Federal Contaminated Site Risk Assessment In Canada Part I: Guidence on Human Health Preliminary Quantitive Risk Assessment (PQRA)
1. Introduction
Risk assessment, whether at the screening level (i.e., preliminary) or more complex, is not an exact science. A wide variety of advice and direction is offered by international, national and provincial/territorial environmental agencies regarding the conduct of risk assessments, and different risk assessors access and rely on the available regulatory advice and direction differently. This results in extensive variability in the estimates of chemical exposure and risk. For example, in 1997, the Canada Mortgage and Housing Corporation (CMHC) commissioned a study whereby nine consulting firms were contracted to estimate the risks posed by a contaminated residential property. The resulting estimates of exposure and risk produced by the different firms varied over nine orders of magnitude for non-cancer endpoints and over 10 orders of magnitude for cancer, despite being given the same site data set. The large variability related primarily to the differing receptors and exposure scenarios assumed by the different firms. Variability was also introduced by the selection of different toxicological reference values (TRVs) for risk characterization.
Likewise, a comparison of 10 preliminary quantitative risk assessments conducted on behalf of Fisheries and Oceans Canada (Risklogic, 2003) revealed widely differing approaches, assumptions, and risk-related conclusions, despite the fact that all 10 sites were similar in land use and public access. The toxicological reference value for just one contaminant, evaluated at all 10 sites, varied by a factor of five among different consulting firms. Numerous other variables and assumptions also varied widely, both among consulting firms, and in one case within the same firm, making it virtually impossible to rely on (at face value) and compare the conclusions among sites and reports with respect to the presence or absence of human health risk, without further analysis and recalculation.
Provincial regulatory agencies across Canada offer differing guidance on many aspects of risk assessment. For example, definitions of acceptable cancer risk vary (BC, Alberta and the Atlantic provinces accept an incremental lifetime cancer risk of 1 x 10-5, while Ontario targets 1 x 10-6). When characterizing the risks posed by exposure to non-carcinogenic substances, British Columbia accepts a Hazard Quotient of 1, while Alberta and Ontario target 0.2. Provinces also differ in their preferred statistics for exposure calculations, varyingly prescribing the maximum contaminant concentration, the 95% upper confidence limit of the mean concentration, or the 90th percentile or 95th percentile of the concentration data distribution.
Based on the above observations, it became apparent that standardized guidance was required at the federal level to assist with the consistent assessment of risks posed by contaminated sites under federal custodianship across the country.
1.1 Background
In 2003 the federal government established the Federal Contaminated Sites Accelerated Action Plan (FCSAAP), a new contaminated sites initiative to assist in identifying, assessing and managing the risks at contaminated properties under the custodial care of Canadian federal government departments.
A major emphasis of the FCSAAP is to give priority for remediation or risk management to those sites and properties posing the greatest risks. The purpose of a preliminary quantitative risk assessment (PQRA) is to quantify the degree of potential human health risk posed by the presence of contamination at a subject site. The results of a PQRA for federal sites/properties may be used by Health Canada to rank and prioritize the subject site for remedial funding under the FCSAAP. As a result, with the current disparity in risk assessment methods, there is a need for standardized risk assessment guidance that will ensure that all federal sites are evaluated for that priority on an equal and defensible basis.
Preliminary quantitative risk assessments generally prescribe methods and assumptions that ensure that exposures and risks are not underestimated. In this way, if negligible or acceptable risks are indicated using these conservative methods, then actual site use patterns and conditions will almost certainly present negligible or acceptable risks. However, the converse is not necessarily true; where PQRA suggests a potential for unacceptable risks, this does not immediately indicate that actual site conditions are unacceptable. Often, further assessment may be necessary to resolve conservatism and uncertainty in the PQRA process before the actual extent of the health risk can be fully quantified and defined.
When risk management strategies are implemented on the basis of the results of a PQRA, the remediated or managed site conditions will almost certainly achieve a reduction in health risk that was greater than might have otherwise been necessary if the on-site risks had been more extensively and accurately ascertained. It becomes a question of cost and feasibility of risk management action when deciding whether to implement remediation on the basis of a PQRA or to further reduce risk assessment uncertainties at a given site before defining the most suitable risk management strategy.
1.2 Purpose
The purpose of this guidance document is to prescribe, to the degree possible, standard exposure pathways, receptor characteristics, toxicological reference values, and other parameters required to quantitatively assess the potential chemical exposures and risks at federal contaminated sites.
The standard PQRA approach presented herein is designed specifically for the assessment of sites that are to remain the properties of federal agencies, properties for which greater consistency in risk assessment methods and interpretation of results is required. For properties being divested to a private party or to provincial governments, or for assessments that address risks from off-site migration of contamination (to an adjacent provincial water body or neighbouring private property, for example), risk assessments may have to be completed in accordance with local provincial/territorial regulatory requirements. Local regulatory requirements may differ from the standardized methods described in this guidance document. When the methods being employed in such cases differ significantly from those presented in this document, risk assessors should identify those assumptions, methods, and interpretations required by provincial agencies that differ from this method, and discuss the implications for the custodial department.
At first glance this guidance may seem overly demanding. However, the length of this document stems predominantly from the inclusion of explanatory text to ensure that the guidance is understood. In other words, an attempt has been made to describe why the methods are requested, not just to delineate those methods.
Most risk assessors have standard spreadsheets containing the various equations, assumptions, TRVs, etc., that they routinely use for risk assessments. The primary requirement for federal sites is to ensure that those spreadsheets comply with the prescribed equations, assumptions, TRVs, etc., outlined herein. Health Canada is flexible on the format and presentation of data and results, as long as the key components described below are included.
Although the guidance offered here is prescriptive in nature, it is not designed or intended as a substitute for the sound professional judgement of a qualified and experienced risk assessment practitioner. It is recognized that many sites will present unique situations not specifically addressed here. Risk assessors are encouraged to ensure that their assessments are complete and that they address all relevant risks. The methods delineated below should not be viewed as a "black box" of equations and assumptions that negate the need for sound professional judgement. However, where possible and appropriate, the guidance provided here should be used. Where alternate or unique approaches have been determined to be necessary, these must be sufficiently documented and described to enable peer review, and must be evaluated for their impact on risk estimates relative to the application of the standard methods prescribed below.
The guidance that follows is organized according to subject areas that Health Canada wants included in the final report. However, it is recognized that different writing styles or corporate standard formats may differ somewhat from those of the outline presented below. Alternate formats are acceptable as long as all of the requested information is presented.
|
