Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Meta-Analysis
. 2020 Dec:145:106111.
doi: 10.1016/j.envint.2020.106111. Epub 2020 Sep 21.

Bayesian hierarchical dose-response meta-analysis of epidemiological studies: Modeling and _target population prediction methods

Bruce Allen  1 , Kan Shao  2 , Kevin Hobbie  3 , William Mendez Jr  3 , Janice S Lee  4 , Ila Cote  4 , Ingrid Druwe  4 , Jeffrey S Gift  4 , J Allen Davis  5
Affiliations
Meta-Analysis

Bayesian hierarchical dose-response meta-analysis of epidemiological studies: Modeling and _target population prediction methods

Bruce Allen et al. Environ Int. 2020 Dec.

Abstract

When assessing the human risks due to exposure to environmental chemicals, traditional dose-response analyses are not straightforward when there are numerous high-quality epidemiological studies of priority cancer and non-cancer health outcomes. Given this wealth of information, selecting a single "best" study on which to base dose-response analyses is difficult and would potentially ignore much of the available data. Therefore, systematic approaches are necessary for the analysis of these rich databases. Examples are meta-analysis (and further, meta-regression), which are well established methods that consider and incorporate information from multiple studies into the estimation of risks due to exposure to environmental contaminants. In this paper, we propose a hierarchical, Bayesian meta-analysis approach for the dose-response analysis of multiple epidemiological studies. This paper is the second of two papers detailing this approach; the first covered "pre-analysis" steps necessary to prepare the data for dose-response modeling. This paper focuses on the hierarchical Bayesian approach to dose-response modeling and extrapolation of risk to populations of interest using the association between bladder cancer and oral inorganic arsenic (iAs) exposure as an illustrative case study. In particular, this paper addresses the modeling of both case-control and cohort studies with a flexible, logistic model in a hierarchical Bayesian framework that estimates study-specific slopes, as well as a pooled slope across all studies. This approach is akin to a random effects model in which no assumption is made a priori that there is a single, common slope for all included studies. Further, this paper also details extrapolation of the estimates of logistic slope to extra risk in a _target population using a lifetable analysis and basic assumptions about background iAs exposure levels. In this case, the _target population was the general United States population and information on all-cause mortality and incidence and mortality from bladder cancer was used to perform the lifetable analysis. The methods herein were developed for general use in investigating the association between any pollutant and observed health-effects in epidemiological studies. In order to demonstrate these methods, inorganic arsenic was chosen as a case study given the large epidemiological database that exists for this contaminant.

Keywords: Bayesian dose-response; Hierarchical model; Lifetable analysis; Meta-analysis.

PubMed Disclaimer

Figures

Figure 1:
Figure 1:. Analysis Flow Chart
Figure 2:
Figure 2:. Dose Pre-Analysis and Uncertainty Flow Chart in Relation to “Best,” Low-end, and High-end Dose Sets
1 High group means minimized or maximized subject to constraint that −2*(LL – MLL) ≤ 2.706 (a 95% bound on the high-group mean). LL is the log-likelihood for the lognormal distribution for the candidate parameter vector; MLL is the maximum log-likelihood. When a published study reports the mean or median values for each group, those values are used directly as the group-specific dose values, with no lognormal fitting. 2 The terminology “low-end,” “high-end,” and “best” estimates are used to avoid confusing the values with credible (or confidence) interval bounds having a specific numerical value (e.g., 95%). Combining the log-likelihood bounds for group-specific means, with percentiles from the Monte Carlo analysis does not allow determination that the bounding estimates have any identifiable associated “confidence level.” They do, however produce reasonable semi-quantitative limits on how uncertain the resulting estimates are.
Figure 3.
Figure 3.. Posterior Distributions for Pooled and Study-specific Logistic Slope Parameters Using the MLE Dose Estimates.
Black vertical lines indicate means of posterior distributions. 95% credible intervals for the logistic slope parameters are highlighted in blue
Figure 4.
Figure 4.. Comparison of Observed RRs and ORs from Individual Studies and Model-predicted Values for All Studies using MLE Dose Estimates
Figure 5.
Figure 5.. Extra Lifetime Bladder Cancer Risk due to Oral iAs Exposure, using MLE Dose Estimates
Figure 6.
Figure 6.. Forest Plot of Extra Lifetime Bladder Cancer Risk at 10 μg/L iAs Exposure, using MLE Dose Estimates
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Allen BC; Shao K, Hobbie K, Mendez W, Lee J, Cote I, Druwe I, Gift J, Davis JA Bayesian Hierarchical Meta-Analysis of Epidemiological Studies, Part 1: Dose and Response Pre-Analysis. xxx 2020; yyy. - PMC - PubMed
    1. ATSDR (Agency for Toxic Substances and Disease Registry). (2007). Toxicological profile for arsenic (update) [ATSDR Tox Profile]. Atlanta, GA: U.S. Department of Health and Human Services, Public Health Service. http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=22&tid=3
    1. ATSDR (Agency for Toxic Substances and Disease Registry). (2016). Addendum to the toxicological profile for arsenic [ATSDR Tox Profile]. Atlanta, GA: Agency for Toxic Substances and Disease Registry, Division of Toxicology and Human Health Sciences. https://www.atsdr.cdc.gov/toxprofiles/Arsenic_addendum.pdf
    1. Bagnardi V, Zambon A, Quatto P and Corrao G, 2004. Flexible meta-regression functions for modeling aggregate dose-response data, with an application to alcohol and mortality. American journal of epidemiology, 159(11), pp.1077–1086. - PubMed
    1. Bates MN; Rey OA; Biggs ML; Hopenhayn C; Moore LE; Kalman D; Steinmaus C; Smith AH Case-control study of bladder cancer and exposure to arsenic in Argentina. Am J Epidemiol 2004;159:381–389 - PubMed

Publication types

LinkOut - more resources

  NODES
Association 2
INTERN 2
Project 1
twitter 2