Development of a guide to applying precaution in local public health

Abstract

Purpose:

The precautionary principle (PP) urges actions to prevent harm even in the face of scientific uncertainty. Members of Toronto Public Health (TPH) sought guidance on applying precaution.

Methods:

We searched five bibliographic databases (yield 60 articles from 1996 to 2009 and 8 from 2009 to 2011) and Google (yield 11 gray literature sources) for material relevant to local public health. From these sources, we extracted questions until saturation was reached (n = 55). We applied these questions retrospectively to eight case studies where TPH felt precaution was applied. We ranked questions for their importance in applying precaution.

Results:

Our final guide included 35 questions in five domains: context, assessment, alternative interventions, implementation, and monitoring and evaluation. Importance rankings varied across cases, but the role of stakeholders in driving precautionary action was consistent. Monitoring and evaluation components could have been strengthened across cases.

Conclusion:

The TPH guide can assist municipal environmental health practitioners in applying precaution in a more transparent manner.

Introduction

Role of local public health agencies in environmental health

The primary focus of local public health agencies is the promotion and protection of the health of the population and the prevention of illness. Public health agencies within local governments have historically responded to a range of environmental hazards using risk assessment approaches to address known or emerging threats. Municipalities and regional health authorities are well positioned to respond to local issues and often form successful partnerships with community stakeholders.¹ Addressing environmental hazards locally can have positive influences at higher levels of government and globally. In the province of Ontario, Canada, local public health units conduct environmental health assessments guided by provincial and federal standards and protocols.^2,3 However, in situations where the evidence of harm and/or exposure for a particular hazard is unknown or evolving, guidance on how to proceed under uncertainty is lacking.

Nature of the precautionary principle (PP)

The precautionary approach that some local authorities have used when dealing with environmental hazards is a ‘principle of public decision-making that requires decision makers, in cases where there are “threats” of environmental or health harm, not to use “lack of full scientific certainty” as a reason for not taking measures to prevent such harm’.⁴ However, the Stakeholder Advisory Group on extremely low frequency electromagnetic fields⁵ as they reviewed its use noted that ‘there is no internationally accepted single definition of the precautionary principle and international agreement or treaties all adopt different formulations of the “precautionary principle” to reflect their particular circumstances’.⁵

There is growing consensus to place precaution at the forefront of environmental and public health policy development.⁶ Greater application has been urged in developing countries,^7,8 with some examples of application. For example, South Africa’s inclusion of the PP in its Water Quality Guidelines has paved the way for graywater use in small-scale agriculture and gardens.⁹

Several definitions of the PP exist. All include a component that urges acting despite uncertainty, when there are reasonable indications to do so.¹⁰ The principle’s focus is on avoiding risk or reducing hazards at the outset of a project or activity by assessing alternatives along with stakeholders.^11,12 Reasonable indications vary with a society’s chosen level of protection against risk, highlighting the essentially normative basis of decision-making under uncertainty.¹³ Some jurisdictions have argued that precautionary thinking should be set through participatory decision-making processes, enabling representation of the more vulnerable.^3,14 Several writers emphasize the flexibility in application of the PP when evidence on harms and exposures is limited or contested, making risk assessment calculations challenging.¹⁵ Other authors argue that decisions must be based on precaution rather than on risk because of inherent and fundamental failures in the current risk assessment paradigm.¹¹ To correct the focus on maximizing the ‘average good’ implicit in utilitarian models of cost–benefit analyses, the greater risk of vulnerable populations must be recognized,¹⁶ as must persistence, bioaccumulation, and exposures to complex chemical mixtures.¹¹ In addition, the distribution of risk and its reversibility should be considered when evaluating alternatives.¹⁷ A UK inter-departmental group on risk assessment emphasized the need to revisit decisions made by applying precaution whenever new information arises.¹⁸

Use of PP in local settings

Several cities in the US have formally incorporated the PP into their policies. In 2003, San Francisco produced a white paper detailing how the PP was contained in its Environment Code.¹⁹ Seattle followed closely in 2004 as did Berkeley, California.^20,21 Examples of application include integrated pest management programs, bans of arsenic-treated wood, and enactment of resource-efficient building ordinances.¹⁶ In the UK, the PP has been cited in local government authority comments on marine-protected areas.²² The World Health Organization (WHO) European Region has issued guidance in local environmental planning processes to use precautionary approaches to prevent environmental health risks.¹ Climate change has pushed a number of cities to consider precautionary approaches to assessment and mitigation of the consequences of extreme adverse weather events e.g. a municipal action plan for Cape Town, South Africa.²³

The precautionary approach is referenced in the City of Toronto 2000 Environmental Plan.²⁴

Toronto Public Health’s (TPH's) environmental research and policy team has drawn upon the PP to inform policy and program activities.

We have used it typically to address emerging issues that are characterized by high community and stakeholder concerns that are not always aligned with each other or with the public health concerns that arise from emerging evidence. However, we have found that while many public health practitioners agree with using precaution in situations of uncertainty, there is ongoing debate about how to operationalize the PP. To our knowledge, there were no existing frameworks or guidance documents that were created specifically for the application of the PP at the local level. Often, local jurisdictions face the challenge of applying precautionary measures (e.g. local bylaws) to address potential threats for their particular community. This is often in the absence of guidance from higher levels of government. The political and economic context may also differ from other jurisdictions and reflects the normative process of applying precaution. The degree of precaution used and specific actions taken on the same issue may therefore vary between jurisdictions because of differences in context, stakeholder and institutional values, and available resources. It is the role of local public health to ensure that local values and context are given due consideration and that different objectives are balanced. Because there is often no ‘proponent’ on whom to shift the burden of proof, this responsibility resides with authorities at the local level. Tools to thoroughly document precautionary decision-making can therefore assist greatly in achieving a satisfactory process and acceptable outcome.

Objective

This paper started as a reflection on this practice²⁵ among members of our team who continued to grapple with how best to apply precaution at the local level. We were cognizant of the debate and caveats on the possible misuse of precaution and the sentiment among some experts that it is not adequately evidence based.^26,27 We therefore sought a coherent, explicit, and transparent way to apply precaution in order to ensure comprehensiveness, consistency, and accountability. The paper describes the process involved in consulting the literature, developing a framework, reviewing cases in which TPH applied precaution and iteratively discussing our findings and modifying the framework, until the production of our guide to applying precaution.²⁸

Methods

Literature scan

We initially identified literature from 1996 to 2009 using the keywords ‘precautionary principle’, ‘public health’, and ‘environmental health’ in five bibliographic databases: CINAHL, Academic Search Premier, Health Business Elite, Nursing & Allied Health Collection, and Psychology and Behavioral Sciences Collection. After eliminating duplicates, 60 articles remained. The same search terms used in the Google search engine found 11 relevant gray literature sources. Overarching pieces on applying precaution were included in keeping with realist review approaches.²⁹ Our focus was on articles relevant to a local/municipal context, avoiding specific case studies at other levels with a narrow focus on a particular exposure. We extracted criteria, indicators, questions, or other components until saturation was reached (n = 55). A repeated literature search in 2011 with the same keywords process in Medline (n = 83), PubMed (n = 138), and Environmental Health Perspectives (n = 46) yielded an additional eight articles.

Table 1 summarizes the strengths and limitations of PP frameworks we felt contained useful criteria but which needed adaptation for local environmental health settings. Our decisions on using these in our guide are contained in the first section of the Results.

Table 1. Strengths and limitations of selected precautionary principle (PP) frameworks.

Framework or criteria	Summary	Strengths	Limitations	Application to local public health settings
Tickner et al.12	One of the first PP frameworks publishedSix steps:(1) Identify the threat and characterize the problem(2) Identify what is known and what is not known about the threat(3) Reframe the problem to describe what needs to be done(4) Assess alternatives(5) Determine the course of action(6) Monitor and follow up	- Systematic evaluation of risk and uncertainty- Seeks to reframe the problem and assess alternatives- Monitors the effects of an action over time	- Suggests that the proponent of a potentially harmful activity bears the burden of proof. This is difficult when multiple stakeholders are involved or when it is difficult to attribute potential harms to any single proponent (or proponents)	- Broad framework can be applicable to local settings
Commission of the European Communities30	This communication from the European Commission establishes guidelines on how to use the PP.(1) Factors triggering recourse to the PP(2) Measures resulting from reliance on the PP(3) Guidelines for applying the PP- Implementation- Triggering factor- General principles of application	- Explores the factors that trigger the PP- Outlines five general principles for application (e.g. non-discrimination, consistency)- Considers society's ‘acceptable’ level of risk	Approach is tied to the current conventional and structured approach of risk assessment, identifies PP as most often applied at the risk management stage	- Broad framework can be applicable to local settings
Government of Canada Privy Council Office3	PP framework in two broad sections:(1) Five general principles of application- Includes emphasis on society's chosen level of protection against risk, precaution based on sound scientific base, and transparency with meaningful public involvement(2) Five principles for precautionary measures- Includes reconsideration of actions based on evolving science and societal values, proportionality of measures to risk, consistency of action, and cost-effectiveness	- Strong emphasis on public involvement in decision-making- Precautionary measures are responsive to changes in scientific evidence, technology, and society's desired level of protection	- Emphasis on meeting obligations under legislation and in international agreements, which may not be sufficiently precautionary	- Framework is created from a federal perspective
Weir et al.42	Three-part application of the PP in public health:(1) Certainty of cause and effect relationship between the exposure and potential harm(2) Nature of the risk and the level of certainty of harm(3) Assessment of precautionary measures	- Quantifies exposures, harms, and costs of actions/inactions- Appraises the proportionality of actions to level of risk and the comparability of precautionary measures to similar situations	- Use of Bradford-Hill's considerations for assessing causation overly restrictive- Does not take into account equity considerations or impact on vulnerable populations- Unclear how the public or stakeholders are involved in the decision-making process	- Work involved in systematic review of causation best done by groups of health units or larger organizations e.g. provincial public health

Development of the guide

Based on the literature scan, we developed a working definition of applying precaution as ‘an approach to public decision-making on issues in which there is potential environmentally related health harm and a need to take measures to prevent that harm in the face of scientific uncertainty’. The uncertainty can be related either to the nature of the causal relationship between an environmental agent and potential harms or to the likelihood of human exposure to the agent.

Using this working definition, we set out to develop a set of questions to guide the application of precaution, using a transparent and deliberative process. Our approach initially began by developing a draft guide that consolidated the strongest components of other frameworks (Table 1) while incorporating other elements that we felt were important based upon our local experience of applying precaution.

Our first draft had 34 questions in four domains: assessment, alternative interventions, implementation, and monitoring and evaluation. Questions in the initial guide came from several sources in the literature, but were largely based on the Precautionary Principle Handbook.¹² Other questions and the ‘context’ domain were generated by the co-authors, through an iterative process.

Retrospective case application and revision

To refine the guide, we applied the questions to case studies from TPH that we felt used a precautionary approach. Initially, 13 case studies were proposed and as we tracked down documentation on these, some were found to not have enough of a precautionary focus to warrant reflection or further analysis. Others we eliminated because TPH had played only a minor role. We arrived at a set of seven past cases and one case in progress. Table 2 summarizes the areas of uncertainty in each case and the corresponding precautionary action taken. The nature and implications of the cases for guide development are set out in the Results section.

Table 2. Case studies used to refine the Toronto Public Health (TPH) Precautionary Principle Guide .

Case study	Area of uncertainty	Precautionary action
Electromagnetic fields (power corridors) and prudent avoidance policy	Childhood leukemia risk from electromagnetic fields in power corridors	Implemented a prudent avoidance policy requiring assessment of electromagnetic fields (EMF) levels in and around hydro corridors to site amenities such that EMF levels are lower to reduce children's exposure
Radiofrequencies (cell phone towers) prudent avoidance policy	Health impacts of long-term, low-level exposure to radiofrequencies	Voluntary City policy relating to the siting of cell phone towers encourages radiofrequency levels to be <1% of the federal standard (Note that this policy is under review to take new evidence into consideration.)
Air quality health index (AQHI) development	Effectiveness of an existing air quality index (AQI) to accurately convey the health effects associated with air pollution. Research indicated a major proportion of premature deaths and hospitalizations were occurring when the air quality was deemed to be ‘good’ according to the AQI	Proposed a new AQHI intervention to provide an accurate health-based air quality measure so that people can minimize exposure by adjusting their behavior
		Populations at risk (young, elderly, those with cardiac and respiratory conditions) have lower ‘action’ thresholds than the general population
West Nile virus (WNV) integrated pest management program	There was a need to minimize the health and environmental impacts from pesticides used to control the mosquito population while balancing the need to prevent further human WNV cases	Application of a tiered mosquito control program that incorporated integrated pest management principles. Emphasis on the least harmful measures (education, non-spray pesticide control, larviciding) was used to minimize exposure to toxic pesticides (e.g. malathion)
Chromated copper arsenate (CCA) in wooden play structures	Arsenic levels were found to be elevated on surfaces and in the soil below wooden play structures in Toronto parks. No standards on acceptable levels of arsenic in wood surfaces and soil existed to determine exposure levels for children	Targeted mitigation strategy was developed based on levels of arsenic found at 217 wooden play structures. Strategies included accelerated removal of older priority structures, regular application of sealant, remediation of soil/sand, and routine monitoring
Pesticides bylaw	Existing federal and provincial legislation was not deemed to be adequate to ensure health protection during vulnerable life stages, in the context of unnecessary home applications of pesticides on lawns	Advocated for a city bylaw to restrict pesticide applications on all public and private properties. Implementation focused on public education first and then phased in enforcement
Fish consumption advice for women in their childbearing years and for families	Existing federal reference limits on mercury and polychlorinated biphenyls were deemed to be inadequate. Risks needed to be considered with the benefits of fish consumption	Prepared advice on fish consumption to women in childbearing years and for families eating fish that applied mercury and PCB standards that were more health protective than federal limits. Tailored outreach to address the needs of high fish consumers
Urban gardening soil assessment protocol	Benefits of urban gardening are widely accepted, but the risks from contaminants and naturally occurring heavy metals in soil are unknown	Developed a soil assessment protocol to provide information on the risks associated with urban gardening. Used composite soil testing and estimates of uptake into production and consumption patterns

Two or more TPH staff active in a given case applied the draft guide by filling in an Excel worksheet that sought to assess the importance of each question in applying precaution (Table 3).

Table 3. Worksheet for refining the Toronto Public Health (TPH) guide to applying precaution.

Domain (context, assessment, implementation, evaluation)	Substantive answer	Importance in applying precaution (not important, somewhat important, important, very important)	Comments/suggestions	Reference for question
Question

Completion usually took several hours to a day and a half, depending on the complexity of the case, its duration, and the nature of decision-making processes. The team discussed the worksheets and revised questions for greater clarity, eliminated irrelevant questions, and added new questions that were judged important in managing environmental health risks (e.g. adding a context domain). Differences of opinion in the importance ranking of questions were resolved through discussion and consensus. An analysis of the importance rankings was conducted by one author (DCC) and subsequently redundant questions and those consistently ranked ‘not important’ or ‘somewhat important’ were dropped (e.g. generation of negative externalities, grade of evidence). A new iteration of the guide was developed after applying a case study, and subsequent case studies were ‘tested’ using the most updated version of the guide. In total, three questions regarding the evidence of assumptions, effectiveness of alternative options, and the effectiveness of the interventions were included based upon group discussions. One question from the European Commission criteria (2000) on consistency was added.³⁰ Once the guide was finalized, the earliest case studies were reapplied to the final guide for completeness.

Results

Development of the guide

The collation of questions from various PP frameworks enabled us to develop a comprehensive and detailed guide to applying precaution. We relied heavily on the original work by Tickner et al.¹² and felt this to be the most comprehensive and systematic framework out of all that were reviewed. We also referenced components of frameworks from the Government of Canada and the European Commission, amongst others.^3,12,30 We did not include a specific trigger for precaution unlike some frameworks because we felt that recognizing the uncertainty of a potential environmental health threat would warrant using the guide to explore if precaution should be applied and to document information and decision-making at a minimum.

Although contextual factors have been addressed in other frameworks, in our experience, these drivers were vital in considering precaution. We added this domain at the beginning of the framework to reflect this importance. We also included questions on equity, empowerment of affected populations, assumptions about the evidence and others on communication, accountability, and evaluation. The final guide (Table 4) provides a systematic approach for applying precaution to environmental health issues in local public health settings.

Table 4. Toronto Public Health’s (TPH) guide to applying precaution – final set of questions .

	References
1. Understanding context
Context
How did the issue come to your attention?	TPH authors
Drivers for decision-making
Is there public concern?	TPH authors
Are there standards to be met?	TPH authors
Who are known supporters and detractors?	TPH authors
Financial gains or losses; for whom?	TPH authors
2. Assessment
Identifying harm
What are the acute and chronic harms?	Tickner et al.12
What are the inter-generational impacts?	Tickner et al.12
How irreversible are the effects?	Tickner et al.12
Populations most susceptible to harm?	Tickner et al.12
Uncertainty of harm
Lines of evidence available to assess harm?	Stirling and Tickner31
What assumptions are made about the evidence?	TPH authors
Identifying exposure
Exposure pathways?	Tickner et al.12
Populations most exposed?	Tickner et al.12
Exposures voluntary, individually controlled?	Sandman32
Uncertainty of exposure
Lines of evidence available to assess exposure?	Stirling and Tickner31
What assumptions are made about the evidence?	TPH authors
3. Exploring alternative interventions
Alternatives for exposure/harm reduction
How can exposure/harm be minimized/eliminated?	Tickner et al.12
Alternative public health interventions?	Tickner et al.12
Evidence of the effectiveness of alternatives?	TPH authors
Feasibility of alternatives
Political, technical, and economic feasibility?	Tickner et al.12
Trade-offs
What risks (if any) do these alternatives produce?	Hansen et al.33
Unintended consequences of these alternatives?	Tickner et al.12
Consistency
Interventions consistent with existing measures for similar exposures?	Government of Canada Privy Council Office3, Commission of the European Communities30
Proportionality
Interventions proportionate to potential harms?	Government of Canada Privy Council Office3, Commission of the European Communities30
Stakeholder views
Efforts made to consult or involve stakeholders?	TPH authors, Government of Canada Privy Council Office3
Extent to which stakeholder views are considered?	TPH authors
Do alternatives empower affected populations?	TPH authors
Equity
Opportunities to reduce or prevent inequities?	TPH authors
4. Implementing intervention(s)
What interventions (if any) will be implemented?	Tickner et al.12
Plans communicated to the public and stakeholders?	TPH authors
5. Monitoring and evaluation
Frequency and intensity of monitoring?	Tickner et al.12, Government of Canada Privy Council Office3
Who is accountable?	TPH authors
How often will new evidence be assessed?	TPH authors
Evaluating effectiveness of interventions?	TPH authors
Feedback on monitoring given to stakeholders?	TPH authors

Findings in application to case studies

The following analysis highlights questions in each domain that were consistently ranked as important or not important in applying the PP to the case studies.

Context

Among the context factors (Domain 1) set out in Table 4, the most consistently important factor in applying precaution was how the issue came to the attention of TPH staff. Public concern and issues raised by city councillors were commonly key drivers in creating interest in an issue. However, public concern about health impacts varied from somewhat important (EMF from power lines – initially important but lessening over time) to very important (lawn pesticide use). Active stakeholders involved in the issue covered a wide range of public and private entities, typical for environmental health issues that face local public health authorities. In other instances, TPH staff identified a need to improve health protection measures such as advocating for the creation of the air quality health index (AQHI) because research showed potential gaps in existing interventions.

Assessing harms and exposures

Staff had engaged in considerable work to both identify harms and exposures and assess the extent of uncertainty associated with each case (Domain 2, Table 4). The nature of the acute and chronic harms or adverse health effects, and the existence of sub-populations at particular risk were each regarded as very important for applying precaution across all cases. A range of evidence was available in each case, and conclusiveness on either causation of harms or extent of exposure varied in each situation. For example, in the chromated copper arsenate (CCA)-treated wood play structure case, evidence was very good on the harms of arsenic, but exposures were unclear. For the EMF and power lines case, explicit assumptions were very important but were of little importance in others (e.g. lawn pesticide use), indicating that TPH staff used more classic risk assessment tools when feasible but did not require them for reaching decisions on appropriate interventions.

The importance of proportionality in applying precautionary actions seemed to vary with the certainty of harms and exposures. Overall, more intrusive approaches (e.g. a bylaw to limit cosmetic pesticide use) were used when the evidence of harm was stronger. In contrast, when the evidence of harm was uncertain, but evidence of exposure was more certain, approaches were less intrusive (e.g. voluntary policy to reduce radiofrequency exposure from cell phone towers).

In addition, this domain was important for identifying vulnerable populations that may require targeted interventions. This included fish consumption messaging for pregnant women and public education for those at higher risk of air quality health impacts. Most of the case studies involved universal precautionary interventions that took into account subgroups who may be at higher risk.

Exploring alternative interventions

Generating a set of alternative public health interventions was an integral part of applying the PP. The effectiveness of the alternative interventions was sometimes difficult to measure because the consequences often have a delayed onset and multi-factorial causes. For example, reducing radiofrequency exposure from cell phone towers or limiting exposure to heavy metal contaminants in urban gardening and correlating these with decreases in adverse outcomes are difficult to quantify.

In all cases, consultation with stakeholders on the relative merits of alternatives was conducted, with their views playing a major role in decision-making in the pesticide bylaw and EMF prudent avoidance policy. Mechanisms for stakeholder engagement varied across the range of traditional approaches such as using existing partnerships, working groups or committees, public meetings, open houses, workshops, and similar forums. A breadth of approaches ensured ability to foster dialog and capture diverse voices. Mechanisms tended to be more extensive and systematic for issues with broader import and impact, and where there had been a political directive for stakeholder consultation such as the one that occurred with the pesticide bylaw.³⁴ For issues of more circumscribed importance, such as with development of the fish consumption guidance, engagement mechanisms were strategic and targeted (e.g. meetings with expert academics, government ministries, and public health partners or focus groups with women in TPH prenatal nutrition programs). Other mechanisms, such as media stories, fostered opportunistic stakeholder engagement such as in the case of the urban gardening soil assessment guide.

Economic feasibility was very important to consider at a local level, and this factor led to modified interventions based on affordability. In the CCA wood play structures, a triaged approach to remediation occurred, with older play structures being replaced as a priority and other play structures with higher measured levels of arsenic having regular application of sealant to prevent arsenic from leaching.

Considerations for unintended consequences were very important for case studies where the benefits needed to be balanced against the risks. For example, in the West Nile virus (WNV) case, the harmful effects of pesticides on the environment needed to be weighed against the benefits of mosquito control to prevent WNV transmission to humans. Similar dilemmas were faced in counseling on fish consumption for pregnant women and developing urban gardening protocols.

Consistency with other jurisdictions was somewhat important and depended on whether a precedent had been set already (e.g. EMF prudent avoidance policy in other jurisdictions). In other cases, taking a precautionary approach meant using more protective measures than existing provincial and/or federal guidelines. For instance, the voluntary radiofrequency prudent avoidance policy recommended cell phone tower radiofrequency emission levels of ≤1% of the federal standard in areas where people usually spend time.

Equity was considered during deliberations, though not necessarily an explicit or strong driver of action in the application phase for all cases. Many of the interventions were universally applied. The radiofrequency prudent avoidance policy for example took an emission-based rather than a distance-based approach to ensure greater protection for all. Vulnerable populations were, however, a key focus in the AQHI, CCA wooden play structures, EMF policy, urban gardening, pesticides bylaw, and fish consumption case studies. EMF policy allowed expansion of park amenities in hydro corridors in otherwise underserved areas. The precautionary fish consumption advice was geared to ethno-racial groups that are frequent fish consumers.

Implementing interventions

The questions in this section were focused on describing the chosen intervention and documenting how this decision would be communicated to stakeholders and the public. They were ranked highly important in applying precaution in all cases.

Monitoring and evaluation

This dimension was ranked as less important in deciding to use a precautionary approach, likely because monitoring and evaluation takes place after the decisions on applying precaution have been taken. Monitoring and evaluation was however an important component of implementation; it helped build support for proposed actions and was also important when decisions were revisited. This domain was integral, for example, to the program for CCA-treated play structures, with extensive, regular testing results being used to guide remediation activities.³⁵ Monitoring of the pesticide bylaw was a formal requirement of Toronto City Council at the time it adopted the bylaw. Analysis of monitoring data on the non-essential use of pesticides was guided by an explicit logic model and collaboration on data analysis oriented by evaluation research questions.³⁶ Monitoring for compliance was a Toronto City Council request for the radiofrequency policy, but monitoring was not formally applied in the case of EMF and power lines in part because mostly other departments handled this policy and the need to apply it has been sporadic since adoption.

Discussion

The development of our guide to applying precaution to local public health issues and retrospective application has given us insight into the drivers that affect our decisions in using precaution. In particular, the addition of a context dimension filled the gaps in other PP tools to make our guide more adapted to the local context. For example, actions taken for certain cases that were brought to TPH's attention by the media, councilors, and the public (e.g. EMFs, CCA wooden play structures) reflected responsiveness to local concerns and values by a precautionary approach. As stated in the European Commission's 2000 communication on the PP, ‘judging what is an acceptable level of risk for society is a … political responsibility’.³⁰ We took into account stakeholders' values and their risk tolerance while assessing the evidence to determine a response proportionate to the potential harms. For example, it was predicted that the pesticide bylaw would be met with opposition from pesticide companies and business owners; these stakeholders were engaged early on in the process, to ensure that their needs were understood and taken into account, although not at the expense of precautionary action to protect health. This ‘early and continuous’ approach to stakeholder engagement is important for preventing delays that can occur later if stakeholders are not involved upfront in the decision-making process.³⁶ Such an approach recognizes the role of multiple actors in municipal governance and the potential for participatory risk assessment processes. An example of their application is the consideration of urban agriculture-associated health (and social) benefits and risks in African cities.³⁷

In our case studies, the costs of assessment and interventions were largely borne by the municipality, rather than other frameworks that emphasized shifting the burden of proof to industry.¹² This is the case with exposures in the general community that are not easily attributed to a single or identifiable proponent (e.g. air pollutants, mercury in commercial fish, and much wastewater). Furthermore, precautionary innovations are more likely to happen and can sometimes more easily gain political support at the local level. The opportunity to experiment and test new approaches can foster diversity in environmental health policy and later on, may be adopted at higher levels of government. This occurred with Toronto's pesticide bylaw that later became provincial law.³⁴

There are a number of levers that can be applied by local government and public health, although these will differ globally. Canadian municipalities have the general power (via provincial statutes such as the Municipal Act in Ontario) to make bylaws, including pesticide bylaws, regulating health and general welfare of their inhabitants, as was confirmed in a landmark decision by the Supreme Court of Canada in 2001.³⁴ In addition, boards of health and public health departments in Ontario are empowered to investigate and act on health hazard prevention and management through the public health standards set by the province's Ministry of Health and Long-Term Care. That mandate is enshrined in the Health Protection and Promotion Act, the provincial legislation that provides authority to inspect, seize, and close places where health hazards may exist.³⁸ However, we do recognize that in other areas of the world, powerful interests can intervene to squash or constrain such municipal creativity, for example, through pre-emption laws at the state or federal levels, as did the tobacco industry fighting smoking restrictions in public places in the USA³⁹ or with court challenges, as occurred with San Francisco's ban on bisphenol A (BPA).⁴⁰ International trade agreements, such as the North American Free Trade Agreement (NAFTA), have been invoked to challenge regional precautionary action on pesticides as exemplified by the Dow Chemical challenge of the province of Quebec's pesticide ban.⁴¹

Like most frameworks, our guide does not include a particular trigger for precaution. The purpose of our guide is to stimulate discussion and ensure that all factors considered in applying precaution are used consistently and in a transparent manner. We also believe that it would be inappropriate to routinely require use of Bradford-Hill's considerations on causality in situations of such uncertainty, as some authors have argued.⁴² The paucity of evidence for causation of harms or extent of exposure is the reason why precautionary approaches are advocated and needed in the first place.

A key strength in our retrospective application of the guide was the team-based discussions around the case studies and improvements made to the guide and our work. In our case, the team benefited from its multidisciplinary nature; the practitioners had diverse expertise and training ranging from toxicology, community medicine, environmental sciences, biological anthropology, and social sciences, to public health policy, civic engagement, and knowledge exchange. This exercise was a reflection on our collective practice and showed gaps in previous work before our guide was developed. For example, our team realized that for many interventions, there was no formalized monitoring and evaluation plan. Now this component is more routinely incorporated in policy analysis at TPH. Most importantly, the team now regularly applies this systematic guide, along with other tools such as health impact assessment, that assists in assessing evidence and determining the most appropriate intervention. Although there is no specific expertise required to apply this approach, we recommend that practitioners ensure that a diversity of practitioners (and stakeholders) are engaged to ensure a breadth of perspectives provide input.

We specifically focused on the development of a guide to applying precaution for local public health needs, recognizing that direction from higher levels of government may be lacking or absent on emerging environmental health issues. We did not address issues of product regulation or safety for instance, as these issues fall under the responsibility of national governments or international regulatory bodies. Local precautionary action can sometimes stand alongside and complement the regulation of other levels of government with positive results (e.g. pesticide bylaw)⁴³ although in other instances, its applicability and effectiveness can be constrained by a lack of jurisdiction (e.g. radiofrequencies from cell towers).⁴⁴

Limitations of our work included inconsistency in rating importance for domains that describe how to implement precaution, rather than if precaution should be used (e.g. implementation, and monitoring and evaluation). Also, the retrospective ‘testing’ of our guide may have overlooked other factors that need to be included. It may be challenging for some jurisdictions to apply this guide because setting aside time to complete it (from half a day to several days per team member) and achieving consensus may be difficult. Although innovation may take more resources, application in smaller local health departments can be less resource intensive, particularly with sharing of innovations within a community of practice. We acknowledge that this guide will not always lead to a clear answer, particularly in situations where there needs to be a balance between risks and benefits, as in the case of fish consumption advice. Also, it is difficult to evaluate the outcomes of precautionary actions because often no adverse effects are measured.¹ Few of the programs were formally evaluated to assess their contribution to reduction of exposures,⁴⁵ except the pesticide bylaw.³⁴

We have focused on using our guide prospectively now to make decisions on applying precaution in situations of uncertainty. We have shared it with colleagues dealing with areas outside of environmental health. Future evaluations of this guide should examine whether gaps previously identified in our work are improved (e.g. monitoring and evaluation), whether decisions have changed based on using the guide to revisit issues as new evidence accumulates, and on its application in substantially different jurisdictional and development contexts than our own.

As we note elsewhere, our guide will not provide a roadmap or a ‘how to’ for applying precaution; it can provide important structure and transparency in deliberations and decision-making, allow for more systematic identification of the uncertainties, by unpacking the information on hazard and exposure, and assist in revealing possible ways forward on contentious public health challenges.²⁸

Conclusion

We developed a guide to applying precaution that is adapted for local authorities based on frameworks in the literature and reflections of our own practice.⁴⁶ This guide was refined using retrospective application to seven case studies where TPH used precaution, and it will require further modifications as it is prospectively applied. We urge local public health practitioners globally to share experience with this and other guides to improve our collective ability to apply a precautionary approach to complex environmental health issues.

Disclosure Statement

All authors have no conflicts of interest to declare, beyond that indicated in their affiliations.