[Skip to Navigation]
January 1987

The Utility of the Dexamethasone Suppression Test-Reply

Author Affiliations

Department of Psychiatry Tufts University School of Medicine Boston Veterans Administration Medical Center Boston, MA 02130
Department of Psychiatry Harvard Medical School Mailman Research Center McLean Hospital Belmont, MA 02178

Arch Gen Psychiatry. 1987;44(1):95-96. doi:10.1001/archpsyc.1987.01800130107018

In Reply.—  In reflecting on correspondence and conversations following our recent review of the possible utility of the dexamethasone suppression test (DST) in psychiatry, we wish to make the following remarks and `amplifications. Certain evaluations pooled data from several studies to test for interesting trends by the use of ϰ2 statistics.1 Janicak et al2 recommended the use of a more conservative ϰ2 procedure developed by Mantel and Haenszel.3 We adapted their approach for use with a microcomputer and incorporated Yates' correction for small samples (the program is available from M.H.T.). The method computes a total ϰ2 statistic by accumulating data from contingency tables arising from each study and decomposes this overall value into a ϰ2 statistic of association testing the main effect, as influenced by the consistency of results across studies, as well as a statistic (ϰ2 for homogeneity) testing the consistency