duration. Primary prevention can be defined as the action taken prior to the onset of disease, which removes the possibility that the disease will ever occur. It signifies intervention in the pre-pathogenic phase of a disease or health problem. It includes the concept of ‘positive health’, a concept that encourages achievement and maintenance of an acceptable level of health that will enable every individual to lead a socially and economically productive life. Primary prevention may be accomplished by measures designed to promote general health and well-being, and quality of life of people, or by specific protective measures.

While not all of these areas of activity relate to primary care or even specifically general practice, key activities such as lifestyle advice, nutrition and health are as much part of practice as immunisation, safe use of drugs, etc.

Secondary prevention: activities aimed at detecting and treating pre-symptomatic disease. It is defined as an action which halts the progress of a disease at its incipient stage and prevents complications. Secondary prevention attempts to arrest the disease process, restore health by seeking out unrecognised disease and treating it before irreversible pathological changes take place, and reverse communicability of infectious diseases.

There is an obvious and clear link between the role of general practice and primary medical care, and secondary protection. The individual focus of general practice, especially in the UK, places the primary care practitioner in a key position to identify areas for secondary prevention and, where relevant, undertake action.

Tertiary prevention: activities aimed at reducing the incidence of chronic incapacity or recurrences in a population, and thus to reduce the functional consequences of an illness; including therapy, various rehabilitation techniques and interventions designed to assist the patient to return to educational, family, professional, social and cultural life. This level involves the treatment and monitoring of people with disease, and general practice plays a key role in this in the UK. For the purposes of this review we have categorised this as treatment and outside our review.

However, strict demarcation between levels is not always possible, as some activities cut across these levels. For example, protecting others in the community from acquiring an infection can involve providing secondary or even tertiary prevention for the infected and primary prevention for their potential contacts. Some activities do not differentiate between levels so promoting smoking cessation is directed at those without any health problems as well as those already identified as having a specific disease. Froon and Benbassat44 have argued that, given the inconsistencies in the way the different definitions of levels of prevention are used in practice, this approach should be abandoned. They argued that interventions should be defined by their objective, target population and type of intervention.

This refocusing of health promotion frameworks has already been promoted in government public health policy. In Choosing Health: Making Healthy Choices Easier,45 the government set out proposals for more integrated approaches to prevention and health promotion for children through Children’s Trusts and Children and Family Centres, and using schools as settings for health promotion working with children and their families. Primary care was also seen as an integral part of a community-based approach to health promotion. It is not clear, however, how far the shift has permeated current policy thinking in relation to primary care and public health activities of general practice. Current policy appears to have demarcated general practice with a stronger emphasis on the GP role and using consultation opportunities to address individual lifestyle issues. The 2010–15 coalition agreement The Coalition: Our Programme for Government35 announced that the Department of Health (DH) would strengthen the role and incentives for GPs and GP practices on preventative services, both as primary care professionals and as commissioners. As primary care professionals, GPs and GP practices play a critical role in both primary and secondary care prevention. They have huge opportunities to provide advice, brief interventions and referral to targeted services through the millions of contacts they have with patients each year. It is clear that current government policy continues to emphasise the important role of general practice and primary care that cut across the different levels of intervention and domains of practice.

Similarly, there is a longstanding debate about the nature of primary care. 46–49 In the UK, ‘primary care’ generally refers to services provided by general practice, dental practice, pharmacists and opticians. There is little discussion about the range of services that this encompasses. In 1994 the US Institute of Medicine defined primary care as the provision of ’integrated, accessible health care services by clinicians who are accountable for addressing a large majority of personal health care needs, developing a sustained partnership with patients, and practicing in the context of family and community’ (p. 15). 50 Starfield47 has discussed the parameters of primary care, focusing on the role of primary care in dealing with the whole person in a holistic way over the long term and providing co-ordination of care for individuals. Starfield47 identified four unique features of a primary care service: first contact access, person-focused care over time, comprehensiveness and co-ordination. Hogg et al. 51 suggest that other important aspects of primary care include patient–provider relationships as defined by communication, holistic care and an awareness of the patient’s family and culture. They also argue that primary care performance needs to be set within a broader structural environment that recognises the wider health-care system, the practice context and the organisation of the practice. This reflects an increasing acceptance of the role of the health-care delivery system, including issues of governance and accountability, resources, inter-relationships between primary care and other health and social care services, and person-centred care. 52,53 Kringos et al. 53 undertook a systematic review to identify the core dimensions of primary care, identifying 10 core dimensions. The review identified preventative and health promotion activities as elements of primary care and, in particular, how such activities are underpinned by dimensions such as co-ordination of care and equity. However, they concluded that a primary care system can be defined and approached as a multidimensional system contributing to overall health system performance and health. The broad nature of primary care and the difficulty of providing a clear definition were highlighted by Peckham and Exworthy. 48 In the UK, primary care is variously equated with general practice, the primary health-care team and a broader view of a wide variety of community-based services. For the purposes of this review we identified our focus as those health improvement activities that are undertaken in general practice or by staff located in practices but which are delivered outside the practice. This definition was not without its problems, and relating definitions to search terms was complex (see Chapter 3). There are, however, specific aspects of policy and practice that directly relate to the organisation and delivery of health improvement within general practice and primary care settings and are directly relevant to this review, including contractual issues in general practice, pay for performance, its organisation and staffing, and relationship to professional public health practice.

The general practitioner contract

Historically, GPs’ status as independent contractors to the NHS has meant general practice has been relatively free of direct management and monitoring compared with other parts of the NHS and has predominantly provided a demand-led sickness service. Thus, any focus on public health has tended to be more focused on a secondary level of prevention, although GPs have always provided general lifestyle advice, vaccination, etc., as a normal part of good practice. 48,54 GPs’ income was based on allowances, capitation and fees for a limited set of services, and the nature and quality of care was undefined and left to professional discretion. 54 There is a range of general incentives for GPs and the wider primary health-care team to engage in public health activities. Capitation payment systems (such as that for UK GPs) provide a general incentive for health promotion and managing long-term care problems to minimise inappropriate use of primary care services – particularly when compared with fee-for-service payment systems. 55,56 However, practice in the UK falls short of this aspiration. 57,58 In addition, wider influences such as national policy frameworks, guidelines, professional competencies, and practice guidelines and educational programmes seem to influence physician practice and encourage more preventative health interventions and practice, but the evidence is limited. 59

Before the introduction of the latest GMS contract in 2004, practices were given financial incentives to provide health promotion. For example, cervical cancer screening is undertaken predominantly within general practice by GPs and practice nurses. Target payments (for achieving 50% and 80% coverage rates) were introduced in 1990 as part of the revised GP contract and, while provision of this financial incentive increased coverage from 53% to 83% of GPs achieving the 80% target, these improvements bore no relation to local need. 60,61 However, for GPs and practices which found it difficult to achieve even 50% coverage, there was no real incentive to try and maximise the uptake of screening. 60

The new GMS contract of 2004 continued the trend of increasing regulation of clinical practice to improve quality of clinical care, with a pay-for-performance system rewarding practices for the quality of their services. 62 The main principles of the new contract were:

• a shift from individual-GP to practice-based contracts

• contracts based on workload management, with core and enhanced service levels

• a reward structure based on the new QOF and annual assessments

• an expansion of primary care services

• modernisation of practice infrastructure (especially information technology systems).

The contract is not universal, as there are alternative contract types alongside the GMS contract, including Personal Medical Services, which cover about 30% of practices and were developed in the 1990s; Alternative Provider Medical Services aimed at contracting private sector companies to run GP practices; and Specialist Provider Medical Services contracts for specific GP services such as those for homeless people. The core contract includes general patient care, with prevention seen as an integral part of the provision of such care. Over the years, specific aspects of prevention have been identified beyond clinical care, such as health checks for new patients. LESs also provided a way for PCTs to contract for specific areas of health promotion such as drug and alcohol services, extending stop-smoking services, breastfeeding initiatives and dealing with other locally identified public health problems. By 2009, LES expenditure represented a significant investment in locally defined services. 63

The QOF has also been used to target some preventative activities, although to date their extent has been limited. There were recommendations in the Marmot Review29 for greater use to be made of the QOF for reducing health inequalities. Currently, despite the inclusion of key clinical areas and an increasing emphasis on prevention, there is little evidence to support any substantial beneficial impact of the QOF on either preventative activities or reducing health inequalities. 64,65 However, there is some evidence to show that the QOF has contributed to a reduction in the delivery of clinical care related to area deprivation,66 and a systematic review of the impact of pay-for-performance schemes on health inequalities concluded that inequalities have largely persisted since the introduction of schemes. 67 The Healthcare Commission report68 noted that the QOF is unlikely to encourage practices to try and include the most ‘hard to reach’, since no further points are received once 90% coverage has been achieved. Typically, QOF points are awarded to reflect workload and consist of both a minimum performance threshold (payments beginning once 40% of eligible practice population have been treated) and a maximum threshold (e.g. 90% of those eligible) with graduated payments in between.

Few QOF indicators directly address health outcomes and primary prevention. The evidence suggests that the QOF has improved health outcomes for some conditions, but has had a limited impact on others, including adverse effects for population subgroups and non-incentivised activities. Current research also highlights the limitations of the QOF scores awarded in improving health outcomes, due to the indicators’ ceiling thresholds and the suboptimal clinical targets set compared with national clinical guidelines. 69,70 One key exception has been the inclusion of smoking targets. These were included in early versions of the scheme and was associated with the universal uptake by GPs for recording the smoking status and providing smoking cessation advice for patients with various comorbid factors [coronary heart disease (CHD), stroke or transient ischaemic attack, hypertension, diabetes, chronic obstructive pulmonary disease (COPD) or asthma and, from 2008, chronic kidney disease, schizophrenia, bipolar disorder or other psychoses]. Notably, the original QOF threshold for intervening with smokers without comorbidity factors was simply that status was recorded at any time. This was changed in 2006, when recording smoking status in non-morbid patients was required periodically (every 27 months), rather than ‘ever’, to attract payment. However, only one indicator is solely focused on the achievement of an actual health outcome (i.e. the number of epilepsy patients who have been seizure free in the last 15 months), whereas the remaining intermediate outcomes relate to targets which are an indirect measure of one’s health (e.g. cholesterol < 5 mmol/l). The QOF points available are also weighted towards particular conditions such as diabetes (88 points) and CHD secondary prevention (69 points), compared with scores available for COPD (30 points), depression (31 points) and other mental health conditions (40 points).

The QOF also allows GPs to ‘exception report’ to ensure that they are not penalised by patients who do not attend reviews of their medication or where patients are unable to tolerate medication, possibly because of side effects. Exception reporting is supposed to ensure that GP performance is measured only against a viable practice population. However, it is possible for reporting to be ‘gamed’ to exclude high-risk patients or those for whom GPs have missed targets, thereby decreasing the patient pool and maximising QOF scoring. 71

The organisation of preventative activities in primary care settings

A Cochrane review72 on the secondary prevention of ischaemic heart disease stated it succinctly: while the benefits of individual medical and lifestyle interventions are established, the effectiveness of interventions which seek to improve the way secondary preventative care is delivered in primary care or community settings is less so. A review by The King’s Fund18 examining the quality of general practice identified that a key challenge to improving the quality of public health and ill-health prevention is that there is a lack of evidence about the effectiveness of interventions carried out by primary care practitioners. The authors concluded that:

Evidence about the effectiveness and cost-effectiveness of public health interventions is growing . . . However, more needs to be done to help understand how general practice can effectively tackle ill-health prevention. General practice, public health practitioners and academics all have the responsibility to work together to improve this evidence base.

p. 3118

While it is clear that there is a substantial commitment, in terms of both policy and practice, for general practice to be engaged in public health activities, the literature tends to discuss this in one of three ways. The first way is in the focus at a conceptual level examining the links between primary care and public health. The second is the focus on the public health role of primary care professionals. Finally, the literature has examined the role of the practice as a location for public health activity. We examine each of these areas in turn.

Conclusion

The discussion of current issues raises a number of concerns about examining the public health role of general practice and primary care. These relate to two broad themes. The first is methodological. While there has been a substantial number of research papers and reviews on health promotion in general practice, the focus has been on the intervention. Very little attention appears to have been paid to the way interventions are delivered or organised. Reviews, in particular, have not addressed this organisational context. The second is that many reviews draw on international evidence and this raises important questions about their relevance to UK general practice. While not unique in terms of its structure and organisation, UK general practice has a particular organisational form, tradition and funding structure which are different from most other countries. Such differences – the QOF being a good example – are likely to create different incentives and contexts within which health promotion takes place.

It is also clear that much of the literature and research on health promotion focuses on the relationship between the practitioner and the individual patient. UK general practice operates within an individualised medical model. 48 Developments in recent years have tended to strengthen this focus, with an emphasis on individual, clinical (predominantly pharmaceutical) preventative interventions. Broader, more community-based approaches – such as those advocated by Julian Tudor-Hart,14 for example – tend not to be supported by policy or professional and regulatory guidelines. Following the Marmot Review29 there has been more policy interest in developing the public health role of general practice, and new QOF criteria to promote preventative activities are being developed by NICE. However, there remains a degree of uncertainty within general practice itself about its public health role and how preventative interventions are best delivered.

Introduction

Our methodological approach was based on the standard methodology for scoping reviews. 108,109 It consisted of two components: (1) literature review and (2) consultation with key stakeholders. The findings from these components were integrated to establish how well current knowledge mapped onto priorities and issues identified by stakeholders, to identify gaps in knowledge and future priorities.

The aim was to conduct a synthesis of the research evidence and wider literature on public health and primary care. Reviews of health improvement interventions pose numerous challenges due to multicomponent interventions, diverse study populations, multiple outcomes measured, the wide range of approaches and study designs used, and the effect of context on intervention design, implementation and effectiveness. 110 An initial scoping of the literature identified in excess of 20,000 papers that may be relevant to the topic area, and we developed strategies designed to manage paper selection and data extraction. The aim was to undertake a review of the extant literature on health improvement and primary care, focusing on the contribution of activities undertaken in or associated with general practice, wider practice and the primary health care team. However, following changes to the NHS and developments in public health service provision in England, the scope of the review was extended to include non-practice-based health improvement activity. This included exploring the roles of primary health care team members in the community and schools, and community pharmacists. However, the focus remained primarily on those activities associated with general practice, and our search strategies were based upon this central premise.

The review focused on examining the health improvement role of general practice in England, Scotland, Wales and Northern Ireland. We initially intended including English-language papers on health improvement roles in health systems with similar general practice organisational arrangements (e.g. Holland, Sweden, Canada, New Zealand), however, the initial literature searches identified a substantial number of papers and, given the specific nature of UK general practice and the need to produce findings relevant to the delivery and organisation of public health activities in UK general practice, it was decided to restrict the searches to UK studies only.

This study was not a systematic review of the literature but, in order to provide a comprehensive synthesis of the evidence, the team approached the review drawing on systematic methods. The conceptual complexity of both public health and primary care raised a number of initial problems in identifying key words for the literature search. In addition, the team also recognised that a focus on service delivery and organisation would create difficulties in identifying relevant literature. It was decided to approach the review in a number of stages including a hand search of selected primary care/public health journals and NHS public health reports. This was to be followed by an initial search of key databases and then broadened to cover all relevant databases. Dependent on the number of papers found, we would then develop a strategy for paper selection.

Ethics

Formal NHS ethical approval was not required, as the study sought information only on current service delivery. The study was classified as service evaluation as defined by current National Research Ethics Committee guidelines: Defining Research, published by the NHS Health Research Authority. 111 The research was conducted in line with ethical standards relevant to applied research and had approval from the London School of Hygiene and Tropical Medicine ethical committee.

Preliminary hand searches

To facilitate the development of search terms for the literature search, preliminary hand searches of key journals and PCT public health reports were undertaken. The purpose of this preliminary work was to create a more useful list of keywords for our search. We selected two recent director of public health reports from each strategic health authority: one from a rural PCT and one from an urban PCT (based on calculated population density). The reports were reviewed for references to ‘GP’, ‘general practice’, ‘primary care’ and ‘primary health care’. Each report was also skimmed to make sure relevant references were identified. We recorded any discussion of health problems in relation to general practice and any mention of general practice interventions as well as noting the common words or phrases used.

Journals included in the hand search were the British Journal of General Practice, Health Promotion International, the Journal of Public Health and Primary Health Care Research and Development. Tables of contents from 1990 to the present were skimmed for titles relating to health promotion or disease prevention services delivered in general practice or by general practice staff. Broad selection criteria were used, and all study types were considered. Full-text papers were downloaded and skimmed and, if the paper still fitted our topic, three pieces of data were recorded for each paper: relevant keywords used, health problem discussed and type of intervention discussed.

In addition to these hand searches, the research team met with public health experts, local public health managers and staff in general practices. The aim of these discussions was to help focus the review and provide a rationale for paper selection. As a result it was decided that we would categorise papers that were selected for review into one of four categories:

• include for evidence synthesis of approaches to delivery/organisation of health improvement

• include because tells us about health improvement activities undertaken in primary care

• mark as a paper ‘of interest’

• reject paper.

While the evidence synthesis focused on the first category, we also decided to include papers in the second category in the review, as they provided useful information about the range of activities being undertaken in general practice.

Preliminary results

There was a wide variation in the amount that GPs were mentioned in PCT public health reports. Several reports did not mention general practice at all while others listed innovative measures that general practices were implementing to tackle public health problems.

Four journals were searched for papers on general practice provision of public health services between 1990 and 2010. Numbers of relevant papers from each journal were 75 from the British Journal of General Practice, six from Health Promotion International, 15 from the Journal of Public Health and six from Primary Health Care Research and Development.

Based on this list of terms, we undertook a MEDLINE search, which yielded over 100,000 hits. A rapid review of a small selection of abstracts showed that few papers were likely to be relevant. This result suggested that our search strategy was too wide. However, we also found that a number of the papers that we uncovered in our hand search had not been detected by this broad search strategy. Our next step was the practical one of narrowing our search finding to a manageable number of papers. We developed a second search strategy that identified specific keywords that appeared to generate substantial relevant hits from within our original list of terms. We also restricted our search to the UK to try to identify a manageable pool of references. See Appendix 1 for the literature search strategy.

Initial literature search

Thus, an initial search of Ovid MEDLINE, informed by the keywords obtained from the hand searches, was conducted by combining synonyms for primary care, public health and the UK. The search strategy is listed in Appendix 1. Results were limited to academic papers published in the English language between 1990 and 2010. The search yielded 14,948 references. Bearing in mind that the final literature search was planned to be conducted on four databases, this high number of papers from one database was deemed to be unmanageable for paper selection. Steps were therefore taken to further refine this search.

Refinement of literature search

The initial search was refined by replacing the public health search terms with synonyms for three areas of public health highlighted in the Healthy Lives, Healthy People White Paper. 37 These were ‘prevention’, ‘health improvement’ and ‘health promotion’. Thus the search was a combination of synonyms for these three areas of public health, primary care and the UK. The search was run on four databases, namely MEDLINE, EMBASE, Global Health and CINAHL Plus, using strategies designed specifically for each database, described in Appendix 2. Results were limited to academic papers published in the English language between 1990 and 2010. This search yielded 7209 results in total after duplicate references were removed using EndNote X4 software (Thomson Reuters, CA, USA).

The discrepancy between 7209 results from four databases in the refined search and 14,948 results from one database in the initial search led to concerns that the refined search may have missed relevant papers. In order to investigate this, a sample of references published in June of each year from 1990 to 2010 which were retrieved by the initial search but not by the refined search were examined. From this, eight additional areas to supplement the refined search were identified. These were alcohol and drug addiction, brief interventions, exercise therapy, immunisation, lifestyle, risk reduction, screening and smoking cessation. Synonyms for the eight areas were combined with search terms for primary care and the UK, and limited to academic papers published in English between 1990 and 2010, as before. The search strategies for each of the four databases are shown in Appendix 2.

The final literature search results consisted of references retrieved by combining the search strategies used in our refined search (see Appendices 1 and 2). Duplicate references were removed using EndNote X4 software. The process was repeated on 12 May 2012 to identify papers published in 2011 and early 2012. The total number of references retrieved for the years 1990–2012 was 16,791.

Paper selection

Paper selection was conducted in three sequential stages: by title, by abstract and by full text. At each stage, the following inclusion criteria were used:

• Is it focused on primary care?

• Is it health improvement (defined as primordial, primary, secondary or tertiary prevention, and also encompassing health promotion)?

• Is it, or could it be, related to service delivery or organisation?

• Does it report research findings or, alternatively, does it contain a description of health improvement activities undertaken in practice?

If the answers to all four questions were yes, the paper was progressed to the next stage. Where abstracts were unavailable or did not provide sufficient information to assess the paper’s relevance, the full text was obtained in order to assess relevance.

The 16,791 references were divided so that, in the title selection stage, the title of each paper was reviewed separately by three reviewers. If any one reviewer considered a paper to be relevant, it was progressed to the abstract selection stage. In the abstract stage, references were divided so that each abstract was reviewed separately by two reviewers. If both reviewers considered a paper to be relevant, it was progressed to the full-text stage. In the full-text stage, data from each paper were extracted by individual reviewers. There was also an opportunity to discard papers at this stage, if considered to be irrelevant upon examination of the full text. The overall search and paper selection process is shown in Figure 1.

FIGURE 1.

Flow diagram of review process.

Supplementary searches

In addition to our database searches we undertook a search of The Cochrane Library to identify potential reviews that might have relevance to our review. Searches of the title, abstract or keywords of reviews in The Cochrane Library were performed in January 2012 using the search terms listed below:

• promotion – 54 results

• promoting – 261 results

• preven* (search restricted to title field only) – 871 results

• ‘primary care’ – 96 results

• ‘general practice’ – 14 results

• ‘general practices’ – 2 results

• cessation – 119 results

• smoking – 134 results

• diabetes – 202 results.

The results were manually reviewed to identify potentially relevant reviews.

We then examined each of the final results for any information relating to public health service delivery and organisation in primary care. We identified 27 potentially useful reviews (see Appendix 3). In addition there were three reviews in progress that were possibly relevant and we also identified a review on Interventions to implement prevention in primary care that was registered in 2001 but then withdrawn in 2005 owing to a lack of resources available to undertake the review.

The results of these reviews were incorporated into the findings from our literature search. While there is inevitably some overlap of papers, we conducted a brief cross-check. Given the differences in search strategies, review focus and quality assessment, this was limited.

We also examined NICE guidance on public health and primary care but this was more limited. NICE started undertaking public health reviews only recently. However, where relevant, the guidance is referred to in this review.

Finally, we searched for other reviews that focus on public health and primary care. Two key reviews were identified. The first was undertaken in Canada23 and the other was by the Committee on Integrating Primary Care and Public Health Board on Population Health and Public Health Practice, US Institute of Medicine. 24 Both these reviews examined the relationship between primary care and public health. The first of these had a very broad focus and had little direct relevance to general practice or actual health improvement activities. The US review was very specific on the US context and the potential for developing public health in primary care. Neither review added anything substantial to our own searches but they are referred to where relevant.

Data extraction

Given the large volume of relevant papers identified, a decision was made to perform data extraction on a subset of papers published from 1996 to early 2012. We identified that older papers tended to be selected less and decided that we would therefore restrict full data extraction to after 1995. We retained the selected references for 1990–5 and, where relevant, examined and included papers in the review where they added additional information. The topic ranges for these papers focused on a small number of themes including health checks for older people, smoking cessation, health promotion payments (under the 1990 GMS contract) and CVD. There was a large number of papers describing activities in general practice and the community. There has been a substantial number of changes to the funding and organisation of primary care since the mid-1990s, and therefore we focused mainly on the more recent papers. Details of selected papers are given in Chapter 4.

Data were extracted from each relevant paper into an Access 2010 (Microsoft Corporation, Redmond, WA, USA) database using a standardised form, which is included in Appendix 4. The form had been designed by the research team and contained fields to capture information on the type of study, outcomes, population targeted, service provider and main results. For intervention studies there were fields for describing the intervention and how it related to service delivery or organisation. Designing a single data extraction form for use across all types of studies was complex and the final format did not provide a single structured way to extract data from qualitative studies. The limitations of the data extraction form were:

1. The depth and nuances of the information extracted from qualitative studies may have been limited by all researchers extracting data not having a ‘rule’ about which findings to extract from qualitative papers, other than ‘a summary of the results and conclusions’.

2. It relied on researchers using individual judgement to interpret the ‘outcome assessment (risk of measurement bias)’ criterion of the quality assessment section about how appropriately and reliably data were collected in the qualitative studies to address their aims.

To counter this we undertook extraction on a group of papers and compared results between researchers to obtain some cross-researcher consistency. In addition, when drafting results sections authors returned to a number of key full texts to ensure correct data interpretation. During data extraction, papers were grouped in terms of whether they simply described health improvement activities in primary care, or they reported research findings which could be included for evidence synthesis. Quality assessment was conducted on the latter group of studies.

Quality assessment

Several tools exist for determining the quality of studies for reviews. However, given the large volume of papers included in this review, and the diversity of study designs which were both quantitative and qualitative, it was not possible to find one tool that would be sufficient for the task. Instead, a general assessment of the quality of each study was formed by critiquing five areas:

• quality of the sampling process (including the sample size)

• appropriateness of the analysis

• comparability of the intervention and comparison groups

• quality of the intervention delivery (including whether consistency of the intervention was measured or there was risk of contamination)

• outcome assessment.

For each area, reviewers stated in the data extraction form whether or not the approaches taken in the study on these five criteria were applicable to the study design. The reviewers indicated, under each area, whether they had no concerns, minor concerns or major concerns. Studies were graded as:

• A, high quality, if there were no concerns across the applicable areas

• B, good quality, if there were one or two minor concerns

• C, lower quality, if there were major concerns or multiple minor concerns

• D, poor quality or not assessable, if there were major concerns or insufficient detail.

The goal of our search strategy was to identify a body of literature on the delivery and organisation of health promotion and disease prevention undertaken in general practice. We expected problems in identifying relevant papers, but the main problem was that we found many unrelated results while missing out some relevant ones. Several challenges were identified that make it difficult to (1) define the topic of general practice and public health and (2) identify the literature on this topic. These challenges can be categorised as ‘practical challenges’ and ‘conceptual challenges’ as follows.

Practical challenges

In our search strategy, papers needed to have a GP or primary care-related term in the title or abstract in order to be picked up. This was clearly necessary to keep findings relevant to our topic, but this strategy misses the body of research on clinical interventions that are delivered in a different setting rather than in general practice. For example, some studies on smoking cessation or exercise on prescription were not done in general practice, so these papers did not have a ‘GP term’ in the keyword list, title or abstract. However, these interventions are in reality often delivered by GPs, so these studies were relevant to our scoping study. This problem was discovered when investigating papers from the hand search that did not appear in the literature search results, and accounts for some of the papers we ‘missed’ in our database search.

Conversely, many of the studies that do involve general practice are surveys of GP knowledge and attitudes, audits of GP patient lists, or surveys of patient reactions to GP-provided services. These qualitative studies are useful for defining a problem or gaining patient perspective, but seldom provide evidence on outcomes of interventions. The abundance of this type of study became clear after a quick scan of the results from our database search, and it partially explains the large number of studies we recovered with our search terms that were not relevant to service delivery.

There are several types of interventions and diseases that are particularly common in the literature of GPs and health promotion, for example behaviour modification and CVD. It would be useful to do broad searches that include these terms to ensure that we have covered these topics, as they relate to our questions, in their entirety. However, the terminology used for these and many other topics is not straightforward. For example, behavioural interventions use a variety of phrases such as ‘health behaviour’, ‘lifestyle management’ and ‘behaviour modification’. Relevant studies on CVD use the terms ‘cardiovascular disease,’ ‘coronary heart disease’ and ‘coronary risk factors’ among others. This use of different keywords to describe the same disease or same type of health interventions makes it difficult to be sure of the completeness of a search. This is a problem for those trying to do review studies, but it also makes it difficult for practitioners at every level (from GPs to PCT public health officials) to find information quickly on a given topic.

As in many fields, the hand search showed that research priorities in general practice and public health have changed over time. These changes in research reflect changes in disease prevalence (as in the case of the rapid increase in obesity-related studies) as well as changes in policies and GP contracts. For example, studies on health checks for over-75-year-olds that were introduced in the 1990 GP contract have been largely replaced by studies of health checks for over-40-year-olds. These changing priorities over time raise questions for our research focus and whether to focus more on the more ‘current’ issues or to give equal weight to all topics that have been researched since 1990.

Finally, a key challenge was to identify papers that specifically addressed service delivery and organisational aspects of health improvement in general practice. It was not possible to use organisational criteria in the search so it was necessary to examine each paper for any reference to delivery and organisational issues. Most papers focused on the intervention and its efficacy or describing the intervention but provided less detail on how it was delivered.

Conceptual challenges

Public health is, by definition, about the health of populations. A good example of what this means for research is displayed on the Cochrane Public Health Group website, which states that: ‘The CPHG [Cochrane Public Health Group] facilitates the production of systematic reviews of the effects of public health interventions to improve health and other outcomes at the population level, not those targeted at individuals’ (their emphasis). 112 But GPs usually deal with health by treating one individual at a time, which seems to conflict with the concept of ‘public health’. However, the collective action of primary care and some individual interventions – such as immunisation – can be seen as having a population effect. Because our research is focused on delivery methods and organisational factors, it attempts to look beyond the individual patient interaction, but research on interventions delivered in general practice ultimately comes down to a one-on-one interaction. This definitional conflict raises questions for our inclusion criteria and how to determine whether or not a study truly qualifies as being about ‘public health’.

The prevalence of chronic diseases and the new focus on their prevention has complicated notions of primary, secondary and tertiary prevention. For example, in the case of a person who has a poor diet and is physically inactive, then goes on to become overweight, and then develops high blood pressure and abnormal cholesterol levels, where does the disease process start? Is primary prevention the prevention of becoming overweight, developing hypertension or developing CVD? Similarly, many titles retrieved in our search focus on self-care for chronic diseases such as diabetes or asthma. Technically, these are about managing an ongoing condition, but the goal of self-care is to prevent progression of disease or hospitalisation, so they could also be thought of as prevention. With topics in the ‘prevention’ literature ranging from recommending healthy diets to managing patients after a heart attack, it becomes difficult to assemble the evidence or even decide what to include in a review of prevention interventions.

Two related questions that have revealed themselves both in the hand search and in a quick review of our search results are (1) where, physically, does general practice end and (2) who constitute the general practice team? Considering the physical space of the general practice gives a different answer from considering the staff of a general practice, as health visitors work with a GP surgery but by definition practise outside the surgery door. And if we extend our focus to all primary care staff the picture becomes even more clouded. Who exactly constitutes a primary health-care provider? Our literature review brought up a large number of results about dentists, school nurses and community pharmacists. While we can easily draw our own boundaries and eliminate certain types of providers or locations of service from our review, the results of our searches caused us to reconsider our initial definitions of general practice, as it brought up types of providers that we had not previously considered.

Consultations with stakeholders

Our original plan was to hold regional stakeholder events and invite practitioners to these to discuss public health issues. However, the period during which the research was undertaken was one of substantial upheaval in the NHS. Following discussion with key stakeholders and with NIHR Evaluation, Trials and Studies Coordinating Centre SDO Programme staff, this approach was amended to one whereby we would undertake more individual discussions with practitioners in general practice and with other key informants. In addition, we were able to draw on interviews conducted for a study on early developments in CCGs and some group discussions with GPs.

The research team undertook a number of discussions with key general practice stakeholders. Initial discussions were held in 2011 with NHS public health, general practice and health promotion staff in England and Wales:

• two primary care health promotion advisors in Wales

• a director of public health (London)

• two practice visits in Wales involving interviews with four GPs, two practice nurses and one practice manager

• two practice visits in England involving interviews with three GPs, three practice nurses, one community nurse and two practice managers.

These visits and interviews were exploratory and helped shape the initial scoping of the literature. The researchers explored key issues as identified by the staff members. Key issues and themes were noted by the researchers. We were also able to capitalise on research being undertaken on emerging CCGs by the Policy Research Unit in Commissioning and the Healthcare System, which is directed by Professor Stephen Peckham. This involved interviews with staff in PCTs, emerging CCGs and practices in eight areas in England and were conducted in 2012. These interviews explored public health roles and responsibilities and the potential impact of organisational changes currently under way in England. Research team members were also involved in concurrent research on primary care, and the opportunity was taken to ask GPs and practice staff in a number of practices in London, the Midlands and the North West to discuss health improvement activities in their practices. In total we drew data from 45 interviews undertaken with practitioners. These included 11 GPs and 11 practice nurses interviewed within the context of their practice, 16 CCG board members (15 GPs and one nurse), three lay members and four public health specialists/consultants.

In addition to individual interviews, two group sessions were held with GPs in two locations to discuss public health roles. One group consisted of 38 GPs in a 3-hour session on public health and general practice. GPs in this group included GP principals, salaried GPs and some trainees. In addition, research team members held a group meeting of eight GP principals, including one GP trainer. Finally, during the research one researcher was involved in the King’s Fund review of quality in general practice,18 which included two discussions with a wide range of stakeholders (GPs, public health specialists, nurses and patient organisations) on health promotion in general practice. Given that the aim of this element of the research was to inform the scoping review and our analysis, interviews and discussions were not recorded and we relied on researcher notes.

The data from these discussions and interviews were in the form of interview extracts (from the research on CCGs), field notes and observations from group meetings, and notes from individual discussions with practitioners. Researchers discussed the notes and interview extracts, and used these to develop key themes. These were used in refining search terms (described earlier in this chapter) and in interpreting literature findings. The discussions and interviews also highlighted the significant impact of the changes in the English NHS that occurred during the period of the research. These data have informed the discussion of this report and the key themes arising from the interviews and discussions are reported in Chapter 4.

Evidence synthesis

For this review a key challenge was to identify evidence on the delivery and organisation of health improvement interventions rather than the effectiveness of a specific intervention irrespective of the context within which it is delivered. As Rychetnik et al. 113 have argued (p. 119):

Public health interventions tend to be complex, programmatic, and context dependent. . . . The evaluation of evidence must distinguish between the fidelity of the evaluation process in detecting the success or failure of an intervention, and the relative success or failure of the intervention itself.

In particular they argued that, where an intervention is unsuccessful, it is important that the evidence should help to determine it was inherently faulty as a result of the failure of intervention concept or theory, or simply badly delivered. It is also important that there be sufficient descriptive information on the intervention and its context to allow for the proper interpretation of the evidence.

In our review we were reliant not only on whether or not contextual issues had been included in the study but also on whether or not these details were reported in the published papers – which was not always the case. Given the breadth and diversity of the topics and papers, and the limited amount of data on service delivery and organisation, the research team agreed that a narrative synthesis would be the most appropriate approach. 114,115 All the papers were allocated topic tags and a member of the research team then examined groups of papers (e.g. on smoking, screening, CVD, child health) and key findings were reviewed to identify any common themes and if there was any evidence to support particular approaches to the delivery and organisation of health improvement activities. Given the range of material, we adopted a textual narrative synthesis, trying to group studies and data into more homogeneous groups. 116 Our approach was to develop an interpretive synthesis of the literature. 117,118 We developed the synthesis along two broad themes:

• what kinds of activities were described in the papers

• what evidence there was for effective approaches to service delivery and organisation.

Drawing on these data we then explored themes around the ways interventions were organised, who delivered them and where they were delivered to identify common issues.

Limitations of the review

This was a complex review to undertake because of problems about definition of terms and also because of the diversity of the topic. The research team made substantial efforts to identify appropriate search terms and develop inclusion and exclusion criteria. We had to take an approach that required only brief scanning of our original search results to reduce the numbers of potential papers. We adopted a checking system for papers initially discarded to ensure completeness. However, we recognise that our eventual strategy produced a particular set of results and that it is possible that other search strategies could have produced a fuller set of results in some areas of public health activity. However, without some initial scoping research in practices and across the literature, identifying these specific areas would have been difficult. For example, with emphasis on primary care/general practice we are aware that a number of more community-based health initiatives that might be usefully classified as primary care were not picked up in our search.

The rejection of papers identified as treatment or tertiary prevention may also have limited the final paper selection. Having completed the searches, we are aware that some topics may be better reviewed as a single group incorporating all activities relating to prevention, including all levels of prevention and treatment. Based on the experience of this review it is possible that in undertaking such a review there would still be very few data on the delivery and organisation of health improvement activities. It is probable that further primary research would be required to identify these aspects of how such interventions are undertaken in practice.

Our conclusions about the delivery and organisation of health improvement activities are also limited by the information provided in the published papers. Generally this was very limited, with the main focus placed on the intervention. There were papers that examined different interventions by different practitioners; for example, screening by nurses or GPs, pharmacists undertaking health checks, smoking advice given by GPs, nurses or health promotion specialists. There were also papers that discussed the provision of activities in different specific locations such as the practice, schools or pharmacies. However, very few papers made any comparison between different practitioners or locations. Most comparisons examined different interventions; for example, offering screening or screening and advice, or comparing verbal and written information but often without details of who did what.

While the selection of papers was undertaken by multiple researchers, the final selected papers were read by only one researcher and data extracted using the standard form. While we adopted a checking process comparing how all researchers read a selected set of papers to ensure consistency of approach and use of the data extraction form, it is possible that there were still differences in interpretation of papers between researchers – especially given the broad range of material reviewed. However, it is unlikely that such differences would have had a significant impact on the eventual conclusions in this review, as all researchers faced similar problems about the lack of detail and data irrespective of topic.

Presentation of findings

The results of the analysis of the literature are presented in three sections. In order to provide an overview of the range and type of literature and research, we have provided a descriptive analysis of the papers in the next chapter (see Chapter 4). Here the aim is to examine what topics and issues research has addressed, when it was published, and key characteristics in terms of quality and focus. We then present our analyses of the data in the papers on what kinds of activities are undertaken (see Chapter 5) and then what evidence there is of effectiveness (see Chapter 6).

Introduction

This chapter provides an overview of the results of the literature search. The aim of this chapter is to provide some detail of the literature review results. It does not examine the evidence itself but rather provides an indication of the areas in which academic and research activity has been focused. The chapter also includes a brief resume of the key points arising from discussions with public health/health promotion staff in PCTs and local health boards, and primary care practitioners in general practice.

In the process of undertaking the review it was clear that there were important trends and features of the literature that are important. The first section of this chapter provides an overview of the search results by publication year and then the extracted papers are examined in relation to their assessed quality, type of prevention activity and details of the type of intervention, where it was provided and by whom. While 658 papers from our initial pool of 12 papers were extracted for more detailed examination, a number of these provided insufficient detail of interventions or were general discussion papers. Initially it was hoped to include data from these in general discussion sections, but generally these papers were not of specific use for describing either interventions or how they are organised. The number of references retrieved by the final literature search by year of publication is shown in Figure 2.

FIGURE 2.

Number of hits from literature search by publication year.

Figure 3 provides details of the numbers of papers extracted for review. What is interesting is the leap in numbers of papers extracted after 2006. We also looked at papers published between 1990 and 1996. Of the 1639 search hits, only 365 were selected for further review and of these 262 were considered for full-text review and inclusion in the study, but only a small proportion – approximately 20% – were research studies that provided evidence relevant to the review. However, examination of these papers revealed that many dealt with issues that were relevant only to the circumstances at that particular time. For example, a number of papers examined the development of health promotion interventions developed in response to the introduction of health promotion banding payments in the 1990 GMS contract. Others specifically examined the impact of the payments on activity. It was not clear, therefore, that these papers would provide relevant data for inclusion in a discussion about current practice. It was decided, therefore, to exclude papers from before 1996 unless they had directly relevant findings that could add to the general body of knowledge.

FIGURE 3.

Number of papers selected for data extraction by publication year.

It was clear from the search that there were key journals in which papers were published. Figure 4 shows the top 20 journals in terms of the number of hits from the final literature search, and Figure 5 shows the numbers of papers in key journals selected for extraction. These demonstrate an enormous leap in the number of papers selected for review after 2006. This may indicate an increase in papers that address issues of delivery and organisation aspects of health improvement. One driver may be papers examining issues related to the QOF or a growing interest in health promotion research more generally.

FIGURE 4.

Top 20 journals in terms of the number of hits from the literature search. BMC, British Medical Council; BMJ, British Medical Journal.

FIGURE 5.

Top 20 journals in terms of the number of papers selected for data extraction. BMC, British Medical Council; BMJ, British Medical Journal.

We then analysed the papers by topic. Each paper was allocated a topic tag based on assessment of the paper. These were categorised into primary topic and subtopics. A number of studies did not fit into topic categories and these were labelled ‘No specific topic’ or ‘Other specific topics’. Papers categorised as no specific topic were health improvement interventions that were either broad in scope or not aimed at a specific health problem. Examples of these types of interventions were community interventions and welfare advice. In addition, some papers focused on a practitioner role in relation to health promotion, such as school nurses or community nurses. Papers in the ‘Other specific health topics’ category were topic areas where only one or two papers were identified in our search. For example, these included papers examining hepatitis B screening, skin cancer, kidney disease, glaucoma and haemoglobin disorders screening. The main areas where papers were located were for CVD (15%), obesity (10%), diet and exercise (10%) and smoking (10%). This provides a broad indication of where the focus of activities, and to a certain extent research, is in relation to health improvement activities in general practice. However, following a review of the papers only 137 were selected for the evidence synthesis. The distribution of topics in these papers is shown in Figure 6. We then examined the numbers of papers selected for the evidence synthesis by topic area as shown in Figure 7.

FIGURE 6.

Distribution of topics of papers selected for the review.

FIGURE 7.

Distribution of topic of papers included in evidence synthesis.

Comparing Figures 6 and 7 shows how the range of activities discussed in the literature does differ from what topics tend to be most commonly studied. Of the 347 papers selected for evidence synthesis, we graded them on the basis of an assessment of the quality of the research methods (see Chapter 3):

• A, high quality (no concerns): 153

• B, good quality (one or two minor concerns): 132

• C, low quality (multiple minor concerns or major concern): 50

• D, poor quality or not assessable (major concerns or lacked sufficient detail): 12.

The proportion of papers within each assessment grade is shown in Figure 8.

FIGURE 8.

Quality assessment of papers included for evidence synthesis.

The selected papers encompassed a wide variety of research designs. Where details were provided we indicated the research design in the data extraction form. However, we did not rank study quality on a hierarchal basis as used in systematic reviews or meta-analysis. We assessed each paper as far as possible for rigour in the design used and whether or not the methods supported confidence in the reported outcome. The range of study designs is shown in Figure 9. It is important to note that the study design and quality assessment relate to the paper as a whole and they do not indicate whether or not we were able to make confident statements about the effectiveness of any particular approach to service delivery or organisation reported in the paper unless this was the specific subject of the research being reported on.

FIGURE 9.

Distribution of study designs.

The biggest groups of papers were for qualitative methods studies and cross-sectional studies or surveys. Cross-sectional studies and surveys were primarily used to examine patient or practitioner knowledge or behaviour about, or as a result of, a health improvement intervention. The majority of these were in specific topic areas such as screening, smoking, and weight or exercise interventions (Figure 10). These tended to focus on the views or perceptions of patients/users and practitioners. None specifically compared groups and few provided before-and-after data related to an intervention.

FIGURE 10.

Distribution of topics of cross-sectional and qualitative studies. Note that total may be greater than 100% due to rounding.

We were particularly interested in those papers classified as evaluating an intervention or evaluating a service delivery. The majority of papers identified as evaluating an intervention had little information on the context or organisation of the way in which the intervention was delivered. For example, interventions included a 10-week physical activity course delivered in schools but with no details of the actual programme or who ran the programme. Similarly, many educational interventions lacked details of who delivered it, where, in what way, etc. Generally papers contained few specific data on the content and context of interventions. There were 89 papers that evaluated the service delivery and organisation of interventions or aspects of practice. There was a mix of research designs, with many studies merely examining patient or practitioner views. Some studies – for stop-smoking interventions and weight/exercise – did look at outcomes but often the descriptions of the delivery of the intervention were limited in the published paper. Lack of detail or of focus on service delivery and organisation aspects of interventions was a general feature of the papers included for review.

We initially attempted to distinguish between three broad areas of health improvement activity: screening, primary prevention and secondary prevention. We excluded tertiary prevention from the study, as we classified this as treatment and outside the scope of the study. Defining papers and categorising them was not without problems. The distribution of papers selected for review and then for the evidence synthesis is shown in Figure 11.

FIGURE 11.

Distribution of topics in selected papers for review and for synthesis. Note that some papers cover more than one area.

Description of papers on screening activities

Overall, 148 papers provided details of some aspects of screening, although only 84 papers specifically examined research relating to a screening programme. Other mentions of screening were included in papers examining other interventions. Of the papers that were specifically examining aspects of screening, the topics covered are shown in Figure 12.

FIGURE 12.

Range of topics within papers examining screening. HIV, human immunodeficiency virus.

Description of primary prevention papers

Based on the original concept of levels of prevention,119 primary prevention is the action taken prior to the onset of disease, which removes the possibility that the disease will ever occur. Two approaches to primary prevention fall under the broad approaches of health promotion and specific protection. In the review of papers it was often difficult to identify whether or not the activities being described were primary prevention. While appearing to be primary prevention in nature, they could have been applied in practice to patients with an existing health issue, for example CVD. There was often insufficient detail to allocate papers between primary and secondary prevention. Unless specifically stated, papers that addressed primary prevention activities but did not refer to patients with existing disease were classified as primary prevention and are considered here. The data presented here reflect the papers selected for this review and may not provide an exhaustive description of all primary prevention activities undertaken in general practice or primary care more generally. The aim is to present what literature we found where there is evidence relating to the effectiveness of interventions.

In total we extracted 326 papers to include in the review of primary prevention activities. Of these, 164 were included in the evidence synthesis. Of these 164, the majority related to child health (50 papers), smoking (49 papers), exercise and weight (40 papers) and vaccination (22 papers). In addition there were a number of papers that related specifically to CVD (8 papers). The remainder deal with falls, diabetes and mental health. Some of the categories overlapped, providing higher numbers of papers for some topics such as child health and smoking where, for example smoking related to young people or where diet, nutrition and exercise were specifically related to young people. In addition, we included 18 papers on community participation and a further 11 papers on welfare advice services in general practice. The distribution of the primary topics is shown in Figure 13.

FIGURE 13.

Distribution of topics of papers classified as primary prevention.

We also examined the 64 papers that were unclassified by the level of prevention (which included welfare advice and community initiatives) but were mainly interventions that addressed particular population groups without clearly defined problems; see Figure 14.

FIGURE 14.

Unclassified papers by population focus.

There were also 176 papers that provided additional and relevant information about public health activities undertaken in general practice. The majority referred to aspects of primary prevention or screening although some were not clear about their target population. Of these, 162 were reviewed in relation to primary prevention but did not provide significant detail of interventions or their impact.

Few papers addressed issues such as cancer or alcohol misuse. Many papers that discussed exercise and weight loss were in relation to people with an existing health problem; they were defined as secondary prevention and are included in the description of secondary prevention activities below. Similarly, stop-smoking interventions are both primary and secondary, and potentially tertiary. Primary prevention focuses on smoking prevention and smoking cessation programmes before any disease, secondary prevention limits the sequelae of a disease and would fall within smoking cessation programmes, and tertiary prevention keeps those who are disabled from experiencing more severe effects, such as in pulmonary rehabilitation programmes.

Within this group of papers we also identified those that examined community-based activities related to general practice and services, such as welfare and employment advice where undertaken, in general practice. Whereas these may be seen as primordial prevention activities, the rationale here is that they are provided within the general practice context as a primary intervention, such as stopping smoking, and many of the papers are concerned with the health impact or at least direct impact on the patient or health services. It is also true that some of the patients referred to such services have pre-existing conditions and thus such services could also be classified as secondary prevention interventions.

Description of secondary prevention papers

We identified 126 papers within the evidence synthesis that were secondary prevention interventions. The most frequent area researched was CVD, with 43 papers addressing this area specifically (including health checks), but there were also 25 papers addressing weight and exercise for people with symptomatic disease. Figure 15 provides an overview of secondary prevention topics.

FIGURE 15.

Overview of secondary prevention paper topics.

It is not clear how far these should all be classified as secondary prevention, as information about patient/population groups and whether they were symptomatic or asymptomatic was not clear. Many of these were therefore seen as primary prevention activities whether they included symptomatic patients or not. Most remaining papers were then related to CVD and pharmaceutical interventions although there were some lifestyle interventions.

Given that much of general practice prevention activity – such as that supported in the QOF – relates to clinical interventions, it is surprising that there were not more papers included in the evidence synthesis. Reviewing the topics of papers identified in the search and selected for review, but not included in the evidence synthesis, does show a high proportion of papers examining clinical prevention interventions. However, many of these had no information or did not address delivery and organisation aspects of interventions, but were merely clinical trial data or evaluations of the clinical intervention. For example, a number of papers examined the use of statins to reduce cholesterol levels without any discussion of the process. Similarly, papers about the effectiveness of clinical measures to reduce blood pressure were identified but examined only the pharmaceutical effectiveness of the drug, not how it was provided or by whom. Descriptions of secondary prevention activities or how these are related to broader health improvement activities were significantly under-represented in our final paper selection. These were more likely to be categorised as chronic disease management or tertiary prevention and outside of our inclusion criteria.

Description of interventions

The papers selected for review encompassed not only a wide range of topics but also the type of intervention, where it took place and who delivered it. This type of data was not available in all papers and, given that some papers examined patient or practitioner views, it was not possible to identify the organisation and delivery of many types of interventions. For example, a number of papers examined brief interventions. However, these papers included those delivered by GPs,120,121 pharmacists and nurses, and covered smoking, alcohol, exercise, weight management and sexual health. Some papers examined the brief intervention itself, others reported on studies that combined brief interventions with screening and other forms of advice or treatment, and others examined training and support for professionals to deliver brief interventions and advice and support. 121–124 While the majority of brief interventions appeared to be delivered in practices by GPs and nurses, some were delivered in the community or pharmacies. 121,122 Taking just one topic area – brief interventions for alcohol – there were papers that combined interventions,122,123 examined the link between screening and a brief intervention,124 studied training for practitioners giving advice122 and investigated a pharmacy-based intervention. 121 It was not possible, therefore, to draw substantive synthesised findings about these intervention types. However, more information about where and who delivered interventions was identifiable from the studies.

Description of papers by location of health improvement activity

We examined the location of the delivery of the health improvement activity in the 483 papers that were selected for the review process. The main location of intervention delivery was general practice (Figure 16). While this may have been affected by the search strategy, the fact that we specifically looked for studies in pharmacies and activities by primary care professionals in the community, the distribution of papers suggests that the main focus of research interest is in relation to interventions delivered in general practice locations (Figure 17). The papers selected for the evidence synthesis show a similar distribution, although proportionally more papers that examined community, the home and pharmacy-based interventions were excluded from the synthesis.

FIGURE 16.

Where the interventions in papers selected for the review were delivered.

FIGURE 17.

Where the interventions in papers selected for evidence synthesis were delivered.

One key problem was a lack of detail in papers clearly defining where activities were performed. For example, papers identifying general practice as the location of the intervention do not necessarily distinguish between whether the activity was in the consultation with a GP or nurse, part of a group or clinic, or some other part of the practice. Similarly, the designation of community is very broad and in most cases did not include a specific venue. Often a number of venues were referred to without any reference to whether or not location was an important element of the delivery and organisation of interventions.

Description of papers by who undertook the intervention

We also examined who delivered the interventions. The majority were delivered by a GP or nurse, and this was reflected in papers selected both for review and for the evidence synthesis (Figures 18 and 19). There is a substantial other category which includes specialist health promotion practitioners, smoke-stop counsellors, other community staff, welfare advisers, etc. However, the findings do underline the key roles of GPs and practice nurses in delivering health improvement activities, but also show that many other professionals contribute to the prevention and health promotion activities related to primary health care and general practice itself.

FIGURE 18.

Who delivered the interventions in papers selected for review.

FIGURE 19.

Who delivered the interventions in papers selected for evidence synthesis.

Key themes arising from discussions with practitioners

As described in the previous chapter, during the study we interviewed a number of primary care and public health practitioners in practices, PCTs and local health boards in London, the Midlands, North West of England and in South Wales. We also held two group discussions with GPs in the south of England to discuss their views about public health. The purpose of these interviews and group discussions was to identify issues; they were not designed or intended to provide specific data on activities undertaken in practices. Early interviews were used to help shape the literature review and raise issues to inform the data synthesis and analysis. Interviews and group meetings were not recorded but researchers took notes and wrote up filed observation notes. In addition, we were able to obtain data from a concurrent study looking at the development of CCGs, which, thanks to the involvement of researchers from this study in the research team, included questions about public health issues. Researcher notes were shared between researchers and thematically analysed by two researchers to derive a list of key themes.

Key themes that arose from the interviews and group discussions were:

• There appears to be a dichotomy between practitioners about whether the focus of general practice should be on more clinical interventions for individuals, and thus prioritising secondary prevention, or general practice should have a more holistic perspective within which health improvement and prevention is a key focus.

• There were mixed views regarding the QOF and its impact and potential for supporting health promotion in general practice. Some GPs did not see that they should ‘interfere’ directly with some areas of lifestyle advice for patients, although most recognised that a general level of advice was appropriate.

• Views varied about the usefulness of screening activities, but many GPs felt that current policies were placing more and more responsibilities on them to screen and advise patients, which was difficult in normal patient consultations.

• GP interviewees generally appeared keen to take on health improvement roles but acknowledged that there is a wide variation in activity and motivation between practices and between practitioners. GPs and other practice staff reported a wide range of health promotion activity ranging from information to proactive approaches.

• Key areas of concern remain, however, specific clinical conditions such as diabetes and CVD, although smoking and obesity were also highlighted by respondents. GPs tend to identify themselves as proactive in their approach to prevention but focus mainly on individual contacts. Some practices have adopted a range of health promotion activities for their patients. This appears unrelated to recent initiatives and incentives in the QOF but is part of how these practitioners view their practice and the services they offer.

• More common activities are in clinical areas such as cancer screening, CHD and diabetes in response to key national priority areas and also responding to local issues.

• There were concerns about what contact practices will have with public health in the future.

• Practices were receiving specific health promotion support from their PCTs or local health boards, including education and training, additional resources and involvement in wider community-based projects. This was undertaken by dedicated staff in local health boards and PCTs and there was a concern that this support and help would not be there in the future.

• GPs involved in CCGs were particularly concerned about the future relationship with public health specialists in terms of getting public health advice and also advice specifically for commissioning. There were concerns that good links developed through PCTs will be lost with the move of public health departments to LAs.

• GPs did not express much interest in or awareness of non-financial incentives. One or two said they would value extra ring-fenced funding to employ extra staff or to invest in their premises. GPs in group discussions generally viewed the additional income from the QOF as a way of investing in new, often nurse-led, activities. However, capacity was clearly an issue regarding whether practices can run health promotion clinics or prevention activities are undertaken during individual consultations.

• There was a recognition and an acceptance that the QOF had been important, not specifically because it was paid but because it had created a standardised system. However, it was clear that how this standardised system was implemented in practices varied considerably.

• Practice nurses generally welcomed the additional responsibilities regarding risk assessment, screening and monitoring patients. However, while some practices have a lot of nurse-led clinics, other practices still embed this activity in individual patient consultations.

• There was a lack of knowledge about wider health improvement issues and services that were not practice-based. GPs had variable knowledge about other services in the local community that provide or support health improvement, such as smoke-stop service, community groups, activities in local schools, etc.

• Discussions with GPs and public health practitioners – particularly as part of the research undertaken with the Policy Research Unit in Commissioning and Primary Care – identified key concerns about arrangements after April 2013. There was concern about the future relationship of public health departments and CCGs with general practice more generally when public health departments move to LAs.

Conclusion

In this chapter we have provided a brief analysis of the papers identified in the review and those selected for inclusion in the review and evidence synthesis. The process of reviewing papers reduced the initial 1143 papers to 483 that provided some information or evidence on the delivery and organisation of health improvement interventions in general practice or by primary care practitioners. In the end only 347 papers were included in the evidence synthesis and many of these had few details about the type, process and location or who provided the intervention. Generally, our review has demonstrated that little attention is paid in the literature to examining the impact of the organisational context on the way services are delivered or how this affects the effectiveness of health improvement interventions in general practice. Often even those papers describing activities had little detail about the implementation and delivery of interventions.

Key themes arising from discussions with practitioners demonstrated that in practice there are a wide variety of views and concerns about health improvement. There are also substantial variations in the way that health improvement activities are undertaken in the practice, with differences in who does them and how they are organised. This reflects the findings from the literature.

Introduction

The aim of this chapter is to describe the types of health improvement activities undertaken in general practice or by practice staff in other locations. This includes activities undertaken in community pharmacies and in schools by primary health-care staff. The information in this chapter is drawn from all papers selected for the review including those that were marked only as ‘of interest’.

Variation in practice and activity was evident in the literature. As discussed earlier in this report, we identified a substantial number of papers that described specific activities or interventions undertaken in general practice. However, closer examination of papers provided less information about how activities were organised. The data presented here provide a general overview of health improvement approaches and are grouped broadly into specific topic areas. The selected papers included a wide range of prevention activities and intervention types. As discussed earlier in this report, it was difficult to distinguish between levels of prevention, with many papers not distinguishing between interventions with asymptomatic and symptomatic individuals. In addition, examining the papers for details of the organisation of the delivery of these interventions was not straightforward but primarily related to who delivered the service, where it was delivered or some other aspect of its organisation. For ease of presentation the findings have been summarised around key areas of activity:

• screening

• child health

• exercise promotion and weight reduction (including interventions aimed at the prevention of specific diseases)

• smoking cessation

• CVD

• community activities

• welfare benefits advice.

Types of screening activities undertaken in general practice

We identified 148 papers that examined the delivery and organisation of screening services. Most of these papers addressed specific screening programmes undertaken in general practice or examined the involvement of practice and primary care staff in national screening programmes. Others focused on screening as part of child health programmes, cardiovascular prevention or treatment, or addressed health checks for people with learning disabilities or the new NHS Health Check programme.

Systematic screening

Several activities aimed at identifying health conditions, or risk factors which predict ill health, are delivered systematically to a target population. The role of general practice in systematic screening programmes includes performing cervical smears, child health surveillance and providing health checks as part of the NHS Health Check programme.

Opportunistic screening

Opportunistic screening can be said to occur when people attend a service provider for a matter related or unrelated to the condition being screened for. The distinction between systematic and opportunistic screening is not absolute, as opportunistic approaches may be used in general practices to increase the screening uptake of systematic programmes. 126,137 While identifying smoking status is a key role for practices and has been included in the QOF, a discussion of smoking is included later in this chapter alongside interventions on smoking.

Screening for other conditions

Our literature review also identified papers on a wide range of other screening activities that can be conducted in general practice. These included antenatal screening for haemoglobin disorders,91,186,187 screening for dementia,188,189 falls,190–192 osteoporosis,193,194 atrial fibrillation,195 depression,196 domestic violence,197 adult hearing,198 hepatitis C,199 human immunodeficiency virus,200 oral cancer,201 syphilis202 and tuberculosis. 203

Support of screening programmes delivered outside general practice

Aside from performing screening activities, general practice also has an information provision and advocacy role with respect to screening programmes delivered outside practice sites. This applied to breast screening, which is usually provided at units run by hospital staff, and bowel screening using home testing kits, which is organised centrally and conducted by post. General practice involvement included GPs verbally recommending screening,204–207 practices sending GP-endorsed screening invitation letters or information leaflets,208–212 and interventions to promote opportunistic reminders for people to attend screening. 210,211 One RCT investigated the effectiveness of nurse visits to women who had not attended breast screening. 213

Child health

There were a large number of papers (50) addressing issues in primary prevention among the child population. These spanned a range of topics relating to breastfeeding, accident prevention, weight management and exercise. Many of the studies focused on the activities of community nurses, school nurses and health visitors with activities taking place outside the practice: in the home, community or schools. The majority of the studies were rated as of good methodological quality, although some were qualitative and descriptive rather than studies of effectiveness. There are gaps in both knowledge of activities actually undertaken and evidence to support interventions. For example, a review on the effectiveness of school nurse interventions89 found little research of acceptable quality and little could be said about its effectiveness.

Most research papers examined changes in patient behaviour as a result of the intervention but few examined clinical outcomes. Studies fell into three broad categories. The first group examined interventions within practices or by primary health-care team staff to identify children at risk of health problems, accidents, etc. In this area of activity, interventions were often targeted at specific groups of children. The second group were interventions aimed at intermediary organisations such as changing school policies or practices. This included training and educational interventions for teachers. 214,215 The third group examined interventions or activities in the community or in schools. 216–219 With respect to childhood and adolescent obesity, some studies suggest that GPs are less inclined to blame their young obese patients for their condition and to hold less negative and stigmatised views towards them, but, as with their obese adult patients, there is a sense of futility and lack of confidence evident in GPs’ approaches to treatment. 220 Generally, however, GPs are less involved in addressing obesity in childhood than other primary health-care staff and teachers. There is some potential for GPs to adopt a more population approach to childhood obesity and play a more active role.

Breastfeeding promotion is largely conducted by health visitors and community-based midwives. One of the few papers that addressed the context of a breastfeeding intervention policy221 discusses the variance in outcomes of a RCT of a community based breastfeeding support group. Through ethnographic enquiry of the complex intervention, they found there was variation in the way that services were organised, which could be explained through a Hierarchy of Service Attributes Model. In the study sites where breastfeeding rates increased, the model was underpinned by good personnel resources and organisational stability, and demonstrated reflective action cycles at its apex. In localities where breastfeeding declined, the authors found that managers focused on solving the problems within the model, such as staff shortages, rather than ways to deliver the policy. The model provides a useful approach to understanding and explaining delivery of public health interventions in primary care and could be usefully applied to other scenarios.

Provision of childhood vaccinations is a key component of public health for children and adolescents. Proactive campaigns by general practices do seem to increase uptake of vaccinations. 222–226 GP recommendation appears to improve uptake compared with non-primary care and other primary care staff. 223

Exercise, weight reduction and diet

Most of the papers on exercise targeted the general patient population although one referred to people with mental health problems and another three to older people. The majority of papers on weight loss were focused on people who already had risk factors or a health issue, primarily CVD. There were few very good-quality studies; most were of medium quality with some mainly minor concerns about sampling, analysis and outcome assessment. Previous reviews227,228 have also found few good-quality studies and described research as suboptimal. Interventions ranged from lifestyle advice to specific interventions such as exercise or weight loss classes. There was a mixture of interventions delivered by practice and community nurses, GPs, specialists (e.g. nutritionists, dieticians), community workers and health promotion specialists.

In 2009, the QOF included physical activity for the first time under a ‘cardiovascular risk assessment and management’ indicator. 229 The indicator set a target that between 40% and 70% of newly diagnosed hypertensive patients should be ‘given lifestyle advice in the last 15 months for: increasing physical activity, smoking cessation, safe alcohol consumption and healthy diet’. The QOF has been a major driver in the development of more systematic approaches to public health interventions. 65,230 While this appears to have stimulated such advice, with practices introducing clinics and ‘lifestyle reviews’, the evidence on its effectiveness is limited. This area of activity is also being driven by the introduction of the NHS Health Check.

The prevalence of overweight and obesity is rapidly increasing in both the developed and the developing worlds, while tobacco use is largely declining or has levelled off in the developed world. A 2007 study of 168,000 patients in 63 countries showed that 64% of men and 57% of women were overweight or obese. 231 In England in 2007, 24% of all adults were classified as obese, an increase of 37.5% since 1993, along with 16.5% of children aged 2–15 years, an increase of 11.5% since 1995. 232 We found 16 papers that examined primary care interventions that addressed weight loss and obesity. Much of the research is of poor quality, with only one study rated as high quality.

Overweight people rarely turn first to their GP or the formal health-care system for help in losing weight. They try various self-help approaches first, despite consistently citing their GP as the most trusted source of advice on matters regarding diet and nutrition. 233 GPs regard obesity, both its treatment and its genesis, as largely the responsibility of the patient, and believe that their capacity to effect positive change in their obese patients’ weight status is seriously limited,234–238 despite having higher than average frequency of contact with their obese patients. 239 Clearly this attitude stands in stark contrast with the move by public health and obesity experts to treat obesity as a chronic disease. The biggest obstacle is health professionals’ inability to recruit patients. This may relate to their limited knowledge of potential benefits. 240 There is also likely to be a resulting underutilisation of pharmacotherapy and weight loss services, as fewer than 40% of GPs view these options as effective, despite their evidence-based inclusion in practice guidelines for both adults and young people. 234 GPs (and practice nurses) make relatively few referrals in a rather haphazard way owing to lack of time, lack of feedback, medicolegal concerns or their view that patients are unlikely to go/respond, and because exercise is not a priority during consultations. They are uncertain about their own influence. 241 As with offering a smoking cessation intervention, and for similar reasons relating to lack of faith in efficacy, GPs are hesitant to raise the issue of weight loss with obese patients if they feel it will negatively impact on their relationship with the patient. 235 Many GPs believe that obesity does not belong within the medical domain. 234

Smoking cessation

There is a vast body of literature on issues related to smoking cessation, ranging from the effectiveness of various pharmacotherapy and behavioural interventions to the role of public policy and government. This review focuses on what GPs are currently known to do when promoting smoking cessation and what the literature says about the effectiveness of various smoking cessation approaches. It was not always possible to distinguish between stopping-smoking services that targeted all smokers and those focusing on smokers with an existing health problem. Services were often provided generically.

General practitioners in England have access to much smoking cessation guidance and numerous best practice documents, including those produced by NICE and the Health Education Authority. 242,243 There is also an emphasis on GPs and other primary care professionals raising smoking in consultations and visits, with brief interventions highlighted as particularly important. 244 Brief interventions are generally defined as lasting between 5 and 10 minutes and include one or more of the following: asking the patient to stop smoking; assessing the patient’s willingness to stop; offering pharmacotherapy and/or other behavioural support; providing self-help materials; and referring the patient to specialist counselling.

There is good evidence that services for helping people quit smoking are cost-effective and the evidence suggests that a belief in clinical effectiveness and cost-effectiveness is an important influence over whether or not GPs offer smoking cessation advice or interventions. 245–247 Smoking is a major contributor to health inequalities, and cessation services do help reduce these. 248–250 However, few studies examined specific population groups although one study suggests a lack of services for ethnic minority groups despite some evidence of the benefit of community outreach services. 251,252 Identifying strategies to find and support smokers from disadvantaged groups is, therefore, of key importance. 253 This systematic review identified 48 papers. Some papers were of poor quality, most were observational studies and many did not report findings for disadvantaged smokers. Nevertheless, several methods of recruiting smokers, including proactively targeting patients on GP registers, routine screening or other hospital appointments, were identified. Barriers to service use for disadvantaged groups were identified, and providing cessation services in different settings appeared to improve access. Overall, the evidence on the effectiveness of some interventions in increasing quitting behaviour and access in disadvantaged groups is limited. While many studies collected socioeconomic data, very few analysed their contribution to the results. 253

Many studies demonstrated that there was poor recording of smoking status in practice patient records and more can be done to identify patients who smoke. 254,255 While recording improved following the introduction of the new GMS contract in 2004 (for patients with various comorbid factors: CHD, stroke or transient ischaemic attack, hypertension, diabetes, COPD or asthma254,256), a 2008 study showed that levels of recording were still low in some practices. This study of practices in Nottingham found a range of recorded smoking status between 39% and 100%. 255 However, of key importance was the fact that, while in many practices the ascertainment of smoking status was incomplete and/or inaccurate, it was the failure to intervene appropriately on known status that remained the biggest challenge. 255

A GP’s query regarding a patient’s smoking status and subsequent recording of this status are the first steps in an effective smoking cessation intervention, as indicated by NICE, the Health Education Authority/Thorax and the US Department of Health and Human Services guidelines. 242 Virtually all English GPs (98%) report following these steps. 257 It is clear that in the UK the universal uptake by GPs of the QOF process258 and its reimbursement for recording smoking status and providing smoking cessation advice have had an impact on levels of recording smoking status, although rates of recording do vary. 254–256 Smoking cessation is a good example of the QOF changing the way in which practices work, by providing prompts and reminders which have helped to ensure practices try and reach people. In group discussions GPs also referred to systems and prompts, but there was some disagreement among GPs about whether these were helpful or not.

However, knowing a patient smokes does not routinely trigger a smoking cessation intervention by the GP. Fewer than half of all GPs consistently advise patients to stop smoking. 257,259–261 The reasons for this are complex. In addition to not believing that the patient is motivated enough to quit smoking, many GPs report concern about harming the doctor–patient relationship by broaching the topic of smoking cessation with a potentially unreceptive patient, deciding, on balance, that protecting this relationship is more important than providing a smoking cessation intervention with questionable odds of succeeding. 262–267 Pilnick and Coleman267 argue that smoking cessation is complex and that offering help on how to quit is difficult. They also suggest that the benefits of nicotine replacement therapy may be due not simply to the clinical effects but also to the legitimisation of the view that stopping smoking is an appropriate problem for medical intervention.

A randomised controlled study of 74 GPs in England has shown some success in overcoming GPs’ concerns about harming the doctor–patient relationship by providing GPs with a simple desktop resource which triggers a smoking cessation intervention. 263 Furthermore, English GPs are more likely to initiate discussion regarding smoking when the patient presents in the surgery with a smoking-related problem. 268,269 Training GPs and nurses in effective delivery of stop-smoking interventions results in better outcomes. 94 Other research supports or encourages such action, including evidence from a Cochrane review. 259,270–272 A lack of knowledge among GPs about how to affect behaviour change when working with addiction, even among GPs who regularly initiate smoking cessation interventions, presents evidence of the need for further specialised training for GPs. 273

Overall, few papers examined the delivery and organisation of smoking cessation services and only two compared different service approaches. There is a mix of activity in general practice and communities more widely but there appears to be little evidence on how best to organise services despite an increase in the recording of smoking status and incentives to promote smoking cessation. 255,256,274–277 In practices, GPs and nurses provide smoking cessation interventions and, while more nurses may be involved, they have restricted prescribing rights. 278 Lack of time in consultations remains a key barrier to services, as do concerns about how to raise smoking issues within the consultation. 279 However, increasing awareness of smoking status, targeting smokers and provision of training to health-care professionals increase cessation interventions in primary care. 261,280–283 There is some evidence to suggest that patients prefer specialist smoking cessation services but that not all GPs make appropriate referrals for these services. 284

Cardiovascular disease

Papers discussing prevention activities related to CVD were one of the most common groups found in our initial literature search and were the largest group of papers included in the evidence synthesis. Within these, by far the most common activity taking place in primary care with regard to the prevention of CVD is the attempt to modify lifestyle risk factors in symptomatic and asymptomatic patients who are deemed to be in the ‘at risk’ category. The risk factors that are commonly targeted are obesity, poor diet, lack of exercise, alcohol use and smoking. Other key areas were the monitoring of, and strategies to control, blood pressure in symptomatic patients, especially those patients with peripheral arterial disease, patient compliance with medications (especially statins), strategies to improve patient outcomes for people with angina and the cost-effectiveness of nurse-led clinics for those with existing CVD. In addition there were a number of studies looking at GP and practice nurse knowledge and behaviour with regard to guidelines for CVD.

Ten papers directly addressed lifestyle risk factor improvement in asymptomatic patients. 285–294 Five papers285–289 looked specifically at trying to alter patients’ behaviour regarding the amount of exercise they took, their diet and other risk factors (e.g. smoking). This was delivered through short interventions which employed a wide range of activities, such as tai chi, exercise classes, guided walks, counselling sessions and, in one case, a home exercise kit. Two papers looked specifically at the effect of risk factor awareness on asymptomatic patients’ behaviour. 290,291 The first of these studies looked specifically at the level of awareness of risk factors for cancer and CVD. 290 It found that there was a good level of understanding for the lifestyle risk factors for CVD but that risk factors for cancer were less well understood. One study focused on lifestyle risk factors with particular reference to patients with high blood pressure where patients were given information and advice about diet and weight loss to help lower blood pressure. 292

Some of the papers provided details about who delivered the interventions. The role of the GP in giving advice concerning lifestyle risks was considered specifically in two papers. 293,294 The first of these examined the use of ‘pop-up’ reminders on practice computers to improve GPs’ awareness of patients with CVD risk factors. The ‘nudge’ screen would prompt the GP to offer advice and give information about lifestyle risk factors for the patient. The second study examined tailored practice care plans which included ‘motivational interviewing’ and target setting for lifestyle change delivered in a primary care setting.

There were five studies on the effectiveness of nurse-led clinics for secondary CVD prevention in a primary care setting. 295–299 All of the studies looked at clinics where practice nurses monitored lipid levels, blood pressure, medication use and other risk factors such as diet, exercise and smoking. Three studies300–302 looked at the effect of nurse-led interventions in primary care on patient compliance with drug regimens for high blood pressure.

Three studies looked specifically at the management of patients with peripheral arterial disease. One study303 looked specifically at the patterns of management of patients with CVD and peripheral arterial disease. It found that, while most patients were given advice concerning smoking cessation and blood pressure, fewer than half were given advice about weight loss or exercise. The second study304 looked specifically at the treatment received by patients who presented with either peripheral arterial disease alone or peripheral arterial disease and CVD. It found that those patients who had peripheral arterial disease alone were given suboptimal care. The third was a simple description of prescribing practices for peripheral arterial disease patients in Ireland. It found that prescribing patterns varied widely and that patient compliance was less likely when they were required to take a number of different medications. It was also found that prescribing patterns for this group of patients was influenced strongly by budgetary concerns, as some of the medications were expensive.

Two papers looked specifically at ethnicity and CVD. 305,306 Murray et al. 305 looked at a strategy to reduce health inequalities between the white ethnic majority and ethnic minorities in Wandsworth, London. It was a 10-year study using blood pressure and cholesterol measurements as proxy markers for CVD. It was difficult to assess whether or not a significant narrowing of inequalities between ethnic groups had been achieved and there were also insufficient data in the paper to assess the approach. An additional problem was that small numbers of some ethnic minorities and South Asians from different countries were pooled. Previous studies have found that CVD risk is not necessarily uniform among the South Asians population, and there are important differences between Indians, Pakistanis and Bangladeshis. The second study306 looked at the feasibility of using screening services to reach ethic minority groups in Sandwell. It was a series of community-based screening events for CVD and diabetes. The screening events were very popular with the ethnic minorities who attended them, but it unclear if the strategy actually resulted in any lifestyle risk factor change, as this does not appear to have been assessed.

Two papers looked at sex and CVD. 307,308 One study used computerised data from primary care practices to investigate how CVD is treated. 308 The results showed that considerable variation existed between practices, though women were consistently given suboptimal care with regard to prescribing and diagnostic tests. Also, doctors did not always follow the guidelines and procedures. The second study307 looked at the clinical management of patients with stable chronic angina pectoris. Women received a lower level of risk factor assessment, secondary prevention therapy, cardiac investigation and coronary revascularisation.

There was just one paper that looked specifically at deafness and screening for cardiovascular risk. 309 The study was carried out in Sandwell and was part of a community-based programme to provide health promotion advice in a deaf community. The approach in the study did not reduce estimates of CVD risk after 6 months among deaf people determined to be at ‘high risk’, despite the high-risk group of deaf people being given advice and being referred to their GP. A speculative link with language barriers was identified (e.g. a lower reading ability preventing access to information on health promotion, and the discovery that there was no sign for cholesterol in British Sign Language).

Community participation

Community participation in general practice and the role of primary health-care teams in tackling population health beyond the surgery door have long been championed. However, the research literature betrays a focus in favour of a biomedical and technical approach to public health. Most papers concern what might be called ‘preventative medicine’ rather than more complex community-based interventions. Regarding this subset, three immediate observations can be made.

Firstly, there was a paucity of papers – only 18 papers in this category – reflecting the focused nature of our original search strategy. Secondly, few studies focus on outcomes: most papers described case studies and used interviews with key stakeholders to illustrate the potential benefit of participatory approaches. Few studies attempted to measure health processes or outcomes. Thirdly, methodological quality was poor. The evaluation of complex community interventions is a study in itself. Unfortunately, beyond rather woolly references to ‘action research’ and the complexity involved, few studies were of the quality required to ground firm conclusions. In many instances, the initiative had apparently progressed without the necessary resources to undertake detailed evaluation.

Six papers examined health promotion initiatives from the vantage of different groups involved. 310–315 For example, Daykin and Naidoo310 explored the views of health professionals on promoting health in areas of poverty with low-income clients. They questioned the assumption that primary health care is the best setting for health promotion; several constraints are identified as limiting its effectiveness. Firstly, the emphasis on individual lifestyle change that often informs government strategy for health promotion causes frustration among health professionals. Characteristics of primary health care limited its scope in addressing socioeconomic inequalities and engaging in multiagency initiatives. Their interviews indicate that, despite these constraints, workers in primary care demonstrate high levels of awareness and commitment to a social model of health promotion. On the other hand, Abbott and Riga311 exposed limited understanding of the links between deprivation and ill health among health-care professionals in east London. Better professional education on equality, diversity and public health approaches, and improved patient education to overcome knowledge barriers were proposed solutions but their views seemed to be anecdotal rather than evidence-based. Both studies suggested a need for resources and training.

Conceptual uncertainties are mirrored in the views of deprived residents. Parry et al. 312 found that they conceptualised health in both physical and non-physical (e.g. being happy, peace of mind) terms. Sense of place affected health through three key levers: physical structures (e.g. quality of housing), social structures (e.g. friendliness of neighbours) and provision of services (e.g. transport). The interplay between these factors was complex and fear was a common node in many of the pathways residents described linking place with health.

A survey of faith groups in Dundee313 found they were already involved to different degrees in health promotion and ready to engage in more. They constitute an untapped resource whose potential has been evidenced in the USA. Kennedy and colleagues314,315 have explored the benefits arising from lay involvement in community-based nutrition programmes in poor neighbourhoods. These include increased service coverage and the ability to reach populations, personal development and enhanced social support.

A similar gulf in expectations characterised health professional views on health visiting in an orthodox Jewish community. 316 The community only partially understood or valued the health visitors’ role, while health visitors’ awareness of aspects of Jewish culture was limited. Closer partnership is required to develop a service that the community finds valuable.

Participatory approaches to health-needs assessment have gained credence as an effective substitute for traditional, epidemiological approaches. A comprehensive literature review found that research on African and Afro-Caribbean populations was limited and uneven, focusing on specific particular diseases. Okereke et al. 317 used a triangulation approach to identify the community’s perspectives on local priorities.

A narrative review of published research on community capacity to manage hypertension among black groups identified 27 relevant studies. 318 However, collaborative methods yielded insufficient evidence of effectiveness in terms of quantifiable outcomes. Intrinsic assumptions about the homogeneity of communities and how they are represented were challenged.

The Community Health Educator is a model developed through a series of participatory action research projects in UK cancer screening during the 1990s. However, little is known about its effectiveness, or how researchers/practitioners engage with communities. 319 Researchers and practitioners need to be flexible in accommodating the ambiguity implicit in the concepts of ‘community’ and ‘ethnicity’. The efficacy of social networks has similarly been questioned. 320 The prevailing assumption that the community health educators’ personal networks are ethnically and neighbourhood-bounded has influenced intervention design but may not reflect the complexity of urban communities. Concepts such as ‘social capital’ and ‘embeddedness’ are notoriously difficult to operationalise.

Welfare advice services

We evaluated eight papers that described or reported research findings on welfare advice services in general practice. These papers constituted a focused assessment of the organisation, role and impact of advice services for practice populations. The general conclusion of these papers was that welfare advice services are an excellent strategy by which primary care organisations address the social, economic and environmental influences on the health of their population. 321,322 In addition, studies show that services are valued by patients and practice staff323,324 and lead to positive social and health outcomes for patients. 325–328

While a number of the studies were small, the quality of the research was mainly rated as A or B (see Chapter 3 for definitions) and the cumulative findings suggest that there are measurable health benefits for those patients who achieve additional income and high levels of patient satisfaction. There were no cost-effectiveness studies of these services. However, most services were delivered by existing welfare services, with practices merely providing premises.

Organisational issues

The focus of most studies we examined in the review was on general practice, although a number of papers described or evaluated activities linked to the primary health care team but delivered in community settings, people’s homes or schools. Generally those activities outside the practice seldom involved GPs and were predominantly delivered by community nurses or other professionals (e.g. health promotion specialists, pharmacists). The papers reviewed identified a number of important issues related to the location of health improvement activities. For example, there is evidence to support the use of community and school-based locations for health improvement activities with people from ethnic minorities. Evidence shows that schools are more likely to be the most appropriate setting for family-based interventions for ethnic minority groups because of easier access and availability, and the relationship between healthy lifestyle and curriculum. Places of worship are more likely to engage in a dialogue around traditional food practices. Places of worship also provide access to the wider family, including grandparents, and offer opportunity for culturally and family-specific support. 329

In group discussions with GPs, the variation in approach and how services were organised and delivered was very evident. Some practices had invested in training and developing their practice nurses and employing health-care assistants to undertake routine screening and monitoring. In some cases this activity was undertaken by running clinic sessions or group sessions while in others it was based on individual patient recall systems. Younger GPs tended to focus more on the impact of the QOF on these sorts of activities. There was much debate about the impact of the QOF not only on preventative interventions but on the way the practices organised activities and patient care. There are some new issues that have been forged by the QOF such as a more embedded emphasis on systematised models of care through the use of disease registers, computerised working methods and greater use of clinical templates. 93 The literature on the impact of the QOF also suggests that it has focused on individual interventions with patients and more depersonalised care. 93,330

While there has been a particular focus on GPs, it is clear that primary care approaches to public health involve all workers in the practice, as well as the wider primary health-care team. There has been an increasing involvement of practice nurses in patient care, particularly for routine screening and review activities, although the majority of consultations are still with the GP. The introduction of the QOF has been important in opening up new opportunities for practice nurses, although these are a continuation of pre-QOF trends (increasing employment of health-care assistants, delegation of clinical work from GPs to nursing staff). 331 These approaches are also reflected in general practice approaches to prevention and health promotion. 65

While this review describes a wide range of activity, there does not appear to be any consistency in how services are organised between practices. From the papers examined in this review it does seem that many of the interventions have been successful in improving record keeping in practice. In areas such as CVD, diabetes, etc., there is an emphasis on identifying risk factors and on education, most of which seems to have limited benefit. Interventions aimed at changing lifestyle (e.g. for weight management, exercise and CVD) of patients who are asymptomatic seems to be limited in terms of any long-lasting impact. Studies tend to be short in duration and rely on self-reported behaviour change. Studies suggest that there are huge variations in prescribing practices, and the amount of information and advice given by GPs. A number of papers report that GPs feel ill-equipped to give that sort of advice. Practice nurses are used more often to provide information and monitoring of patients, both those ‘at risk’ and symptomatic patients. Generally speaking, patients seem to like this, but with disappointing outcomes with regard to patient benefit. However, it is also not clear how far lifestyle advice is either delivered or remembered by patients. In a study of patient views on health promotion and lifestyle advice in general practice, only 6% of patients recalled receiving advice on diet, 4% on exercise and smoking and only 2% on alcohol. 332 Few papers examined services for specific groups of people or reducing health inequalities between groups.

More attention needs to be paid to skill mix in general practice, although the nature of general practice varies. In this sense, although the systematisation of activity provided by QOF has been seen as restrictive for many practices, it has also been helpful. 65,93,331,333

General practitioners also referred to other financial support for public health activity in practices. Some GPs recognised that the principle of capitation funding should provide some incentive to explore prevention activities but felt that limited time in consultations and the demands of clinical practice made this difficult. Some PCTs have been using other elements of the GMS contract such as LES to support health promotion activity based on financial incentives. However, some respondents questioned if such activity being incentivised was already covered either within the core contract or through the QOF. 63 In a more recent study on the early development of CCGs, the researchers identify a number of uncertainties in this area and report mixed views among CCGs and public health staff about the relationships between GP practices and public health. 334 Data from this study also suggest that there is considerable uncertainty and concern about how public health will work in the new system.

Conclusion

The focus of attention is mainly on individual prevention approaches, with practices engaging in both primary and secondary prevention. The focus is on medical conditions related to prevention of specific diseases such as diabetes or CHD. GP activity appears to be driven by specific contractual incentives and conditions. However, the wider literature also suggests that general practice is also affected by other factors such as peer pressure, relationships with public health departments, education and training. While direct incentives such as contracted standards or the QOF do influence practice, there is some debate about whether or not GPs simply react to maximise their income and do not prioritise activities towards those most in need. The use of thresholds can provide perverse incentives and there is some evidence that GPs use exception reporting to maximise their points score and income. However, there is some evidence that the wider primary care team, such as health visitors, are engaged in health improvement interventions such as breastfeeding support where variance in outcomes have been explained from a service delivery perspective. 221

There appears to have been little shifting from the position described by Lawlor et al. ,79 who argued that GPs do not adopt a population approach as they focus on high-risk patients and adopt a predominantly medical approach. They also argued that GPs felt that giving lifestyle advice interfered with the doctor–patient relationship. However, it is clear that many GPs see health promotion as an integral part of practice, whether as individual approaches to primary or secondary health improvement or as a practice approach to improving the health of their patients.

There is little evidence to show that GPs undertake wider public health roles in terms of population surveillance beyond contractually defined screening and monitoring. These do not target major areas of health concern (such as mental health, eye problems and oral health), which receive less attention despite the potential of general practice to play a leading preventative role. While there is good evidence, for example, that GPs address eye health for people with diabetes, less is known about their wider role in screening for eye problems. It is important that GPs link mental and physical health issues, as people with mental health problems, particularly severe mental illness, are at increased risk of a number of physical health problems. 335

Introduction

The evidence on the impact and overall effectiveness of prevention activities was limited in both scope and quality. We included 347 papers in the evidence synthesis. A list of reviewed papers is included as Appendix 5. The extent to which selected papers discussed the organisation and delivery of health improvement activities, or the interaction of general practice and primary health-care team staff with services such as screening that are offered through national programmes, varied considerably. Therefore, not all papers are directly referred to in this discussion; references are given where specific or detailed comments have been made, or where papers support general statements. It is important to note that the focus of the discussion is not necessarily on the effectiveness of the intervention per se. Our aim was to examine whether or not there was evidence of effectiveness of different ways of organising and delivering intervention programmes, for example differences in who delivered an intervention, where it was located or how something was provided. Very few studies provided comparisons of approaches and many studies lacked sufficient detail in the papers to make judgements about effectiveness.

Where they are of value we have referred to relevant Cochrane reviews. We found few data on service delivery and organisation in the reviews but where relevant this is incorporated in the discussion. The chapter is split into sections by area of activity. Key points relating to the organisation and delivery of activities are summarised at the end of the chapter in relation to evidence on:

• cost-effectiveness

• interventions with patients

• interventions with practitioners

• location of the intervention.

Child health

With regard to targeted interventions to identify children at risk of ill health and provide support to them, the evidence suggests that the most effective approach involves consistent and ongoing engagement with families and parents. 218,336,337 Approaches to identifying children at risk of ill health did not lead to significant improvements or changes in behaviour, and the provision of advice by nurses or GPs did not have any impact on behaviour change. 338,339 The evidence on the effectiveness of interventions aimed at intermediary organisations and interventions undertaken in the community and schools suggests that these generally have a positive outcome, although such impacts were small. For interventions with children and their parents, location of activity is important. For the promotion of breastfeeding, delivery needs to be in premises seen by mothers as suitable. 221 Interventions also need to engage a range of professionals and the community. 340–342

While schools were found to be good locations for conducting health promotion, there is little evidence of any substantial change to children’s diet, exercise levels and weight loss. With regard to pre-school children, a systematic review of weight loss initiatives in children aged under 5 years found little robust evidence, and no significant evidence of effect was found. 217

The evidence on the benefit of interventions to promote and support safety advice to reduce accidents suggests that such schemes do not reduce overall numbers of childhood accidents. 337–339,343 One key area of current concern relates to the prevention of obesity in children. Evidence from a recent Cochrane review344 which examined studies from a range of countries in a variety of settings (including schools, home-based and community settings) concluded that child obesity prevention interventions can be effective, especially when assessed using body mass index as an outcome measure. However, the finding should be interpreted cautiously because of ‘unexplained heterogeneity and the likelihood of small study bias’ (p. 2). 344 It was also difficult to know if the outcomes of short-term interventions can be sustained over the longer term (studies were included if they evaluated interventions that were in place for a minimum of 12 weeks). Delivery of effective interventions tended to focus on schools and home settings with input from primary care professionals such as school nurses, health visitors or community support workers rather than within GP practices.

Delivery and organisation of screening services

Despite the importance of general practice as a location of service delivery for a range of screening activities, only a small number of studies systematically compared or explored alternative approaches to performing these activities. While 83 papers were identified for inclusion in the evidence synthesis, 36 were graded as A and 34 as B, with a substantial number of cross-sectional surveys and few trials or evaluations.

Nurse-led cardiovascular screening was one of the few areas in which quantitative evidence on service organisation and delivery could be drawn together, albeit from a few studies with evidence of limited effectiveness of the interventions overall. This has relevance to the NHS Health Check programme, for which little evaluative research has been published yet, owing to its relatively recent introduction. For most other screening services delivered in primary care, our review did not find literature on the effectiveness of different modes of organisation. Instead, the papers addressing particular screening services often focused on different issues. For example, in the case of cervical screening, three studies explored women’s views on non-attendance,129–131 two studies investigated informed consent,126,345 one study examined women’s views on result communication,132 one study assessed the effect of an intervention to improve screening uptake346 and one study examined general practice factors associated with uptake. 347 For some areas, including promoting the uptake of cervical and chlamydia screening, papers provided indications of important considerations for effective service delivery and organisation when considered collectively. In other areas, namely nurse-led systematic screening for atrial fibrillation, alcohol screening and health risk appraisal for older people, evidence was limited to the results of single studies.

One central debate about the delivery of screening relates to targeted, opportunistic and universal screening for diabetes and CVD. These are particularly pertinent in relation to the introduction of the NHS Health Check. 151,348–350 At the time of this review, no substantive evaluations of the NHS Health Check had been completed. However, Dalton et al. 133 found that uptake was only 45% and that uptake was variable between different population subgroups. While many South Asian and younger men in their study had higher than average uptake, many people at higher risk were not screened, which may unintentionally lead to increasing health inequalities. Internationally, it has been shown that screening in other health-care environments can be an effective way of reaching a wider population. 351–354 However, such approaches are not without problems in determining cost-effectiveness, identifying appropriate screening tests and integrating screening into business practices such as pharmacies. 136,352 One pilot project that used opportunistic screening in a supermarket did succeed in reaching specific at-risk groups, although again the ability to do a full range of screening tests limited the overall success of the project. 355 However, extending accessibility to screening may not be a sufficiently effective approach to reducing CVD risk, and studies suggest that multiple risk factor interventions in primary prevention, comprising counselling, education and drug treatments, are more likely to be effective in high-risk groups than in the general population. 356

The organisation and delivery of weight management, diet and exercise interventions

There is some evidence to support strategies which have shown promise for incorporation into GP weight management practices, including referral to commercial slimming services375 – supported by evidence that such services may have better long-term outcomes than commonly reported376 – and partnering with other health-care providers such as dieticians. 236,238,377 However, there is no evidence that the effectiveness of weight management in general practice is long lasting. A review217 on weight management programmes for children found no good-quality evidence of a beneficial effect.

Similarly, in relation to exercise referral schemes (ERSs), the evidence suggests that there may be limited initial benefit but no strong evidence for long-term effects. 228,378–380 There are problems with non-attendance; younger people in particular are less likely to attend. Practice nurses and health educators get better responses than GPs. 380–382 The evidence suggests that there is some benefit in discussing exercise in general practice and that systematic approaches to identifying individuals and to providing sustained advice and support – especially by nurses and health educators – may have some beneficial impact. Lawton et al. 383 have shown in a RCT that physical activity promotion can improve behaviours in general practice care when co-ordinated with exercise on referral, which is widely accessible in the UK. However, the impact appears to be short-term and limited, although one RCT384 suggested that differences can be sustained for 12 months. A systematic review found that lifestyle counselling interventions delivered by primary care providers in primary care settings to patients at low risk (primary prevention) appeared to be of marginal benefit. 39 Resources and time in primary care might be better spent on patients at higher risk of CVD, such as those with existing heart disease or diabetes.

There was very limited evidence to support the use of brief interventions for exercise. 228 Brief interventions (3–10 minutes) or simple pedometer-based programmes delivered through health professionals can lead to substantial increases in patients’ activity levels (by ≈ 30%). 385 Marginal improvements may, however, be cost-effective. 386 Most studies were based in general practice but one study in a breast-screening clinic found that women were interested in obtaining advice and information on weight loss. 387 While evidence on the effect of brief interventions for exercise and physical activities in primary care appears mixed, they have been recommended as an effective intervention in the USA. 385 Current NICE guidance also supports brief interventions and recommends that primary care practitioners ‘should take the opportunity, whenever possible, to identify inactive adults and advise them to aim for 30 minutes of moderate activity on 5 days of the week (or more)’ (p. 4). 388

Weiler and Stamatakis389 have argued that regular intervention built on existing, longstanding primary care relationships may have a significant impact on patients. There is, however, only limited support for this approach. 390 In addition, there is little strong evidence to support the benefits, especially in the long term, of ERSs. A Health Technology Assessment review in 2011391 concluded that ‘there remains very limited support for the potential role of ERS for impacting on PA [physical activity] and, consequently, public health . . . Although ERS programmes in our review aimed to increase medium to long-term PA, they were typically based on only a 10- to 12-week leisure centre based period intervention’ (p. xi). The cost-effectiveness of such schemes is also questionable. 386

There have been three relevant Cochrane reviews392–394 that directly examined evidence for general exercise and weight loss interventions. In addition, there were a number of reviews that specifically addressed diet, exercise and weight loss for the prevention of specific conditions including diabetes395–397 and CVD. 398 All the reviews drew on the findings of international literature and therefore their direct relevance to UK general practice cannot be assumed. The reviews concluded that it is not possible to draw any firm conclusions about the effectiveness of interventions aimed at changing the behaviour of health professionals or the organisation of care to promote weight reduction in overweight and obese adults. 393 Neither is there any consistent evidence to support the hypothesis that multicomponent community-wide interventions are effective at increasing levels of physical activity. 394 The third review found that professional advice with ongoing support has only a moderate effect on increasing self-reported physical activity in the short to mid-term. 392 However, in the studies for this review most participants were volunteers, limiting its applicability to a general practice population. Generally, however, the evidence supports systematic approaches working through disease registers and targeting patients most in need. Systematic programmes of support/advice241,399 are more likely to be effective than opportunistic approaches. 400

The success of smoking cessation interventions is generally supported by good evidence and they are seen as cost-effective. 401–404 There is some variation in reported success rates of smoking cessation interventions. 405 Quit rates in the range of 10–20% of smokers exposed to a smoking cessation intervention are commonly cited. 405–410 It should be noted that there is no universally accepted definition of ‘quit rate’, although most studies use a minimum period of 6 months of abstinence from tobacco use – with the definition of ‘abstinence’ itself being the subject of some disagreement. Brief smoking cessation interventions within general practice settings with patients who smoke are marginally effective. 403 Unassisted 12-month quit rates in the general population are in the range of 2–3%. While brief interventions by GPs may increase this level by an additional 1–3%,411,412 this could lead to a further 75,000–92,000 patients per year in England who could quit smoking if GPs increased the number of smoking cessation intervention initiatives by 50% and accompanied them with nicotine replacement therapy or other pharmacotherapy. 242,413,414 Motivational interviewing by GPs over a number of multiple sessions may be more effective but there is no evidence to identify the optimal number of sessions required. 404 Brief interventions by nurses were not found to be as effective as those by primary care physicians. 402 Community pharmacists who provide support to patients to quit smoking may, when trained, have a positive impact but the evidence is limited. 401 The evidence on community-based interventions lacked sufficient quality and detail to provide guidance on who should deliver interventions or how they should be delivered.

There is also debate within the literature regarding the optimal degree of brevity versus intensity of these interventions. 410,411 Research shows that the move towards guidance favouring ‘intensive’ brief interventions is misguided,415 but no apparent disagreement exists regarding their utility and cost-effectiveness, regardless of intensity. 242 The use of pharmacotherapy, such as nicotine replacement therapy, varenicline or bupropion, with or without a brief GP intervention, increases quit rates considerably, by upwards of 1.5 to 3 times. 407,409,411,413,416–419 Following a brief GP-led smoking cessation intervention, referral of patients to specialist programmes, such as those offering behavioural support, or even guidance to telephone help lines, has also been shown to enhance rates of smoking cessation. 257,415

Overall, the evidence suggests that smoking cessation advice provided by trained, specialist advisors delivered over a number of sessions with follow-up support provides the most effective interventions. Brief interventions by GPs supported by pharmacological support may also have an important impact on smoking rates in absolute terms.

Cardiovascular disease

Of the 10 papers that provided evidence on interventions aimed at lifestyle risk factor improvement in asymptomatic patients, five285–289 examined altering patients’ behaviour regarding the amount of exercise they took, their diet and other risk factors (e.g. smoking). All these papers reported improved behaviour after the study period was complete, and one288 reported a significant improvement in waist circumference at the end of the study. Three of the papers suggested that the intervention programmes were worthwhile and cost-effective (see Evidence on cost-effectiveness). However, there were concerns about the quality of some of the evidence. For example, one of the studies288 was conducted over a very short period (12 weeks), limiting its claim of significant improvement, one was a pilot study285 and one relied on self-reported levels of behaviour change. 289 One study289 pointed out that the programme had ‘failed to engage’ with the ‘hard to reach’ population, especially people from deprived areas and men. In relation to lifestyle risk, some papers described providing information to patients but the effect of risk factor awareness on asymptomatic patients’ behaviour was shown in only one study. 291 This concluded that, even when patients are fully informed of their personalised disease risk factors such as lack of exercise, this knowledge is in itself unlikely to change their behaviour. In another study, where patients with high blood pressure were given information about diet and weight loss to help lower blood pressure, it was found that after 6 months there was no improvement in the study group, either in their risk factors or their blood pressure readings. 292 As discussed earlier in this chapter, in the OXCHECK study the nurse-led approach, which targeted whole families and offered intensive follow-up, was more effective in terms of increasing life expectancy, but less cost-effective, than the less intensive approach of the OXCHECK study, highlighting the need for more intensive and consistent intervention programmes. 357

The use of information technology systems to prompt practitioners to offer advice had no impact on whether or not physicians offered such advice to patients, but it did improve practice record keeping. 294 Another trial, examining motivational interviewing and care plans provided by practitioners, found that they did not lead to benefits in terms of blood pressure, cholesterol or health-related quality of life, although it found that there was a reduction in the number of hospital admissions. 293 Studies that examined the effectiveness of clinics where practice nurses monitored lipid levels, blood pressure, medication use and other risk factors such as diet, exercise and smoking suggest that these led to improvements in some aspects of secondary prevention except for smoking cessation and exercise levels. However, what small improvements there were may be misleading, as they relied on patients self-reporting. 295–299 In a study that looked at the effect of nurse-led interventions in primary care on patient compliance with drug regimens for high blood pressure, Schroeder et al. 300 found that there was no effect on patient compliance, while another study302 found that there was some improvement. Bane et al. 301 concluded that compliance was more likely if patients felt that they had been ‘listened to’ by their GP in the initial consultation.

Murray et al. 305 looked at a strategy to reduce health inequalities between the white ethnic majority and ethnic minorities in Wandsworth, London. It was a 10-year study using blood pressure and cholesterol measurements as proxy markers for CVD. Although the intervention had resulted in a lowering of the risk factors across the board, there were methodological problems such as the small numbers of some ethnic minorities and the pooling of South Asians from different countries, which limit an assessment of effectiveness. Two papers looked at the effect of educational programmes on outcomes for patients with angina. 420,421 The first examined the cost-effectiveness of a personalised health education programme for people with angina in Belfast. Those in the intervention group had three visits per year from a health visitor, whose brief was to discuss ways of living more easily with their disease and ways in which risks of further events might be reduced. The results showed significant improvements in survival and self-assessed quality of life in the study group. The second looked at a similar programme of education for angina patients. The intervention group received personalised health promotional advice from a practice nurse four times a year for two years. The results showed that the study group had improved quality of life and physical mobility, and there were significantly fewer deaths in the study group than in the control group. It also showed that it was important that GPs gave advice to patients concerning physical mobility but that GPs were reluctant to do this because they felt ill-equipped to give it, and that they lacked the training.

Community-based approaches

Despite growing recognition of the potential of community-based efforts to improve public health, the costs and ‘added value’ of such approaches remain unclear. The most substantial systematic review sought studies that included controls of suitable comparative groups and measured both outcome and costs. 422 Of 4405 references screened, 152 records were potentially relevant, but just eight studies met the inclusion criteria. Seven of these reported positive findings, but none was designed to evaluate impact or cost-effectiveness. The review found tentative evidence that community engagement is part of a multifaceted approach to health promotion, may have positive effects and could be cost-effective. As discussed earlier in this chapter, Cochrane reviews of community interventions for exercise and weight loss, and for smoking found little robust evidence of positive effect on smoking prevalence or sustained weight loss. 394,414 The studies also either said little about who delivered the interventions or found little difference in who delivered interventions or how they were organised.

A further uncontrolled study looked at the impact of a training programme in community-oriented primary care on 11 practices. This detailed benefits to staff, for example learning new skills, and to patients in terms of intermediate health outcomes. Costings suggested the projects undertaken were likely to be cost-effective. 423 Overall, the dearth of evidence suggests the desirability of community engagement at all stages of project development to capture the community’s priorities and perspectives, and assess appropriately the costs and benefits of engagement.

A large literature described case studies located in different parts of the country. These predominantly involved urban, deprived populations but rural communities were also covered. These provided testimony to the impact of the programmes under consideration, for example through interviews and focus groups, often supplemented by process data (e.g. attendance rates). However, other measures of impact were seldom forthcoming. The aim was to share good practice and provide learning for others.

Taket and Edmans,424 for example, listed key elements in a framework for community involvement in health and regeneration:

1. work at the most local level

2. involve community in decision-making at all stages

3. local recruitment to paid posts involved, wherever possible

4. ensure that work involved supports local economy in all sectors

5. build capacity

6. train

7. give specific attention to planning and management of change

8. specific attention to building partnership between organisations

9. work towards building trust and openness

10. adopt innovative and multiple ways/processes of working

11. recognise the importance of facilitation

12. adopt a participatory action research approach

13. adopt a holistic approach

14. keep a high profile

15. recognise the need to work creatively on constraints/challenges imposed by funding conditions

16. see regeneration as intimately involved with changing, where necessary, the way that mainstream services operate

17. recognise, respect and work with diversity, in an empowering fashion

18. recognise that it takes time

19. build in external evaluation.

Such learning may also identify barriers. 425,426 Inadequate resource allocation and poor continuity of structures/personnel are frequently barriers to community-based partnership programmes. Ritchie et al. 425 made recommendations to address them (e.g. the need for contingency plans and funding to deal with indeterminate aspects). Studying a social exclusion partnership in Scotland, Carlisle426 identified rival relationships between key partners and the sometimes conflicting priorities of health promotion versus community development as potential barriers.

The literature on community participation identified some important methodological issues and highlighted the inconclusive nature of much of the research. The specific role of primary health-care professionals was seldom described and the contribution of GPs was notably absent. Given our search terms, it is likely that this absence of a GP contribution accurately reflects the literature in this area. Previous reviews of public health and primary care have also noted the limited engagement of general practice and GPs in community-based public health activities, although community nurses have been more engaged in such activity. 73

Summary of the evidence on health improvement in general practice

The evidence synthesis reported here is not designed to provide a comprehensive assessment of the effectiveness of health improvement interventions undertaken in general practice or by primary health team practitioners. Our interest was rather to explore the nature of the evidence, its scope and limitations and to specifically identify what evidence there is on the way health improvement interventions are organised and delivered.

Conclusion

Our review of the literature suggests that there is little robust evidence to demonstrate positive behaviour change or clinical outcomes for preventative activities in primary care. Most papers examining primary prevention were either descriptions of services, commentary about primary prevention or research studies that were rated as of low quality. Few high-quality studies were identified in all key areas of primary prevention. The main exception was in relation to smoke-stop services, where services to support quitting are viewed as cost-effective and approaches that provide ongoing support for smokers wanting to stop, or those who have quit, delivered in a targeted way, are more likely to be effective. 252,439–442 In addition, there is still potential for better identification of smokers and raising smoking awareness by a range of health and other care professionals. The evidence to support interventions for increasing exercise or reducing weight was limited and suggested that such approaches in primary care or in schools have little impact and that any benefit is generally only short-term. In particular, the cost-effectiveness of primary preventative lifestyle interventions is difficult to determine because of the diverse nature of the interventions (type and organisation, different target groups) and research methodology. 442,443 There is good qualitative evidence from a large study of breastfeeding support to demonstrate the health services attributes that might be necessary to deliver interventions effectively. 221

Clearly advice from the GP is still rated highly by patients and may be acted upon. In terms of secondary prevention, there is evidence that more is being done but it is not clear how effective this is. For example, in relation to CHD, general practice has been identified as an ideal setting for delivering secondary prevention. However, the evidence indicates that provision is not as effective as it could be and, while the provision of secondary prevention can be improved by using specific disease management programmes, the optimal mix of their components remains uncertain. 444,445 There is insufficient evidence to make clear decisions about the balance between universal, opportunistic and targeted health promotion interventions. A key point, however, is that multiple risk factor interventions in primary prevention comprising counselling, education and drug treatments were more likely to be effective in high-risk groups than in the general population. 356

Finally, the current interest in incentivising health improvement practice through performance payment systems such as the QOF has had an important impact on both the areas of activity that are prioritised and how services are organised. There is good evidence that practices have responded in different ways to these incentives in terms of how they organise practice. What is clear from the evidence and discussions with practitioners is that there has been a shift to employ nurses and health-care assistants to do much routine prevention work including screening. The QOF has also led to improved recording of the health status of patients in their practice. Interestingly, a Europe-wide survey of GPs in other countries highlighted the lack of reimbursement as a key reason for not engaging in health promotion activities. 446 As the discussion on CVD demonstrates, the use of financial incentives is more likely to lead to GPs following guideline advice. However, the focus on single disease risk factors has been widely criticised and can lead to a number of problems or skew practice to the neglect of non-incentivised areas of practice. 437,438,447,448

Introduction

The coalition government (2010–15) policies since 2010 highlighted the potential for greater GP involvement in public health. The changes to the structure of the NHS and to public health oversight introduced in April 2013 were designed to strengthen local public health, although the extent to which this will support and increase GP involvement is not clear. The combined implications of the government’s proposals for public health in England set out in Healthy Lives, Healthy People37 (and subsequent consultation papers) and the changes contained in the Health and Social Care Act 201221 on the structure and delivery of public health in England have been enormous. In this chapter we discuss those implications which have a direct impact on the public health role of general practice. In the White Paper and other public statements, the government outlined an increased role for GPs and general practices in public health. GPs and their practice teams have a crucial role in promoting health and preventing disease. Each year there are over 300 million consultations with primary care professionals – the majority with GPs but increasingly with practice nurses and other practice-based healthcare staff – with every consultation an opportunity to detect early warning signs that prevent illness and disease. 18 This was recognised by the coalition government of 2010–15 in its White Paper on public health arguing that GPs ‘have huge opportunities to provide advice, brief interventions and referral to targeted services through the millions of contacts they have with patients each year’ (p. 61). 35 In addition, since January 2012 there has been a call to NHS professionals to ‘Make every contact count’449 in a campaign to improve public health via every contact between a NHS professional and the public. As well as all aspects of primary and secondary prevention, there has also been an effort to incorporate the guidance into wider aspects of public health such as homelessness. 450 Furthermore, in 2011 the DH published the Health Visiting Implementation Plan,98 which proposed an increase in the health visitor workforce by 4500 new health visitors by 2015, with the concomitant emphasis on public health outcomes for children and families at all levels.

However, how these roles are developed, supported and sustained within the new public health and commissioning structures of the NHS in England is still not clear. The development of CCGs, new public health commissioning and delivery roles for LAs, and an expanded national public health role through PHE as well as a public health commissioning role for NHS England create a new and potentially more complex commissioning and service delivery environment for many public health activities in primary care. 36,98,451 There are particular concerns about the fragmentation of public health functions and service delivery agencies, the distribution of resources between responsible agencies and how proposals for commissioning and organising public health will provide a cohesive and co-ordinated public health system with a clear general practice role. The aim of this chapter is to set the context of this review within the recent changes to commissioning and public health introduced in April 2013. The chapter provides a critical analysis of the organisational changes and policy implications of the Health and Social Care Act 201221 as they relate to the organisation and delivery of public health within general practice.

The changes

Until April 2013 public health was delivered through a number of different mechanisms:

• DH – setting policy and funding a number of health intelligence programmes including the observatories and cancer registries

• government offices

• National Treatment Agency for Substance Misuse – drug treatment monitoring

• strategic health authorities – strategic oversight and performance of each region

• PCTs – commissioning of programmes to deliver health service outcomes/joint commissioning with LAs

• provider trusts – delivery of public health programmes, for example community services

• LAs – a number of different services provided by local government that have an impact on the public’s health such as environmental health, leisure, planning, housing

• Health Protection Agency – health protection services.

Following the Health and Social Care Act 2012,21 most of these organisations either were abolished or have taken on significantly different responsibilities. In addition, new commissioning and public health organisations have been established. The key strategic changes are as follows:

• Strategic health authorities were abolished at the end of March 2013.

• PCTs were abolished at the end of March 2013.

• PHE was established in April 2013, encompassing the National Treatment Agency and Health Protection Agency, and is responsible for a number of health intelligence functions including the cancer registries and regional observatories, and former government office functions.

• Ring-fenced budgets for public health have been allocated to LAs: unitary authorities and upper-tier authorities (i.e. county councils, not district councils). Directors of public health are employed within these authorities and are joint appointments between PHE and the appointing LA.

• CCGs are now responsible for the commissioning of secondary and community-based health care.

• NHS England (a national commissioning organisation) is accountable for CCGs and oversees the commissioning of specialised services and primary care services (GMS contract).

• A range of provider organisations (‘any qualified provider’) including foundation trusts, charities, independent sector and social enterprise.

In Healthy Lives, Healthy People,37 the government proposed key changes to the provision of public health. The two key changes involved the establishment of PHE in April 2013 and the transfer of local NHS public health responsibilities to LAs. PHE is responsible for funding and ensuring the provision of a wide range of services such as health protection, emergency preparedness, recovery from drug dependency, sexual health, immunisation programmes, alcohol prevention, obesity, smoking cessation, nutrition, health checks, screening, child health promotion (including those led by health visiting, school nursing and general practice), some elements of the GP contract (including parts of the QOF such as those relating to immunisation), contraception and dental public health. It does not have direct commissioning responsibility but will grant funding to LAs and it will need to work through NHS England to commission services, such as screening services, and the relevant elements of the GP contract.

The 2010–15 coalition government announced its intention in The Coalition: Our Programme for Government35 that the DH would strengthen the role and incentives for GPs and GP practices on preventative services, both as primary care professionals and as commissioners. There was a recognition that primary care professionals, GPs and GP practices play a critical role in both primary and secondary care prevention. Key areas highlighted by the government were practitioners’ opportunities to provide advice, brief interventions and referral to targeted services through the millions of contacts they have with patients each year. 37

Healthy Lives, Healthy People37 (para. 451) sets out specific ways by which the DH intends to strengthen the public health role of GPs. These involve a mix of support and incentives:

• PHE and the NHS Commissioning Board will work together to support and encourage CCGs to maximise their impact on improving population health and reducing health inequalities. This includes looking specifically at equitable access to services and outcomes.

• Information on achievement by practices will be available publicly, supporting people to choose their GP practice based on performance. By increasing transparency about how effective different GP practices are in giving public health advice, PHE will enable local communities to challenge GPs to enhance their performance.

• Incentives and drivers for GP-led activity will be designed with public health concerns in mind, for example in terms of prevention-related measures in the QOF. To increase the incentives for GP practices to improve the health of their patients, the DH has proposed that a sum at least equivalent to 15% of the current value of the QOF should be devoted to evidence-based public health and primary prevention indicators from 2013. The funding for this element of the QOF will be within the PHE budget. However, in 2013/14 this has not translated into many additional or different indicators.

• PHE will strengthen the focus on public health issues in the education and training of GPs, nurses and health visitors as part of the DH’s development of a workforce strategy.

These changes have had a significant impact on general practice and its public health role. In particular, five changes potentially increased both GPs’ needs for public health knowledge and practices’ need to increase their provision of health promotion and disease prevention services. The changes also provide significant challenges.

• The now abolished PCTs were responsible for the GMS contract and supporting local practices. The PCT also provided public health support to local practices and delivered some health promotion services such as specialist smoke-stop and sexual health services for young people. In many areas, PCTs developed additional prevention services in collaboration with practices funded through LES within the GMS contract. It is not clear how the new system will continue such support.

• With the abolition of PCTs, newly created LA Health and Wellbeing Boards (HWBs) now have to make decisions about public health priorities for their geographical area. GP representatives from CCGs sit on these boards, and they will probably be expected to contribute to the understanding of what the local health demands are, as well as the ways that general practice and other primary care services can improve health outcomes in these areas.

• General practitioner-led CCGs have taken over the commissioning role of PCTs. This could include some public health commissioning, although the details of which will depend on what NHS England chooses to ‘pass down’ to them. Commissioning public health services will require knowledge of how and where these services are best delivered.

There is a new Commissioning Outcomes Framework which will be used by NHS England to assess CCG performance. The aim of the framework is to allow NHS England to identify the contribution of CCGs to achieving the priorities for health improvement in the NHS Outcomes.

• The QOF will be adjusted so that 15% is made up of evidence-based public health and primary prevention indicators by 2013. 36 While the substance of these changes remains unclear and no change has yet occurred, GPs should be preparing for a shift towards payments for prevention.

• Although funding for public health will be ‘ring-fenced’, overall budget cuts may increase the focus on prevention rather than expensive treatments, a shift which is likely to extend to general practice. Whereas in the past such pressures within PCTs have tended to squeeze out public health expenditure, the protected LA budget will not be at risk of being shifted towards clinical services.

General practice and the new commissioning arrangements

The emphasis on CCGs being responsible for the well-being of their whole community and the stronger role of the HWBs was recommended in the NHS Future Forum report449 and ensured that public health became a shared responsibility across local commissioning organisations. However, the splitting of public health resources across general practice, CCGs, LAs and PHE/NHS England presents new challenges for the organisation and delivery of public health in England. For example, drug, alcohol and mental health services will all be commissioned via LAs. From 2015, LAs will also commission the health visiting services, which may have the effect of separating health visiting from the general practice population. These important public health challenges often overlap and are best dealt with by joint services, which may include GPs and community health professionals in shared-care arrangements if locally appropriate. These services are a good example of how, by commissioning similar public health services by one body (in this case, LAs), there can be streamlining of service delivery and inclusion of GPs where beneficial, but this grouping together will be effective only if there is enough funding to support these joint endeavours.

Commissioning has been split between local and national commissioners. While NHS England has local offices, co-ordination between commissioners is important. It is also be important to ensure that CCGs and GPs more generally are linked to broader public health commissioning arrangements. For example, public health services for children aged under 5 years, which include pre-established programmes and services that fall outside the NHS, are contracted centrally by NHS England, but GPs play a huge role in the care of under-5s and work collaboratively with local health visiting teams and other community staff, who also work closely with LA and third sector organisations – all of which are commissioned by different bodies. It is not clear how the necessary linkages are being made between GPs and community services (and local council services) when it comes to the local organisation and delivery of care and prevention programmes to under-5s.

Some degree of co-ordination will possibly be achieved through CCG representation on HWBs. The boards are seen as key to co-ordinating the local health system:

The joint local leadership of CCGs and local authorities through the health and wellbeing board will be at the heart of this new health and social care system . . . [and] enable greater local democratic legitimacy of commissioning decisions, and provide an opportunity for challenge, discussion, and the involvement of local representatives.

p. 115451

Clinical Commissioning Groups have an avenue, therefore, to be involved in local decisions about public health resource allocation – an important way of maintaining GP involvement in public health. Concerns about CCG access to public health support were raised in discussions with GPs and CCG staff, and also identified in research on the early development of CCGs. 334

However, the most dominant role of the CCG is in commissioning secondary care services, which provides a new set of challenges as well as opportunities for improvement in public health and ill-health prevention. Research on practice-based commissioning found that GPs focused more on preventing ‘unnecessary’ hospital admissions than on primary prevention. 452 Analyses of previous primary care commissioning similarly found GPs used traditional models of general practice and did not address key public health problems. 57,76,258,453 However, this is not a reason to take public health out of GP commissioning responsibilities. In fact, researchers have suggested that GP budgets for commissioning health services be aligned with budgets for commissioning public health. 454 On a more positive note, it is possible that, by giving CCGs responsibility for standard health service commissioning as well as some public health services, they may be forced to think more broadly about their communities. CCGs should have responsibility for, or at least be directly involved in, commissioning those public health activities that most closely relate to the ones they provide themselves via the GP contract (such as contraception services and cervical cancer screening). Aligning these responsibilities puts GPs’ existing knowledge of service provision to work, and helps to ensure that patients have access to the most streamlined pathways for these services (as well as the best-quality ones).

One key area of public health activity is the support for commissioners. Systems developed within PCTs for supporting GPs with epidemiological analyses are likely to have been substantially disrupted with changes to public health departments. New relationships, systems and processes need to be developed to provide what is critical support for the new commissioning bodies. Additionally, this is being done at a time when new organisational structures for public health are being established in LAs, placing further strains on developing relationships.

There is continuing tension around the relationship between general practice commissioning and public health. The evidence from previous approaches to primary care-led commissioning suggests that public health has not been a priority for GP commissioners. 453 The development of primary care groups and, more specifically, PCTs did start to embed a more public health perspective in commissioning, but current changes to both public health and health-care commissioning may simply exacerbate what has often been a troublesome relationship between GP commissioners and public health. 77,453

The General Medical Services contract and public health

Currently many public health activities in primary care are supported by the GMS contract. There are three funding strands:

• core standard tasks – expected in normal practice (advice, information, etc.)

• aspects of the QOF

• Local Enhanced Service elements of the contract.

Where clinical services are closely linked to public health activities such as screening, immunisation, obesity and sexual health services, part of the service is being provided by GPs as part of the GP contract (cervical cancer screening, childhood and elderly vaccinations, and contraception services), but other, similar services are being commissioned by LAs or NHS England (i.e. additional cancer screening, booster vaccinations, sexually transmitted infection screening and treatment, and general screening such as the NHS Health Check). This could lead to a situation where similar services that can be most effectively provided side by side (most obviously, contraception and sexually transmitted infection services) may be available from different providers because they are commissioned by different groups. Greater co-ordination of commissioning and service delivery will be essential. Thought needs to be given to arrangements for lead commissioning with the flexibility such as that available between the NHS and LAs under the 1999 Health Act455 for partnership funding for the commissioning of all aspects of key prevention services by one commissioner.

Local Enhanced Services have been particularly effective in involving GPs in locally driven public health efforts supporting a wide range of evidence-based public health activities, such as identifying CVD risk and providing long-acting contraceptives, and in 2009/10 they accounted for some £370M. 34 Having the option of LESs in the contract has provided a way for GPs to reduce preventable morbidity, and it could continue to do so in the future. This option would be especially helpful in the context of a more diverse provider landscape. However, many of the activities currently funded through LESs, including sexual health, smoking cessation, prevention and treatment of alcohol misuse, falls prevention and mental health promotion, are now commissioned by LAs, with the GP contract the responsibility of NHS England. LESs are commissioned by the local area teams of NHS England, yet the teams do not have any real levers or necessarily the knowledge needed to do this. This appears to be a key weakness in the new system.

One key element of the contract where it is proposed to incentivise more primary prevention in general practice is through the QOF. In the White Paper the government proposed that NICE adjust QOF to ensure that 15% is devoted to ‘evidence-based public health and primary prevention indicators’ (p. 62). 35 Currently, QOF continues to have only two indicators that it designates as ‘primary prevention’; otherwise it focuses mainly on secondary prevention and uses proxy or process outcomes. QOF has had a major impact on how practices undertake public health and other activities leading to more systemisation of public health activity – particularly through use of protocols and special clinics, but mainly with a secondary and medical focus. 65,456 While having some impact on primary and secondary prevention by stimulating GPs to run ill-health prevention clinics for screening and monitoring blood pressure etc., QOF has mainly supported secondary prevention activities. 32,65,93 To date, the evidence that the QOF has improved health outcomes or promoted a public health approach is very limited. 18,65,330 However, financial incentives that are effective in changing practice and more outcomes-based contracts, rather than activity-related incentives, could encourage a more proactive approach. 32 The impact of proposed changes by NICE will, to a large extent, dictate how much GPs are involved in improving the health of their patient population, which will, in turn, be affected by the future organisation of public health commissioning and service delivery.

There are, however, concerns about the extent to which NICE’s rigidly evidence-based approach is relevant for the development of many public health interventions and also whether or not such approaches are relevant to primary care itself. 457–460 In particular, there are concerns about what the contract encourages in the way of public health activity in general practice. There is an important trade-off here: RCT-based ‘certainties’ around expensive technical interventions versus ‘riskier’ but potentially more impactful interventions such as, for example, brief interventions for smoking, which have low impact but are very low cost and, while having a small impact, affect a large number of people. 460 Similarly, many areas of activity such as community-oriented activities or even welfare advice are not currently rewarded.

The implications of the fragmentation of public health services

Not only is the commissioning of similar services undertaken by different bodies, but there is also a more general likelihood of fragmentation of delivery of public health services due to the ‘any qualified provider’ model. The idea behind this approach to commissioning is to increase quality of care via a wider field of competition, on the presumption that increased competition will drive out poor providers and reduce the total number of providers. However, there is a risk that the overall effect of this policy will be that public health-related care is divided over a larger number of providers. The effect on GPs may be twofold. The first is simply that services that were once provided by GPs will be provided elsewhere, such as NHS Health Checks. The second effect, which is likely to be the more common one, is that patients will have to seek out many geographically separated providers for services such as sexually transmitted disease treatment, cancer screening and nutrition advice. This may be confusing and frustrating to patients as well as GPs, requiring good communication between different agencies to minimise this confusion. GPs must be kept aware of which providers are providing which services so they can properly advise their patients on where to seek care. Additionally, GPs must be able to keep track of what care their patients have received from other providers. Co-ordinating with other providers and keeping detailed records of their patients’ care, particularly when it comes to routine screening and immunisation, are key roles of the GP. It will be crucial to build in a way for all commissioners and providers of public health services to share their knowledge and records with GPs so that GPs can continue to do their jobs effectively. For example, there has been a long tradition of health visitors being ‘GP attached’, and parents and children within GP practices will lose this direct relationship with their GP services when health visitors are commissioned by the LA to provide a community-based service.

Ultimately, this will lead to postcode variation in provision or a variant of the ‘inverse prevention’ law if commissioning is more effectively supported and undertaken in healthier, wealthier areas (as, historically, has tended to be the case).

The opening up of public health delivery and practice to a more diverse range of providers calls into question the future role of entire categories of providers already working in public health, including health visitors, midwives and school nurses. Investment in health visiting is discussed in the White Paper37 and is welcome, but other key members of the local primary health care team who deliver public health are not mentioned. For example, the crucial role of school nurses is not mentioned in the documents and it is not clear where this service will need to focus its attention or how it will be integrated into the rest of public health delivery.

One mechanism for prioritising the general practice role in public health is via the Public Health Outcomes Framework. 37 As it currently stands, the Public Health Outcomes Framework indicates the few places where responsibility for achieving indicators is shared with the NHS. Just eight indicators in the entire list refer to the NHS, and most of them are around reducing premature death in people with chronic diseases rather than disease prevention or health promotion. It is not clear what this shared responsibility refers to: provision of funding or delivery/planning of services. Looking at the entire list of public health indicators it is clear that GPs, and therefore the NHS, can be instrumental in delivering many services beyond the ones singled out – in services relating to sexual health or smoking cessation, to name just two areas. When LAs and local HWBs are considering how to best respond to these indicators (which they are driven to do by the ‘health premium’ payments), they should keep GP services in mind for many indicators beyond those with designated NHS involvement. Additionally, the boards may find that GPs are particularly well primed to the idea of indicators, having now been working with the QOF for over 6 years. While these two sets of indicators are very different in character, GPs may be comfortable with the idea of indicators and may have ideas for how to incorporate this new set into their practices.

While HWBs have been given a key co-ordinating role, it is questionable how far they will be able to fulfil all the expectations placed upon them. At the current time there remains considerable ‘fuzziness’ around the exact role of HWBs. The lack of statutory powers given to the boards means that it is only by developing good local relationships that they will be able to fulfil their potential. In recent research on the development of CCGs, the development of a strong ‘co-ownership’ model, where CCGs, LAs and HWBs saw themselves as joint owners of the developing strategy, with all partners being actively involved, was noted in some areas. However, in others, CCGs were developing quite separately from their HWBs, with CCG representatives attending meetings but contributing little. 63,334 Marks et al. 63 have highlighted the difficulty of aligning priorities across large geographical areas and where organisations were not coterminous. This may prove a problem again, as many of the 211 CCGs cover smaller populations than their HWB, and the ability to respond to local concerns and problems will be important as HWBs develop. Ultimately, despite HWBs having ‘strategic influence over commissioning decisions across health, public health and social care’ as one of their main roles,461 there is a danger that, as new local public health systems develop, they may become more fragmented in terms of the relationships between LAs, CCGs and HWBs as well as having to cope with a new national context with PHE and NHS England. This may create additional complexities for local co-ordination by the HWBs and raises important questions about how public health activities in general practice will be incentivised, managed and supported in the future. This uncertainty was of concern to many of the GPs we talked to during the research.

Current developments

In line with the government’s coalition agreement (of 2010–15) policy focused on the health improvement role of general practice. 35 In addition to proposals to revise the QOF to include more primary prevention criteria, the recent Mandate for the NHS Commissioning Board highlights prevention within the concept of making ‘every contact count’, ‘in focusing the NHS on preventing illness, with staff using every contact they have with people as an opportunity to help people stay in good health – by not smoking, eating healthily, drinking less alcohol, and exercising more’ (p. 8). 462 This approach is now reflected in NICE Public Health Guidance with the publication of the draft revision of physical exercise guidelines (currently out for consultation). The proposed revised guidelines specifically target primary care practitioners proposing that they:

Assess the physical activity levels of all adults in contact with primary care services and identify those who are not currently meeting the UK physical activity guidelines. This could be done:

1. opportunistically during a consultation with a GP or practice nurse (or while people are waiting)

2. as part of a planned session on management of long-term conditions run by a practice nurse

3. as part of a consultation with a pharmacist.

p. 8463

With this guidance, primary care practitioners are being encouraged to identify inactive patients, assess their level of activity and deliver a brief intervention to encourage them to increase levels of activity. Concerns have been raised about whether or not it is appropriate for more and more elements to be placed on GPs and nurses given the limited consultation time in practice. These questions have also been raised in connection with a pilot study that examined adding cancer-screening questions into the NHS Health Check. The pilot study involved a screening survey of 4250 patients with high cardiovascular risk undergoing NHS Health Checks. The study, reported at the October 2012 Society of Academic Primary Care conference in Glasgow, found that the process involved substantial nurse and GP time but led to only four cases being identified. 464

Conclusion

The Royal College of General Practitioners argues that GPs should be proactive in carrying out public health activities and interventions, and it is expected that GPs should possess a wide range of skills related to ill-health prevention and public health. 7 Since 1990, GPs in particular, have been encouraged to carry out more public health activities through changes to their contract and the potential of the primary care public health role was highlighted in the Wanless Report on public health. 3,48 However, research continues to find that the relationship between public health and general practice in England focuses primarily on secondary prevention, and many GPs state they lack the skills needed to deliver effective health promotion. 432 The King’s Fund Inquiry on quality in general practice concluded that there is enormous potential for general practice to take a more proactive role in ill-health prevention and public health. 18,465 In fact, a key criticism of much of general practice (when it comes to public health) is that it focuses on either secondary prevention or just information and advice. While both of these activities are useful, other interventions can be more effective. All of these considerations and criticisms are of particular importance when it comes to the QOF, which is a key driver of GP practice.

The Public Health White Paper37 outlines a number of key changes to the organisation and delivery of public health in England. However, while the public health role of general practice has been the subject of much debate over the past few decades, the government does not appear to have grasped how this role can be integrated, developed and supported within the proposed changes. Relying on changes to the QOF is both short-sighted and overly narrow. The fragmentation of the commissioning and delivery of public health is likely to lead to problems of co-ordination. It is not clear how HWBs – responsible for local public health co-ordination – will build integrated approaches with both CCGs and PHE as well as the wide range of local delivery organisations, including general practices. Previous approaches to primary care-led commissioning have not demonstrated that GPs can work closely with LAs or other agencies. 453 A further problem may be the attitude of many GPs themselves, as studies suggest that they are more comfortable managing illness than promoting health. How effective the new proposals will be at engaging GPs and other primary care staff in public health activities is still open to question, as for many GPs there is still a significant amount of distrust. 334

In this chapter we have identified a number of potential problems that may result from current policy developments and organisational changes. While an emphasis on health improvement is welcome, there are a number of potential threats to the ability of primary care, and general practice more specifically, to deliver public health. First, commissioning approaches need to be aligned across the different agencies that will be responsible for aspects of public health. In particular, commissioning agencies need to urgently identify how services from existing groups of primary care-based public health providers (midwives, school nurses, etc.) are commissioned and supported within the new structures. Thought also needs to be given to how health improvement activities that have formerly been funded through GMS, PMS and other local contractual arrangements can be continued, and how local variations in funding and service delivery can be incorporated into the new systems and structures. Developments in health improvement in general practice have benefited from local negotiation and relationships with the public health department in the PCT. Concerns have been raised about the ongoing links between public health and general practice now that LAs have taken on public health responsibilities. While policy is increasingly emphasising public health roles being embedded in primary care practice, many of the current organisational changes are creating potential problems for the commissioning and delivery of many important health improvement activities.

Introduction

The aim of this study was to identify the current extent of knowledge about health improvement activities in general practice and the wider primary health-care team. The primary focus of the review was on activities that were specifically related to activities undertaken in the practice and by primary health-care staff linked to the practice. With the publication of the White Paper on public health,37 this focus was extended to include the role of pharmacies and prevention activities undertaken by primary health-care teams in settings outside the practice such as schools and in the community. Thus activities by school nurses and health visitors were included, but wider public health and health education activities in schools – by the schools themselves – were excluded.

The original focus of the research was to undertake a scoping review of health improvement activities undertaken in general practice or by the primary health-care team to determine the nature and extent of research evidence about these. Through discussions with practitioners we sought to identify how such activities were organised and delivered. An overall aim was to define areas of practice that needed further research. We were also requested by the NIHR SDO programme to focus the research on producing an evidence synthesis rather than merely scoping the research. The key objectives of the review were, therefore, revised in order to provide an overview of the range and type of health improvement activities being undertaken, assess the strength of evidence for such activities where possible and identify gaps in knowledge and areas for further empirical research. As a result, the stakeholder consultations were more limited than originally envisaged and were primarily used to inform the literature review and interpretation of findings.

Key questions for the review team were:

• What was the range and type of health improvement activity undertaken in general practice?

• What evidence was there on effective approaches to the delivery and organisation of health improvement interventions in general practice settings?

• What gaps were there in the evidence base?

• What are the priority areas for future research?

Essentially our focus was twofold, involving:

• an assessment of the efficacy of approaches to the delivery and organisation of health improvement interventions to identify areas of public health that can be delivered in a general practice setting (including identifying the roles of general practice and the staff working in general practice and methods of delivery and organisation)

• identifying research gaps and priorities for further research.

This chapter draws on the findings of the review and discusses them in the context of current developments in the NHS. The chapter summarises our key findings and sets this within a framework that structures the relationship between public health and primary care, and how health improvement activities are delivered and organised.

General practitioners and their practice teams have a crucial role in promoting health and preventing disease. Every consultation is an opportunity to detect early warning signs that prevent illness and disease. GPs and the wider primary health-care professionals are, therefore, ideally placed to ‘make every contact count’. 449 While this rhetoric pervades the NHS and public health discourse, the evidence that primary care consultations can effectively deliver this is less clear from our review. The Royal College of General Practitioners agrees that GPs should be proactive in carrying out public health activities and interventions, and it expects GPs to possess a wide range of skills related to ill-health prevention and public health, as reflected in the Member of the Royal College of General Practitioners curriculum. 466 However, research continues to find that the relationship between public health and general practice in England focuses primarily on secondary prevention,18 and many GPs state they lack the skills needed to deliver effective health promotion. For example, currently GPs respond to requests for support in smoking cessation more often than they proactively engage their patients. Therefore, there is enormous potential for general practice to take a more proactive role in ill-health prevention and public health. NICE public health guidance advises primary care professionals, such as GPs, to opportunistically carry out activities such as brief interventions. 244,388 The White Paper on public health acknowledges this role, stating that GPs have huge opportunities to provide advice, brief interventions and referrals to targeted services. 35

The evidence

The evidence synthesis and review of activities undertaken in primary care focused on how health improvement interventions and activities have been, and are, organised and delivered. We undertook a broad search for relevant literature with the aim of identifying both the extent and the quality of evidence supporting approaches to organising and delivering health improvement interventions. Given the breadth and huge number of potential papers identified in our original searches, we refined our search criteria to focus specifically on papers describing or evaluating UK practice.

While we identified 1140 papers from our original literature search, when they were examined for relevance and then for quality we found that only 347 (30%) met our inclusion criteria for the evidence review. We assessed the quality of each paper in terms of whether or not the research methods supported the outcomes in the paper. We used a grading system based on critiquing five areas:

• quality of the sampling process (including the sample size)

• appropriateness of the analysis

• comparability of the intervention and comparison groups

• quality of the intervention delivery (including whether consistency of the intervention was measured or whether there was risk of contamination)

• outcome assessment.

For each paper we identified whether there were no concerns about the methods or data presented, minor concerns or major concerns. Studies were described as of high quality (A) if there were no concerns across the applicable areas, good quality (B) if there were one or two minor concerns and lower quality (C or D) if there were major concerns or multiple minor concerns.

As detailed in Chapters 4–6, the range and diversity of papers was substantial. In order to examine selected papers we grouped them into broad topic areas. However, even within these there was significant variation between papers. For example, papers classified as child health covered childhood screening, breastfeeding, accidents, diet, sexual health, weight and exercise, etc. Assessing papers presented a number of key difficulties for reviewers. We foresaw that identifying data on service delivery and organisation would be difficult and we expected to exclude many papers because they did not address these issues at all. However, even amongst selected papers in the review it was difficult to summarise the evidence or draw any substantial conclusions from the evidence.

In addition, there is a widely recognised problem about the relevance of much research to general practice. The increasing importance being placed on the role of primary care within health systems raises questions about how an appropriate evidence base for practice can be supported, developed and applied. A simple view of knowledge production and use would see this as a linear process, whereby scientists find out new things about the world that are published in academic journals. Then people who want that knowledge search for it and put it into use. In this classic knowledge-driven model, research generates knowledge, which impels action; thus, where knowledge is generated and by whom is less relevant than ensuring the production of high-quality research findings. 467 This is an oversimplified view of a complex process. 468,469 An interactive, social model of knowledge is more useful. Gibbons et al. 470 make the case for two ‘modes’ of knowledge production. As Sweeney et al. 457 have argued, not only does research needs to be statistically and clinically relevant but there should also be recognition of personal significance and the importance of applying research evidence to the individual within the social context.

Producing evidence on health improvement interventions in general practice raises important questions about the type of evidence required and how it is produced. As Rychetnik et al. 113 argue, evidence needs to be clear not only about whether the intervention is successful or has failed, but also about how it is delivered – especially if failure of an intervention is a result of bad delivery. Furthermore, De Maeseneer et al. 458 have argued that ‘[c]ontextual evidence is necessary to assist doctors to address the challenge of how to treat a particular patient in a given situation’ (p. 1316). Where service delivery needs to be supported by research that is both appropriate and context-relevant – as is the case for much health improvement activity – it is best conducted within primary care. The study by Hoddinott et al. 221 has cast important light on how analysis of the context of service delivery can be explanatory regarding success and failure of interventions. Hospital-based clinical research is not always relevant to the primary care context, and the findings of clinical trials based in secondary care cannot easily be translated into primary care practice. In addition, primary care practice involves an interest in the appropriateness of care and low-technology interventions such as many health improvement activities that require being responsive to individual patient circumstances. In an analysis of NICE guidelines for GPs the researchers concluded that as few as 2% were directly relevant for general practice. 471 The need for a different approach to primary care research is set out in the recent Society for Academic Primary Care Paper Blue Sky Research for Primary Care: A Discussion Paper. 472 Watt concludes that:

The contribution of generalist clinical practice to health improvement is achieved via the sum of care provided for all patients. . . . Such activity is an important determinant of population health, although it is often excluded from public health discourse and from consideration of public health research needs.

p. 11472

From the analyses in our review we conclude that more relevant research needs to be developed that adequately reflects both the context of primary care and other settings where health improvement interventions are undertaken as well as ensuring that service delivery and organisational issues are researched as well. In addition, research needs to compare different methods of delivery and organisation; for example, comparing the role of different professionals and different locations as well as different delivery methods in order to determine the effectiveness of not just the specific intervention but also how it is delivered.

What is done, by whom and where

The selected papers encompassed a wide variety of research designs across a broad range of topic areas. Generally papers had little information on the context or organisation of the way in which the intervention was delivered. There were a substantial number of papers that examined individual interventions without any comparison or were simply studies of patient or practitioner views and perceptions. Much prevention activity in general practice relates to clinical interventions and it was surprising that more such papers were not included in the evidence synthesis. However, we found that a high proportion of papers examined clinical prevention interventions yet had no information about, or did not address, delivery and organisation aspects of interventions but were merely clinical trial data or evaluations of the clinical intervention. There were some data on the type of practitioner and location of the intervention delivery but there were few data that allowed statements to be made about the comparative effectiveness of different locations, or person, delivering the intervention.

The problem of identifying service issues is a common problem in this area of research. For example, a recent cost-effectiveness analysis for brief interventions in general practice for physical exercise argued that the main reason for a lack of evidence was that

it was not possible to disentangle the effectiveness (and hence efficiency) of brief advice interventions from either more intensive counselling, exercise delivery programmes (n = 7) or from other lifestyle advice (n = 10). The problem of disentangling was a function of study design as well as choice of outcome measures.

p. 31473

It is understandable, therefore, that we found little robust evidence to demonstrate positive behaviour change or clinical outcomes for preventative activities in primary care. The evidence on the effectiveness of different approaches by different practitioners and in different locations was very limited. For example, most papers examining primary prevention were either descriptions of services or commentary about primary prevention or research studies that were rated as of low quality. The main exception was in relation to smoking cessation services, where services to support quitting are viewed as cost-effective and approaches that provide ongoing support for smokers wanting to stop, or for those who have quit, when delivered in a targeted way, are more likely to be effective. In addition, there is still potential for better identification of smokers and raising the subject of smoking with patients by a range of health and other care professionals. We found little similar evidence for other areas of prevention activity. The evidence to support interventions for increasing exercise or reducing weight was limited and suggests that such approaches in primary care or in schools have little impact and that any benefit is generally only short-term. In particular, the cost-effectiveness of primary preventative lifestyle interventions is difficult to determine because of the diverse nature of the interventions (type and organisation, different target groups) and research methodology.

The papers examined in this review also appear to show that GPs often feel ill equipped to give prevention and health promotion advice to patients. There is clear evidence that practice nurses are increasingly being used to provide prevention and health promotion and also to monitor symptomatic and asymptomatic (but deemed at risk) patients. While patients seem to like this, there is little evidence of positive outcomes or other patient benefit. There is little evidence to suggest that any lifestyle advice is recalled by patients, let alone acted upon, which may contribute to lower levels of effectiveness. 332 In addition, patients may prefer alternative sources of information for such advice, such as magazines and television. 332 Generally, the evidence is only from short studies, with methodological problems, and benefits demonstrated were only marginal. However, a recent international review of the literature on the role of primary care in improving health literacy among patients suggests that there are effective measures that may help improve people’s knowledge about health and contribute to improved health outcomes. Its authors conclude that:

group and individual interventions . . . in both primary health care and community settings may all be useful in supporting sustained change in health literacy for change in behavioural risk factors. There may be scope for some tailoring of the site and type of interventions depending [on the] risk factor.

p. 10474

The authors conclude that less-intense interventions may be as effective as more intensive ones but caution that more research is needed on the specific intervention type. They note that different interventions produced different results and that similar interventions produced different results depending on the health issue being targeted. Similarly, different locations produced different results as did different degrees of intensity of the intervention.

There is some evidence to suggest that short interventions are effective in general practice, particularly where the GP has done these (e.g. smoking, alcohol, older people), although effects are very small. However, concerns have also been raised about adding more and more advice interventions into the limited consultation time. Making sufficient time available for health promotion and prevention remains a crucial issue. Limited consultation time is seen as a key problem restricting how health issues – such as smoking and lifestyles – can be raised. In a review of the relationship between the GP consultation length, process and outcomes, Wilson and Childs475 concluded that a consistent finding from studies was that GPs with ‘longer than average consultation lengths prescribed less and were more likely to include lifestyle advice and preventive activities’ (p. 1018).

The role of primary health care in public health

There is no doubt that general practice and the wider primary care team play an important role in the delivery of health improvement interventions with local populations. Reviews that have examined the relationship between primary care and public health support a clear, common core of shared principles. 1,23,24 Our review has demonstrated that general practice and those that work in, or with, general practice undertake many key public health functions. While the relationship between primary health care and public health has long been recognised, systems of primary care in different countries do not necessarily fulfil all the elements of primary health care as reflected in the Alma Ata declaration. 46 Concerns about the nature of UK general practice have been identified for some years. 46,48

General practice fulfils many of the key elements of primary health care. However, this review has identified that much prevention activity is undertaken in general practice itself by primary care practitioners. Other activities that are clearly associated with the principles of primary care are delivered by other professionals or workers either in general practice or in other locations, including homes, the community, pharmacies and schools. This suggests that, in developing public health practice and policy in this area, consideration needs to be given to a wider concept of primary care. Policies in the Public Health White Paper37 and also in many guidance documents tend to focus on the role of the general practice. We would argue that this is too narrow a focus.

This review supports the role of primary care professionals as having an important public health function. GPs have always viewed prevention and health promotion as a core part of their role. In a survey of GPs in one English region in 1990, a large majority of respondents claimed to discuss health-related topics with their patients when indicated. However, fewer made specific efforts to discuss smoking habits (64%), alcohol intake (26%), diet (12%) or exercise (11%) as a matter of routine with all their adult patients. 479 Most GPs said they usually offered simple advice, leaflets or other aids when they had identified a problem, but few said they would refer these patients to the practice nurse. However, at that time few GPs undertook systematic screening for risk factors and lifestyle advice for all patients. The introduction of the 1990 contract with specific health promotion banding payments and the introduction of guidelines on screening, the QOF and now Public Health Guidance from NICE has pushed practices towards more systematic prevention and health promotion interventions.

With 30 million patient contacts a year and registered practice populations, it is clear that general practice provides an important location for health improvement initiatives. More importantly, as services such as welfare advice demonstrate, practices provide a location for other services that can also have an impact on health. Conversely, there is good evidence to show that some interventions are best delivered outside the practice. Yet general practice does, along with many other locations, provide opportunities to promote health. There has been criticism of the increasing individual and specific disease focus that is driven by measures such as the QOF and clinical guidelines. In particular, there is concern that practices serving more deprived communities are disadvantaged by such systems but also that many of the primary focuses of policy and practice guidelines – such as smoking, weight loss and cholesterol measurement – do not address wider underlying problems faced by patients, whether health problems, such as alcoholism, or wider socioeconomic issues. 480 Reporting on the experiences of the Scottish Deep End project, Norbury et al. argue that:

practices in severely deprived areas can improve patient’s health and narrow health inequalities [only] by increasing the volume and quality of the care . . . but . . . lack time in consultations to address the multiple morbidity, social complexity and reduced expectations that are typical of patients living in severe socio-economic deprivation.

p. 3481

The way public health is conceptualised in general practice currently tends to focus on specific interventions by specific professionals. Changes in the way that general practice and its relationship with patients is conceptualised has important implications for much health improvement activity. Marteau and Kinmonth482 have questioned the top-down public health approach to issues such as screening and lifestyle advice, arguing for a more nuanced approach to working with patients, and Getz et al. 483 have questioned the ethical basis of opportunistic prevention in practice. Drawing on these kinds of criticisms of health improvement has led to the development of different ways of conceptualising the role of general practice in public health. Watson15 has argued that the practice should not be seen as just a location for public health, and develops proposals for a health-promoting practice. We suggest that this concept be taken one step further, with the practice being seen as a setting for public health. To view general practice as a setting requires a focus on key central concepts that involve the desire to act on policies, reshape environments, build partnerships, and develop empowerment and ownership as well as developing personal competencies. 106,107 In public health the concept of settings has been used successfully in relation to schools and work places, for example. In developing the practice as a setting, the focus is not on what practitioners can provide as interventions but examines what kinds of interactions can be undertaken with those who work in, work with and use the practice. Attention would then be paid to what should be delivered in the practice (and by whom) and if there are circumstances where the practice may simply direct people to other services.

In the international nursing field the concept of ‘positive practice environments’484,485 has been developed to enable nursing, frequently undertaken in under-resourced communities internationally, to be delivered in settings that support good human resource practice and empower the staff to develop positive attitudes and change their own behaviour. The concept could be considered and applied to primary care settings in England. Key to developing such an approach will be to strengthen relationships within the primary health-care team – especially with health visitors and other public health-oriented staff. Whether or not the development of public health within LAs can support such developments is not yet clear and this is an area that HWBs are unlikely to prioritise. Similarly, the new CCGs are not focused on health improvement issues in the short term given the scale of other commissioning priorities that need to be addressed. 334

It is important that GPs and the wider primary health-care team stay involved in public health, and to encourage an even greater level of participation. However, as discussed in Chapter 7, we are concerned by several aspects of the government’s proposals and the impact this could have on the public health role of general practice and the wider primary health-care team’s involvement in health improvement activities. These include the breaking down of relationships between public health and general practice, which are already fairly tenuous in many places, and the need to support improved education and training for staff – especially for newly developing roles such as health-care practitioners, who undertake a substantial amount of routine screening and monitoring work in practices.

The new NHS organisation and structure in England presents a number of challenges to the provision of public health by general practice and the broader primary health-care team. Up until April 2013, PCTs held contracts with primary and community care services and provided public health services and advice. From April 2013, commissioning responsibilities that affect public health services are split between CCGs, the NHS Commissioning Board and LAs.

The issue of clarifying who is a primary care provider and who is responsible for providing public health services, needs to be established as the new commissioning structure for public health is established. For example, school nurses and pharmacists appear in the ‘primary care and public health’ literature and carry out vital public health services, but they did not feature in the public health White Paper37 at all. Traditional services may also have to compete with other public-health service providers, in line with the ‘any willing provider’ policy, to continue to provide public health services. For example, currently NHS Health Checks are provided mainly by general practice but also by pharmacists in community pharmacies. Since April 2013, LAs can commission other organisations to undertake health checks, raising issues of continuity of care. This issue is further complicated by the fact that much of the public health commissioning is at the local level, by LAs and GP consortia. The lack of any consensus on who should be providing services may mean local variation in the future roles of school nurses, community pharmacists and other practitioners who have been providing public health services.

The proposed changes to the QOF to increase the public health focus will affect almost all GPs. The White Paper on Public Health37 states that by 2013 15% of the QOF will be dedicated to evidence-based public health and primary prevention indicators. There are currently just two indicators in the QOF that are categorised as ‘primary prevention’, both related to patients who have been diagnosed with hypertension, and there are six recommended indicators that are categorised as ‘public health’, which are related to obesity and exercise promotion. It is not clear how the increase in primary care and public health indicators will happen and the extent to which primary care practitioners will be involved in developing the indicators. If new indicators are developed, where along the spectrum of a chronic disease will they target their effect?

Another consideration for the QOF public health indicator development is how the indicators will take into account the way public health research priorities have changed over time. Obviously the evidence base should be as current as possible, but, if the indicators just reflect the most recent research trends, they will probably encompass a different set of diseases and types of interventions from if they took a more long-term approach to the evidence. This question applies not just to the QOF, but also to the new proposed public health outcomes framework, which will reward LAs for meeting local public health targets. 37

In Healthy Lives, Healthy People,37 the government acknowledged the need for more research on public health practice. One key development has been the establishment of a new NIHR School for Public Health Research. This new school needs to invest in research that examines the question of the role of public health research in relation to general practice. As we saw in our research, there are some health intervention studies that could apply to general practice and are not carried out in a general practice setting, and there are many general practice studies that are not focused on actual intervention delivery. We found very little research that addressed the context of delivery and organisation in general practice. However, it is often very difficult to actually set research in general practice, when GPs and practice nurses have limited time for treating patients. Infrastructure and research support needs to be given to general practice, and public health research needs to be prioritised to ensure any guidance is context relevant and likely to produce real health gains. It is also necessary to develop a system that allows GPs, practice nurses and other primary care practitioners to be involved in more research in their natural setting, which may help to improve the quality of research on public health interventions in general practice. The concept of positive practice environments may be worth considering as a way of promoting settings for public health interventions, although the evidence for this is largely based on international health-care systems. 485

Conclusion

This review provides a broad overview of the literature on UK health improvement practices in general practice and by the broader primary health-care team associated with the practice. We also included some aspects of school- and pharmacy-based interventions. The aim of the review was not to systematically review all the evidence on health improvement interventions but rather to identify the nature and extent of the evidence on the delivery and organisation of the health interventions. As expected in our original research proposal, we found that the evidence was mixed in terms of detail and methodological quality, and that identifying sufficient information about the context and organisation of the delivery of health improvement interventions was very limited. Given this, undertaking any systematic approach to even a narrative evidence synthesis was not possible.

In the review we have therefore highlighted some key areas where there are evidence strengths and specific shortcomings in the evidence. While it was foreseen at the start of this process, it is clear that further scoping work and specific research is required to fully understand the health improvement role of general practice and identify effective approaches to the delivery and organisation of health improvement in general practice/primary care settings and by primary care practitioners.

Our key conclusion is that there is currently insufficient evidence to strongly support many of the health improvement interventions undertaken in general practice and primary care more widely. This is not to conclude that there are no interventions that provide population health improvements; brief interventions for stopping smoking are a good example. There is, however, some evidence to support specific interventions being undertaken with some patient groups and in some locations. Further research is needed to strengthen these areas to provide more supportive and clear evidence.

What are the priority areas for future research?

Research on health improvement in primary care needs to move beyond clinical research to include delivery systems, and here there is an overriding imperative for such research to be conducted in a primary care context to ensure that research is relevant and hence more likely to be transferred to practice. One important driver for primary care physician involvement in research is the need to improve quality of care. 486 More needs to be done to support the development of research within and by primary care.

Currently the major areas of research tend to be driven by clinical interest and be associated with specific disease groups. There needs to be more discussion about the type and range of disease burdens in primary care – perhaps moving away from areas such as CVD, where rates are declining, to other areas. Attention needs to shift to examining areas such as other chronic disease burdens: mental health of young people, cancer, dementia and other disabilities of old age.

The type of research needs to change. We found substantial numbers of cross-sectional studies examining patient and professional views but less research that examined what was being done and how. Currently there is little substantive research that examines whole areas of health improvement. Studies tend to be grouped around specific aspects of diseases such as identifying risk factors or provision of information. Further reviews could be commissioned that examine the whole prevention pathway for health problems that are managed within primary care drawing together research from general practice, pharmacy, community engagement, etc., to provide a rich, in-depth picture and analysis of the role of primary care in prevention. In some areas this would involve primary research informed by initial reviews. For example, while for stopping smoking it is possible to provide good evidence about the relative roles of general practice and specialist interventions and also about the intensity and delivery of services – including cost-effectiveness – there are few other areas where this is possible. Thus studies would examine interventions in a specific condition (e.g. cancer), for a population group (e.g. older people) or an intervention type (e.g. welfare advice).

Contribution of authors

Professor Stephen Peckham: project lead and corresponding author; involved in all aspects of the research and responsible for the drafting of the final report.

Ms Jane Falconer: librarian; conducted literature searches and retrieval of full-text papers.

Professor Steve Gillam: coinvestigator; refined methods, reviewed papers and commented on the report.

Dr Alison Hann: coinvestigator; refined methods, reviewed papers, drafted report sections and commented on the final report.

Professor Sally Kendall: coinvestigator; refined methods, reviewed papers and commented on the report.

Dr Kiran Nanchahal: coinvestigator; refined methods, reviewed papers.

Mr Benjamin Ritchie: researcher from October 2011; worked with the librarian on searches, managed the data extraction process, reviewed papers, drafted report sections.

Dr Rebecca Rogers: intern researcher until June 2011; worked with the librarian on searches, managed the data extraction process.

Dr Andrew Wallace: researcher until July 2011; worked with the librarian on searches.

Disclaimers

This report presents independent research funded by the National Institute for Health Research (NIHR). The views and opinions expressed by authors in this publication are those of the authors and do not necessarily reflect those of the NHS, the NIHR, NETSCC, the HS&DR programme or the Department of Health. If there are verbatim quotations included in this publication the views and opinions expressed by the interviewees are those of the interviewees and do not necessarily reflect those of the authors, those of the NHS, the NIHR, NETSCC, the HS&DR programme or the Department of Health.

Ovid MEDLINE

Date range searched: 1990–2013.

Date of search: June 2013.

URL: https://ovidsp.ovid.com

Ovid MEDLINE

Date range searched: 1990–2013.

Date of search: June 2013.

URL: https://ovidsp.ovid.com

Ovid EMBASE

Date range searched: 1990–2013.

Date of search: June 2013.

URL: https://ovidsp.ovid.com

Ovid Global Health

Date range searched: 1990–2013.

Date of search: June 2013.

URL: https://ovidsp.ovid.com

EBSCOhost Cumulative Index to Nursing and Allied Health Literature Plus

Date range searched: 1990–2013.

Date of search: June 2013.

URL: https://www.ebscohost.com/

Notes on derivation of search strategies

• Alcohol and drug addiction strategy derived from the Cochrane Drugs & Alcohol Review Group strategy. 487

• Brief intervention strategy derived from NICE Guideline PH1: Brief Interventions and Referral for Smoking Cessation. 488

• Exercise therapy strategy derived from NICE Guideline PH17: Promoting Physical Activity for Children and Young People. 489

• Immunisation strategy derived from NICE Guidance PH21: Reducing Differences in the Uptake of Immunizations. 490

• Tobacco cessation strategy derived from NICE Guideline PH1: Brief Interventions and Referral for Smoking Cessation. 488

Ovid MEDLINE

Date range searched: 1990–2013.

Date of search: June 2013.

URL: https://ovidsp.ovid.com

Ovid EMBASE

Date range searched: 1990–2013.

Date of search: June 2013.

URL: https://ovidsp.ovid.com

Global Health

Date range searched: 1990–2013.

Date of search: June 2013.

URL: https://ovidsp.ovid.com

Cumulative Index to Nursing and Allied Health Literature Plus

Date range searched: 1990–2013.

Date of search: June 2013.

URL: https://www.ebscohost.com/

Buckley BS, Byrne MC, Smith SM. Service organisation for the secondary prevention of ischaemic heart disease in primary care. Cochrane Database Syst Rev 2010;3:CD006772.

Stead LF, Bergson G, Lancaster T. Physician advice for smoking cessation. Cochrane Database Syst Rev 2008;5:CD000165.

Rice VH, Stead LF. Nursing interventions for smoking cessation. Cochrane Database Syst Rev 2008;8:CD001188.

Sinclair HK, Bond CM, Stead LF. Community pharmacy personnel interventions for smoking cessation. Cochrane Database Syst Rev 2004;1:CD003698.

Secker-Walker R, Gnich W, Platt S, Lancaster T. Community interventions for reducing smoking among adults. Cochrane Database Syst Rev 2002;2:CD001745.

Carson KV, Brinn MP, Labiszewki NA, Esterman AJ, Chang AB, Smith BJ. Community interventions for preventing smoking in young people. Cochrane Database Syst Rev 2011;7:CD001291.

Lumley J, Chamberlain C, Dowswell I, Oliver S, Oakly L, Watson L. Interventions for promoting smoking cessation during pregnancy. Cochrane Database Syst Rev 2009;3:CD001055.

Lai DTC, Cahill K, Qin Y, Tang J-L. Motivational interviewing for smoking cessation. Cochrane Database Syst Rev 2010;3:CD006936.

Kaner EF, Beyer F, Dickinson Ho, Pienaar E, Campbell F, Schlesinger C, et al. Effectiveness of brief alcohol interventions in primary care populations. Cochrane Database Syst Rev 2007;2:CD004148.

Baker PRA, Francis DP, Soares J, Weightman Al, Foster C. Community wide interventions for increasing physical activity. Cochrane Database Syst Rev 2011;4:CD008366.

Bosch-Capblanch X, Abba K, Prictor M, Gamer P. Contracts between patients and healthcare practitioners for improving patients’ adherence to treatment, prevention and health promotion activities. Cochrane Database Syst Rev 2007;2:CD004808.

Smeeth LL, Iliffe S. Community screening for visual impairment in the elderly. Cochrane Database Syst Rev 2006;3:CD001054.

Flodgren G, Deane K, Dickinson HO, Kirk S, Alberti H, Beyer FR, et al. Interventions to change the behaviour of health professionals and the organisation of care to promote weight reduction in overweight and obese adults. Cochrane Database Syst Rev 2010;3:CD000984.

Jacobson Vann JC, Szilagyi P. Patient reminder and recall systems to improve immunization rates. Cochrane Database Syst Rev 2005;3:CD003941.

Norris SL, Zhang X, Avenell A, Gregg E, Schmid CH, Lau J. Long-term non-pharmacological weight loss interventions for adults with prediabetes. Cochrane Database Syst Rev 2005;2:CD005270.

Nield L, Summerbell CD, Hooper L, Whittaker V, Moore H. Dietary advice for the prevention of type 2 diabetes mellitus in adults. Cochrane Database Syst Rev 2008;3:CD005102.

Orozco LJ, Buchleitner AM, Gimenez-Perez G, Roque I Figuls M, Richter B, Mauricio D. Exercise or exercise and diet for preventing type 2 diabetes mellitus. Cochrane Database Syst Rev 2008;3:CD003054.

Renders CM, Valk GD, Griffin S, Wagner EH, Eijk JT, Assendelft WJ. Interventions to improve the management of diabetes mellitus in primary care, outpatient and community settings. Cochrane Database Syst Rev 2001;1:CD001481.

Boyle R, Solberg L, Fiore M. Use of electronic health records to support smoking cessation. Cochrane Database Syst Rev 2014;3:CD008743.

Gillespie LD, Robertson MC, Gillespie WJ, Sherrington C, Gates S, Clemson LM, et al. Interventions for preventing falls in older people living in the community. Cochrane Database Syst Rev 2012;9:CD007146.

McClure RJ, Turner C, Peel N, Spinks A, Eakin E, Hughes K. Population-based interventions for the prevention of fall-related injuries in older people. Cochrane Database Syst Rev 2005;1:CD004441.

Hooper L, Bartlett C, Davey SG, Ebrahim S. Advice to reduce dietary salt for prevention of cardiovascular disease. Cochrane Database Syst Rev 2004;1:CD003656.

Ebrahim S, Taylor F, Ward K, Beswick A, Burke M, Davey Smith G. Multiple risk factor interventions for primary prevention of coronary heart disease. Cochrane Database Syst Rev 2011;1:CD001561.

Hillsdon M, Foster C, Thorogood M. Interventions for promoting physical activity. Cochrane Database Syst Rev 2005;1:CD003180.

Waters E, de Silva-Sanigorski A, Hall BJ, Brown T, Campbell KJ, Gao Y, et al. Interventions for preventing obesity in children. Cochrane Database Syst Rev 2011;12:CD001871.

Oringanje C, Meremikwu MM, Eko H, Esu E, Meremikwu A, Ehiri JE. Interventions for preventing unintended pregnancies among adolescents. Cochrane Database Syst Rev 2009;4:CD005215.

Dyson L, McCormick F, Renfrew MJ. Interventions for promoting the initiation of breastfeeding. Cochrane Database Syst Rev 2005;2:CD001688.