Research report summaries 2005–2006

This Web page has been archived on the Web.

Archived Content

Information identified as archived on the Web is for reference, research or recordkeeping purposes. It has not been altered or updated after the date of archiving. Web pages that are archived on the Web are not subject to the Government of Canada Web Standards. As per the Communications Policy of the Government of Canada, you can request alternate formats on the Contact Us page.

Contractors' reports are only available in the language in which they are submitted to the Canadian Nuclear Safety Commission (CNSC).

RSP-0194 – Accidental exposures in radiotherapy survey

Highlighted below are the key findings of the CNSC Accidental Exposures in Radiotherapy Survey (2005). The survey was administered in April 2005. Thirty-nine CNSC licensees (hereafter the radiation therapy cancer centres) across the country were asked to participate.

  • Manufacturer's training is utilized when new equipment is acquired; hardware/software is upgraded, as well as when new staff is recruited or upon request of the radiation therapy cancer centre.
  • In completing the questionnaire, the respondents reported that, in general, an internal training is organized once a training need arises (e.g., when staff is rotated to new units). Only a few radiation therapy cancer centres coordinate periodical internal training sessions.
  • Auditing practices are in place in most of the radiation therapy cancer centres. Each centre follows up cases where an error has occurred through a formal procedure of investigating, reporting, reviewing, communicating recommendations/policy revisions (if any) and training on medical events.
  • As reported in the questionnaires, medical events are predominantly attributable to process as opposed to machine malfunction. Commonly, they are result of under-dose (90 cases or 0.2 percent of the total number of EBRT or 0.008 percent of total number of fractions completed in 2004) and treatment of healthy tissue (77 cases or the 0.2 percent of total number of EBRT or 0.007 percent of total number of fractions completed in 2004) rather than overdose (57 cases or 0.1 percent of the total number of EBRT or 0.0069 percent of total number of fractions completed).

While all of the centres included in this study have identified and implemented some quality control protocols to detect and prevent medical events, there is no standardized systematic protocol to ensure that errors are minimized and tracked for quality improvement purposes.

Top of page

RSP-0195 – Applicability review of fuel and fuel-channel models in thermalhydraulics computer codes for CANFLEX 43-element fuel – Phase 1

This project was initiated to provide a critical review of the models or correlations defining the constitutive relationships in TUF and CATHENA and give an opinion as to the applicability of the use of these models or correlations for safety analysis in scenarios with CANFLEX fuel. The first task was to review the CATHENA and TUF theory manuals to glean from them a list of references thought to be required to meet the overall objectives of the contract. A reference list was provided to the CNSC and as many of the documents as possible were provided. Any reference received was reviewed. Rarely did the cited reference give the information sought by this review. The references did cite many tier-two references which could provide the desired information and some of these references were requested and reviewed. Time did not allow for this type of exploration to identify the empirical basis of all models or correlations in the codes.

Top of page

RSP-0196 – A report on performance demonstration of NDR techniques

The CNSC initiated this project to determine a consistent set of requirements for demonstrating the combined performance of inspector, inspection technique and inspection equipment to reliably inspect CANDU plant components. The CNSC intends to use these requirements as a basis for making recommendations to Canadian nuclear power plant licensees on strengthening inspection requirements, fitness-for-service and plant life management.

Two main methodologies for qualification of inspection systems are used in the nuclear sector worldwide. These approaches are:

  • application of the performance demonstration requirements described in ASME Boiler and Pressure Vessel Code, Section XI, Appendix VIII
  • application of the inspection qualification guidelines produced by the European Network for Inspection Qualification (ENIQ)

An understanding and comparison of these methods is needed to allow accurate assessment of inspection qualification methodologies adopted by Canadian licensees. In order to make a meaningful comparison, it is important to recognize that the inspection requirements of ASME Section XI, Appendix VIII are quite specific and are mandatory in the United States, while the ENIQ process is a set of non-mandatory, more general guidelines which have been interpreted and implemented differently by various European countries.

The scope of the project included:

  • a literature search of certification and regulation requirements in Europe and North America to enable a comparison of the ASME and ENIQ approaches to performance demonstration
  • a survey of nuclear power industry participants in Europe and North America to benchmark the use of performance demonstration regimes based on the ENIQ and ASME methodologies
  • a comparison of the underlying principles and actual application of the ENIQ and ASME performance demonstration methods to identify how either approach could be incorporated into the Canadian certification and regulatory frameworks
  • a comparison of the application of both ENIQ and ASME NDE (nondestructive examination) qualification methodologies to the inspection of power plant components, and
  • determination of an appropriate infrastructure for implementing either an ENIQ and/or ASME performance demonstration program certification process in Canada
This report documents the results of this work. The assessment concludes an inspection qualification process based on the ENIQ approach is the method by which Canadian nuclear power plant licensees can best meet the performance demonstration requirements embodied in CSA N285.4. A consistent approach by all utilities provides for most effective use of regulatory and utility resources and increased opportunity to share qualified NDE systems amongst plant owners.

Consistent with this conclusion the following recommendations are proposed:

  1. The CNSC should endorse the adoption of the ENIQ methodology by the Canadian nuclear sector as a viable means of achieving the performance demonstration requirements in CSA N285.4.
  2. All parties involved in implementing the inspection qualification process should recognize that its existence will be evolutionary. It is more useful to the industry to implement a draft process soon rather than wait for the process to be thoroughly designed before implementation. As the process is utilized and experience is gained (through one or more pilot studies) in qualifying inspection systems, the process will change and evolve.
  3. The CNSC should commission further work to assess and clarify its preferred role in the inspection qualification process. This should be done in consultation with the nuclear utilities and other interested parties and include a detailed assessment of the various regulatory frameworks for inspection qualification employed in European jurisdictions that follow the ENIQ approach.
  4. The CNSC should endorse the creation of an Inspection Qualification Organization (IQO) by the licensees, as described in section 4.4.2 of this report. Inspection qualification bodies formed under the IQO should initially be set up in accordance with ENIQ recommended practice #7, Recommended General Requirements for a Body Operating Qualification of Non-destructive Tests. As the inspection qualification process matures, the need for certification of the IQBs to a recognized standard, such as ISO 17020, should be assessed.
  5. The CNSC should conduct an assessment of the COG proposal for inspection qualification. The assessment should be conducted by individuals with detailed knowledge in, and experience with, the use of the ENIQ methodology.
  6. Several pilot studies using the ENIQ methodology have been performed by the industry. The CNSC and industry should retain the services of individuals with detailed knowledge in, and experience with, the use of the ENIQ methodology to evaluate these pilot studies with a view to identifying possible improvements in the methodology used.

Top of page

RSP-0197 – Assessment of LBB applicability to CANDU primary heat transport piping

OPG's reliance on leak-before-break (LBB) for Darlington was the first and only time that a licensee has used the technology in support of application for a license. CNSC staff has expressed its views on LBB in memos, letters, and license conditions. This report recommends that the CNSC needs documents that are more formal.

The CNSC first permitted leak before break in the licensing of Darlington, to relieve the requirements for protection against pipe whip. This was the first time it had been used to license a reactor anywhere in the world. That application identified certain key elements in the process, the most important of which is the ability to detect and react to leaks. Its use and, in particular, the fracture analysis, has spread to include fitness for service of a number of degrading components such as pressure tubes, feeders, steam generators, erosion/corrosion of piping and IGSCC in stainless steel welds. In the future, licensees will likely use it to justify the reuse of old pressure boundary equipment for plant life extension.

With degrading components, the issues for using leak before break as a part of regulation are different. Ageing can change material properties, not only in the sense of uniform behaviour such as strain ageing or embrittlement but also in a localised way such as with IGSCC, pitting, and FAC near discontinuities. The use of leak before break to justify continued operation with ageing components is more difficult to regulate, because degradation inevitably leads to increasing risk. Unless the regulator accepts leaking as a tolerable end to the degradation regulatory control must be based on some assumed risk. Determining levels of risk for leaking and rupture is complicated and highly speculative.

In this report, I examined the governance. How LBB has been used worldwide to relax the requirements of regulatory bodies; and suggest how these can be integrated into the rules, for pressure boundary integrity, of the CNSC.

Top of page

RSP-0198 – Development of regulatory guidelines on the effectiveness of NPP ageing management programs

This project required the contractor to: perform a survey of international practices for effective ageing management of nuclear power plant systems, structures, or components (SSC); identify and summarize fundamental components of ageing management strategies and programs; and recommend possible regulatory guidelines to facilitate CNSC evaluations of the effectiveness of NPP AMPs within the framework of the CNSC's compliance program.

The report is organized in parts according to the scope. Part I presents a survey of international practices on aeging management consisting of a review of guidance provided by international organizations and a review of regulatory and utility approaches to aeging management of nuclear power plants. Part II recommends regulatory requirements for NPP ageing management programs supplemented by recommendations for meeting these requirements presented in Annex A. In addition, the report identifies and discusses interfaces with related technical areas, i.e., equipment qualification, design basis reconstitution and long term operation. Finally, the report provides for consideration recommendations for future follow up work.

Top of page

RSP-0199 – Regional public opinion surveying on CNSC, nuclear regulatory and nuclear safety awareness

In March 2004, the CNSC commissioned Ipsos-Reid to conduct a telephone survey to assess Canadians' general knowledge, perceptions and attitudes towards nuclear regulation and safety. The primary objective of the present study is to build on the Ipsos-Reid research by targeting the research towards populations most affected, economically and environmentally, by their proximity to large nuclear operations.

Decima conducted a telephone survey with 2,006 respondents between September 21 and 26, 2005, targeting six geographical areas located near nuclear power plants and substantial mining operations in Canada: Point Lepreau, NB; Becancour, QC; Chalk River, ON; Darlington/Pickering, ON; Port Elgin, ON; and North Saskatchewan.

Research results – Comparisons to 2004 general population data

Opinions of residents in the six communities is similar to the opinions of Canadians in general in 2004:

  • Overall, 36 percent of those living near a power plant or mining operation are informed with Canada's nuclear regulation procedures, compared to 33 percent of Canadians in general in 2004.
  • Close to two-thirds (65 percent) of residents in the six communities surveyed are confident that Canada's nuclear industries are effectively regulated for safety, versus 59 percent of Canadians in general in 2004.
  • Approximately 4 percent of residents in the six communities mention the CNSC as being responsible for the regulation of Canada's nuclear industries, versus 2 percent of the general population in 2004.

2005 issues

Three quarters (75 percent) of residents (outside of North Saskatchewan) are either 'very confident' (30 percent or 'somewhat confident' (45 percent) in the safety of the nuclear plant closest to them. A slightly lower proportion (69 percent) feels that they live either 'very close' (26 percent) or 'somewhat close' (43 percent) to a nuclear power plant.

Familiarity with a local plant issue is highest in Point Lepreau, where three quarters (76 percent) of residents are either 'very' (23 percent) or 'somewhat' (53 percent) familiar with the issue presented. It is lowest in Chalk River (34 percent) and Becancour (36 percent). Familiarity levels in Northern Saskatchewan are relatively similar across the three issues presented to these residents, with roughly four in ten indicating they are familiar with each topic.

Approximately one quarter (24 percent) of respondents are familiar with the Canadian Nuclear Safety Commission. Among those who are familiar (n=477), nearly eight in 10 (77 percent) have a positive impression (58 percent).

Top of page

RSP-0200 – Review of the 5-kg corium commissioning test completed in the MFMI Facility

A generic action item (GAI) on the Canadian nuclear sector (AECL, Hydro-Québec, New Brunswick Power, Bruce Power, and Ontario Power Generation) has been raised by the CNSC, on the issue of potential Molten Fuel Moderator Interactions under some postulated accidents in CANDU reactors. The action arises from a long-standing difference between the Canadian nuclear sector and the CNSC. The industry's licensing analysis is based on a model developed by its consultant. The model treats the molten fuel as being finely fragmented as it is ejected at 10 MPa from a ruptured fuel channel, and evaluates the energy transfer from the fine molten particles to the moderator to derive the pressurization inside the calandria vessel. This model is commonly termed the "forced interaction" model. CNSC staff, however, have noted that it is conceivable not all the ejected melt will be finely fragmented as postulated in the forced interaction model. This position was supported by the CNSC's consultant, who contended a steam explosion (i.e., thermal detonation) cannot be ruled out.

To resolve this difference, the Canadian nuclear sector has proposed an experimental program with the objective to confirm the dominant mode of molten fuel-moderator interaction following a severe flow blockage accident in a CANDU reactor.

This project required the contractor to:

  • review the first set of test results obtained from the MFMI tests being carried out by the licensees at Chalk River Nuclear Laboratories (CRNL)
  • explain the test results and their implications regarding the validity of the licensees' model of Molten Fuel Moderator Interaction (MFMI)
  • recommend further actions if any

Top of page

RSP-0201 – Statistical regression model for setting site-specific selenium water quality objectives

Data collected over the period 1990–2004 have been used to assess selenium bioaccumulation in fish of the Athabasca basin of northern Saskatchewan, in particular for sites that are thought to be affected by uranium mining, milling and waste management operations. A dataset comprised of 358 co-located water and fish tissue selenium records has been screened, pre-processed, and statistically analyzed to develop a statistical regression model for setting site-specific water quality objectives. The data have been analyzed for species and site effects and for representativeness. Fourteen recommendations for future sampling have been made. Real-world complexities that affect model predictive power are identified and discussed.

The model projects the water selenium concentration required at a specific site to meet a tissue selenium concentration threshold. The distribution of tissue selenium concentrations measured at project area reference sites has been derived. The upper tail of the reference site distribution (99th percentile =7.2 mg/kgdry wt) or the maximum concentration in the reference site dataset (10.9 mg/kgdry wt) are possible values for the tissue selenium concentration threshold. Alternatively a toxicologically based threshold (e.g., the US EPA draft criterion of 7.91 mg/kgdry wt) or a tissue threshold selected by other means could be used. The water selenium concentration required at a specific site to meet a tissue selenium concentration threshold is referred to as a "site-specific water quality objective." A lower limit of 0.5 g/L has been set for the site-specific water quality objective based on evidence of selenium homeostasis at lower concentrations.

The model has been used to produce a look-up table and automated lookup tool. Either of these tools can be used, with site-specific water and tissue selenium concentration data, to project site-specific water quality objectives. The lookup tool will automatically account for any future updates to the database, model or tissue threshold. The look-up table would require manual updating to account for model or threshold changes.

Key assumptions in the model for projecting site-specific water quality objectives are (1) that site-specific data accurately reflect the site's sensitivity to selenium bioaccumulation in fish, and (2) that site sensitivity will remain constant over time. If actions taken to help meet the tissue selenium concentration threshold reduce the site's sensitivity to selenium bioaccumulation – i.e., by reducing selenium bioavailability or exposure of organisms in the aquatic food chain to bioavailable selenium – the model will set the site-specific water quality objective at a lower concentration than would be needed to meet the tissue selenium concentration threshold. The accuracy of site-specific bioaccumulation estimates hinges on sampling design issues, addressed through sampling recommendations provided in this report.

Four key technical management issues are identified and discussed at the end of the report. They are: (1) establishing the tissue threshold selenium concentration, (2) selecting the fish species to be sampled for comparing sites to the tissue threshold, (3) choosing a sampling statistic to use for estimating site sensitivity to selenium bioaccumulation and (4) anticipating and managing for changes over time in site sensitivity to selenium bioaccumulation.

Top of page

RSP-0202 – Review of thermal mitigation technologies for nuclear generating stations

The CNSC is responsible for regulating nuclear power plants in Canada. There are eight Canadian nuclear power facilities located in Ontario, Québec and New Brunswick. Each facility uses once-through cooling water to discharge waste heat from power production. Thermal effluent, caused by condenser cooling water discharge, is subject to Canadian and provincial laws and regulations (e.g., it may considered to be a deleterious substance under the Fisheries Act or may cause an adverse environmental effect under the Canadian Environmental Assessment Act). Thermal effluent discharge and technology options to control thermal discharges have recently emerged as regulatory considerations for CNSC staff.

Consequently, the CNSC has determined that it requires an independent source of information that can be used as an evaluation tool to support its regulatory position regarding discharge of thermal effluent and potential technologies that could be used to avoid or mitigate potential effects. Golder Associates Ltd. was retained to prepare this report, intended to be a reference document, which could be used as a tool to evaluate licensee submissions on proposed mitigation measures for thermal discharges from Canadian nuclear facilities.

A preliminary identification of potential remediation technologies was completed in the initial literature search conducted prior to the site visits. In the next phase of the project, a more detailed literature search and review was conducted to provide further information into each of the potential remediation technologies. One outcome of the site visits and meetings was identification of relevant laws and regulations applicable to the various facilities. Golder staff obtained copies of the various laws and regulations, and reviewed this material. The findings from the review are presented in section 4 of this report. More detailed information on potential technologies was obtained by literature searches on the internet and from Golder contacts. This information was used to prepare the evaluation contained in section 5 of this report. Golder staff in the United States provided references, data and information regarding thermal control technologies and U.S. regulatory requirements. Similarly, information was collected to provide an overview of experience at facilities internationally and presented in section 6 of this report.

A comparative evaluation matrix was developed (see Table 7.3.4-1, for example) to compare and identify best available technologies for the remediation of thermal discharges at Canadian nuclear facilities. Based on an understanding of site-specific conditions, regulatory requirements, potential environmental effects and potential remediation technologies, the authors first identified a long list of potential remediation options. Next a screening was conducted to eliminate options that were not considered to be technically or economically feasible. Next the authors conducted a comparative evaluation of potential remediation options for each of the eight facilities. Using technical, environmental and economic criteria, the authors rated each identified option for suitability at each facility. Criteria were rated "high", "medium" or "low". Finally, the best available technologies were identified based on the above-mentioned rating.

Top of page

RSP-0203 – Definition of a human performance program to support regulatory activities for the Canadian Nuclear Safety Commission

The overall objective of this project was to provide the basis of a regulatory framework to enable the CNSC to assess the adequacy of licensee's human performance programs (HPPs), and their effectiveness in implementing these programs. To meet this objective, definitions were drafted of human performance and human factors suitable for use by the CNSC.

A literature survey was conducted regarding human performance programs implemented by the nuclear sector and other high reliability industries, regulatory frameworks and international best practices used to regulate human performance programs for nuclear facilities, and to identify elements of a human performance program that are suitable for nuclear facilities. Structured interviews were also carried out with subject matter experts (SMEs) in the definition or administration of Human Performance Programs.

Based on the literature survey and the structured interviews, recommendations were made regarding human performance and human factors definitions, HPP elements, a suitable regulatory framework, and an approach for developing human performance indicators. Fifteen HPP elements were recommended, categorized by four functional groupings:

Policy and organisation

  • human performance policy and guidance
  • accountability structure for human performance
  • effective communication
  • effective procedures

Human performance monitoring and response

  • assessment and monitoring of system performance
  • human performance problem identification
  • human performance problem investigation
  • corrective action program
  • performance improvement initiatives

Human performance traps and tools

  • awareness of human performance traps
  • human performance toolset
  • human performance toolset implementation

Organisational learning and knowledge management

  • knowledge management
  • succession planning
  • change management process

For each of the fifteen HPP elements, it was recommended that anchored rating scales be developed to evaluate the elements of human performance programs where each scale value is described by specific examples of relevance to the element under consideration. Anchored rating scales have been successfully used to assess capability maturity, e.g., of software development processes, safety culture and human factors, whereby increasing maturity is characterised by increasing rigour, depth and consistency of practice and performance. These anchored rating scales will need to undergo verification and the framework should undergo validation by field testing.

Top of page

RSP-0204 – Uranium in effluent treatment process

Technologies for removal of uranium from uranium mine and mill effluents have been reviewed. The results of current practice and the potential for process and technology improvements are discussed.

Uranium was identified in the 2003 Environmental and Health Canada Priority Substance List report as having the potential for environmental effect at the reported concentrations in discharges from some of the older operating mine facilities in Canada. The CNSC has initiated investigations to determine whether it is possible to lower release concentrations.

Uranium removal has not typically been a specific target of water treatment strategies in the uranium industry. The focus has been on radium-226, acidity, and with the development of the high-grade mines in Saskatchewan, on nickel, cobalt, molybdenum and arsenic. The reduction of uranium in process and waste management streams to lower levels in effluents has been a side benefit of the lime neutralisation of acidity, the use of ferric ion to remove arsenic and the removal of suspended solids.

Control at source is a key strategy in achieving low levels in effluents. Uranium in mill process discharge is reduced to a level as low as economically achievable by reducing soluble losses to a minimum. Soluble losses primarily originate from two process streams – (1) unrecovered soluble uranium from CCD (counter current decantation) circuits and (2) residual uranium in the solvent extraction (SX) barren. Both streams typically contain uranium in the range of µg/L to mg/L. Further reduction in uranium levels in either stream would involve the installation of additional stages in the CCD and SX circuits and that is usually not economical or physically feasible.

Other potential sources of uranium in final effluents – mine water, tailings drainage and waste rock seepage can contain mg/L levels of uranium (and other contaminants) but there are few opportunities to reduce the uranium content at source.

This study reviewed the available information from Canada and from selected foreign uranium mine facilities in order to gauge uranium-in-effluent levels and what technology has been developed and applied to remove uranium. The potential target concentrations of uranium-specific technologies are undefined at this time, but as a reference point for this study, it was assumed that the total measured uranium in effluents should be reduced to less than 100 µg/L U or as low as reasonably achievable (ALARA) given technical and economic considerations.

It has been determined that use of calcium hydroxide (lime) in combination with ferric sulphate and flocculants offers the best basis for achieving a consistently low level of uranium in mine effluents. For each mine facility, the process would be adapted to local conditions, the presence of carbonate and/or bicarbonate, suspended solids and other constituents to be removed - such as radium-226, nickel, molybdenum, arsenic and ammonia.

For low dissolved salts streams, reverse osmosis (RO) can be used to produce very low levels of uranium in discharges. This technology is in place at Key Lake treating water containing uranium and nickel from mine perimeter dewatering wells. Ion exchange (IX) offers a treatment option for water streams containing mg/L levels of uranium and containing difficult to treat uranyl carbonate complexes. Cameco is currently investigating the potential of IX to remove uranium at the Rabbit Lake facility. However, lime and ferric sulphate treatment technology will also be needed at the same mine site to treat acidity and remove other contaminants.

Reports of results from other unconventional (to the mining industry) treatment technology such as sorption, reductive and biological precipitation do not appear to indicate significant potential for these technologies at present.

Top of page

RSP-0205 – Eldorado nuclear epidemiology study update – Eldorado uranium miners' cohort: Part I of the Saskatchewan uranium miners' cohort study

A cohort study has been completed in which records for uranium miners and processors who worked for Eldorado Nuclear Limited have been linked to national mortality records (1950–1999) and national cancer incidence records (1969–1999). This study, thus, updates an earlier study in which mortality in the cohort between 1950 and 1980 was ascertained.

This report presents the results of the statistical analysis of a cohort of 17,660 individuals known to have worked for Eldorado somewhere between 1930 and 1999. Based on a total of 5,332 deaths between 1950 and 1999, and 2,355 individuals who developed at least one cancer between 1969 and 1999, several types of analyses have been conducted.

The first analysis was a comparison of the mortality of the cohort with the mortality of the general Canadian population between 1950 and 1999. Lung cancer was elevated not only in the whole cohort, but in various sub-cohorts defined by gender, site of working, underground and mill workers and sub-cohorts defined by first working date for Eldorado. There can be no doubt that much of this excess is attributable to exposure to radioactive radon decay products (RDP) as discussed in more detail subsequently.

For most other the causes of death, the cohort as a whole and the various sub-cohorts had reduced risks relative to the population. This probably represents the healthy worker effect, a supposition which is supported, for example, by the major decrease in ischaemic heart disease to reflect the fact that the risk of heart disease is lower in the cohort being studied than in the population at large. The latter condition would probably prevent people working in a strenuous physical occupation such as mining.

The most notable exceptions for causes of death where rates in the Eldorado cohort were elevated relative to the population include hypertensive causes and external causes such as homicides, suicides and both traffic and non-traffic-related accidents.

Comparisons of the cohort with the general Canadian population with respect to cancer incidence rates between 1969 and 1999 are also reported. The only cancer that is consistently elevated is lung cancer, thus mirroring the mortality results discussed above.

The second analysis presented in the present report is that of mortality from lung cancer with respect to RDP exposure. This is primarily based on 618 lung cancer deaths amongst men in the cohort. This compares with previous analyses of the Eldorado cohort when the total number of such deaths was 122. Thus, the present analysis represents a substantial increase in the power of the study and, thus, should produce more precise estimates.

There is a strong positive monotonic increase in risk of lung cancer death with increasing RDP exposure which is highly statistically significant. This increase generally manifests itself for the three main sites (Port Hope, Port Radium and Beaverlodge sites), although it is likely that the exposure of the Port Hope sub-cohort comes primarily from exposure received at other sites. However, fitting a simple linear excess relative risk model to the data although, again, demonstrating a strong relationship with RDP exposure does provide excess relative risk estimates that are inconsistent across the sites.

Application of the BEIR VI-type risk model which allows for effect modification by time since exposure, exposure rate and age at risk reduces this statistical heterogeneity in terms of the RDP effect and it is clear that these modifying factors contribute to the apparent heterogeneity seen in the simple linear excess relative risk model. Using the same parameterization as chosen by the BEIR VI Committee leads to coefficient estimates that are very similar in the present study as reported by the BEIR VI Committee.

The lung cancer mortality analysis detected no effect of gamma ray exposure on risk of lung cancer mortality and all estimates did not change by including or excluding those with non-Eldorado work histories or those with zero exposure to RDP.

An analysis of lung cancer incidence amongst males was also conducted. The results mirror those of the mortality analysis and as relative measures are used, this is hardly surprising. It should also be noted that the two analyses are not independent in that a substantial proportion of cancer cases contributed to the corresponding death analysis.

The final type of analysis conducted was that examining mortality and cancer incidence for diseases other than lung cancer with RDP exposure and gamma ray exposure. In summary, there is no meaningful evidence of any causal relationship between RDP exposure and increased risk of any of these other diseases; nor was there any meaningful evidence of a relationship with gamma ray dose.

Several limitations should be borne in mind when considering the above results. No data were available on smoking or other possible carcinogens among the cohort. Further, measurement error in exposure estimates could not be taken into account. The implication of these limitations is also discussed in the report.

Exposure to RDP is one of the best-studied carcinogenic phenomena in epidemiology. The results obtained from these studies, primarily of underground miners, are very consistent in showing increases in lung cancer risk from such exposure, but no increase in any other disease.

The present study (which is essentially independent of the data set used by the BEIR VI Committee) further supports these conclusions based on 50 years' mortality experience and 31 years of cancer incidence experience. They certainly support the use of BEIR VI-type models to predict any group's future risk of lung cancer from RDP exposure either from past or current such exposure.

Finally, it is worth noting that as yet only about 25 percent of the cohort has died. Future follow-up and analysis of this cohort with respect to both mortality and cancer incidence should shed further light on our knowledge on the effects of uranium mining and processing in both Saskatchewan and other provinces upon the resulting health of those employed in such occupations.

Date modified: