Methods to account for uncertainties in exposure assessment in studies of environmental exposures.

BACKGROUND: Accurate exposure estimation in environmental epidemiological studies is crucial for health risk assessment. Failure to account for uncertainties in exposure estimation could lead to biased results in exposure-response analyses. Assessment of the effects of uncertainties in exposure estimation on risk estimates received a lot of attention in radiation epidemiology and in several studies of diet and air pollution. The objective of this narrative review is to examine the commonly used statistical approaches to account for exposure estimation errors in risk analyses and to suggest how each could be applied in environmental epidemiological studies. MAIN TEXT: We review two main error types in estimating exposures in epidemiological studies: shared and unshared errors and their subtypes. We describe the four main statistical approaches to adjust for exposure estimation uncertainties (regression calibration, simulation-extrapolation, Monte Carlo maximum likelihood and Bayesian model averaging) along with examples to give readers better understanding of their advantages and limitations. We also explain the advantages of using a 2-dimensional Monte-Carlo (2DMC) simulation method to quantify the effect of uncertainties in exposure estimates using full-likelihood methods. For exposures that are estimated independently between subjects and are more likely to introduce unshared errors, regression calibration and SIMEX methods are able to adequately account for exposure uncertainties in risk analyses. When an uncalibrated measuring device is used or estimation parameters with uncertain mean values are applied to a group of people, shared errors could potentially be large. In this case, Monte Carlo maximum likelihood and Bayesian model averaging methods based on estimates of exposure from the 2DMC simulations would work well. The majority of reviewed studies show relatively moderate changes (within 100%) in risk estimates after accounting for uncertainties in exposure estimates, except for the two studies which doubled/tripled naïve estimates. CONCLUSIONS: In this paper, we demonstrate various statistical methods to account for uncertain exposure estimates in risk analyses. The differences in the results of various adjustment methods could be due to various error structures in datasets and whether or not a proper statistical method was applied. Epidemiological studies of environmental exposures should include exposure-response analyses accounting for uncertainties in exposure estimates.

Organ-specific dose coefficients derived from Monte Carlo simulations for historical (1930s to 1960s) fluoroscopic and radiographic examinations of tuberculosis patients.

This work provides dose coefficients necessary to reconstruct doses used in epidemiological studies of tuberculosis patients treated from the 1930s through the 1960s, who were exposed to diagnostic imaging while undergoing treatment. We made use of averaged imaging parameters from measurement data, physician interviews, and available literature of the Canadian Fluoroscopy Cohort Study and, on occasion, from a similar study of tuberculosis patients from Massachusetts, United States, treated between 1925 and 1954. We used computational phantoms of the human anatomy and Monte Carlo radiation transport methods to compute dose coefficients that relate dose in air, at a point 20 cm away from the source, to absorbed dose in 58 organs. We selected five male and five female phantoms, based on the mean height and weight of Canadian tuberculosis patients in that era, for the 1-, 5-, 10-, 15-year old and adult ages. Using high-performance computers at the National Institutes of Health, we simulated 2,400 unique fluoroscopic and radiographic exposures by varying x-ray beam quality, field size, field shuttering, imaged anatomy, phantom orientation, and computational phantom. Compared with previous dose coefficients reported for this population, our dosimetry system uses improved anatomical phantoms constructed from computed tomography imaging datasets. The new set of dose coefficients includes tissues that were not previously assessed, in particular, for tissues outside the x-ray field or for pediatric patients. In addition, we provide dose coefficients for radiography and for fluoroscopic procedures not previously assessed in the dosimetry of this cohort (i.e. pneumoperitoneum and chest aspirations). These new dose coefficients would allow a comprehensive assessment of exposures in the cohort. In addition to providing newly derived dose coefficients, we believe the automation and methods developed to complete these dosimetry calculations are generalizable and can be applied to other epidemiological studies interested in an exposure assessment from medical x-ray imaging. These epidemiological studies provide important data for assessing health risks of radiation exposure to help inform the current system of radiological protection and efforts to optimize the use of radiation in medical studies.

RadRAT: a radiation risk assessment tool for lifetime cancer risk projection.

Risk projection methods allow for timely assessment of the potential magnitude of radiation-related cancer risks following low-dose radiation exposures. The estimation of such risks directly through observational studies would generally require infeasibly large studies and long-term follow-up to achieve reasonable statistical power. We developed an online radiation risk assessment tool (RadRAT) which can be used to estimate the lifetime risk of radiation-related cancer with uncertainty intervals following a user-specified exposure history (https://irep.nci.nih.gov/radrat). The uncertainty intervals constitute a key component of the program because of the various assumptions that are involved in such calculations. The risk models used in RadRAT are broadly based on those developed by the BEIR VII committee for estimating lifetime risk following low-dose radiation exposure of the US population for eleven site-specific cancers. We developed new risk models for seven additional cancer sites, oral, oesophagus, gallbladder, pancreas, rectum, kidney and brain/central nervous system (CNS) cancers, using data from Japanese atomic bomb survivors. The lifetime risk estimates are slightly higher for RadRAT than for BEIR VII across all exposure ages mostly because the weighting of the excess relative risk and excess absolute risk models was conducted on an arithmetic rather than a logarithmic scale. The calculator can be used to estimate lifetime cancer risk from both uniform and non-uniform doses that are acute or chronic. It is most appropriate for low-LET radiation doses < 1 Gy, and for individuals with life-expectancy and cancer rates similar to the general population in the US.

Fluoroscopy X-Ray Organ-Specific Dosimetry System (FLUXOR) for Estimation of Organ Doses and Their Uncertainties in the Canadian Fluoroscopy Cohort Study.

As part of ongoing efforts to assess lifespan disease mortality and incidence in 63,715 patients from the Canadian Fluoroscopy Cohort Study (CFCS) who were treated for tuberculosis between 1930 and 1969, we developed a new FLUoroscopy X-ray ORgan-specific dosimetry system (FLUXOR) to estimate radiation doses to various organs and tissues. Approximately 45% of patients received medical procedures accompanied by fluoroscopy, including artificial pneumothorax (air in pleural cavity to collapse of lungs), pneumoperitoneum (air in peritoneal cavity), aspiration of fluid from pleural cavity and gastrointestinal series. In addition, patients received chest radiographs for purposes of diagnosis and monitoring of disease status. FLUXOR utilizes age-, sex- and body size-dependent dose coefficients for fluoroscopy and radiography exams, estimated using radiation transport simulations in up-to-date computational hybrid anthropomorphic phantoms. The phantoms include an updated heart model, and were adjusted to match the estimated mean height and body mass of tuberculosis patients in Canada during the relevant time period. Patient-specific data (machine settings, exposure duration, patient orientation) used during individual fluoroscopy or radiography exams were not recorded. Doses to patients were based on parameter values inferred from interviews with 91 physicians practicing at the time, historical literature, and estimated number of procedures from patient records. FLUXOR uses probability distributions to represent the uncertainty in the unknown true, average value of each dosimetry parameter. Uncertainties were shared across all patients within specific subgroups of the cohort, defined by age at treatment, sex, type of procedure, time period of exams and region (Nova Scotia or other provinces). Monte Carlo techniques were used to propagate uncertainties, by sampling alternative average values for each parameter. Alternative average doses per exam were estimated for patients in each subgroup, with the total average dose per individual determined by the number of exams received. This process was repeated to produce alternative cohort vectors of average organ doses per patient. This article presents estimates of doses to lungs, female breast, active bone marrow and heart wall. Means and 95% confidence intervals (CI) of average organ doses across all 63,715 patients were 320 (160, 560) mGy to lungs, 250 (120, 450) mGy to female breast, 190 (100, 340) mGy to heart wall and 92 (47, 160) mGy to active bone marrow. Approximately 60% of all patients had average doses to the four studied organs of less than 10 mGy, 10% received between 10 and 100 mGy, 25% between 100 and 1,000 mGy, and 5% above 1,000 mGy. Pneumothorax was the medical procedure that accounted for the largest contribution to cohort average doses. The major contributors to uncertainty in estimated doses per procedure for the four organs of interest are the uncertainties in exposure duration, tube voltage, tube output, and patient orientation relative to the X-ray tube, with the uncertainty in exposure duration being most often the dominant source. Uncertainty in patient orientation was important for doses to female breast, and, to a lesser degree, for doses to heart wall. The uncertainty in number of exams was an important contributor to uncertainty for ∼30% of patients. The estimated organ doses and their uncertainties will be used for analyses of incidence and mortality of cancer and non-cancer diseases. The CFCS cohort is an important addition to existing radio-epidemiological cohorts, given the moderate-to-high doses received fractionated over several years, the type of irradiation (external irradiation only), radiation type (X rays only), a balanced combination of both genders and inclusion of people of all ages.

Organ Doses from Chest Radiographs in Tuberculosis Patients in Canada and Their Uncertainties in Periods from 1930 to 1969.

This paper describes a study to estimate absorbed doses to various organs from film-based chest radiographs and their uncertainties in the periods 1930 to 1948, 1949 to 1955, and 1956 to 1969. Estimated organ doses will be used in new analyses of risks of cancer and other diseases in tuberculosis patients in Canada who had chest fluoroscopic and radiographic examinations in those periods. In this paper, doses to lungs, female breast, active bone marrow, and heart from a single chest radiograph in adults and children of ages 1, 5, 10, and 15 y in the Canadian cohort and their uncertainties are estimated using (1) data on the tube voltage (kV), total filtration (mm Al), tube-current exposure-time product (mA s), and tube output (mR [mA s]) in each period; (2) assumptions about patient orientation, distance from the source to the skin of a patient, and film size; and (3) new calculations of sex- and age-specific organ dose conversion coefficients (organ doses per dose in air at skin entrance). Variations in estimated doses to each organ across the three periods are less than 20% in adults and up to about 30% at younger ages. Uncertainties in estimated organ doses are about a factor of 2 to 3 in adults and up to a factor of 4 at younger ages and are due mainly to uncertainties in the tube voltage and tube-current exposure-time product.

2004 update of dosimetry for the Utah Thyroid Cohort Study.

In the 1980s, individual thyroid doses and uncertainties were estimated for members of a cohort of children identified in 1965 in Utah and Nevada who had potentially been exposed to fallout from the Nevada Test Site. That reconstruction represented the first comprehensive assessment of doses received by the cohort and was the first large effort to assess the uncertainty of dose on an individual person basis. The data on dose and thyroid disease prevalence during different periods were subsequently used in an analysis to determine risks of radiogenic thyroid disease. This cohort has received periodic medical follow-up to observe changes in disease frequency and to reassess the previously reported radiation-related risks, most recently after a Congressional mandate in 1998. In a recent effort to restore the databases and computer codes used to estimate doses in the 1980s, various deficiencies were found in the estimated doses due to improperly operating computer codes, corruption of secondary data files, and lack of quality control procedures. From 2001 through 2004, the dosimetry system was restored and corrected and all doses were recalculated. In addition, two parameter values were updated. While the mean of all doses has not changed significantly, many individual doses have changed by more than an order of magnitude.

A perspective on public concerns about exposure to fallout from the production and testing of nuclear weapons.

Exposures of the American public occurred nationwide from the testing of nuclear weapons in the United States, the Pacific, and the former Soviet Union. After decades of diminished public awareness on the subject of health risks resulting from exposure to fallout, the release of the National Cancer Institute's 1997 report on nationwide exposure to 131I from the Nevada Test Site (NTS) has led to renewed interest. Public requests for information are focused on individual and family health problems, the right to credible and full disclosure of information, and the need for medical care and assistance for exposure-related health problems. Public concerns have been raised regarding: (a) the lack of information on the potential health risks from exposure to all biologically significant radionuclides in fallout; (b) the lack of independent oversight that includes public participation; (c) governmental portrayal of exposures averaged over very large segments of the population without identification of much larger values for individuals or population subgroups likely to be at highest risk; and (d) a governmental response to known or suspected human exposures that consumes large periods of time and devotes considerable funding to various research-related activities before serious consideration is given to addressing health care responsibilities to exposed individuals. To some extent, these complaints and concerns are rooted in the legacy of government secrecy surrounding the development and testing of nuclear weapons, public distrust of government sources of information about radiation exposures and health risks, and the imposition of past exposures without informed consent. Members of the public participating in the oversight of dose reconstruction projects and epidemiologic studies are requesting information on the total impact from all relevant sources of exposure at each site that might contribute significantly to an individual's risk, including exposure to local releases and to NTS and global fallout. Information is being requested on individual doses and risks from these cumulative exposures, with estimates of uncertainty, including estimates of the absorbed organ dose (as opposed to the effective dose), the risk of disease incidence as opposed to the risk of a cancer fatality, and the chance that a person's diagnosed disease was caused by past exposure (i.e., the probability of causation). This paper attempts to address some of these concerns. We conclude by noting that many individuals exposed in childhood during the 1950's to 131I in fallout from nuclear weapons production and testing would qualify for compensation and medical care if the present rules for the adjudication of claims for atomic veterans and radiation workers at DOE sites were to be extended to the public.

Estimation of internal exposure to uranium with uncertainty from urinalysis data using the InDEP computer code.

The National Institute for Occupational Safety and Health (NIOSH) is currently studying mortality in a cohort of 6409 workers at a former uranium processing facility. As part of this study, over 220 000 urine samples were used to reconstruct organ doses due to internal exposure to uranium. Most of the available computational programs designed for analysis of bioassay data handle a single case at a time, and thus require a significant outlay of time and resources for the exposure assessment of a large cohort. NIOSH is currently supporting the development of a computer program, InDEP (Internal Dose Evaluation Program), to facilitate internal radiation exposure assessment as part of epidemiological studies of both uranium- and plutonium-exposed cohorts. A novel feature of InDEP is its batch processing capability which allows for the evaluation of multiple study subjects simultaneously. InDEP analyses bioassay data and derives intakes and organ doses with uncertainty estimates using least-squares regression techniques or using the Bayes' Theorem as applied to internal dosimetry (Bayesian method). This paper describes the application of the current version of InDEP to formulate assumptions about the characteristics of exposure at the study facility that were used in a detailed retrospective intake and organ dose assessment of the cohort.

Interactive RadioEpidemiological Program (IREP): a web-based tool for estimating probability of causation/assigned share of radiogenic cancers.

The Interactive RadioEpidemiological Program (IREP) is a Web-based, interactive computer code that is used to estimate the probability that a given cancer in an individual was induced by given exposures to ionizing radiation. IREP was developed by a Working Group of the National Cancer Institute and Centers for Disease Control and Prevention, and was adopted and modified by the National Institute for Occupational Safety and Health (NIOSH) for use in adjudicating claims for compensation for cancer under the Energy Employees Occupational Illness Compensation Program Act of 2000. In this paper, the quantity calculated in IREP is referred to as "probability of causation/assigned share" (PC/AS). PC/AS for a given cancer in an individual is calculated on the basis of an estimate of the excess relative risk (ERR) associated with given radiation exposures and the relationship PC/AS = ERR/ERR+1. IREP accounts for uncertainties in calculating probability distributions of ERR and PC/AS. An accounting of uncertainty is necessary when decisions about granting claims for compensation for cancer are made on the basis of an estimate of the upper 99% credibility limit of PC/AS to give claimants the "benefit of the doubt." This paper discusses models and methods incorporated in IREP to estimate ERR and PC/AS. Approaches to accounting for uncertainty are emphasized, and limitations of IREP are discussed. Although IREP is intended to provide unbiased estimates of ERR and PC/AS and their uncertainties to represent the current state of knowledge, there are situations described in this paper in which NIOSH, as a matter of policy, makes assumptions that give a higher estimate of the upper 99% credibility limit of PC/AS than other plausible alternatives and, thus, are more favorable to claimants.