|dc.description.abstract||Objectives: The thesis comprises two parts. Firstly, a systematic review (SR)
systematically examined meta-analytic studies assessing the prevalence of sexual
abuse in childhood. A meta-analysis sought to systematically identify and synthesise
data from studies providing prevalence estimates of contact and non-contact child
sexual abuse (CSA) as self-reported by adolescents within the past 10 years.
Methods: Consistent with Cochrane Collaboration guidelines, the systemic
review adopted a comprehensive search of electronic databases and additional
sources, including communication with authors working in the field, and the use of
ancestry and descendency approaches between February 2013 and March 2014. The
meta-analysis also adopted an equally systematic search conducted from March 2013
to June 2014. Assessment of quality and risk of bias were conducted on the included
studies using PRISMA criteria and STROBE guidelines.
Results: The systematic review identified six meta-analytic studies for review.
The quality of studies and the range of definitional and methodological factors
studied varied; results sections were well covered, whereas, quality and risk of bias
within studies were generally poorly addressed. Prevalence estimates varied
considerably and high heterogeneity was consistent across all analyses.
Nine population studies measuring prevalence of contact and non-contact
CSA were included in the meta-analysis. Meta-analyses were conducted overall and
across male and female populations. Prevalence estimates varied considerably across
studies, with contact CSA ranging from 2% to 39.8% and overall non-contact CSA,
estimates ranged from 1% to 24.6%. Substantial heterogeneity was present across all
analyses and therefore findings should be interpreted with caution. Nevertheless, the
findings were thought-provoking and most likely due to differences in definitions of
CSA and inconsistent use of validated instruments.
Conclusions: Child sexual abuse is an international problem which is highly
pervasive across all societies and populations studied. Females consistently report
higher rates of CSA than males, and some Asian countries, namely China, produce
relatively lower CSA estimates, even controlling for a wide range of methodological
factors and study characteristics.
The considerable heterogeneity was apparent both between and within studies
and it appears likely based on the current findings, and in the context of previous
research, that there are a wide range of methodological and socio-demographic
factors which moderate CSA prevalence estimates. More specifically, the lack of a
universally recognised definition of CSA proves especially problematic for
researchers. The issue is further complicated by use of non-standardised instruments
and inconsistent reporting and dissemination of findings. There is a need for future
epidemiological studies to adhere to universal guidelines using standard definitions,
standardisation of instruments and standardisation of reporting and dissemination to
facilitate development of health policies, resource allocation and prevention
initiatives for clinical and social services.||en