Random-effects meta-regression models for studying nonlinear dose-response relationship, with an application to alcohol and esophageal squamous cell carcinoma

Matteo Rota, Rino Bellocco, Lorenza Scotti, Irene Tramacere, Mazda Jenab, Giovanni Corrao, Carlo La Vecchia, Paolo Boffetta, Vincenzo Bagnardi

Research output: Contribution to journalArticle

60 Citations (Scopus)

Abstract

A fundamental challenge in meta-analyses of published epidemiological dose-response data is the estimate of the function describing how the risk of disease varies across different levels of a given exposure. Issues in trend estimate include within studies variability, between studies heterogeneity, and nonlinear trend components. We present a method, based on a two-step process, that addresses simultaneously these issues. First, two-term fractional polynomial models are fitted within each study included in the meta-analysis, taking into account the correlation between the reported estimates for different exposure levels. Second, the pooled dose-response relationship is estimated considering the between studies heterogeneity, using a bivariate random-effects model. This method is illustrated by a meta-analysis aimed to estimate the shape of the dose-response curve between alcohol consumption and esophageal squamous cell carcinoma (SCC). Overall, 14 case-control studies and one cohort study, including 3000 cases of esophageal SCC, were included. The meta-analysis provided evidence that ethanol intake was related to esophageal SCC risk in a nonlinear fashion. High levels of alcohol consumption resulted in a substantial risk of esophageal SCC as compared to nondrinkers. However, a statistically significant excess risk for moderate and intermediate doses of alcohol was also observed, with no evidence of a threshold effect.

Original languageEnglish
Pages (from-to)2679-2687
Number of pages9
JournalStatistics in Medicine
Volume29
Issue number26
DOIs
Publication statusPublished - Nov 2010

Fingerprint

Dose-response
Alcohol
Random Effects
Meta-Analysis
Regression Model
Alcohols
Cell
Alcohol Drinking
Estimate
Fractional Polynomials
Dose-response Curve
Describing Function
Cohort Study
Polynomial Model
Case-control Study
Random Effects Model
Ethanol
Statistical Models
Excess
Case-Control Studies

Keywords

  • Alcohol
  • Dose-response
  • Esophageal squamous cell carcinoma
  • Fractional polynomial regression
  • Meta-analysis
  • Random-effects models

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability

Cite this

Random-effects meta-regression models for studying nonlinear dose-response relationship, with an application to alcohol and esophageal squamous cell carcinoma. / Rota, Matteo; Bellocco, Rino; Scotti, Lorenza; Tramacere, Irene; Jenab, Mazda; Corrao, Giovanni; La Vecchia, Carlo; Boffetta, Paolo; Bagnardi, Vincenzo.

In: Statistics in Medicine, Vol. 29, No. 26, 11.2010, p. 2679-2687.

Research output: Contribution to journalArticle

Rota, Matteo ; Bellocco, Rino ; Scotti, Lorenza ; Tramacere, Irene ; Jenab, Mazda ; Corrao, Giovanni ; La Vecchia, Carlo ; Boffetta, Paolo ; Bagnardi, Vincenzo. / Random-effects meta-regression models for studying nonlinear dose-response relationship, with an application to alcohol and esophageal squamous cell carcinoma. In: Statistics in Medicine. 2010 ; Vol. 29, No. 26. pp. 2679-2687.
@article{545359b4733d474a833655d26b3eb3d3,
title = "Random-effects meta-regression models for studying nonlinear dose-response relationship, with an application to alcohol and esophageal squamous cell carcinoma",
abstract = "A fundamental challenge in meta-analyses of published epidemiological dose-response data is the estimate of the function describing how the risk of disease varies across different levels of a given exposure. Issues in trend estimate include within studies variability, between studies heterogeneity, and nonlinear trend components. We present a method, based on a two-step process, that addresses simultaneously these issues. First, two-term fractional polynomial models are fitted within each study included in the meta-analysis, taking into account the correlation between the reported estimates for different exposure levels. Second, the pooled dose-response relationship is estimated considering the between studies heterogeneity, using a bivariate random-effects model. This method is illustrated by a meta-analysis aimed to estimate the shape of the dose-response curve between alcohol consumption and esophageal squamous cell carcinoma (SCC). Overall, 14 case-control studies and one cohort study, including 3000 cases of esophageal SCC, were included. The meta-analysis provided evidence that ethanol intake was related to esophageal SCC risk in a nonlinear fashion. High levels of alcohol consumption resulted in a substantial risk of esophageal SCC as compared to nondrinkers. However, a statistically significant excess risk for moderate and intermediate doses of alcohol was also observed, with no evidence of a threshold effect.",
keywords = "Alcohol, Dose-response, Esophageal squamous cell carcinoma, Fractional polynomial regression, Meta-analysis, Random-effects models",
author = "Matteo Rota and Rino Bellocco and Lorenza Scotti and Irene Tramacere and Mazda Jenab and Giovanni Corrao and {La Vecchia}, Carlo and Paolo Boffetta and Vincenzo Bagnardi",
year = "2010",
month = "11",
doi = "10.1002/sim.4041",
language = "English",
volume = "29",
pages = "2679--2687",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "26",

}

TY - JOUR

T1 - Random-effects meta-regression models for studying nonlinear dose-response relationship, with an application to alcohol and esophageal squamous cell carcinoma

AU - Rota, Matteo

AU - Bellocco, Rino

AU - Scotti, Lorenza

AU - Tramacere, Irene

AU - Jenab, Mazda

AU - Corrao, Giovanni

AU - La Vecchia, Carlo

AU - Boffetta, Paolo

AU - Bagnardi, Vincenzo

PY - 2010/11

Y1 - 2010/11

N2 - A fundamental challenge in meta-analyses of published epidemiological dose-response data is the estimate of the function describing how the risk of disease varies across different levels of a given exposure. Issues in trend estimate include within studies variability, between studies heterogeneity, and nonlinear trend components. We present a method, based on a two-step process, that addresses simultaneously these issues. First, two-term fractional polynomial models are fitted within each study included in the meta-analysis, taking into account the correlation between the reported estimates for different exposure levels. Second, the pooled dose-response relationship is estimated considering the between studies heterogeneity, using a bivariate random-effects model. This method is illustrated by a meta-analysis aimed to estimate the shape of the dose-response curve between alcohol consumption and esophageal squamous cell carcinoma (SCC). Overall, 14 case-control studies and one cohort study, including 3000 cases of esophageal SCC, were included. The meta-analysis provided evidence that ethanol intake was related to esophageal SCC risk in a nonlinear fashion. High levels of alcohol consumption resulted in a substantial risk of esophageal SCC as compared to nondrinkers. However, a statistically significant excess risk for moderate and intermediate doses of alcohol was also observed, with no evidence of a threshold effect.

AB - A fundamental challenge in meta-analyses of published epidemiological dose-response data is the estimate of the function describing how the risk of disease varies across different levels of a given exposure. Issues in trend estimate include within studies variability, between studies heterogeneity, and nonlinear trend components. We present a method, based on a two-step process, that addresses simultaneously these issues. First, two-term fractional polynomial models are fitted within each study included in the meta-analysis, taking into account the correlation between the reported estimates for different exposure levels. Second, the pooled dose-response relationship is estimated considering the between studies heterogeneity, using a bivariate random-effects model. This method is illustrated by a meta-analysis aimed to estimate the shape of the dose-response curve between alcohol consumption and esophageal squamous cell carcinoma (SCC). Overall, 14 case-control studies and one cohort study, including 3000 cases of esophageal SCC, were included. The meta-analysis provided evidence that ethanol intake was related to esophageal SCC risk in a nonlinear fashion. High levels of alcohol consumption resulted in a substantial risk of esophageal SCC as compared to nondrinkers. However, a statistically significant excess risk for moderate and intermediate doses of alcohol was also observed, with no evidence of a threshold effect.

KW - Alcohol

KW - Dose-response

KW - Esophageal squamous cell carcinoma

KW - Fractional polynomial regression

KW - Meta-analysis

KW - Random-effects models

UR - http://www.scopus.com/inward/record.url?scp=77958518121&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77958518121&partnerID=8YFLogxK

U2 - 10.1002/sim.4041

DO - 10.1002/sim.4041

M3 - Article

C2 - 20809481

AN - SCOPUS:77958518121

VL - 29

SP - 2679

EP - 2687

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 26

ER -