Including health in environmental impact assessments: is an institutional approach useful for practice?

Abstract

Internationally the inclusion of health within environmental impact assessment (EIA) has been shown to be limited. While health-focused research has focused on the technical provision of health information, policy analysis theory may enable description and explanation of the institutional conditions surrounding health inclusion in EIA. However, whether this framework is considered useful by practitioners has yet to be tested. To investigate this, data were collected via a workshop (n = 22) and the results were analysed using ‘Institutionalist’ units of analysis (ideas, actors, organisations and institutions). These results were then emailed to participants who were asked to undertake a follow-up survey about the analysis and approach (n = 9). The workshop results suggested various influences on how and why health is considered or not in EIAs. Overall the survey respondents agreed that the approach was conceptually and practically useful but that the framework alone is insufficient and further work is needed to convince potential users of the value of health in EIA. The findings support the need for more detailed research.

Keywords::

• environmental impact assessment
• human health
• institutionalism
• policy analysis
• research
• realism

Introduction

Environmental impact assessments (EIAs) of large-scale development projects are undertaken in over 190 countries, with the aim of ensuring environmental sustainability is considered in decision-making and project approvals (Morgan 1998; Glasson et al. 2013). EIA has long been recognised as an important vehicle for considering the human health impacts of development as part of sustainable development (Morgan 2003; Ahmad 2004; Glasson et al. 2013). Potentially EIAs are a tool to manage risk and capture opportunities for sustainable natural and human systems. Empirical investigations of EIAs until the mid-2000s, however, consistently showed that health is not fully considered (Harris et al. 2009), supporting the conclusion that EIA represents a ‘missed opportunity’ for health (Bhatia & Wernham 2008).

The past decade has seen a number of initiatives which support better consideration of health in EIA. There are now international private sector requirements to better consider health in impact assessment (International Council on Mining and Minerals 2010; International Finance Corporation 2012), regulatory developments to reinforce considering health in EIA such as in the recent EU directive, as well local initiatives to the increased use of health impact assessments (HIAs) mainly in the developed world (Winkler et al. 2013) such the USA (Bhatia & Wernham 2008), Canada (Kwiatkowski & Ooi 2003), Australia (Harris & Spickett 2011; Elliot & Thomas 2014) and New Zealand (Morgan 2011). These activities have, potentially, provided the institutional conditions for health considerations to be standardised in EIA practice and project approvals. Yet it is not clear whether and how previous practices are actually changing.

The vast majority of research investigating the inclusion of human health in EIAs has focused on the technical quality and type of health considerations in EIA documentation (environmental impact statements; EISs), largely in Western contexts. The word ‘health’ is rarely included (Arquiaga et al. 1994; Steinemann 2000). Environmental health concerns associated with changes to the bio-physical environment such as air quality, noise and soil contamination are considered (British Medical Association 1998; Alenius 2001); however, the link between these changes and health outcomes is not made. Health as a broader issue – associated with socio-economic impacts or ‘determinants of health’ such as lifestyle, personal circumstances, social influences, availability and access to services and wider economic conditions (International Council on Mining and Minerals 2010) – is included rarely (Chadwick 2002; Cole et al. 2004). Furthermore, EIAs are unlikely to use health data, develop the causal pathway between an environmental trigger and health outcomes, or assess the potential for projects to influence health inequalities (Arquiaga et al. 1994; British Medical Association 1998; Steinemann 2000). Notably in the private industry arena where health in EIA may be undertaken directly driven by international lending requirements, very little is known about the quality of health inclusion or the facilitators for this (Winkler et al. 2013).

This content-focused research has, notably, not systematically investigated the broader institutional conditions and mechanisms surrounding this technical inclusion of health in EISs. They do exist, however, with findings across these studies and articles (Arquiaga et al. 1994; Steinemann 2000; Bhatia & Wernham 2008; Morgan 2008), suggesting definitional confusion of what is meant by ‘health’ and determinants of health, professional and institutional bias, lack of capacity, differing professional expectations, poor cross-disciplinary engagement, data limitations, limited inclusion of health in legislative frameworks and lack of standards, guidance on good practice, tools and training.

Systematically investigating the complex conditions surrounding EIAs and approvals decisions is essential to progress knowledge about how best to include human health. EIAs are nested within complex political decisions and processes (Cashmore et al. 2007) often driven by the sheer size and significance of the projects assessed. It is therefore unsurprising that the content of EIAs is tied to a range of broader legislative requirements as well as policy and political decision-making and additional approvals processes with stakeholders with conflicting agenda (Cashmore et al. 2008).

Recent realist research has suggested the usefulness in unpacking complex conditions surrounding the inclusion of HIA in public policy processes and systems using ‘institutionalist’ policy analysis theory (Howlett et al. 2009). Realist research works by providing a framework for analysis which focuses attention on systems at different levels of analysis as well as using existing theory to describe and explain empirical findings (Pawson & Tilley 1997; Sayer 2000). Notably EIA research has had a long history of using institutionalist approaches to policy analysis (Taylor 1984; Bartlett 1989; Bartlett & Kurian 1999; Morgan 2008) as well as realist research methodology (Cashmore 2004; Pischke & Cashmore 2006; Cashmore et al. 2007). However, this type of research has not been conducted in health-oriented EIA research. We have, therefore, recently developed a comprehensive realist approach to investigating how, why and to what extent human health is included in EIA which is theoretically based on institutional policy analysis. The findings reported here inform broader research design. Importantly our focus is on health within the EIA process which may or may not include HIAs, and not HIAs as conducted separate to EIAs or on policies (see Harris et al. (2009) and Harris-Roxas et al. (2012) for an historical overview of HIA).

A major tenet of realist research designs is for research to be connected to and useful for existing practice (Pawson & Tilley 1997; Sayer 2000). However, the use of theory in realist research may be perceived by some as divorced from the pragmatic realities of impact assessment practice (Haigh et al. 2012). The focus of our research reported here is therefore whether a conceptually broad policy analysis framework can be useful for understanding the practice of including human health in project-level EIA. At the same time we intended the findings to provide some initial insight into the broader institutional conditions surrounding the inclusion of human health in EIA.

Methods

Our purposes and (qualitative) approach are in line with central elements of realist research methodology which empirically unpacks phenomena as well as compares theories for their ability to describe and explain those phenomena (Bhaskar 1978; Sayer 1992). Ethical approval was granted by UNSW Human Research Ethics Advisory Panel I (social/health research – approval number 9_14_025).

Research team and reflexivity

Both researchers are experienced HIA academics and professionals with a longstanding interest in the inclusion of health in EIA (Harris et al. 2009) as well as realist research (Haigh et al. 2012; Harris et al. 2012; Harris, Haigh et al. 2014; Harris, Sainsbury et al. 2014). PH conceived the study design and analysed the workshop data, and both authors designed the survey and analysed the survey data.

Data collection

Data were collected via a workshop and a follow-up survey. In addition, a draft version of this paper was circulated to all participants for accuracy check and comment. There were no comments. The workshop participants (n = 22) self-selected to attend a workshop titled ‘Sharing global experiences with including health in major project EIA’, held at the International Association for Impact Assessment annual conference in Vina Del Mar, Chile, in April 2014. An abstract for the conference was provided in the conference programme and a flyer was handed around prior to the session at the conference.

The workshop took an hour and a half and was designed around the following questions and facilitated by PH. First all participants were introduced to the purpose of the workshop and asked as a large group to define what health is in major project EIAs. Then four small groups were formed, facilitated by a member of the IAIA Health Section who also made notes and wrote up key points on flipcharts, and asked to discuss:

• The value of health for your work: pros and cons (15 min with 10 min feedback).

• Has your organisation included health: how, why, to what extent? (15 min with 10 min feedback)

• What institutional support is required to make health an important consideration in project approvals? (10 min with 5 min feedback)

These three questions correspond to the core units of analysis in institutional policy analysis (Howlett et al. 2009; Nykvist & Nilsson 2009; Marsh 2010) which are ideas (the content of policy, in this case the value of health as an idea), actors (the people and their organisations involved in developing policy) and institutions (the structures, rules and regulations which influence how and why policy is made). Notably we focused on ‘organisations’ as suggested by neo-institutional policy theory (Scott 2005), as the point within institutions where actors (and their ideas and interests) come together.

The 22 workshop participants came from varying countries and regions across the world, including Europe, the USA, Canada, Latin America and South East Asia. Experience ranged from 2 to 20 years. Each identified as professionally involved in EIA either as HIA consultants (n = 6), academics (n = 4), EIA consultants who were interested in health (n = 3), as regulators for EIA (n = 2) or HIA (n = 1) or environmental planning (n = 1), as industry proponents for transnational (n = 2) and national corporations (n = 2), and a multi-national non-government agency (n = 1).

The notes from the workshop were written and provided to all participants who were asked to check the notes for accuracy and to make any additional points.

Participants were then emailed a table containing a summary of the workshop, analysed against ‘Institutionalist’ Public Policy Framework (Howlett et al. 2009) and a link to an online survey developed by PH and FH (see Box 1). There were survey respondents (see Table 1).

Table 1 Survey respondent characteristics.

Sector	Years of experience	Academic background	Region
Government	More than 10 years	Zoology, botany and ecology	Asia
Private (consultant)	More than 10 years	Health assessment, risk assessment, toxicology, EIA	North America
Private (research, consultant)	6–10 years	Environmental scientist epidemiologist	European Union
International public organisation	3–5 years	Public health	South America
Extractive industry	Less than 2 years	Sociology and nursing	European Union
Extractive industry	6–10 years	Management engineering, international healthcare management, economics and policy	European Union
Consultant/academic	6–10 years	Mining engineering/population health	North America
Academic	More than 10 years	PhD, science	European Union
Private (consultant)	More than 10 years	Medicine, epidemiology, public policy	European Union

Results

Workshop

Crucially the workshop suggested that the dimensions underpinning the questions provide a useful framework for thinking about the range of influences and broader conditions involved in including health in EIA. These clearly need to be further developed but the findings do suggest that issues can be categorised in the same manner as the units of analysis in institutional policy analysis: ideas, actors, organisations and (institutional) structures (see Figure 1). This figure was developed by tabulating each of the issues raised in the workshop against each dimension. Each, the workshop suggested, can form an important line of future inquiry and an in-depth analysis for future research. Also crucially just as in policy analysis, these dimensions often overlap, for example meeting company standards straddles both actors (because this facilitates their interpretation of what health is) and institutional rules requirements (for organisations to include health). An interesting finding is that even without an explicit question about actors, participants did discuss issues about actors which can be separated analytically both from their organisations and also from wider institutional rules and requirements.

Figure 1 Institutionalist framework – initial conceptualisation of the conditions influencing inclusion of health in EIA.

Following one theme along this line of analysis serves as an example where the complexity of health became a critical dimension. Health as an idea was seen as (sometimes) valuable because it provides something tangible and measurable concerning the potential impact of projects. However, the cons were that health as an idea can struggle for conceptual clarity because it can overlap with environmental, social and economic concerns which are also considered in EIA. Moving to actors, the broad idea of health became problematic when decision-makers in government had to look at the range of issues in an EIA, although interestingly companies (particularly management) were seen as being more open to this than governments. Turning to organisations, one interesting finding was that the broad view of health can lead to different stakeholders in EIA focusing on different aspects of health that do not necessarily align, even within an organisation or company. Also organisationally, the roles and responsibilities of stakeholder agencies were questioned, with lack of clarity about whether agencies were seen as regulators, advisors or approving agencies. Institutionally the inclusion of health is facilitated by internal regulations in companies but also external laws and regulations. Linking health to policy prerequisites such as economic development or sustainability was also seen as providing institutional support. However, this also required multi-sectoral collaboration, spearheaded through political support and leadership, and crucially ‘trust building’, between the Ministries of Environment and Health.

Survey

Our analysis of the workshop suggested that the institutional policy analysis framework is useful. The survey responses provided additional support for this as well as providing more depth to our understanding of the factors influencing the inclusion of health in EIA. Participants recognised that there is currently a problem how health is (or is not) included in EIAs of major projects. Almost all (one respondent did not answer this question) currently either explicitly or implicitly considered factors that could influence whether or not health will be considered within major projects EIA.

Overall most respondents (n = 8, with one respondent not answering this question) agreed that it was useful to unpack the broader conditions influencing the inclusion of health in decision-making, and none considered it ‘not at all useful’. Everyone who responded found this to be conceptually useful and six of eight respondents also found it practically useful. As one participant stated,

understanding the varying categories and potential groups involved is valuable when looking at health implications for industrial projects. (International public organisation)

Respondents provided some qualitative examples of how the framework could have practical application. A respondent who works in Project HIAs commented how the framework could help with the integration of HIA with EIA and social impact assessment. Another commented on its use in understanding the current resistance to the inclusion of health. However, it was also noted that the framework alone is not enough and that further work is needed to convince potential users of the value of health in EIA.

it could be very useful to standardize the integration with the environmental and social impact assessment. (Consultant)

To better understand the resistance that some stakeholders have all the time and many stakeholders have sometimes to including health issues in project and policy development. (Consultant)

At this point, the[se] outputs will not help raise the profile of health for inclusion. We missed the business case. Given HIA's are primarily voluntary … we need to make this case for major projects. (Consultant)

The comments also revealed that there are differing perspectives on the role of different actors and how those roles are currently being fulfilled. For example, there were contrasting views on the role consultants were seen to have in influencing the inclusion of health in EIA. One government respondent talked about consultants having a role alongside proponents and regulators (environmental protection department):

There is no structured framework to guide decision making for requiring health impact to be addressed in project EIA: very much left to the practitioners (EPD, proponents, consultants) to initiate. (Government)

Whereas another consultant described being powerless:

As a HIA practitioner I am always including ‘health’. But it is not me who is taking the decision on whether health will be considered within major project EIA … (Consultant)

Another consultant talked about their responsibility to demonstrate how they have taken stakeholder input (presumably including health related) into account

Yes, we very clearly state how the regulatory regime and stakeholder inputs are considered. (Consultant)

A policy advisor in an international organisation focused on public sector actors responsible for IA to expand their interests beyond environmental considerations:

…expanding the role of [E]IA public sector professionals to include other sectors is an area that requires further consideration. (Policy adviser – international organisation)

Participants also identified activities required to progress this work. This included more work understanding the factors, developing sector-specific frameworks and linking theory to everyday practice (see Table 2)

Table 2 Suggestions for further research needed.

Further research	Example quote
More work understanding factors	‘A useful start … but really I think we need to unpack the categories a bit more – there are many aspects we could dig into with respect to the different actors’. (Consultant/academic)‘break them down into sub-categories’. (Consultant/academic)
Developing sector-specific frameworks	‘At the national level determining frameworks for dealing with individual sectors as a means to move integrate health would be valuable’. (International public organisation)‘understanding the varying categories and potential groups involved is valuable when looking at health implications for industrial projects’. (International public organisation)
Linking theory to everyday practice	‘useful but oftentimes these are more conceptual in nature and so putting them in practice is challenging’. (International public organisation)

Discussion

Overall our results suggest that the institutionalist policy framework is a potentially useful tool for further conceptualising and unpacking conditions influencing the inclusion of health in EIA. In doing so we identified a range of factors conceptualising how factors interact and play out at the levels of ideas, actors, organisations and institutional levels. This has added depth to our (and, the results suggest, participants') understanding but also enabled us to begin unpacking the potential mechanisms and contextual factors that can support (or not) in inclusion of health in major project EIA. The responses to our survey suggest that the framework has practical application; participants found it a useful conceptual tool.

Our findings substantiates prior research that, concerning health in EIA practice, health as an idea and technical issue in EIA is conceptually varied and requiring better definition to be included meaningfully for proponents and regulators and approvers (Harris et al. 2009; Harris & Spickett 2011). The medical model, focusing on diseases, apparently still dominates thinking, and there remains work for a broader view of health to be accepted as useful in EIA (Kemm 2000). We were also able to demonstrate that inclusion of health in EIA is largely subject to broader conditions to do with actors and their values, the organisations they work in, and regulations and cross-agency collaboration surrounding EIA. The core business of EIA systems is focused on the physical environment, followed by social and, increasingly, economic development (Bhatia & Wernham 2008; Harris et al. 2009). The problem of including health in EIA practice goes beyond whether health is conceptually positioned as an idea concerned either with ‘inside the [project] fence’ impacts or ‘outside the fence’ impacts (Krieger et al. 2010). Rather a more nuanced position is required concerning project approval decisions which are influenced in turn by requirements, stakeholder (including proponents, regulators and communities) decisions and values, consultants reports and broader regulations (Cashmore et al. 2007, 2008; Bhatia & Wernham 2008). Health is, thus, not only a technical consideration in EISs. There is a need for a shift in systems and conditions too. Furthermore, our findings suggest that this is a global issue, notably where lower and middle income countries can struggle with institutional requirements and capacity when compared to developed countries (Abah 2012; Winkler et al. 2013).

Concerning theory, policy analysis literature has long differentiated between research for policy and research of policy (Milio 1987; Kemm 2001; Howlett et al. 2009). Considering how health is technically included – as an idea – in EISs is an example of the former and there is no doubt this is important. However, our findings suggest that research of how and why health is included in EIA – as an idea, by people, in organisations and within institutions – is also worthwhile pursuing. The use of institutional analysis approach suggests that using this within the broader realist design will provide valuable insight for this ongoing problem of how, why and the extent to which health is included in EIA. Taking this line has the possibility of connecting health with existing EIA research both methodologically and conceptually. This adds theoretical depth for the ‘health in impact assessment’ field (Morgan 2012).

There are several limitations. We have reported here, on a small-scale, currently unfunded investigatory piece of research. Our sample size was small and the participants were self-selected. Participants may have had a pre existing interest in health in EIA and in that sense, although our focus at this stage is understanding practice rather than generalisibility, may not be representative of the broader EIA community. However, they did come from a broad range of areas (e.g. consultants, government and so on) and were highly experienced, bringing in-depth knowledge of the field and how it is developing. Our response rate to the survey was, however, only 41%. This and some of the survey findings suggests that there is still some way to go to link academic research with the work of practitioners and regulators. Our findings provide initial links between (empirical experience) and theoretically based lines of description and analysis. Clearly the results are initial only and not detailed enough; to build on this and be able to ultimately develop solutions to the problem, we need much more and deeper levels of data collection and analysis.

Conclusion

Health is, possibly, the least institutionalised dimension of impact assessment practice (Harris-Roxas et al. 2012). Turning attention to how health fits with the rules, players, organisations and ideas which are inherent in EIA systems may provide the conceptual platform for institutional progress. This has the potential to benefit not only impact assessment as a process but also the health of communities impacted upon by an increasingly developed world. This is, however, only the beginning of the story. As one participant commented,

I think digging deeper on this topic is important. Continue … push forward! (Consultant/academic)

Acknowledgements

We acknowledge the valuable input from workshop participants, including the IAIA Health Section, to the ‘Sharing global experiences with including health in major project EIA’, held at the International Association for Impact Assessment annual conference in Vina Del Mar, Chile, in April 2014.