• Open Access

Translating policy into practice: a case study in the secondary prevention of coronary heart disease

Authors


Lindsay Prior PhD MA BSc (Soc) BSc (Maths)
School of Sociology
Social Policy & Social Work
Queen’s University
Belfast BT7 1NN
UK
E-mail: l.prior@qub.ac.uk

Abstract

Background  This paper focuses on the relationships between health ‘policy’ as it is embodied in official documentation, and health ‘practice’ as reported and reflected on in the talk of policy-makers, health professionals and patients. The specific context for the study involves a comparison of policies relating to the secondary prevention of coronary heart disease (CHD) in the two jurisdictions of Ireland – involving as they do a predominantly state funded (National Health Service) system in the north and a mixed health-care economy in the south. The key question is to determine how the detail of health policy as contained in policy documents connects to and gets translated into practice and action.

Methods  The data sources for the study include relevant health-care policy documents (N = 5) and progress reports (N = 6) in the two Irish jurisdictions, and semi-structured interviews with a range of policy-makers (N = 28), practice nurses (14), general practitioners (12) and patients (13) to explore their awareness of the documents’ contents and how they saw the impact of ‘policy’ on primary care practice.

Results  The findings suggest that although strategic policy documents can be useful for highlighting and channelling attention to health issues that require concerted action, they have little impact on what either professionals or lay people do.

Conclusion  To influence the latter and to encourage a systematic approach to the delivery of health care it seems likely that contractual arrangements – specifying tasks to be undertaken and methods for monitoring and reporting on activity – are required.

Translating policy into practice

During recent years there has been considerable interest in problems associated with what has been called translational research. The meaning of the term translational research is somewhat unstable, but in most cases, it refers to the ways in which findings from laboratory and clinical sciences transfer into activities aimed at improving the health of the public.1

Naturally, scientific knowledge and innovation can rarely if ever move directly from ‘bench to bedside’, but rather has to diffuse through a wide range of conduits before arriving in the public domain. In the process of diffusion, health policy and associated health delivery mechanisms will often play a part. Unfortunately, there has been surprisingly little consideration given either to the role of policy or to the potential influence of such policy on clinical practice at the community level.2,3 In an age in which policy is itself expected to be evidence-based and evidence-driven, this constitutes something of an important lacuna.4,5

In this paper, our focus is on policies relating to the prevention of coronary heart disease and in particular, secondary prevention which aims, through early detection and intervention, to prevent disease progression and thereby reduce morbidity and mortality.6 Our geographical context is health policy in the two jurisdictions of Ireland. However, CHD has served as a target for health interventions across the globe, and multiple policies relating to CHD prevention have been published worldwide. In some cases, the policy focus has been on ‘health’ in general whilst in others it has been on single modifiable CHD risk factors (e.g. diet, physical activity, smoking). The latter are especially relevant given that, for example, 50–75% of the decline in CHD mortality in Ireland can be attributed to lifestyle changes.6,7 During recent decades, most countries have witnessed declining mortality from CHD – with only Russia and some central and eastern European countries evidencing negative trends. Thus, age-standardized death rates from CHD declined by around 48% for men and 47% for women in both Ireland and the UK between 1985–89 and 2000–04.8 Nevertheless, despite an established history of policy-making relating to CHD it remains ‘the single most common cause of deaths in the EU’9, p. 7 including both Northern Ireland (NI) and the Republic of Ireland (RoI).10–12 CHD also remains a major cause of disability, associated with impaired quality of life and significant economic costs, including direct health-care expenses, productivity losses and informal care.9 Given those data, questions arise concerning how specific health policies relate to health outcomes, and how such policies get translated into practice. It is this last question that constitutes the focus for this paper.

Two health-care systems

There are several noteworthy differences between the two health-care systems in which this research was conducted.13 In the RoI health care is funded by a combination of general taxation, social and private insurance and ‘out of pocket’ expenses, whilst in Northern Ireland (NI) government taxation alone funds 90% of health care, through the NI version of the UK NHS. Everyone in NI has free access to a NHS general medical practitioner (GP) and at the time of study prescriptions were issued free of charge (90%), or with minimal cost. By contrast, in the RoI, only the third of the population with the lowest income and people aged over 70 years were covered by the publicly funded General Medical Services (GMS) scheme and entitled to free health care, including prescriptions. Non-GMS patients must pay for visits to a GP (35–50 euro per visit) and prescriptions (up to 75 euro per month). At the time of study, GP care was delivered in NI within the NHS Quality and Outcomes Framework (QOF)14 which usually involved at least yearly monitoring of specified risk factors for people with CHD. QOF constitutes a ‘pay-for-performance’ contract that includes payments for GPs for provision of specific aspects of care and is linked to the level to which they provide this care for the patients in their practice. For example, financial rewards relate to the percentage of patients with CHD whose blood pressure is checked and is controlled to a specified target level. GP care in the RoI was largely unstructured, except for practices associated with this study and the 20% of practices participating in a national pilot initiative (Heartwatch) which paid GPs for providing structured reviews.15

Data sources and methods of analysis

The investigation formed part of a larger research programme (the SPHERE study) which included a randomized controlled trial of an intervention designed to improve secondary prevention in primary care.16 Ethics approval was originally granted by a Local Ethics Committee in NI (Queen’s University, Ref. number 20/02) and by the Irish College of General Practitioners. The part of the study reported on here aimed to investigate how ‘policy’ as embodied in policy documents was linked to ‘practice’ as reported by policy-makers, health professionals and patients. To this end an initial search of the websites of the Department of Health, Social Services and Public Safety (DHSSPS) in NI and the Department of Health and Children (DOHC) in the RoI was undertaken for health policy literature relating to CHD, published, during a 10-year period (Jan 1999, to Dec 2009) including aspects of both primary and secondary prevention. Nineteen documents were identified and searched for references to other potentially relevant documents. Five policy documents and six progress reports in which CHD was a key point of focus were selected for detailed study (Tables 1 and 2). Each document was analysed to assess (i) the priority given to secondary prevention, (ii) the priority given to specific modifiable risk factors and (iii) information on how relevant strategies were to be implemented.

Table 1. Republic of Ireland policy documents and progress reports relevant to the secondary prevention of coronary heart disease
Name of documentAbbreviationPublisherYear
  1. DOHC, Department of Health and Children; HSE, Health Service Executive; ICGP, Irish College of General Practitioners.

  2. *Not included in analysis.

Building Healthier Hearts15BHHDOHC1999
Heartwatch Reference Guide. The Initial Implementation Phase of a National Programme in General Practice for the Secondary Prevention of Cardiovascular Disease25HWDOHC2003
Heart Health Task Force First Progress Report July 1999–June200141DOHC2001
Ireland’s Changing Heart; Second Report on the Implementation of the Cardiovascular Health Strategy 1999–200224DOHC2003
Heartwatch Report42ICGP2004
Heartwatch Clinical Report: March 2003 to December 2005: Second Report43DOHC/HSE/ICGP2006
Changing Cardiovascular Health*11 DOHC2010
Table 2. Northern Ireland policy documents and progress reports relevant to the secondary prevention of coronary heart disease
Name of documentAbbreviationSourceYear
  1. DHSSPS, Department for Health Social Services and Public Safety, Belfast.

Investing for Health23IFHDHSSPS2002
Investing in General Practice: The New General Medical Services Contract14GMSDHSSPS2004
A Healthier Future: A 20 Year Vision for Health and Well-being in Northern Ireland 2005–202544AFHDHSSPS2005
Investing for Health Update 200445DHSSPS2004
Investing for Health Update 200546DHSSPS2005

To explore the ways in which policy ‘text’ transferred into practice, key stakeholders (policy-makers, primary care professionals, patients) who had contributed to the documents or were involved in their implementation were invited to participate in semi-structured interviews. Policy-makers (PLs) were selected purposively, from contributors named in publications, and by asking interviewees to identify others who might have additional background knowledge. Thirty-six policy-makers were invited to interview (NI: n = 21; RoI: n = 15) by the researcher (JW). Of these, eight refused the invitation for diverse reasons. For example, two in Northern Ireland stated that others were better placed to represent their views, and one that they were unfamiliar with the policy, whilst five (NI = 1; RoI = 4) gave no reason.

Twenty-four practices that participated in our associated research programme were also purposively selected to reflect practice type (single-handed/group), size (from 1500 to 15 000 patients), regional research centre (North, East, West), study arm (intervention/control) and location (rural/urban, defined by distance from nearest hospital). Within each practice at least one practitioner was selected for invitation to an interview [14 practice nurses (PNs) (NI: n = 5; RoI: n = 9) and 13 general practitioners (GPs) (NI: n = 3; RoI: n = 10)]. One GP (RoI) declined interview because his practice premises were being renovated. To gather a reasonable range of views, patients (NI: n = 6; RoI: n = 8) were selected from the study practices so as to include people of varied age, gender, location, diagnosis, length of time since diagnosis, marital status, educational background, employment status and, in RoI, GMS eligibility. Only one patient (from RoI) declined, because of a personal tragedy. Interviews of 20–70 min’ duration, took place between December 2006 and November 2007, either at government departments, GPs’ premises or at patients’ homes (n = 63) or by telephone (n = 4). JW interviewed policy-makers; JW and MD interviewed GPs, nurses and patients.

Interview schedules included key issues identified from the documents and were modified following the analyses of early interviews, to allow exploration of emerging themes. Questions explored participants’ awareness of specific policies and perceptions as to how each policy prioritized secondary prevention, impacted on practice and affected patient care and well-being. Respondent’s awareness of any initiatives or future strategies relating to secondary prevention was also explored. Interviews were audio-taped with participants’ consent.

In terms of what has been called ‘the argumentative turn’ in policy analysis17– of which narrative analysis forms a part18– considerable emphasis is placed upon the structure and content of text and speech19,20 in producing and sustaining policy drives of all kinds. We therefore examined the key CHD documents so as to identify the main claims and ‘lines’ of argument and placed these within a grid that facilitated comparison of document contents. The interview data were transcribed verbatim, checked for accuracy, coded manually and analysed using a thematic framework.21,22 Emergent themes were categorized and an iterative process that influenced subsequent interviews and analyses was followed. Increasingly similar interview responses were elicited as the study progressed, suggesting data saturation. The nature and degree of correspondence of the claims made in the policy documents and the activities reported in qualitative interviews was assessed.

Documentation and its focus

Policy documents comprised two types which we refer to as ‘strategic’ [Building Healthier Hearts (BHH); Investing for Health (IFH); A Healthier Future (AHF) – see Tables 1 and 2 for the full titles of these documents] and ‘operational’ (HW; GMS contract). The first described the nature of the CHD problem, what needed to be performed to respond to it and a visionary perspective as to how the ‘doing’ would be achieved through efforts of community and health services. The second contained implementation plans for attaining specified goals and targets.

Strategic documents varied greatly in their emphasis on CHD and its prevention. The RoI BHH, for example, focused exclusively on CHD and its management, including secondary prevention, whilst the NI IFH addressed only the wider determinants of health. Indeed, IFH emphasized the need to encourage people to make ‘healthier choices’ but gave little priority to secondary prevention. AHF mentioned the development of seven major Chronic Condition Management Programmes, one of which related to CHD but did not discuss secondary prevention. Of the operational documents, Heartwatch (HW) focused on the management of individual CHD risk factors in primary care; the GMS Contract (applicable to all UK general practices) included management of specific risk factors as one of several clinical areas within its Quality Outcome Framework.

The language of ‘risk’ and of ‘cost’ looms large in policy rhetoric on both sides of the Irish border. It is on the basis of statistical and population based risk assessments that the justification for health interventions is usually based. In this context, NI’s ‘IFH’23 linked home heating, air pollution, physical inactivity, smoking, high blood pressure and high cholesterol as risk factors for CHD and the ROI’s second report on ‘Ireland’s Changing Heart’24 added obesity, and social deprivation to the list. In general, however, the policy documentation of both countries tended to emphasize biomedical risk factors such as blood pressure, cholesterol and diabetes rather than lifestyle factors in the understanding of CHD. Nevertheless, of the three strategic documents (AHF; IFH; BHH) only the last included specific recommendations concerning blood pressure, although it acknowledged cholesterol as a risk factor and recommended that it should be monitored. All three documents acknowledged diabetes as a risk factor: BHH suggested that people with diabetes should be included in prevention programmes in a manner similar to those with cardiovascular disease. AFH suggested a target of reversing the trend towards doubling prevalence of diabetes over 10 years.

Of the operational documents, both HW25 and the GMS contract14 identified specific targets for blood pressure management but only HW described what GPs should do when these were not achieved. Both documents set targets for total cholesterol, but only HW indicated specific targets for its sub-fractions (total cholesterol, LDL, HDL and triglycerides) and a management plan. HW identified targets related to general prescribing practice (e.g. reviewing frequency of prescriptions) while the GMS contract’s targets included specific medications. HW included targets and management plans for body mass index and waist measurement and specific targets relating to blood pressure and glucose control in diabetes; the GMS contract included specific targets for diabetes management but not for anthropometric measures.

As to lifestyle risk factors, smoking cessation was discussed in all the strategic policies (BHH, IFH, AHF). BHH emphasized how this was to be achieved (e.g. by training health professionals). IFH was less specific about the process but specified funding and pledged production of relevant strategy papers. AHF set a target (a specified relative increase in non-smokers) and suggested creating smoke-free public places and workplaces. HW required GPs to record individuals’ smoking status and, for smokers, a specific management plan. In contrast, the GMS contract set targets for GPs only to record smoking status for their practice population.

The RoI’s BHH recommended processes relating to other lifestyle issues (e.g. providing appropriate education materials, including materials for those with low literacy or low incomes). NI’s strategic policies included targets for physical activity (IFH; AHF), obesity and alcohol (AHF). IFH pledged production of relevant strategy papers and funding. HW described specific targets for physical activity but no other lifestyle risk factors. However, the GMS contract addressed only one lifestyle factor: smoking.

In general then, the strategic documents highlighted risk factors for CHD that needed to be tackled at an individual and a population level, but mostly failed to specify exactly what needed to be performed to improve outcomes. The operational documents specified particular monitoring and management strategies for individual patients.

The interview data

Interviews were directed toward obtaining information about how the policy statements in documents connected to and were translated into action at practice level. Themes that emerged from the interview data included a variable awareness of the documentation and knowledge concerning document content; the significance of operational documents in connecting policy to practice; a variable understanding of an approach to the management of ‘risk’, and differing judgements concerning the influence of ‘evidence’ as against politics in the design and implementation of policy.

In terms of knowledge and awareness, our policy-making informants displayed a considerable variability in both of these factors with respect to policy documentation as a whole. For example, although all were familiar with the document with which they had been associated, some policy-makers were uncertain about the definition of secondary prevention. Practitioners (GPs and PNs) also varied in the extent of their awareness of policy documents. NI respondents were conscious of the existence of strategic documents but unfamiliar with their content and could not comment regarding the priority given to CHD and secondary prevention. Respondents often claimed to be ‘aware’ of the documentation, but lacking precise knowledge of document context as the following brief quotes indicate. ‘Haven’t read it… but yes, I’m aware of it’ (GP; NI). ‘I’ve heard of investment for health but … I haven’t heard it in detail’ (PN; NI); ‘I can’t remember that particular one’ (GP; RoI).

However, NI practitioners were familiar with the GMS contract as an operational document with direct relevance to their work. Most could discuss it in depth and showed evidence of having appraised its value. Some reported appreciating its impact on practice by making them ‘…form a register’ (GP; NI), although it created ‘extra workload’ (GP; NI). Nonetheless, GPs and PNs perceived that it prioritized secondary prevention and had a beneficial impact; viz: ‘So I think overall the patients’ care is improved’ (GP; NI). One PN’s comment suggested concerns regarding the contract’s emphasis on achieving targets, without apparent consideration of individuals’ immediate needs and particular circumstances, thus –‘Ultimately it’s going to benefit the patient, if their BP can be below or their cholesterol below [a target value] but it’s the motive behind it isn’t it as well, it’s just crazy’ (PN; NI).

Few RoI practitioners in our sample were aware of HW (none had participated in the HW initiative). A typical response from a RoI GP was ‘Heartwatch…I’m not sure to what extent … I don’t know how many, what percentage of general practices were involved…’. Nevertheless, irrespective of their awareness of policy documents, GPs from both NI and RoI were quick to state that CHD management was a key aspect of their daily work, thus, ‘I mean GPs and nurses and doctors and all health professionals are quite aware of secondary prevention (GP; NI)’. And, ‘if somebody is a smoker you talk to them about their smoking, if they are having difficulties getting their cholesterol down you talk to them about those things’ (GP; RoI). As to patients: in both jurisdictions they demonstrated some awareness of lifestyle issues relevant to their condition but remained unaware of the policy documents.

Whilst awareness of strategic documents was limited, NI and RoI policy-makers and practitioners agreed that policies outlined in the operational documents (HW; GMS contract) were relevant to practice and provided a structure for managing key risk factors effectively. There was also consensus that structured programmes helped avoid gaps in management and provide equitable care for patients. As one practice nurses stated: ‘You’re covering all corners, you are not just seeing people and seeing them for one thing but you are doing the whole lot for them’ (PN; NI). Another recognized that managing patients on an ad hoc or contingent basis was not as good as it was via a structured system. ‘Sometimes things get forgotten or go missing or they just, people fall between two stools...’ (PN; RoI). Yet another informant suggested that the contract had improved record keeping thus: ‘We now have everyone on a recall system which means that we now know who we’re getting in through the door and who we’re not, rather than hoping that we have everyone in but not exactly knowing’ (GP; NI).

However, some interviewees criticized structured programmes of care as representing ‘tick box’ medicine, anti-holistic and against the ethos of general practice, prioritizing the care of CHD compared to other conditions or problems. Thus, said one, ‘It’s a very cut up approach’ (PL; RoI). Another stated: ‘I’m sold on the idea of giving them [the patients] one target but I’m not going to ignore the others, I think they have to be done in parallel’ (GP; RoI). Other practitioners argued that the GMS contract underpinned good quality care. Thus, ‘while it may be an accusation of tick box medicine I actually don’t think it is, I think it’s getting a structure to the care the patient should be getting’. (PL; NI). Both NI and RoI practitioners voiced concerns that structured programmes could impinge on other aspects of general practice, particularly relating to time constraints and the organization of resources, thus. ‘The nurse’s time is the big thing really…’ (GP; RoI). And, ‘I think they’re good but they’re time consuming’ (PN; NI).

As is implied by some of the above statements, the language of ‘risk’ was clearly echoed in interviews with professionals, and it was recognized that risk assessments formed the evidential basis for much of the CHD policy. Thus, one of the RoI policy-makers stated with regard to BHH, ‘Well by the time this was written there were decades of acknowledgement of the main risk factors such as raised blood pressure, raised serum cholesterol and the consumption of tobacco as well as inactivity, diabetes and increasingly raised body weight’. In general, however, policy-makers tended to speak about ‘risk’ in global terms and only one, who had also worked as a GP, commented on specific risk factors. By contrast, the GPs identified specific risk factors that could be modified, and they recognized both ‘medical’ and ‘lifestyle’ risk factors for CHD. However, they appeared to prioritize ‘medical’ risk, such as blood pressure, over lifestyle risk. Thus, as one GP stated, ‘we would have been good at the medical side of it which would be checking their tablets and blood pressure and cholesterols for instance, not looking at regular exercise and stress’ (GP; RoI). And, ‘we’d talk to them about risk with blood pressure, we’d talk to them about how they can help to lower it themselves, what other interventions they can do like exercise and diet and smoking and cutting out salt and things like that’ (GP; NI). Nurses in NI also commented on managing risk factors and counselling patients about lifestyle issues. ‘I would check their medication, their compliance, any side effects, check their weight and height, blood pressure, give them lifestyle advice……….. and take the blood,….if there was any problems then I would refer them on to the GP…’ (PN; NI).

None of the patients spoke of ‘being at risk’ or ‘risk factors’– a not uncommon finding.26 Indeed, when questioned about ‘risk factors’ some showed hesitation, ‘Risks?’ (PT; NI). However, monitoring provided them with reassurance, suggesting that they recognized some factors were associated with CHD risk, thus: ‘I never had a great problem with blood pressure so it was nice to have that sort of ongoing thing to see it was alright every so often you know’. (PT; NI). And ‘you know. Just kept an eye on things…’ (PT; RoI).

In line with contemporary notions of responsibility, there was consensus among all participant groups that patients must be ‘responsible for their own health’. However, NI policy-makers considered that Health Boards, health-care professionals, communities and non-health agencies also had a key role in health promotion and that they (as policy-makers) had a role in facilitating and encouraging such contribution. As one NI policy-maker stated, ‘Health and social care professionals in [the] Board should take much more cognisance of their role in promoting health and preventing disease’. Interestingly RoI interviewees failed to express similar ideas of partnership, suggesting less active integration between health sectors in the RoI.

As intimated above, most professionals regarded ‘evidence’ as central to policy and to action, but they were equally clear about the role of other factors in implementing and sustaining CHD policies successfully. Hence, policy-makers perceived that policy formation usually arose from some ‘political agenda’, rather than any health agenda per se, and that ‘politics’ was important in underpinning implementation of policy in both NI and RoI. Finance and ‘cost’ were also seen to play a part. As one respondent commented: ‘BHH was a very high profile document with a high profile launch, it was launched by the Minister himself with suggestions of tax incentives…funding does arise from political decisions’ (BHH: PL; RoI). Or, again, ‘Politically… it was a good time with the Assembly setting up and different government departments keen to be involved and it got a strong ministerial support that it was something I guess that all political parties were in favour of’ (IFH: PL; NI). And, on the GMS contract, ‘It’s all part of the national negotiations…so the government chose to invest in it big time’ (PL; NI). Cost and finance also emerged as important factors in discussions with GPs in both jurisdictions and references to the public purse, cost effectiveness and maximal health gain echoed throughout the interviews. For example, one (NI) GP commented on the role of lifestyle factors in terms of economics: ‘Lifestyle interventions [are] not going to cost a lot of money compared to the drugs budget’. However, the practice nurses continued to speak of health and health interventions in terms of a welfare goal rather than a marketable ‘good’ and were less likely to refer to financial factors relating to their care of patients.

Connecting policy to practice

Both NI and RoI policy documents give voice to a narrative that emphasises how CHD morbidity and mortality rates for men and women are among the highest in western Europe, but that such rates are modifiable via effective interventions. However, whilst all the documents discussed herein acknowledged a need to tackle the various antecedents of CHD they varied in their emphasis and focus.

Our policy-making informants proved able to recount, in relative depth, the key features of the document-based CHD narrative. Health professionals, however, were unable to offer substantial comment on the content of any of the strategic level documents. Thus, it seems that primary care practitioners’ lack of awareness of strategic documents (BHH; IFH; AHF) reflects a degree of disconnection between the professionals responsible for implementing the policy goals, and the policy itself.

Despite that it seems likely that structured programmes of care supported by financial incentives, as within HW and the GMS contract, can bridge the gap between policy and practice. Many of our respondents argued that the role of finance is crucial here, and in our study, there was clear recognition from both policy-makers and GPs that delivering health care and health interventions is a ‘business’ and that business needs incentives. As one NI policy informant stated, ‘They (GP practices) are independent businesses and you are buying a service from them’. However, whilst there is some evidence that incentives can improve performance in health care in general,27,28 there are those (including some of our informants) who argue that target-oriented financial incentives lead to ‘tick box’ medicine rather than improved quality of care.29 Nevertheless, most of our respondents’ comments indicated support for the idea of financial incentives and that such incentives target improvements in patients’ health and population health gain.30 References to possible incentives for patients were, however, absent.31

At a broader level, discussion concerning the relevance of incentives to improved performance in health care reflects the changing dynamics of health-care systems. It has been suggested that a market-driven approach to health care and the use of performance indicators may help to improve quality – although the evidence is not always clear.32 The language used by our interviewees reflected a greater emphasis on incentive driven health care in RoI. This might have been related to an awareness of a lack of government funding for preventive health care in the RoI compared to NI – where the GMS contact linked financial reward to measured performance. Certainly, baseline measurements in the SPHERE study showed better management of blood pressure and cholesterol for patients with CHD in NI compared to RoI.33 The SPHERE study provided financial support for the implementation of an innovative prevention programme in general practices in both jurisdictions: it was associated with fewer hospital admissions,34 with a differential economic impact between the two health-care systems35 but no added benefit for control of risk factors. The nature and design of incentives and their role in policy implementation warrants further study.

Our interviewees’ responses also reflected on other changing aspects of contemporary health care and, in particular, of the role of professionals vis à vis their clients. Professionals saw their role and responsibility as one of providing support for clients to make healthy choices but perceived that clients held major responsibility for their own health and well-being. This stance assumes that patients have the necessary resources to make lifestyle changes and that when made aware of their health risks and how to manage them, they will change their behaviour.36 It also implies that patients understand the language of risk in the same way as professionals.37,38

Finally, our findings suggest that politics is central to policy implementation and to policy formation – although politics is, in general, often downplayed as a determinant of health care.39 Yet, our informants emphasized how politicians are key to the mobilization of resources that underpin the implementation of health policy, and how they can be pivotal to highlighting ‘problems’ and their ‘solutions’.

Conclusions

Our focus has been on the connections between policy and practice, and at the outset, we broadly framed our discussion in the language of translational research. Translational research is most commonly discussed in a mesh that incorporates ideas of knowledge transfer and knowledge exchange, and as far as policy is concerned, the prime focus is usually on how to get evidence into policy. Yet, health policy is not merely a conduit through which scientific evidence – and opinion – might flow, but is itself an active agent in the health-care system of most if not all countries. The question we have addressed, therefore, is how policy might get into practice.

Several conclusions are evident from our work. First, that strategic policy documents of the kind that we have discussed do not seem to have a detailed and precise impact on what either professionals or lay people do. However, their publication is important in that they serve to signal and highlight health issues that require concerted action. In that context, the ways in which politicians underpin policy documentation (via speeches, launches and ministerial prefaces to the publications) play an important part in amplifying awareness of health ‘problems’ and their solutions.

It is also clear that ‘policy’ changes do in fact diffuse through the culture of both professionals and their clients over time. Thus, a focus on what might be called lifestyle choices – such as smoking and diet – and the emphasis placed on the ‘responsibility’ of patients to look after their own health are instances of wider cultural and ideological changes that have emerged throughout western European and other societies during the last 40–50 years. In a similar way, broad changes in the semantics of health care can serve to re-direct thinking and action about how to identify and manage health problems – as with the focus on ‘risk’ rather than ‘cause’.40 In all of these ways, it might be said that the language of health policy is of the utmost significance. However, it is also clear that policy rhetoric has little impact on the precise focus of interactions between health professionals and their clients unless it is encompassed within operational documents such as the GP contract. Contracts are, above all, documents that ‘do’ things. Naturally, what motivates professionals to action and to fulfil the contract is probably very different from what motivates their clients – financial incentives on the one hand and quality of life issues on the other, perhaps. Nevertheless, it is clear that unless policy is translated into a clearly structured set of activities, practice will remain sporadic, contingent and piecemeal.

Conflicts of Interest

There are no conflicts of interest.

Acknowledgements

The authors wish to acknowledge the support of the SPHERE Study research team in this work, including C Leathem, A Houlihan, M O’Malley, V Spillane, H Grealish, P Ryan (research nurses); M Corrigan, M D’Eath (qualitative researchers). We thank the patients and practitioners in each of the participating practices for their contributions. LP, MEC, MD are co-funded by the Centre of Excellence for Public Health (Northern Ireland), a UKCRC Public Health Research Centre of Excellence. Funding from the British Heart Foundation, Cancer Research UK, Economic and Social Research Council, Medical Research Council, Research and Development Office for the Northern Ireland Health and Social Services and the Wellcome Trust, under the auspices of the UK Clinical Research Collaboration.

Ancillary