Child sexual abuse (CSA) involving sexual contact between an (usually male) adult and a child has been reported by 20% of women and 5-10% of men worldwide (1-3). Surveys likely underestimate prevalence due to underreporting and memory failure (4-6). Although official reports have declined somewhat in the U.S. over the past decade (7), close to 90% of sexual abuse cases are never reported to the authorities (8).

CSA is associated with serious mental and physical health problems, substance abuse, victimization, and criminality in adulthood (9-12). Mental health problems include post-traumatic stress disorder, depression, and suicide (13, 14). CSA may interfere with attachment, emotional regulation, and major stress response systems (15). CSA has been used as a weapon of war and genocide and is associated with abduction and human trafficking (2).

Much of the research on CSA has been plagued by non-representative sampling, deficient controls, and limited statistical power (16). Moreover, CSA is associated with other forms of victimization (17), which complicates causal analysis of its role in adult functioning. However, associations in larger scale community and well-patient samples have been confirmed after controlling for family dysfunction and other risk factors (18, 19), in longitudinal investigations that measure pre- and post- CSA functioning (20), and in twin studies that control for environmental and genetic factors (12, 21).

Most CSA is committed by family members and individuals close to the child (1), which increases the likelihood of delayed disclosure (22), unsupportive reactions by caregivers and lack of intervention (8, 23) and possible memory failure (24, 25, cf. 26). These factors all undermine the credibility of abuse reports, yet there is evidence that when adults recall abuse, memory veracity is not correlated with memory persistence (27, 28). Research on child witness reliability has focused on highly publicized allegations of abuse by preschool operators, and has emphasized false allegations rather than false denials (29, 30). Cognitive and neurological mechanisms that may underlie the forgetting of abuse have been identified (31-33).

Scientific research on CSA is distributed across numerous disciplines, resulting in fragmented knowledge that is often infused with unstated value judgments. Consequently, policymakers have difficulty utilizing available scientific knowledge and gaps in the knowledge base are not well articulated. We recommend interdisciplinary research initiatives and a series of international consensus panels on scientific and clinical practice issues related to CSA. This can promote (a) increased inclusion of CSA education in the curriculum in medical and mental health fields, (b) improved education of the public, the media, and professionals working with alleged CSA victims, (c) greater visibility and improved dissemination of CSA research, (d) increased focus on CSA by researchers in a range of disciplines, and (e) improved cost-benefit analyses of intervention, including prevention efforts.

We call on researchers from social science, medical, and criminal justice fields to gather better information on the prevalence (34), causes, consequences, prevention, and treatment of CSA. A 1996 report from the Department of Justice (35) estimated rape and sexual abuse of children to cost $1.5 billion in medical expenses and $23 billion total annually to U.S. victims. Whereas $2 is spent on research for every $100 dollars in cost for cancer, only $.05 is spent for every $100 dollars in cost for child maltreatment (36). The National Child Traumatic Stress Network is a federally funded network of 54 sites providing community-based treatment to children and their families exposed to a wide range of trauma. The network should be expanded to address the enormous public health consequences of child trauma, and supported to develop new forms of treatment. Even creation of a new Institute of Child Abuse and Interpersonal Violence within the NIH would be justified on the basis of the emotional and economic cost of these problems

Jennifer J. Freyd, Department of Psychology, University of Oregon, Eugene, Oregon