Viewpoint |
The precautionary approach: from Birth to childhood Epidemiology for risk assessment: losing the beginner's confidence
Josep M. Antó** Respiratory and Environmental Health Research Unit, Municipal Institute of Medical Research (IMIM-IMAS), Universitat Pompeu Fabra, Barcelona, Spain
Correspondence: Josep M. Antó, Respiratory and Environmental Health Research Unit, Municipal Institute of Medical Research (IMIM-IMAS), Universitat Pompeu Fabra (UPF), Barcelona, Spain e-mail: jmantoa{at}imim.es
Like most epidemiologists, I learned that epidemiological methods are particularly useful for providing background information for risk assessment as an important tool for health protection and something I had the opportunity to apply from the beginning. In the eighties, using daily observations of emergency room visits for chronic obstructive pulmonary disease and applying Box–Jenkins models, we challenged that the usual air pollution standards at that time protected the health of the exposed populations with a margin of safety as intended.1 Although the publication of the paper was accompanied by a faultfinding editorial, subsequent research proved a wide replication of our results.2 The subsequent translation of the results of time-series studies into more stringent standards for air pollution has, in general, faced two contentious difficulties: the lack of a plausible biological explanation and the difficulties of translating the association between current low levels of air pollution and mortality into a population-based measure of harm such as disability-adjusted years of life lost. The former is now less of a problem following studies showing that the inhalation of fine and ultrafine particles is associated with cardiovascular toxicity.3 The latter is still a difficult issue; a recent risk assessment of air pollution did not use time-series studies for short-term effects and mostly relied on studies of chronic effects.4 During the same period we embarked on a different study of asthma outbreaks in the city of Barcelona, at that time competing to host the 1992 Olympic Games. In separate studies we provided strong evidence that the outbreaks that caused more than 1000 attendances at emergency departments and approximately 20 deaths were due to inhalation of soybean dust spread over the city from industrial operations at the city harbour. These results were sufficiently clear to justify temporarily suspending the unloading of soybean until appropriate filters had been installed in the offending silos.5–7 Paradoxically, these two successful experiences developed under the paradigm of the epidemiological contribution to risk assessment led me to incubate serious doubts about both the scientific robustness and the public health rationality of risk assessment. These doubts were stimulated by several readings. One was a paper published in 1928 describing an asthma outbreak in Ohio as a result of the inhalation of castor bean dust.8 The authors predicted the occurrence of asthma outbreaks due to soybean dust inhalation when soybean was introduced as the substitute for castor bean in the Toledo mills. Figley and Elrood were proved right some 50 years later after repeated asthma outbreaks had occurred in many other places.7 Another paper was by Carl Shy, who argued with Walter Holland about the scientific justification for further reductions in the air pollution standards of the late 1970s. Shy's reasoning was that further reductions far below the lowest observed effect level were necessary to compensate for the lack of knowledge about possible effects on vulnerable populations such as children.9,10 In these two papers I found an indication that something was wrong in risk assessment. Some years later, after reading the proceedings of the Wingspread Conference advocating the need for a new approach in environmental and health protection called the precautionary principle,11 I became convinced that my previous doubts about risk assessment were legitimate and deserved more careful attention.12
 |
The precautionary principle: a relevant new concept for risk assessment and health protection |
|---|
 Top The precautionary principle: a... The precautionary approach: from...References |
|---|
In the seminal Wingspread Conference the precautionary principle was defined as an approach to establishing the required action to control the risk derived from products or technologies before causal scientific evidence has been established, with its strongest formulation being the requirement of absolute proof of safety before allowing the dissemination of new technologies. The notions underlying the precautionary principle had been formulated some years before, both in the environmental policy field and in the academic environment. The Rio Conference pronounced, as its Principle 15, that in order to protect the environment, the precautionary principle approach shall be widely applied by states according to their capability and that where there were threats of serious or irreversible damage, lack of full scientific certainty shall not be used as a reason for postponing cost-effective measures to prevent environmental degradation. Among other studies, one that contributed to the research background of the precautionary principle was the investigation of the environmental effects of the endocrine disruptors by Colborn and others.13 An important underlying reasoning was that the standard risk models for environmental and health protection as used by governmental agencies were too restrictive, leaving room for huge amounts of irreversible harm to occur before any action was taken. Other cases, such as the consequences of human exposure to asbestos, benzenes, or PCBs, are relevant in understanding the need for a new approach.
What is wrong with the current model? Risk assessment has evolved during recent decades, accumulating experience and research to the point of being a relatively robust approach for the quantification of harm owing to a given exposure and for assessing the magnitude of future harm if such exposure is not abated. The information provided by risk assessment is the basis for risk communication and risk management, which together constitute the triad of the risk-based health protection approach. As such it has been applied to a long list of individual contaminants including lead and ozone. In contrast, risk assessment has proven to be too limited to accommodate the complexity of exposures involving mixtures of chemicals, as is the case in air pollution or even more complex environments such as climate change. It has also failed to account for ambiguities such as the lack of evidence for vulnerable populations or uninvestigated effects with the use of a prudence criteria, consisting of the reduction by one or more orders of magnitude of the lowest observed effect level, which has proved clearly insufficient. Although the current model of risk assessment is based on robust scientific principles such as the quantification of dose–response curves and on hierarchies of designs with experimental and cohort studies at the top, the final risk categorization has no formal quantitative way to accommodate the heterogeneity of evidence provided by different disciplines such as toxicology and epidemiology or the animal model and human population-based equivalents. Limitations are perhaps even more important in the area of risk management, a process that has failed to include many non-chemical risks such as allergens and has covered only a limited proportion of the potentially toxic chemicals. In addition, although the process has benefited from the formidable efforts of agencies such as the International Agency for Research on Cancer in Europe and the Environmental Protection Agency and the National Institute for Occupational Safety and Health in the United States, the overall process is scattered and fragmented across types of risks and agencies. Furthermore, even in those cases where the risk-based model may be seen as appropriate—for example, the regulation of lead—one can identify one of its more important limitations: its reactive and delayed nature. It is not difficult to believe that these limitations may have facilitated the occurrence of several public health tragedies, including the current asbestos-related cancer epidemic.
Considering the limitations of the current model of risk assessment, the precautionary principle has several underlying notions that make it a promising new approach. First there is the notion that the lack of evidence should not impede action when the potential damage is sufficiently important; in other words, the precautionary principle is an anticipatory, proactive approach that is very close to the notion of prevention in public health. However, the anticipatory nature of precaution, at variance with prevention, lies in its willingness to act when evidence of harm is not yet available simply because it has had no time to occur. Here, there is an important difference from the current concept of preventive action which throughout the epidemiological tradition has been based on the establishment of a causal relationship between a given exposure and the corresponding outcome as a requirement for future action. Although the latter approach is scientifically robust and minimizes the possibility of taking expensive actions in vain, it maximizes the possibility of no action even when true and relevant harm is occurring. The trade-off between sensitivity and specificity has, in the current model of health protection, been resolved in favour of specificity. The consequence is that when the causal link is established, usually after one or more decades of research, the amount of harm accumulated in the exposed populations can be huge. Although the problem is relatively simple to understand its solution is likely to be extremely difficult. Could the precautionary principle be a useful new approach?
|
The precautionary approach: from birth to childhood |
|---|
|
TopThe precautionary principle: a...The precautionary approach: from...References |
|---|
Fortunately, during the few years since the first proposals of the precautionary principle, the literature exploring potential alternatives and solutions has started to grow, leading the new concept from birth to childhood. A new impetus was given to the field by the Lowell Statement on Science and the Precautionary Principle, which was the result of an international workshop held in 2001 leading to a consensus statement calling for a more prominent presence for precautionary issues in the research and policy agendas.14 In addition, the proceedings of another international conference, which was held in 2002, have recently been published.15 These materials, together with other contributions, have resulted in a substantial number of new ideas and experiences. In the field of epidemiology and public health, Weed has recently discussed several alternatives to the current approach to the epidemiological interpretation of causal evidence, including discounting the contrary evidence, reducing and relaxing the causal criteria, and changing the level of statistical significance.16 Although these alternatives will reduce the possibility of no action in situations involving true harm, their nature is close to the traditional prudential approach of reducing by one or more orders of magnitude the lowest observed effect level to compensate for the uncertainty that frequently affects the risk assessment. From a statistical perspective, Comba et al.17 have compared the Bayesian utilitarian approach with the approach based on the maximin principle as used in the risk management process, suggesting that the core difference between the two approaches has to do with the choice between averaging knowledge or equitably distributing technological risks. An important dimension of the debate around precaution is about widening the range of disciplines, with the simultaneous need both to redefine the roles of the disciplines classically involved in the field such as, epidemiology18 and toxicology,19 and to integrate the contribution of other disciplines such as ecology. Another important contribution of the increasing literature in this field is the expansion of the conceptual background of the precautionary principle, showing that its progression should involve not only empirical research but also theoretical elaboration. One of the more relevant precautionary-related concepts refers to the willingness to act before the occurrence of important irreversible harm. Unfortunately, public health theory has not developed a clear concept for this notion. In modern public health, the concept of prevention has, in general, been based on the demonstration that a relevant amount of harm has been causally produced by a given exposure or pollutant with the two aspects the amount of harm and the causal relationship being equally important. Health promotion may be seen as a more anticipatory action since it is aimed at maintaining and reinforcing health. Some authors have suggested that the concept that best reflects the anticipatory willingness implied in the precautionary principle is ‘foresight’, which was first used in Germany during the earliest attempts to develop a precautionary approach.15 Other concepts that also need a better understanding in this context are complexity, ambiguity, and prudence, since all cover important aspects of the precautionary principle. Also, the concept of risk as used in epidemiology and public health as a measure of harm already produced in the population needs to be distinguished from the notion of risk as a threat that may cause relevant harm in the future, since this distinction has important consequences not only for risk assessment but also for risk communication and teaching.
The final aim and the justification of the precautionary principle is the protection of the environment and human health. Since the Rio Conference, political initiative has been at the forefront of precautionary ideas. In Europe we have had early policy initiatives, especially the important Communication by the European Commission that was released in 2000.20 The Communication included some rules to guide the adoption of precautionary actions, such as proportionality, non-discrimination, consistency, full examination of the cost–benefit function, and close monitoring of scientific developments that could lead to future changes in the adopted actions. Following these rules, authors from the Centre for Technology Assessment in Baden-Buttemberg have developed a comprehensive framework that includes nine risk evaluation criteria, six risk classes, a decision tree, and three management categories and that is aimed at improving the effectiveness, efficiency, and political feasibility of risk management procedures.21 On the basis of the characteristics of different risk types and the problems of complexity, uncertainty, and ambiguity, these authors have distinguished three approaches, labelled as risk-based, precaution-based, and discourse-based management strategies, and emphasis is given to the need to include democratic deliberative processes in the three strategies.21 In the United States, the current Republican mandate is not providing the best political environment for the precautionary approach to be developed. However, thanks to the vigour and professionalism of the public health agencies some US environmental agencies are developing important initiatives. The National Institute of Environmental Health Sciences is developing several initiatives such as the National Center for Toxicogenomics and the National Toxicology Program to develop the precautionary principle in the field of exposures to mixtures of pollutants using the knowledge gained in understanding the multifactorial nature of gene–environment interactions.22 The passage of the precautionary principle from birth to childhood will intensify counter-initiatives from industries and corporations, which in general tend to adopt an antiregulatory approach. In the United States more than in Europe, the legal sentences on precautionary-related litigations may have important consequences for the development of the precautionary field, and in this contexts new reasoning and concepts against the precautionary approach are expected. One of these is the concept of hormesis, which refers to the possibility that low doses of harmful exposures may result in beneficial effects and which is gaining ground among industry-sponsored researchers and analysts.15
The growing experience and understanding of the precautionary principle, as summarized above, is providing a certain momentum for its formal inclusion in public health agendas, including policy, research, and teaching initiatives. In this process it is important that new proposals are built upon the best current experience. The current model of health protection based on the combination of quantitative risk assessment, risk communication, and risk management and managed by specialized agencies constitutes an important achievement of public health. It deserves to be improved, reinforced, and extended. The advocacy of the precautionary principle should not be seen as a threat to the current model of health protection. Instead, the impetus for an increasing application of the precautionary approach should also constitute a stimulus for improving the current model of health protection, one of the strengths of which is its evidence-based nature. However, the current model of health protection is far from being an appropriate strategy for the current needs of public health. Its main limitation lies in the fact that, for many potentially relevant risks, there is no available evidence to perform a quantitative risk assessment, and hence under the current model there is no room for progressing to risk management. There is abundant experience that waiting to accumulate enough evidence leads to the accumulation of serious harm to human health, as in the case of asbestos. Consequently, there is an urgent need for new strategies for developing preventive interventions when evidence of harm is not available either because research about putative harm is still insufficient or, preferably, because harm has not yet occurred. The new precautionary strategies should extend and complement the coverage of the current risk-based strategies. The precautionary principle constitutes a promising framework for the investigation and development of new health protection strategies and may provide an appropriate framework to explore, investigate, and develop a wide range of alternatives. These alternatives should range from modifications to the current risk-based model, facilitating its application at an earlier stage, to the strongest strategy of assessing the safety of new products and technologies before marketing. For such endeavours the current scientific methods are likely to be grossly insufficient and the development of new predictive methods will be needed. This research and development process should be actively promoted with appropriate funding and teaching. The European Commission is now developing the 7th Framework Programme for Research and Development, and this could be an excellent opportunity to launch an ambitious research programme for the precautionary approach. The time is now.
|
References |
|---|
|
TopThe precautionary principle: a...The precautionary approach: from...References |
|---|
1 Sunyer J, Anto JM, Murillo C, Saez M. Effects of urban air pollution on emergency room admissions for chronic obstructive pulmonary disease. Am J Epidemiol 1991;134:277–86; discussion 287–9.
2 Katsouyanni K, Touloumi G, Spix C, et al. Short-term effects of ambient sulphur dioxide and particulate matter on mortality in 12 European cities: results from time series data from the APHEA project. Air Pollution and Health: a European approach. BMJ 1997;314:1658–63.
3 Brook RD, Franklin B, Cascio W, et al. Air pollution and cardiovascular disease: a statement for healthcare professionals from the Expert Panel on Population and Prevention Science of the American Heart Association. Circulation 2004;109:2655–71.
4 Kunzli N, Kaiser R, Medina S, et al. Public-health impact of outdoor and traffic-related air pollution: a European assessment. Lancet 2000;356:795–801.[CrossRef][ISI][Medline]
5 Sunyer J, Anto JM, Rodrigo MJ, Morell F. Case-control study of serum immunoglobulin-E antibodies reactive with soybean in epidemic asthma. Lancet 1989;1:179–82.[CrossRef][ISI][Medline]
6 Anto JM, Sunyer J, Rodriguez-Roisin R, et al. Community outbreaks of asthma associated with inhalation of soybean dust. Toxicoepidemiological Committee. N Engl J Med 1989;320:1097–102.[Abstract]
7 Anto JM, Sunyer J, Reed CE, et al. Preventing asthma epidemics due to soybeans by dust-control measures. N Engl J Med 1993; 329:1760–3.
8 Figley KD, Elrood RH. Endemic asthma due to castor bean dust. JAMA 1928;90:79–82.
9 Shy CM. Epidemiologic evidence and the United States air quality standards. Am J Epidemiol 1979;110:661–71.
10 Holland WW, Bennet AE, Cameron IR, et al. Health effects of air pollution: Reappraising the evidence. Am J Epidemiol 1979;110:527–659.
11 Raffensperger C, Tickner JA, editors. Protecting public health and the environment: implementing the precautionary principle. Washington, DC: Island Press, 1999.
12 Anto JM, Sunyer J, Kogevinas M. Environment and health: the long journey of environmental epidemiology at the turn of the millennium. J Epidemiol Biostat 2000;5:49–60.[Medline]
13 Colborn T, vom Saal FS, Soto AM. Developmental effects of endocrine-disrupting chemicals in wildlife and humans. Environ Health Perspect 1993;101:378–84.[ISI][Medline]
14 Tickner JA. Precaution, environmental science and preventive policy. Washington, DC: Island Press, 2002.
15 Tickner JA. Commentary: barriers and opportunities to changing the research agenda to support precaution and primary prevention. Int J Occup Med Environ Health 2004;17:163–73.[Medline]
16 Weed DL. Precaution, prevention and public health ethics. J Med Philos 2004;29:313–32.[CrossRef][ISI][Medline]
17 Comba P, Martuzzi M, Botti C. Comparison of Bayesian-utilitarian and maximin principle approaches. Int J Occup Med Environ Health 2004;17:193–6.[Medline]
18 McMichael AJ. Prisoners of the proximate: loosening the constraints on epidemiology in an age of change. Am J Epidemiol 1999; 149: 887–97.
19 Silbergeld EK. Commentary: the role of toxicology in prevention and precaution. Int J Occup Med Environ Health 2004;17:91–102.[Medline]
20 Commission of the European Communities. Communication on the precautionary principle. Brussels, 2000.
21 Klinke A, Renn O. A new approach to risk evaluation and management: risk-based, precaution-based, and discourse-based strategies. Risk Anal 2002;22:1071–94.[CrossRef][ISI][Medline]
22 Suk WA, Olden K. Multidisciplinary research: strategies for assessing chemical mixtures to reduce risk of exposure and disease. Int J Occup Med Environ Health 2004;17:103–10.[Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||