Autosomal models of sex effect in bipolar-related major affective illness

Abstract

IN ALMOST every study of major affective illness to date a greater number of women have been ill than men. This observation has been made repeatedly in population studies’-5 andin first admission studies.6-8 The reported sex-ratios vary considerably and appear to depend in part on the type of affective illness studied; with the unipolar (UP) form having a ratio of approximately 1:2.5, men to women, while the bipolar (BP) form has an approximate ratio of 1: 1.5.10 Further evidence for a differential sex effect on morbidity comes from family studies in which it appears that female relatives are at greater risk than male relatives regardless of the polarity of the proband. The basis of this sex effect has been the focus of a great deal of interest especially with regard to the BP form of major affective illness and its reported linkage to the known X-chromosomal markers of protan and deutan color blindness and the Xg blood group.ll-$5 If BP illness is an X-linked dominant trait, then its greater frequency among women could in large part be explained. However, the area remains controversial with family studies of BP illness now appearing which report frequencies of male to male transmission that cannot be accounted for solely using the hypothesis of X-linked transmission.8Jo~l6*19 The purpose of this study is to investigate an alternative hypothesis that BP-related major affective illness is transmitted as an autosomal dominant trait whose phenotypic expression is variable depending upon the individual’s sex. While it is not possible to establish with certainty that autosomal transmission is occurring in the absence of classical Mendelian ratios or a definitive chromosomal linkage study, it is possible to approach the problem by formulating mathematical models which correspond to specific modes of autosomal inheritance. The theoretical prevalences in relatives of major affective illness generated from these mathematical models when compared with prevalence rates actually observed in family studies offer a test of the compatibility of the hypothesis with the data. Using suitable family study data available in the literature this approach was undertaken with two modes of autosomal transmission being considered, that of a single major locus (SML) and that of a multifactorial (MF) mode of inheritance.