Evaluating uncertainty to strengthen epidemiologic data for use in human health risk assessments

Systematic update to the mammalian relative potency estimate database and development of best estimate toxic equivalency factors for dioxin-like compounds

The World Health Organization (WHO) assesses potential health risks of dioxin-like compounds using Toxic Equivalency Factors (TEFs). This study systematically updated the relative potency (REP) database underlying the 2005 WHO TEFs and applied advanced methods for quantitative integration of study quality and dose-response. Data obtained from fifty-one publications more than doubled the size of the previous REP database (∼1300 datasets). REP quality and relevance for these data was assessed via application of a consensus-based weighting framework. Using Bayesian dose-response modeling, available data were modeled to produce standardized dose/concentration-response Hill curves. Study quality and REP data were synthesized via Bayesian meta-analysis to integrate dose/concentration-response data, author-calculated REPs and benchmark ratios. The output is a prediction of the most likely relationship between each congener and its reference as model-predicted TEF uncertainty distributions, or the 'best estimate TEF' (BE-TEF). The resulting weighted BE-TEFs were similar to the 2005 TEFs, though provide more information to inform selection of TEF values as well as to provide risk assessors and managers with information needed to quantitatively characterize uncertainty around TEF values. Collectively, these efforts produce an updated REP database and an objective, reproducible approach to support development of TEF values based on all available data.

Estimating the Prevalence of Rare Diseases: Long-Chain Fatty Acid Oxidation Disorders as an Illustrative Example

Introduction

Determining the epidemiology of disease is critical for multiple reasons, including to perform risk assessment, compare disease rates in varying populations, support diagnostic decisions, evaluate health care needs and disease burden, and determine the economic benefit of treatment. However, establishing epidemiological measures for rare diseases can be difficult owing to small patient populations, variable diagnostic techniques, and potential disease and diagnostic heterogeneity. To determine the epidemiology of rare diseases, investigators often develop estimation models to account for missing or unobtainable data, and to ensure that their findings are representative of their desired patient population.

Methods

A modeling methodology to estimate the prevalence of rare diseases in one such population—patients with long-chain fatty acid oxidation disorders (LC-FAOD)—as an illustrative example of its applicability.

Results

The proposed model begins with reliable source data from newborn screening reports and applies to them key modifiers. These modifiers include changes in population growth over time and variable standardization rates of LC-FAOD screening that lead to (1) a confirmed diagnosis and (2) improvements in standards of care and survival estimates relative to the general population. The model also makes necessary assumptions to allow the broad applicability of the estimation of LC-FAOD prevalence, including rates of diagnosed versus undiagnosed patients in the USA over time.

Conclusions

Although each rare disease is unique, the approach described here and demonstrated in the estimation of LC-FAOD prevalence provides the necessary tools to calculate key epidemiological estimates useful in performing risk assessment analyses; comparing disease rates between different subgroups of people; supporting diagnostic decisions; planning health care needs; comparing disease burden, including cost; and determining the economic benefit of treatment.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12325-022-02186-2.

Using the Matrix to bridge the epidemiology/risk assessment gap: a case study of 2,4-D

BACKGROUND

The Matrix is designed to facilitate discussions between practitioners of risk assessment and epidemiology and, in so doing, to enhance the utility of epidemiology research for public health decision-making. The Matrix is comprised of nine fundamental "asks" of epidemiology studies, focusing on the types of information valuable to the risk assessment process.

OBJECTIVE

A 2,4-dichlorophenoxyacetic acid (2,4-D) case study highlights the extent to which existing epidemiology literature includes information generally needed for risk assessments and proffers suggestions that would assist in bridging the epidemiology/risk assessment gap.

METHODS

Thirty-one publications identified in the US Environmental Protection Agency 2,4-D epidemiology review were assessed. These studies focused on associations between 2,4-D exposure and non-Hodgkin lymphoma (NHL), respiratory effects, and birth outcomes.

RESULTS

Many of the papers met one or more specific elements of the Matrix. However, from this case study, it is clear that some aspects of risk assessment, such as evaluating source-to-intake pathways, are generally not considered in epidemiology research. Others are incorporated, but infrequently (e.g. dose-response information, harmonization of exposure categories). We indicated where additional analyses or modifications to future study design could serve to improve the translation.

DISCUSSION

Interaction with risk assessors during the study design phase and using the Matrix "asks" to guide the conversations could shape research and provide the basis for requests for funds to support these additional activities. The use of the Matrix as a foundation for communication and education across disciplines could produce more impactful and consequential epidemiology research for robust risk assessments and decision-making.

Modelling the impact of MAUP on environmental drivers for Schistosoma japonicum prevalence

BACKGROUND

The modifiable areal unit problem (MAUP) arises when the support size of a spatial variable affects the relationship between prevalence and environmental risk factors. Its effect on schistosomiasis modelling studies could lead to unreliable parameter estimates. The present research aims to quantify MAUP effects on environmental drivers of Schistosoma japonicum infection by (i) bringing all covariates to the same spatial support, (ii) estimating individual-level regression parameters at 30 m, 90 m, 250 m, 500 m and 1 km spatial supports, and (iii) quantifying the differences between parameter estimates using five models.

METHODS

We modelled the prevalence of Schistosoma japonicum using sub-provinces health outcome data and pixel-level environmental data. We estimated and compared regression coefficients from convolution models using Bayesian statistics.

RESULTS

Increasing the spatial support to 500 m gradually increased the parameter estimates and their associated uncertainties. Abrupt changes in the parameter estimates occur at 1 km spatial support, resulting in loss of significance of almost all the covariates. No significant differences were found between the predicted values and their uncertainties from the five models. We provide suggestions to define an appropriate spatial data structure for modelling that gives more reliable parameter estimates and a clear relationship between risk factors and the disease.

CONCLUSIONS

Inclusion of quantified MAUP effects was important in this study on schistosomiasis. This will support helminth control programmes by providing reliable parameter estimates at the same spatial support and suggesting the use of an adequate spatial data structure, to generate reliable maps that could guide efficient mass drug administration campaigns.

Methods for evaluating variability in human health dose-response characterization

The Reference Dose (RfD) and Reference Concentration (RfC) are human health reference values (RfVs) representing exposure concentrations at or below which there is presumed to be little risk of adverse effects in the general human population. The 2009 National Research Council report Science and Decisions recommended redefining RfVs as "a risk-specific dose (for example, the dose associated with a 1 in 100,000 risk of a particular end point)." Distributions representing variability in human response to environmental contaminant exposures are critical for deriving risk-specific doses. Existing distributions estimating the extent of human toxicokinetic and toxicodynamic variability are based largely on controlled human exposure studies of pharmaceuticals. New data and methods have been developed that are designed to improve estimation of the quantitative variability in human response to environmental chemical exposures. Categories of research with potential to provide new database useful for developing updated human variability distributions include controlled human experiments, human epidemiology, animal models of genetic variability, in vitro estimates of toxicodynamic variability, and in vitro-based models of toxicokinetic variability. In vitro approaches, with further development including studies of different cell types and endpoints, and approaches to incorporate non-genetic sources of variability, appear to provide the greatest opportunity for substantial near-term advances.

Uncertainty Analysis of Mobile Phone Use and Its Effect on Cognitive Function: The Application of Monte Carlo Simulation in a Cohort of Australian Primary School Children

Previous epidemiological studies on health effects of radiation exposure from mobile phones have produced inconsistent results. This may be due to experimental difficulties and various sources of uncertainty, such as statistical variability, measurement errors, and model uncertainty. An analytical technique known as the Monte Carlo simulation provides an additional approach to analysis by addressing uncertainty in model inputs using error probability distributions, rather than point-source data. The aim of this investigation was to demonstrate using Monte Carlo simulation of data from the ExPOSURE (Examination of Psychological Outcomes in Students using Radiofrequency dEvices) study to quantify uncertainty in the output of the model. Data were collected twice, approximately one year apart (between 2011 and 2013) for 412 primary school participants in Australia. Monte Carlo simulation was used to estimate output uncertainty in the model due to uncertainties in the call exposure data. Multiple linear regression models evaluated associations between mobile phone calls with cognitive function and found weak evidence of an association. Similar to previous longitudinal analysis, associations were found for the Go/No Go and Groton maze learning tasks, and a Stroop time ratio. However, with the introduction of uncertainty analysis, the results were closer to the null hypothesis.

Bayesian network-based framework for exposure-response study design and interpretation

Conventional environmental-health risk-assessment methods are often limited in their ability to account for uncertainty in contaminant exposure, chemical toxicity and resulting human health risk. Exposure levels and toxicity are both subject to significant measurement errors, and many predicted risks are well below those distinguishable from background incident rates in target populations. To address these issues methods are needed to characterize uncertainties in observations and inferences, including the ability to interpret the influence of improved measurements and larger datasets. Here we develop a Bayesian network (BN) model to quantify the joint effects of measurement errors and different sample sizes on an illustrative exposure-response system. Categorical variables are included in the network to describe measurement accuracies, actual and measured exposures, actual and measured response, and the true strength of the exposure-response relationship. Network scenarios are developed by fixing combinations of the exposure-response strength of relationship (none, medium or strong) and the accuracy of exposure and response measurements (low, high, perfect). Multiple cases are simulated for each scenario, corresponding to a synthetic exposure response study sampled from the known scenario population. A learn-from-cases algorithm is then used to assimilate the synthetic observations into an uninformed prior network, yielding updated probabilities for the strength of relationship. Ten replicate studies are simulated for each scenario and sample size, and results are presented for individual trials and their mean prediction. The model as parameterized yields little-to-no convergence when low accuracy measurements are used, though progressively faster convergence when employing high accuracy or perfect measurements. The inferences from the model are particularly efficient when the true strength of relationship is none or strong with smaller sample sizes. The tool developed in this study can help in the screening and design of exposure-response studies to better anticipate where such outcomes can occur under different levels of measurement error. It may also serve to inform methods of analysis for other network models that consider multiple streams of evidence from multiple studies of cumulative exposure and effects.

Modeling Schistosoma japonicum Infection under Pure Specification Bias: Impact of Environmental Drivers of Infection

Uncertainties in spatial modeling studies of schistosomiasis (SCH) are relevant for the reliable identification of at-risk populations. Ecological fallacy occurs when ecological or group-level analyses, such as spatial aggregations at a specific administrative level, are carried out for an individual-level inference. This could lead to the unreliable identification of at-risk populations, and consequently to fallacies in the drugs’ allocation strategies and their cost-effectiveness. A specific form of ecological fallacy is pure specification bias. The present research aims to quantify its effect on the parameter estimates of various environmental covariates used as drivers for SCH infection. This is done by (i) using a spatial convolution model that removes pure specification bias, (ii) estimating group and individual-level covariate regression parameters, and (iii) quantifying the difference between the parameter estimates and the predicted disease outcomes from the convolution and ecological models. We modeled the prevalence of Schistosoma japonicum using group-level health outcome data, and city-level environmental data as a proxy for individual-level exposure. We included environmental data such as water and vegetation indexes, distance to water bodies, day and night land surface temperature, and elevation. We estimated and compared the convolution and ecological model parameter estimates using Bayesian statistics. Covariate parameter estimates from the convolution and ecological models differed between 0.03 for the nearest distance to water bodies (NDWB), and 0.28 for the normalized difference water index (NDWI). The convolution model presented lower uncertainties in most of the parameter estimates, except for NDWB. High differences in uncertainty were found in night land surface temperature (0.23) and elevation (0.13). No significant differences were found between the predicted values and their uncertainties from both models. The proposed convolution model is able to correct for a pure specification bias by presenting less uncertain parameter estimates. It shows a good predictive performance for the mean prevalence values and for a positive number of infected people. Further research is needed to better understand the spatial extent and support of analysis to reliably explore the role of environmental variables.

Epidemiology: a foundation of environmental decision making

Many epidemiologic studies are designed so they can be drawn upon to provide scientific evidence for evaluating hazards of environmental exposures, conducting quantitative assessments of risk, and informing decisions designed to reduce or eliminate harmful exposures. However, experimental animal studies are often relied upon for environmental and public health policy making despite the expanding body of observational epidemiologic studies that could inform the relationship between actual, as opposed to controlled, exposures and health effects. This paper provides historical examples of how epidemiology has informed decisions at the U.S. Environmental Protection Agency, discusses some challenges with using epidemiology to inform decision making, and highlights advances in the field that may help address these challenges and further the use of epidemiologic studies moving forward.

Radiofrequency Electromagnetic Radiation and Memory Performance: Sources of Uncertainty in Epidemiological Cohort Studies

Uncertainty in experimental studies of exposure to radiation from mobile phones has in the past only been framed within the context of statistical variability. It is now becoming more apparent to researchers that epistemic or reducible uncertainties can also affect the total error in results. These uncertainties are derived from a wide range of sources including human error, such as data transcription, model structure, measurement and linguistic errors in communication. The issue of epistemic uncertainty is reviewed and interpreted in the context of the MoRPhEUS, ExPOSURE and HERMES cohort studies which investigate the effect of radiofrequency electromagnetic radiation from mobile phones on memory performance. Research into this field has found inconsistent results due to limitations from a range of epistemic sources. Potential analytic approaches are suggested based on quantification of epistemic error using Monte Carlo simulation. It is recommended that future studies investigating the relationship between radiofrequency electromagnetic radiation and memory performance pay more attention to treatment of epistemic uncertainties as well as further research into improving exposure assessment. Use of directed acyclic graphs is also encouraged to display the assumed covariate relationship.

Human Health Risk Assessment Applied to Rural Populations Dependent on Unregulated Drinking Water Sources: A Scoping Review

Safe drinking water is a global challenge for rural populations dependent on unregulated water. A scoping review of research on human health risk assessments (HHRA) applied to this vulnerable population may be used to improve assessments applied by government and researchers. This review aims to summarize and describe the characteristics of HHRA methods, publications, and current literature gaps of HHRA studies on rural populations dependent on unregulated or unspecified drinking water. Peer-reviewed literature was systematically searched (January 2000 to May 2014) and identified at least one drinking water source as unregulated (21%) or unspecified (79%) in 100 studies. Only 7% of reviewed studies identified a rural community dependent on unregulated drinking water. Source water and hazards most frequently cited included groundwater (67%) and chemical water hazards (82%). Most HHRAs (86%) applied deterministic methods with 14% reporting probabilistic and stochastic methods. Publications increased over time with 57% set in Asia, and 47% of studies identified at least one literature gap in the areas of research, risk management, and community exposure. HHRAs applied to rural populations dependent on unregulated water are poorly represented in the literature even though almost half of the global population is rural.

Quantitative bias analysis of a reported association between perfluoroalkyl substances (PFAS) and endometriosis: The influence of oral contraceptive use

An association between serum levels of perfluoroalkyl substances (PFAS) and endometriosis has recently been reported in an epidemiologic study. Oral contraceptive use to treat dysmenorrhea (pelvic pain associated with endometriosis) could potentially influence this association by reducing menstrual fluid loss, a route of excretion for PFAS. In this study, we aimed to evaluate the influence of differential oral contraceptive use on the association between PFAS and endometriosis. We used a published life-stage physiologically based pharmacokinetic (PBPK) model to simulate plasma levels of perfluorooctanoic acid (PFOA) and perfluorooctane sulfonate (PFOS) from birth to age at study participation (range 18-44years). In the simulated population, PFAS level distributions matched those for controls in the epidemiologic study. Prevalence and geometric mean duration (standard deviation [SD]) of oral contraceptive use in the simulated women were based on data from the National Health and Nutrition Examination Survey; among the women with endometriosis the values were, respectively, 29% and 6.8 (3.1) years; among those without endometriosis these values were 18% and 5.3 (2.8) years. In simulations, menstrual fluid loss (ml/cycle) in women taking oral contraceptives was assumed to be 56% of loss in non-users. We evaluated the association between simulated plasma PFAS concentration and endometriosis in the simulated population using logistic regression. Based on the simulations, the association between PFAS levels and endometriosis attributable to differential contraceptive use had an odds ratio (95% CI) of 1.05 (1.02, 1.07) for a log_e unit increase in PFOA and 1.03 (1.02, 1.05) for PFOS. In comparison, the epidemiologic study reported odds ratios of 1.62 (0.99, 2.66) for PFOA and 1.25 (0.87, 1.80) for PFOS. Our results suggest that the influence of oral contraceptive use on the association between PFAS levels and endometriosis is relatively small.

Anemia risk in relation to lead exposure in lead-related manufacturing

Background

Lead-exposed workers may suffer adverse health effects under the currently regulated blood lead (BPb) levels. However, a probabilistic assessment about lead exposure-associated anemia risk is lacking. The goal of this study was to examine the association between lead exposure and anemia risk among factory workers in Taiwan.

Methods

We first collated BPb and indicators of hematopoietic function data via health examination records that included 533 male and 218 female lead-exposed workers between 2012 and 2014. We used benchmark dose (BMD) modeling to estimate the critical effect doses for detection of abnormal indicators. A risk-based probabilistic model was used to characterize the potential hazard of lead poisoning for job-specific workers by hazard index (HI). We applied Bayesian decision analysis to determine whether BMD could be implicated as a suitable BPb standard.

Results

Our results indicated that HI for total lead-exposed workers was 0.78 (95% confidence interval: 0.50–1.26) with risk occurrence probability of 11.1%. The abnormal risk of anemia indicators for male and female workers could be reduced, respectively, by 67–77% and 86–95% by adopting the suggested BPb standards of 25 and 15 μg/dL.

Conclusions

We conclude that cumulative exposure to lead in the workplace was significantly associated with anemia risk. This study suggests that current BPb standard needs to be better understood for the application of lead-exposed population protection in different scenarios to provide a novel standard for health management. Low-level lead exposure risk is an occupational and public health problem that should be paid more attention.

Electronic supplementary material

The online version of this article (doi:10.1186/s12889-017-4315-7) contains supplementary material, which is available to authorized users.

Causal inference in cumulative risk assessment: The roles of directed acyclic graphs

Cumulative risk assessments (CRAs) address exposures to multiple chemical and nonchemical stressors and often focus on characterization of health risks in vulnerable populations. Evaluating complex exposure-response relationships in CRAs requires the use of formal and rigorous methods for causal inference. Directed acyclic graphs (DAGs) are graphical causal models used to organize and communicate knowledge about the underlying causal structure that generates observable data. Using existing graphical theories for causal inference with DAGs, risk analysts can identify confounders and effect measure modifiers to determine if the available data are both internally valid to obtain unbiased risk estimates and are generalizable to populations of interest. Conditional independencies implied by the structure of a DAG can be used to test assumptions used in a CRA against empirical data in a selected study and can contribute to the evidence evaluations related to specific causal pathways. This can facilitate quantitative use of these data, as well as help identify key research gaps, prioritize data collection activities, and evaluate risk management alternatives. DAGs also enable risk analysts to be explicit about sources of uncertainty and to determine whether a causal effect can be estimated from available data. Using a conceptual model and DAG for a hypothetical community located near a concentrated animal feeding operation (CAFO), we illustrate the advantages of using DAGs for evaluating causality in CRAs. DAGs also can be used in conjunction with weight of evidence (WOE) methodology to improve causal analysis for CRA, which could lead to more effective interventions to reduce population health risks.

Toward Greater Implementation of the Exposome Research Paradigm within Environmental Epidemiology

Investigating a single environmental exposure in isolation does not reflect the actual human exposure circumstance nor does it capture the multifactorial etiology of health and disease. The exposome, defined as the totality of environmental exposures from conception onward, may advance our understanding of environmental contributors to disease by more fully assessing the multitude of human exposures across the life course. Implementation into studies of human health has been limited, in part owing to theoretical and practical challenges including a lack of infrastructure to support comprehensive exposure assessment, difficulty in differentiating physiologic variation from environmentally induced changes, and the need for study designs and analytic methods that accommodate specific aspects of the exposome, such as high-dimensional exposure data and multiple windows of susceptibility. Recommendations for greater data sharing and coordination, methods development, and acknowledgment and minimization of multiple types of measurement error are offered to encourage researchers to embark on exposome research to promote the environmental health and well-being of all populations.

Mapping Soil Transmitted Helminths and Schistosomiasis under Uncertainty: A Systematic Review and Critical Appraisal of Evidence

BACKGROUND

Spatial modelling of STH and schistosomiasis epidemiology is now commonplace. Spatial epidemiological studies help inform decisions regarding the number of people at risk as well as the geographic areas that need to be targeted with mass drug administration; however, limited attention has been given to propagated uncertainties, their interpretation, and consequences for the mapped values. Using currently published literature on the spatial epidemiology of helminth infections we identified: (1) the main uncertainty sources, their definition and quantification and (2) how uncertainty is informative for STH programme managers and scientists working in this domain.

METHODOLOGY/PRINCIPAL FINDINGS

We performed a systematic literature search using the Preferred Reporting Items for Systematic reviews and Meta-Analysis (PRISMA) protocol. We searched Web of Knowledge and PubMed using a combination of uncertainty, geographic and disease terms. A total of 73 papers fulfilled the inclusion criteria for the systematic review. Only 9% of the studies did not address any element of uncertainty, while 91% of studies quantified uncertainty in the predicted morbidity indicators and 23% of studies mapped it. In addition, 57% of the studies quantified uncertainty in the regression coefficients but only 7% incorporated it in the regression response variable (morbidity indicator). Fifty percent of the studies discussed uncertainty in the covariates but did not quantify it. Uncertainty was mostly defined as precision, and quantified using credible intervals by means of Bayesian approaches.

CONCLUSION/SIGNIFICANCE

None of the studies considered adequately all sources of uncertainties. We highlighted the need for uncertainty in the morbidity indicator and predictor variable to be incorporated into the modelling framework. Study design and spatial support require further attention and uncertainty associated with Earth observation data should be quantified. Finally, more attention should be given to mapping and interpreting uncertainty, since they are relevant to inform decisions regarding the number of people at risk as well as the geographic areas that need to be targeted with mass drug administration.

Impacts of geocoding uncertainty on reconstructed PFOA exposures and their epidemiological association with preeclampsia

Many epidemiology studies have investigated associations of perfluorooctanoate (PFOA) exposures with a variety of adverse health outcomes for participants in the C8 Health Project. The exposure concentrations (i.e., air and groundwater) used in these studies were determined primarily based on participant's residential locations. However, for residential addresses that could not be geocoded to the street level, the exposure concentrations were assigned based on population-weighted ZIP code centroid, which may result in exposure mischaracterization. The aim of this current study is to evaluate the potential impact of mischaracterized exposure concentrations due to geocoding uncertainty on the predicted serum PFOA concentrations and the epidemiological association between PFOA exposure and preeclampsia. For both workplace addresses and incompletely geocoded residential addresses, we used Monte Carlo (MC) simulation to assign alternate geographic locations within the reported ZIP code (instead of population-weighted ZIP code centroids) and the corresponding exposure concentrations. We found that mischaracterization of residential exposure due to population-weighted ZIP code centroid assignment had no significant impact on the serum PFOA concentration predictions and the epidemiological association of PFOA exposure with preeclampsia. In contrast, the uncertainty in workplace exposure moderately impacted the rank exposure among the participants. We observed a 41% increase in the average adjusted odds ratio of preeclampsia occurrence that may be due to differing proportions of cases (64.3%) and controls (54.5%) with workplace address geocodes during pregnancy. This finding suggests that differential exposure mischaracterization can be reduced by obtaining accurate exposure information such as street addresses and tap water consumption, for both workplaces and residences. The analysis we present is one approach for estimating the potential impacts of positional errors in a geocoding-based exposure assessment on exposure estimates and epidemiological study results.

Bayesian networks in infectious disease eco-epidemiology

Globally, infectious diseases are responsible for a significant burden on human health. Drivers of disease transmission depend on interactions between humans, the environment, vectors, carriers, and pathogens; transmission dynamics are therefore potentially highly complex. Research in infectious disease eco-epidemiology has been rapidly gaining momentum because of the rising global importance of disease emergence and outbreaks, and growing understanding of the intimate links between human health and the environment. The scientific community is increasingly recognising the need for multidisciplinary translational research, integrated approaches, and innovative methods and tools to optimise risk prediction and control measures. Environmental health experts have also identified the need for more advanced analytical and biostatistical approaches to better determine causality, and deal with unknowns and uncertainties inherent in complex systems. In this paper, we discuss the use of Bayesian networks in infectious disease eco-epidemiology, and the potential for developing dynamic tools for public health decision-making and improving intervention strategies.

Bayesian Approach to "Healthy Worker Hire Effect" in Standardized Mortality Ratio Analysis

OBJECTIVES

In this study, we address the healthy worker hire effect that arises when people with greater than average health are recruited to work in industrial jobs.

METHODS

Epidemiologists have used both general and working population reference rates to gauge influence of healthy worker hire effect on the standardized mortality ratio. We propose a Bayesian procedure that uses information derived from general and working population reference rates to calculate standardized mortality ratio.

RESULTS

The procedure is illustrated in the context of heart disease and lung cancer mortality analyses of a cohort of workers from a fluoropolymer production facility.

CONCLUSIONS

Application of our method should allow for fuller discussions of the healthy worker effect when one of its components, the healthy worker hire effect, is evaluated quantitatively. Our method can be utilized to improve risk estimates for a cohort with occupational exposure.