Methods for Estimation of Radiation Risk in Epidemiological Studies Accounting for Classical and Berkson Errors in Doses

Kukush, Alexander; Shklyar, Sergiy; Masiuk, Sergii; Likhtarov, Illya; Kovgan, L.; Carroll, Raymond J.; Bouville, André

doi:10.2202/1557-4679.1281

Cited by 16 publications

(21 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A separate estimation of the Berkson and classic errors, which would lead to a more realistic radiation risk analysis (Kukush et al 2011), has not been implemented.…”

Section: Resultsmentioning

confidence: 99%

Thyroid Cancer Study among Ukrainian Children Exposed to Radiation after the Chornobyl Accident

Likhtarov¹,

Kovgan²,

Masiuk³

et al. 2014

Health Physics

Self Cite

View full text Add to dashboard Cite

In collaboration with the Ukrainian Research Center for Radiation Medicine, the U.S. National Cancer Institute initiated a cohort study of children and adolescents exposed to Chornobyl fallout in Ukraine to better understand the long-term health effects of exposure to radioactive iodines. All 13,204 cohort members were subjected to at least one direct thyroid measurement between 30 April and 30 June 1986 and resided at the time of the accident in the northern part of Kyiv, Zhytomyr, or Chernihiv Oblasts, which were the most contaminated territories of Ukraine as a result of radioactive fallout from the Chornobyl accident. Thyroid doses for the cohort members, which had been estimated following the first round of interviews, were re-evaluated following the second round of interviews. The revised thyroid doses range from 0.35 mGy to 42 Gy, with 95 percent of the doses between 1 mGy and 4.2 Gy, an arithmetic mean of 0.65 Gy, and a geometric mean of 0.19 Gy. These means are 70% of the previous estimates, mainly because of the use of country-specific thyroid masses. Many of the individual thyroid dose estimates show substantial differences because of the use of an improved questionnaire for the second round of interviews. Limitations of the current set of thyroid dose estimates are discussed. For the epidemiologic study, the most notable improvement is a revised assessment of the uncertainties, as shared and unshared uncertainties in the parameter values were considered in the calculation of the 1,000 stochastic estimates of thyroid dose for each cohort member. This procedure makes it possible to perform a more realistic risk analysis.

show abstract

“…A separate estimation of the Berkson and classic errors, which would lead to a more realistic radiation risk analysis (Kukush et al 2011), has not been implemented.…”

Section: Resultsmentioning

confidence: 99%

Thyroid Cancer Study among Ukrainian Children Exposed to Radiation after the Chornobyl Accident

Likhtarov¹,

Kovgan²,

Masiuk³

et al. 2014

Health Physics

Self Cite

View full text Add to dashboard Cite

show abstract

“…Then, the perturbations are simulated with a condition of

{normalΣ}_{b = 1}^{B} U_{b} = 0

, where U b is U at the b th iteration, following Kukush et al . (). Parameter estimates are then obtained by naive regression for each data set with the contaminated (error‐inflated) predictors.…”

Section: Methodsmentioning

confidence: 99%

“…We first consider situations under the classical measurement error model. In radiation epidemiology, where the magnitude of dose error is expected to increase with dose, it is common to assume a constant distribution of additive error for the logarithm of dose (Jablon, 1971;Kukush et al, 2011;Pierce et al, 1990). Let X and W be the unobservable true dose and the observed, error prone surrogate respectively.…”

Section: Regression Calibrationmentioning

confidence: 99%

“…Analyses that do not account for such errors can result in bias, distortion of the shape of the association between measured variables and outcome, and loss of efficiency (Schafer and Gilbert, 2006). Numerous references describe measurement error correction methods motivated by applications in radiation epidemiology (Kukush et al, 2011;Kwon et al, 2016;Li et al, 2007;Mallick et al, 2002;Masiuk et al, 2016). Generally, the radiation dose is estimated for individual subjects, some magnitude of error is assumed and regression analyses are conducted using the individual data.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Simulation–Extrapolation for Bias Correction with Exposure Uncertainty in Radiation Risk Analysis Utilizing Grouped Data

Misumi

Furukawa

Cologne

et al. 2017

Journal of the Royal Statistical Society Series C: Applied Statistics

View full text Add to dashboard Cite

Summary.In observational epidemiological studies, the exposure that is received by an individual often cannot be precisely observed, resulting in measurement error, and a common approach to dealing with measurement error is regression calibration (RC). Use of RC, which requires assumptions about the distribution of unknown error-free (true) variables, leads to concern about the possibility of bias due to misspecification of that distribution. The simulationextrapolation (SIMEX) method, in contrast, does not require a distributional assumption. However, analyses of large cohorts may be performed by using grouped or person-year data, and application of SIMEX to grouped data is not straightforward, particularly when there is a mixture of classical and Berkson measurement errors. We compared RC and SIMEX with grouped data analyses to assess robustness of the RC method to misspecification of the true dose distribution. We also applied SIMEX assuming mixtures of classical and Berkson errors and compared the results with those obtained by using RC for classical error only. SIMEX had less bias than RC and performed well regardless of the true dose distribution, whereas RC based on a misspecified true dose distribution showed greater bias than when based on the correctly specified true dose distribution.

show abstract

“…Considering the same problem, Mallick et al [15] proposed a Bayesian method using Markov chain Monte-Carlo (MCMC) techniques and Li et al [16] developed a Monte Carlo expectation-maximization (MCEM) approach. Kukush et al [17] investigated a different measurement error model that also incorporates errors of both types and provided maximum likelihood-based methods for logistic regression. The available methods accounting for the effect of both classical and Berkson measurement errors in regression analysis generally require that the variances of the errors be known or related through a known function.…”

Section: Introductionmentioning

confidence: 99%

Expected estimating equation using calibration data for generalized linear models with a mixture of Berkson and classical errors in covariates

Tapsoba

Lee

Wang

2013

Statistics in Medicine

View full text Add to dashboard Cite

Data collected in many epidemiological or clinical research studies are often contaminated with measurement errors that may be of classical or Berkson error type. The measurement error may also be a combination of both classical and Berkson errors and failure to account for both errors could lead to unreliable inference in many situations. We consider regression analysis in generalized linear models when some covariates are prone to a mixture of Berkson and classical errors and calibration data are available only for some subjects in a subsample. We propose an expected estimating equation approach to accommodate both errors in generalized linear regression analyses. The proposed method can consistently estimate the classical and Berkson error variances based on the available data, without knowing the mixture percentage. Its finite-sample performance is investigated numerically. Our method is illustrated by an application to real data from an HIV vaccine study.

show abstract

Methods for Estimation of Radiation Risk in Epidemiological Studies Accounting for Classical and Berkson Errors in Doses

Cited by 16 publications

References 23 publications

Thyroid Cancer Study among Ukrainian Children Exposed to Radiation after the Chornobyl Accident

Thyroid Cancer Study among Ukrainian Children Exposed to Radiation after the Chornobyl Accident

Simulation–Extrapolation for Bias Correction with Exposure Uncertainty in Radiation Risk Analysis Utilizing Grouped Data

Expected estimating equation using calibration data for generalized linear models with a mixture of Berkson and classical errors in covariates

Contact Info

Product

Resources

About