Bayesian Updating of Model-Based Risk Estimates Using Imperfect Public Health Surveillance Data

ABSTRACT

Model-based estimation of the human health risks resulting from exposure to environmental contaminants can be an important tool for structuring public health policy. Due to uncertainties in the modeling process, the outcomes of these assessments are usually probabilistic representations of a range of possible risks. In some cases, health surveillance data are available for the assessment population over all or a subset of the risk projection period and this additional information can be used to augment the model-based estimates. We use a Bayesian approach to update model-based estimates of health risks based on available health outcome data. Updated uncertainty distributions for risk estimates are derived using Monte Carlo sampling, which allows flexibility to model realistic situations including measurement error in the observable outcomes. We illustrate the approach by using imperfect public health surveillance data on lung cancer deaths to update model-based lung cancer mortality risk estimates in a population exposed to ionizing radiation from a uranium processing facility.

Key Words:

• Bayesian melding
• Bayesian Monte Carlo
• Markov chain Monte Carlo
• health risk assessment
• combining information

ACKNOWLEDGMENTS

We thank John Connor and Paula Holbrook of the Ohio Department of Health, Vital Statistic Unit, for their assistance in obtaining the lung cancer death certificate information. In addition, we thank Drs. Susan Pinney and Ron Freyberg of the University of Cincinnati for providing migration information from the Fernald Medical Monitoring Program.

The opinions expressed in this presentation do not necessarily reflect the opinions of the US Department of Health and Human Services or the Centers for Disease Control and Prevention.

Notes

¹The life table model used to estimate the excess and background risk was run 500 times to develop the Monte Carlo estimates of the marginal uncertainty distributions. Values for the uncertain components of the risk estimation model were sampled for each of the Monte Carlo runs.

²Percentiles of the marginal uncertainty distributions for the excess and background risk under the assumption that the joint uncertainty in these values follows a multivariate lognormal distribution.

¹Based on 10,000 samples of the uncertain components contributing to the probability of capturing lung cancer deaths.

²Percentiles of the uncertainty distribution for the probability of capturing a lung cancer death under the assumption that this uncertainty can be described by a Beta distribution with parameter values v = 20 and w = 9.

¹The life table model used to estimate the excess and background risk was run 500 times to develop the Monte Carlo estimates of the marginal uncertainty distributions. Values for the uncertain components of the risk estimation model were sampled for each of the Monte Carlo runs.

²Percentiles of the marginal uncertainty distributions for the excess and background risk under the assumption that the joint uncertainty in these values follows a multivariate lognormal distribution.