Giovanni Tripepi1, Kitty J. Jager2, Friedo W. Dekker2, Carmine Zoccali1
1CNR-IBIM, Clinical Epidemiology and Physiopathology of Renal Diseases and Hypertension of Reggio Calabria, Reggio Calabria, Italy
2ERA-EDTA Registry, Department of Medical Informatics, Academic Medical Center, University of Amsterdam, Amsterdam, and
Tóm tắt
The internal validity of an epidemiological study can be affected by <i>random error</i> and <i>systematic error</i>. Random error reflects a problem of precision in assessing a given exposure-disease relationship and can be reduced by increasing the sample size. On the other hand, systematic error or bias reflects a problem of validity of the study and arises because of any error resulting from methods used by the investigator when recruiting individuals for the study, from factors affecting the study participation <i>(selection bias)</i> or from systematic distortions when collecting information about exposures and outcomes <i>(information bias)</i>. Another important factor which may affect the internal validity of a clinical study is <i>confounding</i>. In this article, we focus on two categories of bias: selection bias and information bias. Confounding will be described in a future article of this series.
