Abstract
Child sexual abuse is a well-established risk factor for a range of psychiatric disorders and other health problems. Ever since Mullen et al reported that there were significantly higher rates of psychiatric disorder among women who had been exposed to abuse compared to women that hadn't [1], there have been numerous epidemiological studies examining the rates of abuse and the poor health outcomes associated with it. What these studies have clearly demonstrated is that the rate of child sexual abuse (CSA) is disturbingly high and that it increases the risk for psychotic and non-psychotic disorders [2–4].
The origins of child sexual abuse are less clear. In this issue of the journal, Martin et al, [5] in a carefully designed longitudinal study examine the maternal factors (assessed at the time of the child's birth) associated with subsequent child sexual abuse when the children were 21 years old. The rates of sexual abuse in this study, as in many studies, are disturbingly high with 20.2% of females and 11.4% of males reporting non-penetrative abuse, and 10.2% of females and 8.1% of males penetrative abuse. Maternal characteristics associated with abuse fall broadly in the domain of social disadvantage, with some variance according to the type of abuse being examined.
While we are beginning to get a better understanding of the prevalence of child sexual abuse and, more importantly, its long-term health impact, there still remain basic issues that need to be addressed. First, the lack of clarity around the definition of abuse and trauma. Second, the impact of social context on abuse. Third, the mechanisms by which abuse exerts its effect, and finally what do we do about preventing abuse and dealing with it in the clinical setting.
Childhood adversity has long been recognised as a risk factor for psychiatric disorder. However, childhood adversity can take many forms, ranging from having uncaring parents through to CSA, all of which have been grouped together as trauma. Much of the early work on childhood adversity was on childhood separation or neglect, building on the observations of Spitz [6] and Bowlby [7] and the experimental work of Harlow [8]. There is no doubt that CSA is a specific form of childhood trauma, as are other forms of child maltreatment, especially physical abuse, verbal abuse (along with bullying) and neglect. However, when examining the long term consequences of childhood trauma there needs to be more precision in defining the type of trauma such as was done in Martin's study [5] who categorised CSA into penetrative and non-penetrative abuse. It is possible to create more specificity that includes things such as inappropriate talking or touching such as used by Bebbington et al [9]. Other aspects of the perpetration of the trauma, such as the age at which the abuse occurs, the frequency and duration of the abuse, the attitude of the survivor to their experience as well as the acknowledgment, support and treatment offered to that individual may all have an important effect upon long-term outcome, but are seldom examined.
It is clear that not all abused children will have psychological sequelae. In a follow up of the Isle of White study, 45% of the children that had experienced abuse reported no psychopathology in adulthood [10] and in an Australian study following up children that had been abused, 64% did not have adverse psychiatric sequelae (they did not have a psychiatric diagnosis on the Victorian case register) [3]. Perhaps this speaks to the resilience of the individual or possibly the context of the abuse, particularly the family environment. A single act of abuse within a supportive and caring family will have a different effect than abusive experiences in a chaotic and unsupportive family. Collishaw et al [10] found that high levels of family functioning with low levels of family discord, maternal psychiatric disorder or maternal separation from parents, caring parent and the ability to form good interpersonal relationships contributed to resilience. These positive aspects of the social environment possibly have a buffering effect against the impact of CSA.
An understanding of some of the mechanisms that underlie these effects is beginning to be elucidated. Traditional psychological models of understanding point to the effect of CSA on personality formation, especially on the attachment system and formation of a coherent self is seen in the development of insecure, disorganised attachments that increase the risk of developing severe personality disorders such as borderline personality disorder [11]. This is also associated with the use of dissociation as a defence against the repetitive abuse, the consequences of which can be seen in repetitive deliberate self harm or self mutilation, and possibly in a somatisation of response [12].
These psychological effects are clearly underpinned by effects upon the basic neurobiology of our response to stress. Overexposure to glucocorticoids released by a chronically stimulated HPA axis mediated by the actions of a range of neuropeptides and monoamines is associated with structural change in areas of the brain such as the amygdala and hippocampus in animal models of chronic stress [13] and in humans subjected to trauma or gross childhood neglect such as the Romanian orphans. Such long term changes are considered to be the consequences of early life trauma [14] and may be a contributing factor to poor response to treatment.
The epigenetic effects of chronic stress start early in development. Meaney's work in rat models of development points to the interaction of genetic inheritance, maternal behaviour and stress in cortical development and eventual good mothering behaviour [15]. Maternal behaviour can feed back to the level of DNA transcription through methylation or acetylation of histones. This alters the level of gene expression [16] leading to behavioural changes in nurturing and maternal behaviour and levels of anxiety in the offspring that are mediated by alterations in low levels of nerve growth factor-inducible protein A that in turn reduces the expression of glucocorticoid receptors. The interaction of genetic inheritance and early life adversity has perhaps been best demonstrated by Caspi's work [17] on the Dunedin Longitudinal Study in which he demonstrated the interaction of MAO-A polymorphism, level of maltreatment in childhood and subsequent behaviour. Prolonged deprivation as experienced by the Romanian orphans has additional epigenetic effects with methylation contributing to the shortening of telomere length, which in turn is correlated with advanced cellular aging [18].
Finally what can be done to decrease CSA and how should it be treated? First there is a need for public health interventions to deal with social disadvantage, with an emphasis on education in addition to strategies to deal with alcohol and drug misuse and tobacco cessation. In addition to this there is a need for adequate child protection agencies to provide support to vulnerable families. The inability of such agencies to respond to any but the most egregious examples of abuse has undermined the possible benefits of mandatory reporting of child abuse in many parts of Australia. Helping such agencies respond constructively to the family situation and resourcing a network of services that can assist vulnerable families would be an advance. A role for targeted interventions with past victims of CSA may be indicated in the light of high rates of perpetration by survivors of CSA [19]. However it is even unclear who should be treated.
In the clinical situation, a history of childhood adversity should always be sought. This requires tactful and sensitive history taking, with attention paid as to whether the person is still affected by the abusive experience. In the situation where a patient either ruminates about past abusive experiences, or considers himself or herself as still suffering as a consequence then, in the first instance, it will be helpful to allow them to talk about the experience so as to appropriately contextualise this. An essential component of this is to relieve some of the guilt associated with such experiences by pointing out that the perpetrator is the wrong doer, not the patient. It is not clear whether more prolonged psychotherapeutic interventions dealing with abusive experiences are indicated, especially when side effects such as an increase in self-harming behaviour can occur. Structured therapies, such as DBT may be more appropriate.
Very possibly the early identification and intervention for young people who are victims of CSA may hold the key to improved long-term outcomes.
In summary, child sexual abuse is far too common in our society. It can have long-term consequences and contribute to psychiatric disorders in adulthood. The mechanisms of this is truly biopsychosocial. We need to focus on trying to prevent abuse occurring and be mindful of its consequences in our treatment of patients.