Loading...

Matched Studies in Medical Research

1,008 views

Loading...

Loading...

Transcript

The interactive transcript could not be loaded.

Loading...

Loading...

Rating is available when the video has been rented.
This feature is not available right now. Please try again later.
Published on Dec 29, 2011

Presented by Dr. John Klein, PhD, Director and Professor of the Division of Biostatistics, Medical College of Wisconsin.
Matched pairs designs are often suggested for medical research. In these studies subjects are assigned to treatment within pairs in hope of making sure differences between treatment outcome is not due to confounding with other patient characteristics. Pairing of subjects can be biological (twins, litter mates) or based on a set of possible covariates matching subjects in a larger cohort.
In this presentation, we discuss how matched pairs studies can be designed. We examine the efficiency of these designs relative to an unmatched design. We examine how the inference made on paired designs is affected by the statistical method selected for the analysis and we see how incorrect inferences can be made when paired data is analyzed as being independently. We show in examples how to properly analyze paired data. Much of this work is a result of a supplemental grant to the Clinical and Translational Science grant which supports the CTSI of Southeastern Wisconsin.

Loading...

When autoplay is enabled, a suggested video will automatically play next.

Up next


to add this to Watch Later

Add to

Loading playlists...