- Email: [email protected]
Can Health Impact Assessment fulfil the expectations it raises?
Public Health (2000) 114, 431±433 ß R.I.P.H.H. 2000 www.nature.com/ph
Leading Article Can Health Impact Assessment ful®l the expectations it raises? JR Kemm1* 1
Health Impact Assessment Unit, Welsh Combined Centres for Public Health, Llanishen, Cardiff, UK The United Kingdom and other European Governments are increasingly calling for Health Impact Assessment (HIA) of policies in order to predict how they will affect the health of populations. Approaches to HIA can be characterised as broad focus (holistic, sociological, qualitative) or tight focus (limited, epidemiological, quantitative). HIA must add value to decision making and lead to better decisions than would have otherwise been made. The quality of HIA will be judged on its utility, its predictive accuracy and its process. HIA must be closely integrated with the decision making process. HIA may be undertaken in combination with Environmental Impact Assessment (EIA) or separately. HIA does not mean that health should take primacy over other policy goals but does ensure that health is considered. Public Health (2000) 114, 431±433. Keywords: health impact assessment; health policy; environmental impact assessment
The call for HIA It has been increasingly realised that policy decisions in non-health areas (such as economics, employment, education, transport, housing and law and order) have a far greater importance in determining population health than decisions about health services.1,2 This has led many to call for Health Impact Assessment (HIA) to clarify the health consequences of policy options. In England, Our Healthier Nation3 committed the Government to applying HIA to its policies and documents from Wales,4 Scotland5 and Northern Ireland6 have made similar committments. Considerable progress has been made on applying HIA in several European countries including The Netherlands,7 Finland,8 Sweden9 and Germany.10 Articles in the European Union treaties imply a requirement for HIA. Article 129 of the Treaty of Maastricht includes the subparagraph `Health protection requirements shall form a constituent part of the communities other policies' and Article 152 of the Amsterdam Treaty includes the subparagraph `A high level of human health protection shall be ensured in the de®nition and implementation of all Community policies and activities'. The High Level Committee on Health of the European Union has recommended the development of an easy-to-use checklist of steps in policy appraisal for health impact to be used for policy development and during implementation.11 WHO urged the application of HIA to all policies or programmes likely to have an effect on health.12 With this high level of international and governmental interest the description of HIA as `an idea whose time has come'13 seems justi®ed. What is HIA The BMA Board of Science and Education de®ned HIA as `a methodology which enables the identi®cation, prediction and evaluation of the likely changes in health risk, both positive and negative (single or collective), of a policy, programme, plan or development action on a de®ned *Correspondence: JR Kemm, Health Impact Assessment Unit, Welsh Combined Centres for Public Health, Ffynnon Las, Ty Glas Avenue, Llanishen, Cardiff CF14 5EZ, UK. Accepted 25 July 2000
population. These changes may be direct and immediate or indirect and delayed.'14 More succinctly a World Health Organisation working party de®ned it as `a combination of procedures, methods and tools by which a policy, programme or project may be judged as to its potential effects on the health of a population and the distribution of those effects within the population.'15 Sometimes the term is used to describe any discussion of policy which mentions health consequences (see for example the cases cited by Frankish and colleagues16) but it would be better to restrict use of the term to studies which describe the causal pathways linking the intervention to the predicted health outcomes. Some de®nitions of HIA17 suggest it should be limited to interventions which are not primarily intended to in¯uence health. Two general streams which can be categorised as broad or tight focus can be recognised within HIA18 (Table 1). Broad focus HIA takes an holistic view of health, has disciplinary roots in sociology, rarely attempts to quantify risk and attaches great weight to popular and lay concerns. In contrast, tight focus HIA has disciplinary roots in epidemiology and toxicology, emphasises aspects of health which are measurable, or at least observable, and seeks to derive quanti®ed estimates of risk. Of course, there is a continuous spectrum and most practice of HIA falls between these two extremes. HIA tools published from British Columbia (Canada),19 Sweden20 and Merseyside (UK)21 are examples of approaches towards the broad focus end of the spectrum while those from New Zealand22 and Bielefeld (Germany)23 are examples of approaches towards the tight focus end. While HIA, as de®ned in this paper, is prospective, the process of identifying and describing the consequences of an intervention may be undertaken retrospectively or concurrently. Retrospective HIAs are needed in order to furnish the information and understanding required for future prospective HIAs. What are the expectations of HIA? Prior to HIA, policy making did not take place in an intellectual vacuum and advocates of HIA must demonstrate that their procedures lead to better decisions on policies and projects than would otherwise have been
Health Impact Assessment JR Kemm
432
Table 1
View of health Disciplinary roots Ethos Quanti®cation Types of evidence Precision
Broad focus HIA and tight focus HIA compared Broad focus
Tight focus
Holistic Sociological Democratic Vague Key informants, popular concern Low
Emphasis on de®ned and observable aspects Epidemiology, toxicology Technocratic Precise Measurement High
made. It is not suf®cient to point out that construction of a new road will create increased noise and air pollution, or that closing of a factory will produce unemployment which leads to social disruption and distress. Decision makers do not need a new discipline of impact assessors to tell them such facts. The added value that HIAs might claim to bring are the following: Identify factors (harmful or favourable) that would not otherwise have been identi®ed. Quantify the magnitude of harmful and bene®cial impacts more precisely than could otherwise have been done. Clarify the nature of trade-offs in policy making by better identi®cation and description of the elements involved. Allow better mitigation of harmful factors or enhancement of bene®cial factors. Make the decision-making process more transparent and lead to more participation by stakeholders. Change the culture so that policy makers always take health into consideration. Quality criteria for HIA The foregoing list could be used as a set of utility criteria and thus quality criteria for HIA. Another quality criterion could be predictive accuracy, the extent to which predictions on the nature and magnitude of consequences are shown to be ful®lled by subsequent monitoring. A further set of quality criteria relate to process and include items such as the thoroughness with which the scienti®c literature has been searched for information and the extent to which key stakeholders have been involved. Some descriptions of HIA methods21,22 suggest inclusion of a ®nal step described as auditing or evaluating which involves re¯ecting on the quality of the HIA process. It is beyond the scope of this paper to examine all published HIA reports and assess the extent to which they have added value to decisions. Although the authors of HIA reports regard the activity as very worthwhile a cynic could frequently question how much new knowledge the report contained or how it had contributed to the decision-making process. In many cases the real value of the HIA may have been the opportunity to build alliances and raise the pro®le of health matters in departments not usually concerned with it. HIA as part of the decision-making process Five major stages can be discerned in all descriptions of HIA procedures14,21 ± 23 although the detailed steps within them vary: Public Health
1. Screening Ð a preliminary assessment to see if the project is likely to pose any signi®cant health problems and is therefore worth subjecting to further HIA. 2. Scoping Ð broadly outlining the possible hazards and bene®ts and the questions and issues to be addressed in the assessment process. 3. Risk assessment Ð better characterisation of the nature and magnitude of hazards and bene®ts. 4. Decision making Ð choice of options (including the no action option) and considering possible modi®cations and sub options. 5. Implementation and monitoring Ð executing the decision and observing its consequences. In practice this process is not sequential but a series of several iterative steps. Thus preliminary risk assessment may identify further issues requiring a revision of the terms of reference agreed in the scoping process. Decision making after the ®rst risk assessment may lead to modi®cation of the initial proposals and the need for further risk assessment of the revised proposal. Developments during implementation and monitoring may lead to decisions being revisited with the need for further HIA and so on. HIA is not an isolated stage but rather an integral part of the complicated process of making and implementing a decision. The screening step ensures that more intensive consideration of health impacts is only given to decisions which require it and avoids delaying decisions to which HIA has nothing to add. Integrating HIA and EIA HIA has joint conceptual origins in Environmental Impact Assessment (EIA) and in Healthy Public Policy Analysis.24 Many discussions14,22,25,26 assume that HIA will form part of an EIA. In theory, integrating EIA and HIA into a single Environmental Health Impact Assessment has the advantage of saving effort, time and money for all concerned. However there is the risk that, when the two are combined, one element may be overemphasised at the expense of the other. While EIA directs attention to ecosystems and biodiversity it may fail to address impacts on human communities and culture.27,28 Undertaking HIA separately from any other impact assessment ensures that health remains the prime focus. The case for combining HIA with EIA is strongest in the context of development projects and weakest in the context of policy decisions in areas such as ®scal, education and law and order, which have many more implications for health than for the environment. Policy implications of using HIA Policy makers and health advocates may have rather different expectations of HIA. Policy making is a complex
Health Impact Assessment JR Kemm
process, more often incremental than linear and in¯uenced by many interest groups.29 Health is only one of many issues that concern policy makers and they may feel that HIA gives undue emphasis to one aspect of policy. On the other hand, health advocates would argue that HIA merely identi®es how other policy goals impact on health. When policy makers have to balance health against other policy goals, HIA allows them to do so in full knowledge of the consequences. HIA does not mean that health considerations will always take primacy over others but it does mean that the potential impact on health will always be considered. Conclusion This is certainly a time of opportunity for HIA. Policy makers are interested in the health consequences of policy and seeking methods to predict these, which are simple to use. The message that HIA can provide such a method has largely been accepted. This danger for HIA is that, unless it makes good its promise and demonstrates its utility in the near future, it will be discarded as yet another fashion which raised expectations but proved to lack substance. References 1 Benzeval M, Judge K, Whitehead M. Tackling Inequalities in Health: An Agenda for Action. King's Fund: London, 1995. 2 Acheson D. Independent Enquiry into Inequalities in Health: Report. The Stationery Of®ce: London, 1998. 3 Department of Health. Saving lives: Our Healthier Nation. The Stationery Of®ce: London, 1999. 4 Welsh Of®ce. Better Health; Better Wales. The Stationery Of®ce: Cardiff, 1998. 5 Scottish Of®ce. Working Together for a Healthier Scotland. The Stationery Of®ce: Edinburgh, 1998. 6 Secretary of State for Northern Ireland. Well Into 2000. Department of Health and Social Services: Belfast, 1997. 7 Putters K. Health Impact Screening. Ministry of Health, Welfare and Sport: Rijswijk, The Netherlands, 1997. 8 Koivusalo M, Santalahti P, Ollila E. Healthy public policies in Finland. Eurohealth 1998; 4: 32 ± 34. 9 Berensson K. Focusing on health in the political arena. Eurohealth 1998; 4: 34 ± 36. 10 Fehr R. Environmental health impact assessment-evaluation of a 10 step model. Epidemiology 1999; 10: 618 ± 625. 11 Hubel M. Evaluating the health impact of policies: a challenge. Eurohealth 1998; 4: 27 ± 29.
12 WHO Health 21; An Introduction to the Health for All Policy Framework for the WHO European Region. WHO: Copenhagen, 1999. 13 Scott Samuel A. Health Impact Assessment; an idea whose time has come. Br Med J 1996; 313: 183 ± 184. 14 BMA Board of Science and Education. Health and Environmental Impact Assessment. Earthscan: London, 1998. 15 WHO Regional Of®ce for Europe. Gothenburg Consensus Paper: Health Impact Assessment: Main Concepts and Suggested Approach. European Centre for Health Policy: Brussels, 1999. 16 Frankish CJ, Green LW, Ratner PA, Chomik T, Larsen C. Health Impact Assessment as a tool for population health promotion and public policy, 1996. www.hc-sc.gc.ca= hppb=healthpromotiondevelopment=pube=impact=impact.htm. 17 Boothroyd P. Policy Assessment In: Vanclay F, Bronstein DA (eds) Environmental and Social Impact Assessment. Chichester: John Wiley & Sons, 1995. 18 National Assembly for Wales. Developing Health Impact Assessment in Wales. Better Health; Better Wales. National Assembly for Wales: Cardiff, 1999. 19 Population Health Resource Branch. Health Impact Assessment Toolkit. Ministry of Health, British Columbia: Vancouver, 1994. 20 Landstings Forbundet and Svenska Kommunforbundet. Focusing on Health Stockholm, 1998. www.lf.se=hkb. 21 Scott-Samuel A, Birley M, Ardern K. The Merseyside Guidelines for Health Impact Assessment. Liverpool: Merseyside Health Impact Assessment Steering Group, 1998. 22 Public Health Commission. A Guide to Health Impact Assessment. Wellington, New Zealand: Public Health Commission, 1995. 23 Pastides H, Carvalan C. Methods for Health Impact Assessment in Environmental and Occupational Health. Geneva: WHO, 1998. 24 Milio N. Promoting Health Through Public Policy. Canadian Public Health Association: Ottawa, 1986. 25 Ewan C, Young A, Bryant E, Calvert D. National Framework for Environment and Health Impact Assessment. Canberra: National Health and Medical Research Council Australia, 1994. 26 Turnbull RGH. Environmental and Health Impact Assessment of Development Projects: A Handbook for Practitioners. Elsevier Applied Science: Barking, 1992. 27 Joffe M, Sutcliffe J. Developing policies for a healthy environment. Health Prom Int 1997; 12: 169 ± 173. 28 Arquiaga MC, Canter LW, Nelson DI. Integration of health impact considerations in environmental impact studies. Impact Assess 1994; 12: 175 ± 197. 29 Walt G. Health Policy: An Introduction to Process and Power. Zed books: London, 1994.
433
Public Health
Leading Article Can Health Impact Assessment ful®l the expectations it raises? JR Kemm1* 1
Health Impact Assessment Unit, Welsh Combined Centres for Public Health, Llanishen, Cardiff, UK The United Kingdom and other European Governments are increasingly calling for Health Impact Assessment (HIA) of policies in order to predict how they will affect the health of populations. Approaches to HIA can be characterised as broad focus (holistic, sociological, qualitative) or tight focus (limited, epidemiological, quantitative). HIA must add value to decision making and lead to better decisions than would have otherwise been made. The quality of HIA will be judged on its utility, its predictive accuracy and its process. HIA must be closely integrated with the decision making process. HIA may be undertaken in combination with Environmental Impact Assessment (EIA) or separately. HIA does not mean that health should take primacy over other policy goals but does ensure that health is considered. Public Health (2000) 114, 431±433. Keywords: health impact assessment; health policy; environmental impact assessment
The call for HIA It has been increasingly realised that policy decisions in non-health areas (such as economics, employment, education, transport, housing and law and order) have a far greater importance in determining population health than decisions about health services.1,2 This has led many to call for Health Impact Assessment (HIA) to clarify the health consequences of policy options. In England, Our Healthier Nation3 committed the Government to applying HIA to its policies and documents from Wales,4 Scotland5 and Northern Ireland6 have made similar committments. Considerable progress has been made on applying HIA in several European countries including The Netherlands,7 Finland,8 Sweden9 and Germany.10 Articles in the European Union treaties imply a requirement for HIA. Article 129 of the Treaty of Maastricht includes the subparagraph `Health protection requirements shall form a constituent part of the communities other policies' and Article 152 of the Amsterdam Treaty includes the subparagraph `A high level of human health protection shall be ensured in the de®nition and implementation of all Community policies and activities'. The High Level Committee on Health of the European Union has recommended the development of an easy-to-use checklist of steps in policy appraisal for health impact to be used for policy development and during implementation.11 WHO urged the application of HIA to all policies or programmes likely to have an effect on health.12 With this high level of international and governmental interest the description of HIA as `an idea whose time has come'13 seems justi®ed. What is HIA The BMA Board of Science and Education de®ned HIA as `a methodology which enables the identi®cation, prediction and evaluation of the likely changes in health risk, both positive and negative (single or collective), of a policy, programme, plan or development action on a de®ned *Correspondence: JR Kemm, Health Impact Assessment Unit, Welsh Combined Centres for Public Health, Ffynnon Las, Ty Glas Avenue, Llanishen, Cardiff CF14 5EZ, UK. Accepted 25 July 2000
population. These changes may be direct and immediate or indirect and delayed.'14 More succinctly a World Health Organisation working party de®ned it as `a combination of procedures, methods and tools by which a policy, programme or project may be judged as to its potential effects on the health of a population and the distribution of those effects within the population.'15 Sometimes the term is used to describe any discussion of policy which mentions health consequences (see for example the cases cited by Frankish and colleagues16) but it would be better to restrict use of the term to studies which describe the causal pathways linking the intervention to the predicted health outcomes. Some de®nitions of HIA17 suggest it should be limited to interventions which are not primarily intended to in¯uence health. Two general streams which can be categorised as broad or tight focus can be recognised within HIA18 (Table 1). Broad focus HIA takes an holistic view of health, has disciplinary roots in sociology, rarely attempts to quantify risk and attaches great weight to popular and lay concerns. In contrast, tight focus HIA has disciplinary roots in epidemiology and toxicology, emphasises aspects of health which are measurable, or at least observable, and seeks to derive quanti®ed estimates of risk. Of course, there is a continuous spectrum and most practice of HIA falls between these two extremes. HIA tools published from British Columbia (Canada),19 Sweden20 and Merseyside (UK)21 are examples of approaches towards the broad focus end of the spectrum while those from New Zealand22 and Bielefeld (Germany)23 are examples of approaches towards the tight focus end. While HIA, as de®ned in this paper, is prospective, the process of identifying and describing the consequences of an intervention may be undertaken retrospectively or concurrently. Retrospective HIAs are needed in order to furnish the information and understanding required for future prospective HIAs. What are the expectations of HIA? Prior to HIA, policy making did not take place in an intellectual vacuum and advocates of HIA must demonstrate that their procedures lead to better decisions on policies and projects than would otherwise have been
Health Impact Assessment JR Kemm
432
Table 1
View of health Disciplinary roots Ethos Quanti®cation Types of evidence Precision
Broad focus HIA and tight focus HIA compared Broad focus
Tight focus
Holistic Sociological Democratic Vague Key informants, popular concern Low
Emphasis on de®ned and observable aspects Epidemiology, toxicology Technocratic Precise Measurement High
made. It is not suf®cient to point out that construction of a new road will create increased noise and air pollution, or that closing of a factory will produce unemployment which leads to social disruption and distress. Decision makers do not need a new discipline of impact assessors to tell them such facts. The added value that HIAs might claim to bring are the following: Identify factors (harmful or favourable) that would not otherwise have been identi®ed. Quantify the magnitude of harmful and bene®cial impacts more precisely than could otherwise have been done. Clarify the nature of trade-offs in policy making by better identi®cation and description of the elements involved. Allow better mitigation of harmful factors or enhancement of bene®cial factors. Make the decision-making process more transparent and lead to more participation by stakeholders. Change the culture so that policy makers always take health into consideration. Quality criteria for HIA The foregoing list could be used as a set of utility criteria and thus quality criteria for HIA. Another quality criterion could be predictive accuracy, the extent to which predictions on the nature and magnitude of consequences are shown to be ful®lled by subsequent monitoring. A further set of quality criteria relate to process and include items such as the thoroughness with which the scienti®c literature has been searched for information and the extent to which key stakeholders have been involved. Some descriptions of HIA methods21,22 suggest inclusion of a ®nal step described as auditing or evaluating which involves re¯ecting on the quality of the HIA process. It is beyond the scope of this paper to examine all published HIA reports and assess the extent to which they have added value to decisions. Although the authors of HIA reports regard the activity as very worthwhile a cynic could frequently question how much new knowledge the report contained or how it had contributed to the decision-making process. In many cases the real value of the HIA may have been the opportunity to build alliances and raise the pro®le of health matters in departments not usually concerned with it. HIA as part of the decision-making process Five major stages can be discerned in all descriptions of HIA procedures14,21 ± 23 although the detailed steps within them vary: Public Health
1. Screening Ð a preliminary assessment to see if the project is likely to pose any signi®cant health problems and is therefore worth subjecting to further HIA. 2. Scoping Ð broadly outlining the possible hazards and bene®ts and the questions and issues to be addressed in the assessment process. 3. Risk assessment Ð better characterisation of the nature and magnitude of hazards and bene®ts. 4. Decision making Ð choice of options (including the no action option) and considering possible modi®cations and sub options. 5. Implementation and monitoring Ð executing the decision and observing its consequences. In practice this process is not sequential but a series of several iterative steps. Thus preliminary risk assessment may identify further issues requiring a revision of the terms of reference agreed in the scoping process. Decision making after the ®rst risk assessment may lead to modi®cation of the initial proposals and the need for further risk assessment of the revised proposal. Developments during implementation and monitoring may lead to decisions being revisited with the need for further HIA and so on. HIA is not an isolated stage but rather an integral part of the complicated process of making and implementing a decision. The screening step ensures that more intensive consideration of health impacts is only given to decisions which require it and avoids delaying decisions to which HIA has nothing to add. Integrating HIA and EIA HIA has joint conceptual origins in Environmental Impact Assessment (EIA) and in Healthy Public Policy Analysis.24 Many discussions14,22,25,26 assume that HIA will form part of an EIA. In theory, integrating EIA and HIA into a single Environmental Health Impact Assessment has the advantage of saving effort, time and money for all concerned. However there is the risk that, when the two are combined, one element may be overemphasised at the expense of the other. While EIA directs attention to ecosystems and biodiversity it may fail to address impacts on human communities and culture.27,28 Undertaking HIA separately from any other impact assessment ensures that health remains the prime focus. The case for combining HIA with EIA is strongest in the context of development projects and weakest in the context of policy decisions in areas such as ®scal, education and law and order, which have many more implications for health than for the environment. Policy implications of using HIA Policy makers and health advocates may have rather different expectations of HIA. Policy making is a complex
Health Impact Assessment JR Kemm
process, more often incremental than linear and in¯uenced by many interest groups.29 Health is only one of many issues that concern policy makers and they may feel that HIA gives undue emphasis to one aspect of policy. On the other hand, health advocates would argue that HIA merely identi®es how other policy goals impact on health. When policy makers have to balance health against other policy goals, HIA allows them to do so in full knowledge of the consequences. HIA does not mean that health considerations will always take primacy over others but it does mean that the potential impact on health will always be considered. Conclusion This is certainly a time of opportunity for HIA. Policy makers are interested in the health consequences of policy and seeking methods to predict these, which are simple to use. The message that HIA can provide such a method has largely been accepted. This danger for HIA is that, unless it makes good its promise and demonstrates its utility in the near future, it will be discarded as yet another fashion which raised expectations but proved to lack substance. References 1 Benzeval M, Judge K, Whitehead M. Tackling Inequalities in Health: An Agenda for Action. King's Fund: London, 1995. 2 Acheson D. Independent Enquiry into Inequalities in Health: Report. The Stationery Of®ce: London, 1998. 3 Department of Health. Saving lives: Our Healthier Nation. The Stationery Of®ce: London, 1999. 4 Welsh Of®ce. Better Health; Better Wales. The Stationery Of®ce: Cardiff, 1998. 5 Scottish Of®ce. Working Together for a Healthier Scotland. The Stationery Of®ce: Edinburgh, 1998. 6 Secretary of State for Northern Ireland. Well Into 2000. Department of Health and Social Services: Belfast, 1997. 7 Putters K. Health Impact Screening. Ministry of Health, Welfare and Sport: Rijswijk, The Netherlands, 1997. 8 Koivusalo M, Santalahti P, Ollila E. Healthy public policies in Finland. Eurohealth 1998; 4: 32 ± 34. 9 Berensson K. Focusing on health in the political arena. Eurohealth 1998; 4: 34 ± 36. 10 Fehr R. Environmental health impact assessment-evaluation of a 10 step model. Epidemiology 1999; 10: 618 ± 625. 11 Hubel M. Evaluating the health impact of policies: a challenge. Eurohealth 1998; 4: 27 ± 29.
12 WHO Health 21; An Introduction to the Health for All Policy Framework for the WHO European Region. WHO: Copenhagen, 1999. 13 Scott Samuel A. Health Impact Assessment; an idea whose time has come. Br Med J 1996; 313: 183 ± 184. 14 BMA Board of Science and Education. Health and Environmental Impact Assessment. Earthscan: London, 1998. 15 WHO Regional Of®ce for Europe. Gothenburg Consensus Paper: Health Impact Assessment: Main Concepts and Suggested Approach. European Centre for Health Policy: Brussels, 1999. 16 Frankish CJ, Green LW, Ratner PA, Chomik T, Larsen C. Health Impact Assessment as a tool for population health promotion and public policy, 1996. www.hc-sc.gc.ca= hppb=healthpromotiondevelopment=pube=impact=impact.htm. 17 Boothroyd P. Policy Assessment In: Vanclay F, Bronstein DA (eds) Environmental and Social Impact Assessment. Chichester: John Wiley & Sons, 1995. 18 National Assembly for Wales. Developing Health Impact Assessment in Wales. Better Health; Better Wales. National Assembly for Wales: Cardiff, 1999. 19 Population Health Resource Branch. Health Impact Assessment Toolkit. Ministry of Health, British Columbia: Vancouver, 1994. 20 Landstings Forbundet and Svenska Kommunforbundet. Focusing on Health Stockholm, 1998. www.lf.se=hkb. 21 Scott-Samuel A, Birley M, Ardern K. The Merseyside Guidelines for Health Impact Assessment. Liverpool: Merseyside Health Impact Assessment Steering Group, 1998. 22 Public Health Commission. A Guide to Health Impact Assessment. Wellington, New Zealand: Public Health Commission, 1995. 23 Pastides H, Carvalan C. Methods for Health Impact Assessment in Environmental and Occupational Health. Geneva: WHO, 1998. 24 Milio N. Promoting Health Through Public Policy. Canadian Public Health Association: Ottawa, 1986. 25 Ewan C, Young A, Bryant E, Calvert D. National Framework for Environment and Health Impact Assessment. Canberra: National Health and Medical Research Council Australia, 1994. 26 Turnbull RGH. Environmental and Health Impact Assessment of Development Projects: A Handbook for Practitioners. Elsevier Applied Science: Barking, 1992. 27 Joffe M, Sutcliffe J. Developing policies for a healthy environment. Health Prom Int 1997; 12: 169 ± 173. 28 Arquiaga MC, Canter LW, Nelson DI. Integration of health impact considerations in environmental impact studies. Impact Assess 1994; 12: 175 ± 197. 29 Walt G. Health Policy: An Introduction to Process and Power. Zed books: London, 1994.
433
Public Health