DEAFNESS IN DALMATIANS - DOES SEX MATTER

Recent studies on the prevalence of deafness in Dalmatians have report ed vastly different gender effects on the prevalence. The diverse conc lusions in these reports cover all possibilities, higher prevalence in males, no difference between genders and higher prevalence in females . Much of this confusion about the effect of gender on the Dalmatian's hearing status is due to the unsatisfactory statistical interpretatio n of the available data. Careful analysis of a large (n=1234) composit e database on Dalmatians in the UK has provided a reliable assessment of the effect of gender on deafness. The overall rate of deafness in t he tested Dalmatians was 18.4%, of which 13.1% were unilaterally deaf and 5.3% were bilaterally deaf. The overall deafness in females (21.1% ) was significantly higher (p=0.014) than that in males (15.5%). Tn al l subsets of the full dataset [subsets obtained by partitioning by tes ting locations, colour (black or liver spots), parental hearing status (normal or untested) and time (year of test)], the prevalence was hig her in females compared with males. Unlike other studies, the confound ing of the gender effect with other factors, particularly the parental hearing status, was avoided in the large, composite UK study. Because dogs from the same litter might be correlated with respect to their d eafness status, the data were analysed further to take the random litt er effect into account. This further emphasised the difference between genders. We believe that the reason for the prevailing confusion in t he literature is that the unwary referees and readers give the same am ount of credence to 'not significant' results from small scale studies as they give to significant results from large scale studies and do n ot always recognize the absence of sound statistical methods. (C) 1998 Elsevier Science B.V.