University of Cambridge > Talks.cam > Statistics > Case-Control Studies when Cases are Extreme Values of a Continuous Variable

Case-Control Studies when Cases are Extreme Values of a Continuous Variable

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact rbg24.

We typically think of case-control studies as comparing individuals with a disease to a healthy control group. However, sometimes the condition that defines a ?case? is an unusual value of a continuous outcome. For example, neonates are said to have intra-uterine growth restriction (IUGR) if their weight at birth, adjusted for gestational age, is below the 10th percentile. IUGR is associated with a number of negative health effects. Most research on the etiology of IUGR has been retrospective case-control studies in which neonates with IUGR are compared to those above the 10th percentile with respect to potential risk factors.

The fact that IUGR is defined by converting a continuous variable, (adjusted) weight at birth, into a dichotomous variable, suggests that more information should be available by making explicit use of weight at birth and not just its dichotomous version. I will present some simple ideas that show how to take advantage of the continuous outcome itself. These analyses offer much higher power for identifying risk factors. They even make it possible to estimate the effect on the continuous outcome from case-control type sample data and to exploit data from subjects who have not been classified with respect to qualitative covariates. They also suggest how one might choose an efficient sampling strategy.

This research was motivated by a large prospective study on thrombophilia and pregnancy outcomes conducted by Dr. O. Salomon, Prof. U. Seligsohn and Dr. R. Achiron. Part of the statistical research was carried out by Ms. Nadya Krishtal as part of her M. Sc. thesis at Tel-Aviv University.

http://www.math.tau.ac.il/~dms/

This talk is part of the Statistics series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

© 2006-2024 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity