A hierarchical framework for correcting under-reporting in count data

DOI10.1080/01621459.2019.1573732zbMATH Open1428.62480arXiv1809.00544OpenAlexW3106035388WikidataQ128227440 ScholiaQ128227440MaRDI QIDQ5208048FDOQ5208048

Authors: Oliver Stoner, Theo Economou, Gabriela Drummond Marques da Silva 

Publication date: 15 January 2020

Published in: Journal of the American Statistical Association (Search for Journal in Brave)

Abstract: Tuberculosis poses a global health risk and Brazil is among the top twenty countries by absolute mortality. However, this epidemiological burden is masked by under-reporting, which impairs planning for effective intervention. We present a comprehensive investigation and application of a Bayesian hierarchical approach to modelling and correcting under-reporting in tuberculosis counts, a general problem arising in observational count data. The framework is applicable to fully under-reported data, relying only on an informative prior distribution for the mean reporting rate to supplement the partial information in the data. Covariates are used to inform both the true count generating process and the under-reporting mechanism, while also allowing for complex spatio-temporal structures. We present several sensitivity analyses based on simulation experiments to aid the elicitation of the prior distribution for the mean reporting rate and decisions relating to the inclusion of covariates. Both prior and posterior predictive model checking are presented, as well as a critical evaluation of the approach.

Full work available at URL: https://arxiv.org/abs/1809.00544

Recommendations

• Modelling under-reporting in epidemics
• Estimation of under-reporting in epidemics using approximations
• Underreporting in generalized Poisson regression model with an application to demand for medical care data
• A marginal moment matching approach for fitting endemic‐epidemic models to underreported disease surveillance counts
• A BAYESIAN HIERARCHICAL MODEL FOR POISSON RATE AND REPORTING-PROBABILITY INFERENCE USING DOUBLE SAMPLING

zbMATH Keywords

censoringBayesian methodstuberculosisunder-detectionunder-recording

Mathematics Subject Classification ID

Applications of statistics to biology and medical sciences; meta analysis (62P10) Empirical decision procedures; empirical Bayes procedures (62C12)

Cites Work

• Bayesian data analysis.
• Bayesian image restoration, with two applications in spatial statistics (with discussion)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Count data models with selectivity
• Estimating with incomplete count data: A Bayesian approach
• Sparse Bayesian modelling of underreported count data
• A hierarchical framework for correcting under-reporting in count data
• A BAYESIAN HIERARCHICAL MODEL FOR POISSON RATE AND REPORTING-PROBABILITY INFERENCE USING DOUBLE SAMPLING
• Bayesian interval estimation for the difference of two independent Poisson rates using data subject to under-reporting

Cited In (11)

• Dealing with under-reported variables: an information theoretic solution
• A marginal moment matching approach for fitting endemic‐epidemic models to underreported disease surveillance counts
• Testing of Poisson mean with under-reported counts
• The complexity of finding optimal subgraphs to represent spatial correlation
• Under-reporting of COVID-19 in the northern health authority region of British Columbia
• The complexity of finding and enumerating optimal subgraphs to represent spatial correlation
• A hierarchical framework for correcting under-reporting in count data
• Improved inference for areal unit count data using graph-based optimisation
• Multivariate hierarchical frameworks for modeling delayed reporting in count data
• A Bayesian model of underreporting for sexual assault on college campuses
• Bias correction in clustered underreported data

Uses Software

• R
• BayesDA
• nimble

This page was built for publication: A hierarchical framework for correcting under-reporting in count data

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5208048)