October 1990

Abnormal Test Results?

Author Affiliations

Department of Pediatrics The University of North Carolina School of Medicine Pediatric Teaching Service The Moses H. Cone Memorial Hospital 1200 N Elm St Greensboro, NC 27401-1020

Am J Dis Child. 1990;144(10):1069. doi:10.1001/archpedi.1990.02150340011003

Sir.—The report by Charrow and Hedges1 in the February 1990 issue of AJDC contains a conceptual error. The method for calculating posterior probability (PP) of noncarrier and carrier status at a given assay value, as used in the report, actually gives the PP for all values above or below the value, depending on which status one is assessing.

The PP (sometimes called the likelihood ratio) at a given value equals the ratio of the heights of the normal curves at that value.2 These heights, given the means and SDs for both affected and unaffected populations, can be calculated from this formula:

in which SD is the standard deviation, xi is the observed value, and x is the mean for the given population.

Charrow and Hedges did not give sufficient data to calculate the PP for the example levels of hexaminidase A used in their report. By crude

First Page Preview View Large
First page PDF preview
First page PDF preview