Assessing sexual sadism in sexual homicide offenders using behavioural dimensional scales, typological approaches, and categorical diagnoses

Sexual sadism has been central to conceptualizing sexual homicide typologically, diagnostically, and behaviourally. Although about a third of cases involve sexual sadism, its reliable ascertainment is challenging, which is concerning given consequences of making or missing the diagnosis. This study examined 297 sexual homicide cases from Australia and New Zealand recruited from official online legal sources and evaluated the application of seven sexual sadism measures (clinical and research diagnoses, two typological approaches, and three behavioural scales). There were high correlations between all measures except clinical diagnosis. We confirmed scale and inter-rater reliability, convergent and concurrent validity, and external validity of behavioural scales. Rates of sexual sadism ranged from 4% for clinical diagnosis to 45% using the recommended Sexual Sadism Scale (SeSaS) Part 1 cut-off of 4. But other approaches identified between 1 in 5 and 1 in 3 cases. Cut-off scores of 5 on the SeSaS Part 1, 6 on the Sexual Homicide Crime Scene Rating Scale for Sexual Sadism (SADSEX-SH), and 6 on the Massachusetts Treatment Centre Sadism Scale (MTCSS) appeared equivalent and identified around 30% of cases. Offence and offender characteristics correlated with sadism ascertained with the SeSaS and SADSEX-SH were equivalent and as expected from the literature. We conclude that the SeSaS and SADSEX-SH are equivalent scales, but the recommended SeSaS Part 1 cut-off of 4 may be too low in sexual homicide cases. We make practical recommendations for a structured professional judgment approach to diagnosing sexual sadism disorder.