Evaluating Public Health Interventions: 4. The Nurses' Health Study and Methods for Eliminating Bias Attributable to Measurement Error and Misclassification.

Abstract

The Nurses' Health Study and many other large longitudinal cohorts around the world use the food frequency questionnaire to assess dietary intake over time, and to relate diet to health. Controversies concerning this questionnaire's ability to adequately measure diet have led to a flurry of methods for evaluating the magnitude of measurement error and misclassification in exposure assessment, and for correcting the point and interval estimates of effect on the basis of these assessment methods for this error. Nurses' Health Study investigators have been in the forefront of these developments and their applications, although hundreds of other investigators have also used them. This commentary provides an overview of the methods and their uses, and concludes with remarks on their potential applications in the evaluation of public health interventions.