POLICY AND PRACTICE
Participation in health impact assessment: objectives, methods and core values
Participation aux études d'impact sur la santé : objectifs, méthodes et valeurs clés
Participación en la evaluación del impacto sanitario: objetivos, métodos y valores básicos
John WrightI; Jayne ParryII,1; Jonathan MathersI
IHealth Impact Assessment Research Unit, Department of Public Health & Epidemiology, University of Birmingham, Birmingham, England
IISenior Lecturer, Health Impact Assessment Research Unit, Department of Public Health & Epidemiology, University of Birmingham, Edgbaston, Birmingham B15 2TT, England (email: J.M.Parry.1@bham.ac.uk)
ABSTRACT
Health impact assessment (HIA) is a multidisciplinary aid to decision-making that assesses the impact of policy on public health and on health inequalities. Its purpose is to assist decision-makers to maximize health gains and to reduce inequalities. The 1999 Gothenburg Consensus Paper (GCP) provides researchers with a rationale for establishing community participation as a core value of HIA. According to the GCP, participation in HIA empowers people within the decision-making process and redresses the democratic deficit between government and society. Participation in HIA generates a sense that health and decision-making is community-owned, and the personal experiences of citizens become integral to the formulation of policy. However, the participatory and empowering dimensions of HIA may prove difficult to operationalize. In this review of the participation strategies adopted in key applications of HIA in the United Kingdom, we found that HIA's aim of influencing decision-making creates tension between its participatory and knowledge-gathering dimensions. Accordingly, researchers have decreased the participatory dimension of HIA by reducing the importance attached to the community's experience of empowerment, ownership and democracy, while enlarging its knowledge-gathering dimension by giving pre-eminence to "expert" and "research-generated" evidence. Recent applications of HIA offer a serviceable rationale for participation as a means of information gathering and it is no longer tenable to uphold HIA as a means of empowering communities and advancing the aims of participatory democracy.
Keywords: Public health; Public policy; Needs assessment; Consumer participation; Social justice; Evidence-based medicine; Policy making; Decision making/methods; Review literature; United Kingdom (source: MeSH, NLM).
RÉSUMÉ
Les études d'impact sur la santé (HIA) constituent un outil multidisciplinaire d'aide à la décision, qui évalue l'impact des politiques sur la santé publique et sur les inégalités en matière de santé. Elles ont pour but d'aider les décideurs à maximiser les gains en termes de santé et à réduire les inégalités. Le document de consensus de Göteborg (1999) indique les raisons qui justifient d'inclure la participation communautaire en tant que valeur clé dans le cadre de ces études. Selon ce document, la participation aux études d'impact implique la population dans le processus de décision et corrige le déséquilibre démocratique entre le gouvernement et la société. Elle donne aux citoyens le sentiment que la santé et la prise de décision leur appartiennent et qu'ils interviennent personnellement dans la formulation des politiques. Cependant, les dimensions participative et d'encouragement à l'autonomisation des études d'impact peuvent s'avérer difficiles à mettre en pratique. Dans le présent examen des stratégies participatives adoptées dans les applications principales des études d'impact au Royaume-Uni, nous avons trouvé que l'objectif de ces études, à savoir influencer la prise de décisions, crée une tension entre leurs dimensions participative et documentaire. Par conséquent, les chercheurs ont réduit la dimension participative, en diminuant l'importance accordée à l'expérience communautaire d'autonomisation, d'appropriation et de démocratie, et ont élargi la dimension documentaire en donnant la prééminence aux données de l'expertise et de la recherche. Les applications récentes des études d'impact sur la santé donnent une justification utile de la participation en tant que moyen de collecte d'informations mais il n'est plus défendable de soutenir ces études en tant que moyen de renforcer l'autonomie des communautés et de faire avancer la démocratie participative.
Mots clés: Santé publique; Politique gouvernementale; Evaluation besoins Participation consommateurs; Justice sociale; Médecine factuelle; Choix d'une politique; Prise décision/méthodes; Revue de la littérature; Royaume-Uni (source: MeSH, INSERM).
RESUMEN
La evaluación del impacto sanitario (EIS) es un instrumento multidisciplinario de ayuda a la toma de decisiones que determina la repercusión de las políticas en la salud pública y en las desigualdades sanitarias. Su finalidad es ayudar a los decisores a maximizar los beneficios sanitarios y reducir las desigualdades. El Documento de Consenso de Gotemburgo (DCG) de 1999 proporciona a los investigadores una base sólida para establecer la participación comunitaria como un valor central de la EIS. Según el DCG, la participación en la EIS empodera a los implicados en los procesos decisorios y corrige el déficit democrático existente entre el gobierno y la sociedad. La participación en la EIS transmite la sensación de que la salud y la toma de decisiones pertenecen a la comunidad, y las experiencias personales de los ciudadanos se convierten en un factor esencial de la formulación de políticas. Sin embargo, las dimensiones de participación y empoderamiento de la EIS pueden ser difíciles de llevar a la práctica. En esta revisión de las estrategias de participación adoptadas en aplicaciones clave de la EIS en el Reino Unido, observamos que el objetivo de ésta de influir en la toma de decisiones es origen de tensiones entre la línea participativa y la línea de reunión de conocimientos. En consecuencia, los investigadores han reducido la dimensión participativa de la EIS prestando menos importancia a la experiencia de empoderamiento, adhesión y democracia de la comunidad, y han ampliado paralelamente -procediendo para ello a resaltar la labor de los expertos y la evidencia generada por las investigaciones- la faceta de acopio de conocimientos. Las aplicaciones recientes de la EIS brindan una justificación práctica de la participación como medio para reunir información, pero ya no es sostenible la idea de mantener la EIS como una opción para empoderar a las comunidades e impulsar los objetivos de la democracia participativa.
Palabras clave: Salud pública; Política social; Participación comunitaria; Justicia social; Medicina basada en evidencia; Evaluación de necesidades; Formulación de políticas; Toma de decisiones/métodos; Literatura de revisión; Reino Unido (fuente: DeCS, BIREME).
Introduction
Health impact assessment (HIA) explores the effects of policies, programmes and projects initiated in "non-health" sectors (for example, a regeneration programme, or a new transport system) on public health and health inequalities. HIA can be undertaken at a number of decision-making levels: for example, an HIA may be undertaken on proposals for a new leisure centre (project level), a regional transport scheme (programme level), national taxation schemes (national policy level) or even international policies such as the Common Agricultural Policy (supra-national policy level). Early studies of HIA focused on its application to projects and developments, but today, there is growing interest in its application to policy (1).
In 1999, a formal expression of the methods and values of HIA was made in the Gothenberg Consensus Paper (GCP). The GCP argued that that HIA must embrace the broader ideals of social policy or be divorced from the "reality" of the environment in which it hopes to find a place, and asserted the right of people to participate in the "formulation, implementation and evaluation of policies ... both directly and through elected political decision makers" (2). According to the GCP, participation in HIA made the policy process more transparent and democratic. It empowered people within the decision-making process and redressed the democratic deficit between government and society. Participation in HIA generates a sense that health and decision-making is community-owned and the personal experiences of citizens become integral to the formulation of policy.
Early applications of HIA adopted the values set out in GCP and directly involved communities in the identification, assessment, and prioritization of health impacts. However, in the four years since the GCP was drawn up, policy-linked HIA faces serious difficulties. Although advocated at regional, national and supra-national levels, the incorporation of HIA into decision-making has been patchy at best (38). Numerous reasons for this "failure" have been suggested for example, the loss of key supporters of the process, or the existence of other forms of impact assessment (e.g. environmental health assessment, sustainable impact assessment) (9). However, even where the political climate has been broadly supportive, HIA has proved difficult to operationalize its participatory and empowering dimensions in particular. While participation may encourage open, democratic debate about policy options, the time and resource demands of the political process limit the extent to which the community can be engaged, and for decision-makers, time and resource constraints are among the most frequently mentioned barriers to conducting policy research (10). The requirement to operate within the time frame of policy creates tension between HIA's participatory and knowledge-gathering dimensions. In other words, community participation presents HIA researchers with a conundrum: the adherence to the core values of community participation and empowerment threatens the likelihood of being able to influence policy-making processes.
Methods
We have outlined the potential difficulties facing participation in HIA elsewhere (1, 11). In this paper, we review actual experiences with participation in the United Kingdom. In particular, we consider whether HIA should welcome participatory methods of appraisal in the name of empowerment, representation and ownership, or whether participation should be restricted in the name of timely and resource-efficient assessments.
Our examples are drawn from the United Kingdom Health Development Agency web site, a well-known and accessible resource offering an abundance of information on HIA (12). We have selected assessments in which the requirement for public participation might be expected to have been greater than usual. Thus, our examples include assessments undertaken on both policies and projects by key government agencies the Scottish Executive, the London Health Observatory in which public participation was actively courted. Similarly, we also present HIAs of more "contentious" policies and projects, in which levels of public interest were high (the redevelopment of Alconbury airfield, the expansion of Finningley airport, and change of fuel at a cement works in Rugby).
Early approaches to health impact assessment
In the United Kingdom, initial approaches to HIA, building upon on the work of Labonté (13), the advice of key government guidance papers for example, the National Health Service's A resource for health impact assessment (NHSR) (14) and The Merseyside guidelines to health impact assessment (15) advocated the use of participatory techniques as a means of empowering local communities to represent their own interests against the macro design of policies affecting community health.
From late 1997, the Liverpool Observatory/IMPACT group authors of the Merseyside guidelines pioneered the HIA process with a series of assessments on government programmes and projects in the Merseyside area. Typically, these approaches to HIA began with the appointment of a steering group, which conducted a series of consultation exercises with the affected community. These exercises identified a set of health impacts, which were analysed for magnitude of effect, and prioritized by the community and other participants themselves. Although guidance acknowledged that, in many cases, community perceptions of risk and effects were subjective; these subjectivities were maintained to be no less "valid or important" than "precise technical data" (15). Thus, while the Merseyside series emphasized the knowledge-seeking dimension of HIAs, they regarded its democratic and empowering dimension as critically important. For example, in 1997, a project assessment of community safety initiatives recommended that local people "should not just be asked to approve the ideas of others", but should be involved in the "designing" of community safety measures (16). Similarly, in 1999, evaluators defended an HIA of a drug prevention strategy against criticisms that its recommendations had largely been "based on people's opinions", rather than evidence from the literature, claiming that participatory data was more "locality specific" and no less valid than evidence from the literature (17).
However, internal evaluations of early assessments in the Merseyside series were critical of the use of participation. In 2000, an assessment of a proposed sporting centre in Parr noted that previous HIAs had involved an "over reliance on opinion, and not enough on fact" (18). Evaluations of earlier contributions to series had recommended that the views of all participants should be researched and validated, and that future assessments must make a "more analytical use of the literature and other experts' knowledge" (18). Taking these shortcomings into account, the Parr Sports Centre assessment did not advance the use of large community focus groups as a means of gaining a comprehensive picture of public opinion, rather "interviews with two or more respondents tended to work best". Smaller consultation groups offered increased room for compromise, provided alternative viewpoints, and ensured that the issues were better explored (18).
The experience of other organizations
The experience of the Merseyside group has not been unusual: other HIA workers have also encountered difficulties in operationalizing participation. In 2000, the Scottish Executive commissioned a pilot study of HIA through the Scottish Needs Assessment Programme (19). This Programme conducted HIAs on the Edinburgh council's urban transport strategy and the north Edinburgh housing strategy using a framework broadly similar to the Merseyside guidelines (20). For the housing strategy, the assessors completed a community profile, which identified key stakeholders, marginalized groups and affected communities. Although a series of focus groups was undertaken with community groups, the assessors noted that "time and resource constraints dictated that existing groups be consulted rather than attempting to convene new groups for the purposes of the assessment", and expressed concern that the knowledge gained from the groups could not be described as representative of the community (21). Wider participation would have added more weight to the study, but the broad scale of the policy, the time constraints of the policy process, and the size of the affected population made truly representative participation impossible to achieve. Similarly, in their HIA of Edinburgh council's urban transport strategy, the Scottish Needs Assessment Programme departed from a bottom-up participatory approach by prioritizing and estimating the magnitude of the health impacts themselves, because "it was decided that time constraints together with the size of the population affected by the transport policy made meaningful consultation with the community impossible" (21).
Between 2000 and 2002, the London Health Observatory conducted a series of rapid HIAs using workshop-based participation exercises to assess the effects of policy strategies on public health (2230). Although the Observatory used its own model of HIA A short guide to health impact assessment (SG-HIA) (22) it drew heavily from the GCP (2) and Merseyside guidelines (15), and accorded equal worth to the views of stakeholders and experts alike, arguing that HIAs' value depended upon "creating a new sound" by bringing these different voices together. The London Health Observatory offered a different rationale for participation than that envisaged in the GCP. For most of the HIAs conducted by the Observatory, a core team of professionals reviewed the strategy documents and settled the main topics for analysis, which were subsequently analysed at a workshop on rapid participatory appraisal by team-nominated invitees typically, about 40 people drawn from a range of sectors and levels of seniority within the London Health Commission, London Assembly's Environment Committee, community organizations, public health contacts, and Greater London Authority strategy leaders. The workshops reviewed the strategy and its major policy objectives, discussed the HIA process, and considered the relevant public health data, before breaking into smaller facilitated groups to analyse specific areas of the strategy. Following the discussion, facilitators fed main points into a plenary session, after which the core team compiled a report using the notes from the sessions. Recommendations were then formulated on the basis of the workshop discussions and, where the evidence base supported recommendations, special note was made in the report. Ultimately, the report was offered to all participants for comment.
The SG-HIA also demonstrated an acute awareness of the limitations that the policy process imposed upon HIA, recognizing that policy-makers and senior managers might be reluctant to introduce HIA because of lack of resources, "an already overcrowded agenda" or even "a tradition of minimal public or community involvement". Accordingly, the SG-HIA assigned to HIA a new role, that of bringing "health issues into the foreground of policy". The change was subtle, but significant. Involving stakeholders and the local community conveyed "both intellectual and democratic legitimacy" to HIA (22). Thus, participation had become instrumentally valuable, conferring legitimacy. This change of role made it possible for researchers to restrict or expand participation with the intention of drawing the attention of policy-makers to health issues and influencing decisions. Herein, the SG-HIA advised that HIA must be "sufficiently flexible", and address "both the resources and the responsibilities of decision-makers". There could be "no single 'right way' of introducing or implementing HIA". Assessors needed to develop an approach that was "effective in their particular circumstances" (22). Thus, for the London Health Observatory, emphasis on the participatory and knowledge-gathering dimensions of HIA had become negotiable relative to context, circumstance and guided by the more general aim of influencing decision-makers in a way that earlier approaches had not envisaged.
Public participation in other health impact assessments
In September 2000, Cambridgeshire Health Authority conducted an HIA of the redevelopment of a disused American air base at Alconbury (31). The Authority adopted the methodology set out in the Merseyside guidelines. However, where the guidelines had awarded equal weight to HIAs' participatory and knowledge-gathering dimensions, the impetus for the Alconbury assessment was the Cambridgeshire Health Authority's practical need to respond to local concerns about the development. This requirement affected the role that community participation played in the assessment: it became impossible to award equal weight to the participatory and knowledge-gathering dimensions of HIA.
The Alconbury assessment team gathered opinions from the public on issues of concern, while public health specialists assembled expert advice and evidence from the literature in order to confirm, dismiss, or make recommendations about these issues. For example, where the community voiced concerns about "the possibility of freight being moved by air", the assessors refused to investigate the possibility because the developers indicated that, "there is no intention to include flying within this planning application." Similarly, where residents expressed concerns about "flooding from run-off from the increase in roof area due to the development", the assessors indicated that they were "guided by the Environment Agency's conclusions on the adequacy of the developer's proposals", and that "there would appear to be little or no risk to health from flooding". Where residents were concerned that "the increase in road traffic will inevitably lead to more road traffic accidents," the assessors responded: "the risk is comparatively small" and that "improved information and signage for drivers may reduce this risk further". Where the community expressed fears regarding "increased pollutants", the assessors responded: "adverse impact from the development generated air pollution on health will be minimal" (31). More examples could be given, but the point is that the requirement to respond to local concerns about the project inhibited the Alconbury assessment team's capacity to award equal weight to subjective data, and data from the literature. As a result, they dispensed with HIAs' empowering and participatory dimension.
In a similar situation, in August 2002, the University of Birmingham conducted an HIA of a proposal to replace 40% of the coal fuel for a cement kiln in Rugby with chopped, discarded vehicle tyres (32). Health impacts were identified through a documentary review of the proposal, interviews with key stakeholders, open invitation public forums, a non-systematic literature review and discussions with environmental experts. The assessors involved local residents through small group meetings, email correspondence and the "open drafting" (that is, drafts made available to all parties to comment on prior to the final version being published) of the report, but time constraints restricted wider community participation. The use of smaller groups "worked well" permitting a "sharing of understandings and viewpoints", and email correspondence gave the assessors access to information and documents they would not otherwise have had (32). But the purpose of community participation was purely to gain information that would assist in developing the impact pathways.
Participation in HIA is best effected when researchers have ample time and resources. In May 2000, Doncaster Health Authority commissioned an HIA of the planning application for the proposed Finningley airport development. The assessment was conducted over a four-month period during the wider planning approval process, as a collaboration between the Authority, and the Universities of Leicester and Sheffield. The timing was ideal: it was the first time that an HIA had been conducted at the initial planning application stage of a development (33). Adopting a modified version of the Merseyside model, the assessment team emphasized the importance of public participation for addressing local concerns, and for the more "ethical reasons of social justice". In practice, they identified vulnerable groups, and conducted extensive consultations with existing representative bodies using semi-structured interviews and focus groups. The team also consulted parish councils, lobby groups and residents of proximate areas. The Finningley assessment serves as benchmark for the successful application of participatory techniques in HIA (33). Indeed, residents identified the need for a new road to support the airport, which was later incorporated into the plan.
Discussion
Early HIAs had advocated participation in the name of empowerment, redressing the democratic deficit, and local ownership of policy. They sought community participation in the identification, analysis and prioritization of impacts. They placed much emphasis on profiling communities, identifying vulnerable groups and empowering these local people to take control of decisions affecting their lives. However, as experience with the HIA process grew, assessors found that even within "small" project-level HIAs, participation was resource-intensive, difficult to carry out, vulnerable to bias, and often unrepresentative of community views. Moreover, as time, resources and other constraints placed limits on participation, assessment teams nominated consultation mechanisms; they relied on expert advice, and did not always seek out hard-to-reach groups. Earlier HIAs had not anticipated that the community might be difficult to work with, that effective participatory partnerships might take time to build, and that the time and resources required for their construction might not be consistent with the requirements of decision-makers.
But perhaps we should not be too surprised by the inability of HIA to take forward public participation as originally envisaged. Participation may be ingrained within the culture of modern health, development and regeneration policy, but participatory research can often find, for example, that hard-to-reach groups are difficult to involve, that it is not always clear who are the legitimate representatives of communities, and that the poor and disadvantaged are not always interested in contributing to the process, and sometimes prefer that projects are managed by professionals (3437). Indeed, reports of participatory projects in other areas have noted that local people were often too busy going about their daily business to become involved in participatory programmes, and participants who had time to contribute to the process often lacked "legitimacy" or were alienated from their communities through their association with the project. Moreover, communities were either fragmented, too divided along socioeconomic, territorial or interest-based lines to be of any real use, and were unwilling put differences aside for the benefit of the project (38). Furthermore, difficulties are more than just practical: the meaning of participation is constructed in the context of specific social relations (39, 40). Communities have to "learn how to participate" and are taught by "experts", who define what is reasonable (41). Participatory strategies ignore that fact that communities are not only deprived in terms of material and social resources, but also in terms of what they "think" is possible (42, 43). Thus, participation can perpetuate existing dependency structures, and may even be used to ensure that community factions fight within themselves rather than direct concerted attacks on the "experts" (4446).
Conclusions
The constraints of time and resources on HIA mean that the likelihood of bringing about genuine ownership and empowerment is small there simply may be an insufficient amount of time to achieve access to hard-to-reach groups, to consider whether the views of all stakeholders are represented, to allow the public to initiate and direct the assessment and so forth. HIA teams generally apply participatory methods and should acknowledge that they use the community largely to generate information for tracing causal pathways (47). Used as such, participation may extend benefits to some individuals who feel they "have got something out of the process", but this cannot be seen as the main objective of an HIA. Groups conducting HIAs should acknowledge the tension between its knowledge-seeking and participatory dimensions, should be more explicit about the need to trace credible pathways between policy and health, and should be more conscious of the time required to deliver on the policy agenda, and the resources needed to build participatory partnerships with communities. 
Conflicts of interest: none declared.
References
1. Parry J, Stevens A. Prospective health impact assessment: pitfalls, problems and possible ways forward. BMJ 2001;323:1177-82.
2. European Centre for Health Policy. Gothenburg Consensus Paper. Brussels: World Health Organization Regional Office for Europe; 1999. Available from: http://www.who.dk/document/PAE/Gothenburgpaper.pdf
3. Varela Put G, den Broeder L, Penris M, Roscam Abbing EW. Experience with HIA at the national policy level in the Netherlands: a case study. Brussels: European Centre for Health Policy, World Health Organization Regional Office for Europe; 2001 (EUR/00/5019128/4).
4. Wyn Hus D. Health impact assessment in Sweden: putting ideas into practice. Urban Voice 2002;3:6-7.
5. Nilunger L, Schafer Elinder L, Pettersson B. Health impact assessment: screening of Swedish governmental enquiries. Eurohealth Winter 2002/2003;8:30-3. Available from: http://www.who.dk/Document/Obs/EuroHealth8_5.pdf
6. Berensson K. Swedish experience of HIA as a policy tool to promote equity at the local level. Human impact assessment seminar, 24 January 2002, Helsinki, Finland. Available from: http://www.stakes.fi/sva/huia/seminar/berensson.html
7. Lock K, Gabrijelcic-Blenkus M, Martuzzi M, Otorepec P, Wallace P, Dora C, et al. Health impact assessment of agriculture and food policies: lessons learnt from the Republic of Slovenia. Bulletin of the World Health Organization 2003;81:391-8. Available from: http://www.who.int/bulletin/volumes/81/6/en/lock.pdf
8. Banken R. Strategies for institutionalizing HIA. Copenhagen: World Health Organization Regional Office for Europe; 2001. Available from: http://www.who.dk/Document/E75552.pdf
9. Mindell J, Joffe M. Health impact assessment in relation to other forms of assessment. Journal of Public Health Medicine 2003;25:107-13.
10. Bullock H, Mountford J, Stanley R. Better policy making. London: Policy Studies Directorate, Cabinet Office; 2001. Available from: http://cmps.gov.uk/publications/downloads/better_policy_making.pdf
11. Parry J, Wright JSF. Community participation in health impact assessment: intuitively appealing but practically difficult. Bulletin of the World Health Organization 2003;81:388
12. Health impact assessment gateway, Health Development Agency. Available from: http://www.hiagateway.org.uk/
13. Labonté R. Health promotion and empowerment: practice frameworks. Centre for Health Promotion, University of Toronto and ParticipAction, Toronto; 1993.
14. Ison E. A resource for health impact assessment: the case studies. London: NHS Executive; 2000. Available from: http://www.londonshealth.gov.uk/allpubs.htm#Top
15. Scott-Samuel A, Birley M, Arden K. The Merseyside guidelines for health impact assessment. Liverpool: Merseyside health impact assessment Steering Group/Liverpool Public Health Observatory; 1998. Available from: http://www.liv.ac.uk/~mhb/publicat/merseygui/
16. Winters L, Scott-Samuel A. Health impact assessment of the community safety projects Huyton SRB area. Liverpool: Liverpool Public Health Observatory; 1997.
17. Fleeman N. Health impact assessment of alchemy: Southport Drug Prevention Initiative. Liverpool: Liverpool Public Health Observatory; 1999.
18. Fleeman N, Jeffrey V. A health impact assessment of a proposed sports centre in Parr. Liverpool: Liverpool Public Health Observatory; 2000.
19. Scottish Needs Assessment Programme. Piloting the process in Scotland. Edinburgh: Scottish Needs Assessment Programme; 2000. Available from: http://www.hiagateway.org.uk/Resources/completed_hia_database/completedhialist.asp
20. Scottish Needs Assessment Programme. Health impact assessment of the NEAR housing strategy. Edinburgh: Scottish Needs Assessment Programme; 2000. Available from: http://www.hiagateway.org.uk/Resources/completed_hia_database/completedhialist.asp
21. Scottish Needs Assessment Programme. Health impact assessment of the City of Edinburgh Council's urban transport strategy. Edinburgh: Scottish Needs Assessment Programme; 2000. Available from: http://www.hiagateway.org.uk/Resources/completed_hia_database/completedhialist.asp
22. Samson-Barry H. A short guide to health impact assessment. London: NHS Executive; 2000. Available from: http://www.hiagateway.org.uk/Resources/completed_hia_database/completedhialist.asp
23. London Health Commission. Health impact assessment draft air quality strategy, London: London's Health; 2001. Available from: http://www.hiagateway.org.uk/Resources/completed_hia_database/completedhialist.asp
24. London Health Commission. Health impact assessment draft economic development strategy, London: London's Health, 2001. Available from: http://www.hiagateway.org.uk/Resources/completed_hia_database/completedhialist.asp
25. London Health Commission. Health impact assessment draft energy strategy, London: London's Health; 2001. Available from: http://www.hiagateway.org.uk/Resources/completed_hia_database/completedhialist.asp
26. London Health Commission. Health impact assessment draft biodiversity strategy, London: London's Health; 2001. Available from: http://www.hiagateway.org.uk/Resources/completed_hia_database/completedhialist.asp
27. London Health Commission. Health impact assessment ambient noise strategy, London: London's Health; 2002. Available from: http://www.hiagateway.org.uk/Resources/completed_hia_database/completedhialist.asp
28. London Health Commission. Health impact assessment draft london plan, London: London's Health; 2002. Available from: http://www.hiagateway.org.uk/Resources/completed_hia_database/completedhialist.asp
29. London Health Commission. Health impact assessment The Mayor's draft strategy for municipal waste management, London: London's Health; 2001. Available from: http://www.hiagateway.org.uk/Resources/completed_hia_database/completedhialist.asp
30. London Health Commission. Health impact assessment The Mayor's draft transport strategy, London: London's Health; 2001. Available from: http://www.hiagateway.org.uk/Resources/completed_hia_database/completedhialist.asp
31. Lilley, ME. Alconbury health impact assessment report. Cambridge: Cambridgeshire Health Authority; 2000. Available from: http://www.hiagateway.org.uk/Resources/completed_hia_database/completedhialist.asp
32. Cook A, Kemm J. Health impact assessment report on proposal to substitute chopped tyres for some of the coal as fuel in cement kiln. Birmingham: Health Impact Research Unit; 2002. Available from: http://www.hiagateway.org.uk/Resources/completed_hia_database/completedhialist.asp
33. Abdel-Aziz M I, Radford J, McCabe J. Health impact assessment, Finningley Airport. Doncaster: Doncaster Health Authority; 2000. Available from: http://www.doncasterfinningleyairport.co.uk/
34. Abelson J, Forrest P, Eyles J, Smith P, Martin E, Gauvin F. Deliberations about deliberative methods: issues in the design and evaluation of public participation processes, Social Science and Medicine 2003;57:239-51.
35. Docherty I, Goodlad R, Paddision R. Civic culture, community and citizen participation in contrasting neighbourhoods. Urban Studies 2001;38:2225-50.
36. Atkinson R. Discourses of partnership and empowerment in contemporary British urban policy. Urban Studies 1999;36:59-72.
37. Boyce WF. Disadvantaged persons' participation in health promotion projects: some structural dimensions. Social Science and Medicine 2001;52:1551-64.
38. Glicken J. Getting stakeholder participation "right": a discussion of participatory processes and possible pitfalls. Environmental Science and Policy 2000;3:305-10.
39. Cattell V. Poor people, poor places, and poor health: the mediating role of social networks and social capital. Social Science and Medicine 2001;52:1501-16.
40. Jewkes R, Murcot A. Community representatives: representing the "community". Social Science and Medicine 1998;46:843-58.
41. Litva A, Coast J, Donovan J, Eyles J, Shepherd M, Tacchi J, Abelson J, Morgan K. "The public is too subjective": public involvement at different levels of health-care decision making. Social Science and Medicine 2002;54:1825-37.
42. Monnikhof RAH, Edelenbos J. Into the fog? Stakeholder input in participatory impact assessment. Impact Assessment and Policy Appraisal 2001;19:29-39.
43. Zakus JDL. Resource dependency and community participation in primary health care. Social Science and Medicine 1998;46:475-94.
44. Zakus JDL, Lysack CL. Revisiting community participation. Health Policy and Planning 1998;13:1-12.
45. Cornwall A, Jewkes R. What is participatory research? Social Science and Medicine 1995;41:1667-76.
46. Coburn D. Income inequality, social cohesion and the health status of populations: the role of neo-liberalism. Social Science and Medicine 2000;51:135-46.
47. Arnstein S. A ladder of citizen participation. Journal of the American Institute of Planners 1969;35:216-24.
(Submitted: 10 December 2003 Final revised version received: 19 April 2004 Accepted: 19 April 2004)
1 Correspondence should be sent to this author.