Multiple-bias modelling for analysis of observational data

Sander Greenland, John Copas, David R. Jones, David Spiegelhalter, Kenneth Rice, Ben Armstrong, Stephen Senn, James Carpenter, Mike Kenward, Bianca De Stavola, Dorothea Nitsch, D. Nitsch, Colin R. Muirhead, James S Hodges, Andrew Gelman, David Draper, Paul Gustafson, Lawrence McCandless, Donald B. Rubin

Research output: Contribution to journalArticle

318 Scopus citations

Abstract

Conventional analytic results do not reflect any source of uncertainty other than random error, and as a result readers must rely on informal judgments regarding the effect of possible biases. When standard errors are small these judgments often fail to capture sources of uncertainty and their interactions adequately. Multiple-bias models provide alternatives that allow one systematically to integrate major sources of uncertainty, and thus to provide better input to research planning and policy analysis. Typically, the bias parameters in the model are not identified by the analysis data and so the results depend completely on priors for those parameters. A Bayesian analysis is then natural, but several alternatives based on sensitivity analysis have appeared in the risk assessment and epidemiologic literature. Under some circumstances these methods approximate a Bayesian analysis and can be modified to do so even better. These points are illustrated with a pooled analysis of case-control studies of residential magnetic field exposure and childhood leukaemia, which highlights the diminishing value of conventional studies conducted after the early 1990s. It is argued that multiple-bias modelling should become part of the core training of anyone who will be entrusted with the analysis of observational data, and should become standard procedure when random error is not the only important source of uncertainty (as in meta-analysis and pooled analysis).

Original languageEnglish (US)
Pages (from-to)267-306
Number of pages40
JournalJournal of the Royal Statistical Society. Series A: Statistics in Society
Volume168
Issue number2
DOIs
StatePublished - Jan 1 2005

Keywords

  • Bayesian statistics
  • Confidence profile method
  • Confounding
  • Epidemiologic methods
  • Leukaemia
  • Magnetic fields
  • Meta-analysis
  • Meta-statistics
  • Monte Carlo methods
  • Observational data
  • Odds ratio
  • Relative risk
  • Risk analysis
  • Risk assessment
  • Sensitivity analysis

Fingerprint Dive into the research topics of 'Multiple-bias modelling for analysis of observational data'. Together they form a unique fingerprint.

Cite this