5 Health Impact Assessment in a policy context

on Medicare and the money needed to provide illness services. Similarly, in the UK, health is synonymous with the National Health Service (NHS).

The majority of the population thinks of health as being far more than access to the services needed because of ill health. For planners seeking to develop healthy communities, consideration of health does not equate to healthcare or health systems. For health professionals who are striving to get policy makers to consider the health implications of their actions, the meaning ascribed to health falls well outside the healthcare system. Health policy increasingly needs to be thought of as ‘all policy’. The reality is that policy and practice both within health and outside it are still confined to ‘silos’.

To adopt a problem oriented approach to health policy, therefore, the traditional boundaries of discussion need to be blurred. To be of value and create new ways of thinking about policy problems, one must first ‘problematise the problem’. HIA is a process that seeks to get health considerations onto the policy agenda and thus by necessity must focus on attempts to breakdown ‘silo thinking’.

ANALYSIS OF HEALTH IMPACT ASSESSMENT

An analysis of the origins of HIA shows that it is a highly contested concept. It has a range of meanings and is intended to fulfil a variety of roles depending upon local, regional or even historical origins. Australia’s involvement in the development of HIA was formalised in the 1990s with the publication of national guidelines called the National Framework for Environment and Health Impact Assessment (Ewan et al. 1994). These guidelines were designed for use within the environmental health policy context and were based on the thinking that underpins Environmental Impact Assessment (EIA). They were subsequently developed into the enHealth Health Impact Assessment Guidelines (2001).

In the late 1990s and early 2000s, the Department of Heath and Ageing (DHA) funded several other pieces of research into HIA. These included work on its potential role as a public health policy tool (Mahoney et al. 2002), and the legislative and administrative dimensions of HIA at state and territory level (National Public Health Partnership [NPHP] 2005). Another study focused on strengthening equity considerations in policy making, which included the development of a suite of analytical tools based on HIA (Mahoney et al. 2004).

HIA has a history of approximately 50 years and can be described as a concept still being developed. Analysis of early writings on it show that one of the prime motivators for its development was as a mechanism for rectifying perceived problems in the policy or planning processes of government. Depending on the context in which its introduction was advocated, these problems tended to relate to, but were not exclusively limited to, environmental management, population health, or government agendas seeking to reduce inequalities or disparities within the population. These early writings show that HIA is not a fixed concept with a unitary form based on a set of clearly defined and unanimously agreed principles, but is a highly contested field within and between countries and disciplines.

In the latter half of the 1990s, interest in HIA was rekindled with the publication of papers in leading health journals in England and Canada. These described new policy problems and advocated a new form of HIA focused on public health and policy development. The common theme within them was the argument that increasing attention needed to be paid to the potential role that public policy plays in determining population health outcomes. HIA, it was argued, would help governments to assess the level and severity of the impact that policy proposals would have on health outcomes. This form of HIA has become known as policy HIA or strategic applications of HIA (Abrahams et al. 2002).

As a consequence of this new interest in HIA, a great deal of effort was put into developing a consensus view of the areas of commonality within it. The Gothenburg Consensus Paper (GCP) was produced after extensive international consultation and it defines HIA as:

a combination of procedures, methods and tools by which a policy, program or project may be judged for its potential effects on the health of the population, and the distribution of those effects within the population.

(European Centre for Health Policy [ECHP] 1999 p 4)

The GCP conceded that there are problems with achieving a unified understanding of HIA, arguing that the ‘ramifications of the HIA process are so broad that consensus around aspects … must be built up gradually’ (ECHP 1999 p 2). It calls for agreement on tools for the screening policies for their potential impacts, consideration of the implications of participation in decision-making processes of government, processes for communicating risks, and mechanisms for negotiating changes to proposals so that potential negative or detrimental impacts on health can be avoided prior to the decision being made, as well as a strengthening of the potential positive impacts. The GCP sets out four broad elements that have, in effect, come to describe the attributes important to and underpinning virtually all forms of HIA. These include:

• consideration of evidence about the anticipated relationships between a policy, programme or project and the health of the population (understood as total population and groups within the population);
• consideration of the opinions, experience and expectations of those who may be affected by the proposed policy, programme or project;
• provision of more informed understanding by decision makers and the public regarding the effects of the policy, programme or project on health;
• proposals for adjustments/options to maximize the positive and minimize the negative health impacts.

(ECHP 1999 p 5, bold and italics in original)

Despite striving for a consensus, subsequent writings indicate that there are still inherent differences in people’s understanding of HIA depending upon the perspective they hold, the purpose they ascribe to it, and the context in which it is to be applied. For example, many people describe HIA purely in functional terms, that is, as a tool or instrument (Lock 2000); others describe it through its possible structure, that is, as a framework or approach (Berensson 1998); and others describe it in terms of the innovative potential that it offers for addressing problems differently, that is, as a concept or way of thinking (Scott-Samuel 1998).

Much attention has been paid in the research to practical dimensions such as the development of tools and frameworks, techniques for accessing evidence of likely impact and reporting on the outcomes of completed studies that have used HIA. More recently, attention has been directed to evaluative studies, which examine the value that HIA offers policy makers in influencing decision-making processes (Department of Health and York Health Economics Consortium 2006; Taylor et al. 2003). Little attention has been paid to the implications of the multiple and contested meanings of HIA or the fact that it has emerged as a response to a range of quite different policy problems.

This chapter focuses on the role of HIA in addressing two specific policy problems and explores the implications of its application in these contexts in terms of Australian health policy in it broadest sense. The term ‘policy’ is used generically to describe any new or modified policies, projects, programs, or services that are developed by governments at all levels and which have the potential to impact upon health, defined broadly.

IDENTIFICATION OF THE PROBLEM

Bacchi (1999) in her book, Women, policy and politics: the construction of policy problems, argues that it is impossible to separate the objects or targets of a policy from the way they are spoken about or represented. Any issue is an interpretation and, as such, involves judgment and choice. Policy proposals, therefore, cannot be separated from interpretation, they contain them, they frame the way(s) in which the issues are responded to and these become interventions, which have problematic outcomes. Bacchi argues that we give a particular ‘shape’ to social problems through the way we speak about them and the way(s) we recommend they be dealt with. There are very strong parallels between her work and the development of HIA that have direct relevance to both the policy context and to this chapter. Her specific interest is in the implications of problematisation, or problem representation, rather than the problem per se. Specifically she is interested in three broad areas: ‘What’s the “problem” represented to be; what presuppositions are implied or taken for granted in the problem representation and what effects are connected to this representation of the problem’ (Bacchi 1999 pp 1–2).

With HIA there are multiple ways of understanding or interpreting: What’s the problem represented to be? what are the presuppositions implied and taken for granted within these representations, and the effects or implications of the identified solution(s) to resolving the problem? The framing of policy problems results in the development of new tools or approaches to respond to these areas. HIA is therefore both a product of representations of particular policy problems and the subject of these representations. It is beyond the scope of this paper to examine all of the possible representations of policy problems and thus the forms that HIA can take. These can be sliced according to form, function, intended outcome, context or discipline base. To use an analogy, they can be likened to the multiple ways in which a cake can be sliced as well as the various types of cake that might warrant different types of slicing.

This chapter focuses on the two most common representations of HIA based on two policy problems:

Bacchi also cautions that there are no assumptions in the use of this approach or in the language that is used. There is no one reality that stands outside representation. That is, there is no form of reality or truth against which claims can be judged. All are representations, and, as such, are not right or wrong, better or worse. ‘The goal in taking this approach is to examine the ways in which public policy problems achieve their reality in language’ (Bacchi 1999 p 37) and as such they illustrate the inherent tensions that will always lie within the political and policy development process. For HIA, the implication of this is that despite the development of a body of literature and methodological tools to support the policy development process, there will never be one agreed form of HIA that can or should be applied to all contexts. As a form of intervention that supports perceived social, political, or environmental problems, HIA will always be context bound. This section introduces each of the problem representations and their links to HIA. The problematisation of each of these is examined in the subsequent section.

FRAMING THE PROBLEMS AND LINKS TO HIA

HIA as a solution to achieving healthy public policy

In the late 1980s and early 1990s, the concept of healthy public policy was introduced by health professionals who were seeking to extend the scope of health policy beyond the traditional medical health paradigm. The basic tenet underpinning healthy public policy is that health is influenced by a range of factors largely outside the control of the individual and the healthcare system and that the focus of activity needs to be directed at a population rather than on an individual level (Milio 1981). According to the World Health Organization (WHO), healthy public policy is characterised by an explicit concern for health and equity in all areas of policy and by accountability for health impacts (WHO 1998). The main goal of policy framed this way is to create environments where people can lead healthy lives and, in order to do this, it is necessary to address inequity and disadvantage at population levels. So the healthy public policy agenda sought to be transformative, emancipatory, political and health promoting (Baum 1998; Simpson et al. 2004), rather than health protecting.

In framing the problem that led to the development of HIA linked to the achievement of healthy public policy, the social model of health is based on the assumption that health status is not only determined by the health services people have access to, and by their own behaviour, but by a range of factors outside their control, including government policies. Health policy and the health sector, therefore, cannot deliver improved health outcomes alone and so health becomes everyone’s business. Policy makers in all sections of government must either be required to account for, or encouraged to consider, the likely impacts of their actions on the health of the population in order for public policy outcomes to be considered healthy.

To achieve this, a new approach is required which can assist the decision maker to identify the health impacts that are likely to occur as a consequence of their actions, thus ensuring that healthy outcomes are created. HIA is represented as a support tool for decision makers that can be used to identify potentially negative impacts and ameliorate these as well as to identify and strengthen actions, which can have positive or beneficial health outcomes. The potential health impacts might be positive, negative or unknown and they are not spread evenly or consistently across the population. The use of HIA in a policy context allows for health considerations to be included where they currently are not, for evidence of likely effects to be factored into the decision making, and for the trade-offs likely to arise to be considered prior to a commitment to proceed. The driver for its use is thus the avoidance of something accompanied by a desire to ensure that health considerations are given high priority within government.

As the focus of action rests both outside the health sector and inside it, proponents of healthy public policy have advocated the need for intersectoral working to support the achievement of what is essentially ‘health for all people’. In this problem representation, health practitioners must accept that they do not control the achievement of health outcomes or the mechanisms by which health is ‘delivered’ to the population. All policy becomes health policy and the achievement of positive health status requires interventions or actions from a variety of sources, largely outside the control of the health sector. The risks associated with this representation are that the concept of health is so broadly defined that responsibility for it rests everywhere and nowhere simultaneously.

HIA as an extension of EIA

In 1969 the National Environmental Policy Act (NEPA) was introduced in the US. Within 2 years of its introduction EIA processes were established. One of the purposes of this legislation was the promotion of effort ‘which will prevent or eliminate damage to the environment and biosphere and stimulate the health and welfare of man’ (Banken 1999 p S27). Despite prior work, it seems clear that public health started to be formally integrated into EIA processes in the 1980s as a consequence of the release of the WHO report on the health and safety components of EIA. This report recommended the use of the risk assessment and management processes within EIA focusing on toxicological aspects (WHO 1987). This was one of the principal drivers for the development of HIA.

As stated above, Australia has a strong tradition of this form of HIA. Other countries with a similar tradition include New Zealand, Germany, Canada, UK, Thailand and countries in receipt of World Bank or (former) International Monetary Fund (IMF) development funding, including countries in Africa. Countries with a strong tradition of using EIA processes within environmental decision making have procedures enshrined in law and have clearly established protocols to guide the use of EIA or HIA, as well as clear expectations of the types of outcomes required by law. The goal of this form of HIA is the protection of human health. It focuses on specific threats to community or public health and it seeks to forecast the likely unintended consequences of changes to the physical environment. The enHealth HIA guidelines define it as:

The process of estimating the potential impact of a chemical, biological, physical, or social agent on a specified human population under a specified set of conditions for a certain time frame.

(enHealth 2001 p v)

Here HIA is concerned with identifying the gross and direct physical impacts of an action on health and it does not seek to identify the positive effects on health that the proposal might potentially lead to. Generally, it is characterised by the following features: it is legislated, scientific, and analytical, based on proven dose–response or cause–effect risk assessment theory and methodology. In order for an EIA (containing a health component) or a separate HIA to be undertaken, an assessment of a possible risk to human health is needed. That is, a problem or concern needs to have been identified and this prompts or triggers an HIA or EIA. The impact of this is that impacts are perceived to be embedded in the proposal and as such HIA is a reactive process.

There are two schools of thought on the reasons why it developed in this way. The first is that it was a natural extension or progression of EIA, building on the pre-established base of impact assessment methodology and responding to a gap in the existing EIA processes. The second is that it arose out of a major weakness within the application of EIA.

In the first of these interpretations, there is an assumption that the development of HIA was logical, linear and a natural extension of existing work. The development of Social Impact Assessment (SIA) is construed in much the same way. In this representation, HIA is depicted as riding on the successes of EIA. For instance, Berensson (1998) attributes EIA as the inspiration for the development of HIA in a policy context in Sweden with HIA tools. Lock (2000) also depicts the development of HIA as a natural extension of both EIA and SIA and based on similar principles. Lock describes the first documented HIA in the UK. It was based on a submission to a public inquiry on the proposed second runway at Manchester airport. This HIA was undertaken prospectively using methods based on EIA processes. The study was limited by a lack of quantitative data but still proved to be a powerful lobbying tool, resulting in the implementation of changes to the planning proposals, including increased provision of public transport and noise reduction schemes (Lock 2000 p 1395).

Within Australia, where EIA is required by federal, state and territory governments in slightly different ways, there is an onus on an organisation proposing any new development project to complete an Environmental Effects Statement (EES) when requested by the minister responsible for planning. These requirements vary from state to state and are designed to draw out the health risks associated with any new development and to identify the measures that will be taken to ameliorate these risks. The approach is highly regulatory and is essentially an administrative process to ensure compliance based on scientific proof and the controls of legislation. Guidelines set down for the applications to HIA, therefore, focus on explaining preconditions. HIA in this form is a highly regulated instrument for health protection applied by experts and on whom responsibility rests. Within this form of HIA, decision making is perceived to be effective because there is increasing evidence that the work conducted has been rigorous and able to demonstrate that if changes are made to specific proposals based on the findings of the EIA/HIA then health risks will be avoided.

The alternative view is that this form of HIA developed because of shortcomings in EIA and SIA processes, especially practitioners’ failure to take adequate account of health considerations or to limit the interpretation of health to narrow physical impacts, especially those related to exposure to chemicals. Lock argues that, ‘In practice very few environmental assessments are carried out and they rarely consider human health’ (2000 p 1395).

In summary, this form of HIA is largely an action-forcing device (Banken 1999 p S29) where the threat of a risk to health is sufficient to warrant changes to a proposal. It is based on the precautionary principle of do no harm and arises out of the problems government faces in seeking to protect the population from potential risks. In this form, people are depicted as having things happen to them rather than as active players within policy or decision-making contexts. For example, disadvantaged groups are depicted as being exposed to more health hazards and thus are more susceptible to these hazards. Harm is seen to be avoidable and controlled through regulatory procedures. Quantitative modelling is a technique that can be used to predict the overall impact of a decision on population health because cause–effect relationships can be established. There is little, if any, uncertainty within this form of HIA. There is also little, if any, concern for other policies of government unless they pose a threat to human health.

THE POLICY RESPONSES: THE IMPLICATIONS OF REPRESENTATIONS

The previous section identified the links between two different problem representations and the development of two quite differing forms of HIA as a response. It explored both ‘What is the problem represented to be?’ and the presuppositions implied or taken for granted within each representation. The following focuses on the effects of the identified solution(s) (in this case, HIA) to resolving the policy problems.

It is not the intention to create a discussion based on binaries by presenting only two of the possible problem representations. Early writings on HIA portrayed these two forms of HIA as the only ones, locating them on a continuum with each end possessing specific attributes linked to methodological origins (Kemm 2000). The properties at each end of this continuum were compared and the implications for policy makers of each end were explored. This approach is unhelpful because it is based on the assumption that the form that HIA takes is fixed. Because HIA is a response to the way a policy problem is represented then its form and qualities will be derived by that framing and by the solution required for that problem. HIA will, therefore, always be tenuous or volatile, context specific and liable to change as, and when, the representation of the policy problem changes. When viewed as a solution to a problem, it is easier to understand HIA as dynamic. As such it is not only the way that the policy problem is represented that has the potential to create tensions for policy makers, but also as a solution to a problem, HIA possesses inherent tensions in its own right. This is the main point of confusion within much of the writing about HIA.

HIA as a solution to achieving healthy public policy

It is not difficult to identify the policy problem in this representation or the presuppositions underpinning it. Journal articles, which focused specifically on the rationale for the introduction of healthy public policy, were supported at the time by an international momentum including specialist conferences and initiatives such as WHO’s Health for All or Healthy Cities. The description of the problem was accompanied by calls for a mechanism to address it. HIA, as the potential solution, was developed quickly and guidelines for its application were circulated internationally. These were accompanied by reports of case studies that had successfully applied HIA, as well as by writings which encouraged the application of HIA to the link between policies and the problems of reducing inequalities and inequities in population health.

The introduction of HIA within the healthy public policy context offers new possibilities but also creates many new problems for policy makers. The representation is premised on changing the expectations of policy makers, of their role and their current processes. This form of HIA is founded on the following assumptions:

• that all decision makers should scrutinise their processes for potential health outcomes
• that people should be consulted on decisions that are likely to affect their health
• that a culture of openness and transparency is required within work practices
• that integrated work practices should be in place and are important for strengthening decision-making practices within government.

To be an effective solution to the problem of achieving healthy public policy, HIA needs to be accepted by all policy makers as a credible tool, everyone needs to be committed to its use and evidence of the links between the policy and the potential health impacts needs to be accessible to all. The following discussion focuses on the inherent tensions underpinning several of these issues.

On one hand, within all levels of government there is a consistent lack of scrutiny by health professionals of the policies under their control. There appears to be a belief that because health policies and practices are well intentioned then they will not contain unintended, potentially negative or inequitable impacts. On the other hand, while health professionals argue that it is vital that the health impacts of government policy are identified, many departments do not consider that it is their responsibility to take action linked to health advancement for the population. Their willingness to adopt a process like HIA will limit both the government’s ability to achieve healthy public policy as well as HIA’s potential to be an effective solution.

Demands for proof of the added value that HIA brings to the policy process can often run counter to the premises upon which the principles of this form of HIA are based. For example, a recent study in the UK Department of Health and York Health Economics Consortium (2006) has sought to ascertain the cost effectiveness of HIA in improving decision making. This form of HIA, however, is premised on two principles:

A causal connection cannot be established between HIA and the range of other activities occurring both inside and outside government that seek similar goals. To try to separate HIA from these other activities and attribute costs and benefits seems to run counter to the goals of this form of HIA and is more akin to the requirements of the other form of HIA described in this paper (i.e. those of certainty, scientific rigor and cause–effect relationships). The use of evaluative approaches such as cost–benefit analysis necessitates quantitative evaluation tools that try to establish causal connections between a range of complex factors that cannot be measured and which run counter to the intended role of HIA in encouraging greater consideration of the health consequences of policy processes. The report on the study shows that it was extremely problematic to ascribe a value to the activity because there was a lack of consistency across the completed HIAs and the individual effects that the studies had could not be quantified.

Discipline-based judgments about what constitutes evidence of effect also create tensions. In public health, hierarchies or gold-standard approaches to valuing different forms of evidence are prioritised. Evidence from external sources about the potential impact of an action on health (defined broadly) will not necessarily be available or taken as seriously by health professionals. HIA has necessitated the development of new ways of understanding the role of evidence in policy processes. It has shown that much policy making might be drawn on evidence to frame the problem, but will not necessarily draw on evidence to indicate how the proposed intervention will achieve the desired outcome.

HIA as an extension of EIA

Given the clear history of HIA arising from EIA, less has been written about the long-term implications of its achievement within the changing policy area of environmental management. It is easy to conceive of the problem of protecting public health as relatively static and of its solution as trustworthy and predictable. Despite the predictability and rigor of HIA in this context, there are still tensions with its application. The greatest of these is that it is not used often enough as either a component of EIA or as a separate process. There is one other important over-riding tension with this representation that is of direct relevance – it is changing because the problem upon which it is premised is changing. Unlike the change which occurred through the introduction of the healthy public policy representation of HIA, this change is subtle.

In 2004, a conference on the role of HIA in environmental management (enHealth 2004) explored the tensions faced by environmental health and public health planners and policy makers in seeking to increase its use, especially for complex issues such as economic development, resource management, land use planning, and spatial planning. The list of tensions identified by representatives from every state and territory in Australia was virtually identical.

The tensions indicated that beneath the routine administrative and bureaucratic frustrations the participants experienced in trying to protect public health, there was a series of new problems. These were: new, complex and interrelated issues to deal with; pressure to work with people and organisations differently; and increased levels of public and community awareness of issues linked to decision making, health impacts and environmental management. What participants were describing was a shift in the way the problem was defined. Not only was the ‘What is the problem represented to be?’ subtly changing, but so too were the presuppositions that underpinned it. It is clear that their discussions focused on the tensions of dealing with a solution, which no longer fitted the problem because the problem was being reframed by a range of external influences. Participants were the subjects of a changing policy environment rather than the instigators of it. Some were able to describe the changing environment in which they were operating, some argued that the solution could not solve the problems they had to address, others argued that the solution was not effective because people who should be using it would not. What was clear was that the discussions never focused on the changing nature of the problem or the new way it was being understood. The discussion focused on the solution and the tensions experienced by it not fitting rather than on the problem it was aimed at solving.

Additionally, because many participants viewed the tensions as coming from outside, there was resistance to change either because they did not see the need to or because it was threatening. Being ‘within’ the profession made it difficult for them to move outside their familiar representation of the policy problem, to notice the shifts, to understand the reasons for them or to consider the opportunities for reframing both the problem and the solution.

Koontz et al. (2004), in their book on environmental management and the role of government, challenge the presuppositions underpinning the framing of the environmental management problem and provide a new way of thinking about the solution that is required:

the diffusion and availability of environmental information, growing emphasis on deliberative democracy, and the place-based nature of many environmental problems that resist command and control strategies have made collaborative environmental management an increasingly popular means for addressing environmental problems.

(Koontz et al. 2004 p 174)

This quotation sums up the tension being expressed by the conference participants. Regulation and command and control strategies such as HIA (framed from the traditions of EIA) are examples of bureaucratic structures that keep the power firmly vested within government. Expert driven, regulatory approaches have, and will continue to have, an important role to play but there is an increasing need for policy processes to adopt a more innovative stance that moves beyond regulation.

Policy makers appear to be facing constant pressure to balance environmental management and economic growth. Such decision making results in both direct and indirect health outcomes and the pressure to understand these is constantly growing. Increasingly, environmental and natural resource activities such as ecosystem management, land use planning, water management and quality improvement systems, involve complex relationships between multiple stakeholders. Framing the problem simply in terms of protecting public health is too simplistic. Similarly, a solution that cannot accommodate complexity will not be popular. A repertoire of approaches will be needed because the problems are too complex for a simple or single solution and HIA will need to be reframed in the light of this. Koontz et al. (2004) for instance, identify three key roles for government – those of following, encouraging and leading in environmental management.

As a solution to the shifting representation of the policy problems, the highly regulated, quantitative application of HIA using a narrow definition of health linked to direct health outcomes, is insufficient. Complex social, economic and environmental issues need complex multi-layered approaches involving a range of disciplinary perspectives and new ways of valuing evidence of indirect or cumulative impacts on health. For example, it is virtually impossible to find conclusive evidence for the direct effects of inadequate or poor community planning in health, or to be able to establish a causal link between inadequate public transport and the range of factors that determine health status, such as low educational achievement, low income and poor-quality housing. Similarly, it is virtually impossible to include consideration of the impacts of developments or policies on mental or psychosocial health. The approach to HIA therefore needs to change to become more participatory, less expert driven and more able to accommodate a range of research perspectives

FUTURE DIRECTIONS FOR POLICY DEVELOPMENT

There are clear opportunities for HIA to make an effective contribution to the resolution of some of the problems facing policy makers. When viewed from the perspective that all policy is a representation of a problem, HIA’s complexity can be reduced and its value understood. Until HIA can be understood as context specific there will be constant resistance to it having multiple meanings, and multiple possible applications or flexible applications. This reframing will take many years and a great deal of movement by people both inside and outside the health policy sector.

Related posts:

8 Municipal public health planning policy in Victoria 14 The problem of failing to provide culturally and linguistically appropriate healthcare 11 Conflicting values in health information policy 13 Dilemmas in end-of-life care: the Maria Korp case 20 Implementing post-injury rehabilitation policy 6 Federalism and health

Analysing Health Policy A Problem-Oriented Approach