Advertisement

Misclassification

  • Bryan Kestenbaum
Chapter

Abstract

  1. 9.1

    Misclassification results from errors in measuring the study data.

     
  2. 9.2
    Non-differential misclassification describes errors that occur randomly or to a similar degree across a study population.
    1. 9.2.1

      Non-differential misclassification of the exposure occurs when errors in exposure measurements are similar among people with and without the outcome.

       
    2. 9.2.2

      Non-differential misclassification of the outcome occurs when errors in outcome measurements are similar among exposed and unexposed people.

       
    3. 9.2.3

      Non-differential misclassification of the exposure typically leads to attenuation of the observed relative risk.

       
     
  3. 9.3
    Differential misclassification describes systematic errors in study measurements.
    1. 9.3.1

      Differential misclassification of the exposure occurs when errors in exposure measurements differ appreciably among people with and without the outcome.

       
    2. 9.3.2

      Differential misclassification of the outcome occurs when errors in outcome measurements differ appreciably across levels of the exposure.

       
    3. 9.3.3

      Differential misclassification may exaggerate the size of an association, dilute an association, or create false associations.

       
     

Reference

  1. 1.
    Smith NL, Psaty BM, Heckbert SR, Tracy RP, Cornell ES. The reliability of medication inventory methods compared to serum levels of cardiovascular drugs in the elderly. J Clin Epidemiol. 1999;52(2):143–6.CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Bryan Kestenbaum
    • 1
  1. 1.Division of Nephrology, Department of MedicineUniversity of WashingtonSeattleUSA

Personalised recommendations