rss
Occup Environ Med 64:562-568 doi:10.1136/oem.2006.026690
  • Education

Bias in occupational epidemiology studies

  1. Neil Pearce1,
  2. Harvey Checkoway2,3,
  3. David Kriebel4
  1. 1Centre for Public Health Research, Massey University Wellington Campus, Wellington, New Zealand
  2. 2Department of Environmental and Occupational Health Sciences, University of Washington, Seattle, Washington, USA
  3. 3International Agency for Research on Cancer, Lyon, France
  4. 4Department of Work Environment, University of Massachusetts at Lowell, Massachusetts, USA
  1. Correspondence to:
 Professor N Pearce
 Director, Centre for Public Health Research, Massey University Wellington Campus, Private Box 756, Wellington, New Zealand; n.e.pearce{at}massey.ac.nz
  • Accepted 6 October 2006
  • Published Online First 19 October 2006

Abstract

The design of occupational epidemiology studies should be based on the need to minimise random and systematic error. The latter is the focus of this paper, and includes selection bias, information bias and confounding. Selection bias can be minimised by obtaining a high response rate (and by appropriate selection of the control group in a case-control study). In general, it is important to ensure that information bias is minimised and is also non-differential (for example, that the misclassification of exposure is not related to disease status) by collecting data in a standardised manner. A major concern in occupational epidemiology studies usually relates to confounding, because exposure has not been randomly allocated, and the groups under study may therefore have different baseline disease risks. For each of these types of bias, the goal should be to avoid the bias by appropriate study design and/or appropriate control in the analysis. However, it is also important to attempt to assess the likely direction and strength of biases that cannot be avoided or controlled.

Footnotes

  • Published Online First 19 October 2006