Steven J. Collings

The Long-Term Effects of Contact and Noncontact Forms of Child Sexual Abuse in a Sample of University Men.

The long-term effects of child sexual abuse were examined in a sample of 284 university men. After controlling for the effects of dysfunctional parenting behaviors, a history of abuse involving physical contact was found to be associated with elevated scores on all subscales of the Brief Symptom Inventory. The comparison involving noncontact forms of abuse produced no significant abuse related effects. These findings are discussed in terms of their implications for research, intervention, and social policy.

Development, Reliability, and Validity of the Child Sexual Abuse Myth Scale

The Child Sexual Abuse (CSA) Myth Scale was developed to offer a means of reliably measuring acceptance of child sexual abuse myths and stereotypes. Scale items were developed using data from the popular and professional literature on child sexual abuse. The CSA Myth Scale was administered to a sample of 146 men and 259 women drawn from the general population. Three factors emerged from the factor analysis of the total sample, which supported the hypothesis that social attitudes to child sexual abuse constitute a multidimensional construct. The CSA Myth Scale yielded a Cronbach alpha of .764 and a test-retest reliability coefficient of .874. The three CSA Myth Scale factors showed good convergent and discriminant validity, in that factor scores were highly correlated, in expected directions, with scores on Burts Rape Myth Scale and with scores on the Jackson Incest Blame Scale.

Patterns of disclosure in child sexual abuse

This study examined patterns of disclosure in a sample of 1 737 cases of child sexual abuse (1 614 girls and 123 boys) reported in the North Durban policing area of KwaZulu-Natal, South Africa, during the period January 2001 to December 2003. A content analysis of disclosure patterns identified two broad dimensions of disclosure (Agency: child-initiated disclosure versus detection by a third party, and Temporal duration: an event versus a process); with these disclosure dimensions defining four discrete categories of disclosure: purposeful disclosure (30% of cases), indirect disclosure (9% of cases), eyewitness detection (18% of cases), and accidental detection (43% of cases). A multinomial logistic regression analysis revealed that disclosure patterns were independently predicted by the victims age, the nature of the victim-perpetrator relationship, the offenders age, the frequency of abuse, and reporting latency. The implications of the findings for primary prevention, forensic interviewing practice, and future research are discussed in detail.

Childhood Sexual Abuse in a Sample of South African University Males: Prevalence and Risk Factors

Using a retrospective questionnaire approach, the current study explored the prevalence of child sexual abuse in a sample of 284 South African university men. The definition of child sexual abuse used in the study included all unwanted contact or non-contact sexual experience involving a child of 17 years or younger. Eighty-two respondents (28,9%) reported a total of 91 experiences of contact (36,3%) and non-contact (63,7%) sexual abuse which met this definition. However, different definitions of abuse yielded markedly different prevalence rates. After controlling for definitional and methodological sources of variation, the prevalence rate obtained for the South African sample did not differ significantly from those reported in studies of American college men. Factors which were found to increase the risk of victimization included respondents race (i.e. being black), being raised by parents who were punitive or emotionally rejecting, and separation from the natural father for a major portion of childhood.

Unsolicited interpretation of child sexual abuse media reports

OBJECTIVE The purpose of the study was to investigate the extent and nature of spontaneous attributional activity elicited by a newspaper report of child sexual abuse. METHOD One hundred and seventy-six respondents, who were recruited through appeals placed in the letters column of a local newspaper, were presented with a newspaper report that described either a stereotype-congruent (rape by stranger in a public place) or a stereotype-incongruent (indecent assault by father at home) abuse incident. Respondents were asked to provide written descriptions of their thoughts and feeling about the abuse incident, which were analysed for attributional content. RESULTS Ninety-three percent of respondents provided one or more attributional statement, with attributional statements comprising 27% of all statements. Statements implying offender culpability were the most frequently employed attributional category (78% of all attributional judgments) while statements implying victim culpability comprised less than 2% of attributional judgments. Respondents who read the stereotype-incongruent description made more unsolicited causal and moral attributions than did respondents who read the stereotype-congruent description. CONCLUSIONS The results suggest that newspaper reports of child sexual abuse do elicit spontaneous attributional activity, that statements implying offender culpability are the most frequently employed attributional category, and that attributional activity is inhibited by stereotype congruent depictions of abuse. Implications for research and for prevention are discussed.

HIV post-exposure prophylaxis for child rape survivors in KwaZulu-Natal, South Africa: Who qualifies and who complies?

OBJECTIVE Our objective was to audit the provision and utilization of HIV post-exposure prophylaxis (PEP) to child rape survivors in the Province of KwaZulu-Natal, South Africa. METHODS A prospective design was used to collect data from a convenience sample of 200 consecutive cases of child rape referred for medico-legal assessment to a state hospital located to the north of the City of Durban (South Africa) in the period October to December 2004. For each case, information was obtained regarding demographic characteristics of the child, the childs HIV status at presentation, any excluding factors for PEP therapy, and the extent of adherence to the antiretroviral therapy regimen. RESULTS One hundred and twenty children (60%) qualified for PEP provision, with children being excluded because they refused initial HIV testing (1.5%), tested HIV positive at presentation (6.5%), or because of delayed (>72hour) presentation (32%). Of the 113 children who were followed through the PEP provision system, 40 (35.4%) returned for the full 28-day course, and only 4 (3.5%) returned at both 3- and 6-month follow up. CONCLUSIONS Adherence rates for PEP therapy following child rape in South Africa are extremely low, with low adherence rates being associated with a theoretical risk of both reduced efficacy and drug resistance. In this context, there is a need for further research designed to identify the reasons for such non-adherence as well as a need for secondary prevention programs designed to both encourage more immediate reporting of child rape and to increase adherence to the PEP regimen.

Sexual Aggression: A Discriminant Analysis of Predictors in a Non-Forensic Sample:

The prevalence of sexual aggression was examined in a non-forensic sample of 393 university males. A history of sexually abusive behaviour was reported by 25% of respondents, with six per cent reporting an act of sexual aggresssion that met the legal definition of rape or attempted rape. The results of a discriminant analysis indicated a significant relationship between sexually abusive behaviour and attitudes supportive of sexual dominance and aggression. These findings are discussed in terms of their implications for research and primary prevention.

The Impact of Contextual Ambiguity on the Interpretation and Recall of Child Sexual Abuse Media Reports

Previous research indicates that approximately 40% of child sexual abuse media reports areambiguous in the sense that information relating to the nature of abuse, the identity of the perpetrator,and/or the location of abuse is not specified. This research examined the impact of suchcontextual ambiguity on the recall and interpretation of child sexual abuse media reports. The189 respondents were recruited through media appeals and presented with one of three versionsof a child sexual abuse press report (stereotype congruent, stereotype incongruent, or contextuallyambiguous). Respondents were asked to provide written descriptions of their thoughts andfeelings about abuse incidents, which were analyzed for attributional content. Recall wasassessed after a 2-week interval. Results indicate that stereotypes affect the recall and interpretationof child sexual abuse media reports in the presence of contextually ambiguous individuatinginformation but not in the presence of nonstereotypic information that is conte...

Development and preliminary validation of a screen for interpersonal childhood trauma experiences among school-going youth in Durban, South Africa.

This paper describes the development and preliminary validation of the Developmental Trauma Inventory (DTI), which is a 36-item, retrospective, self-administered screen for interpersonal childhood trauma experiences developed specifically for the South African context. Preliminary validation of the inventory was conducted using a sample of 720 school-going adolescents attending a high school in the Durban Metropolitan area (South Africa). Factor analysis produced the best fit for a 10-factor model (emotional abuse, community assault, domestic assault, poverty, witnessing community violence, witnessing domestic violence, indecent assault, domestic neglect, rape, and domestic injury). Contrary to expectations, items relating to loss and separation (e.g. death of a parent) did not produce a clear factor structure. Identified scales had good internal consistency (0.70 to 0.81), low factor inter-correlations, and high concurrent criterion-related validity in the sense that all scales were significantly correlated with scores on clinical measures of post-traumatic stress disorder (PTSD) and/or complex PTSD. These findings provide preliminary support for the utility of DTI in the South African context.

An Emic Perspective on the Dynamics of Non-Fatal Suicidal Behaviour in a Sample of South African Indian Women

We aimed to obtain an emic perspective on the dynamics of non-fatal suicidal behaviour in a high risk population group in South African, viz. adolescent females of Indian origin. Interviews were conducted with 10 respondents admitted to a state hospital in the greater Durban area, following their intentional overdose of prescription medication. Interviews were tape recorded, transcribed, and thematically analysed in an attempt to identify influences on suicidal behaviour. Intrapersonal influences on suicidal behaviour included poor communication and problem-solving skills. Interpersonal influences included family conflict, rigid and inflexible parental attitudes, a family history of attempted or completed suicide and high parental expectations. Socio-cultural influences related largely to conflicting social roles and values in the context of contemporary acculturation pressures. The study findings are discussed in terms of their implications for prevention and future research.