Risk factors and sexual re-offending

Sexual re-offending is associated with deviant sexual interests and antisocial orientation/lifestyle instability (Hanson & Morton-Bourgon, 2004). The authors remark that (male) sexual offenders are more likely to reoffend with a non-sexual offence. Their meta-analytic review in relation to whether predictors of sexual recidivism are substantially different from predictors of non-sexual recidivism included mainly studies of male sexual offenders. Their study confirmed the strongest predictors for sexual recidivism and the importance of conflicts in intimate relationships and emotional identification with children associated with sexual recidivism. For female sexual offenders Beech, Fisher and Thornton (2003) point to two areas that deserve careful assessment: the woman's ability to resist pressure from male co-offenders (if the offending happened under coercion) and their role as caretakers of children.

Research data on recidivism of female sexual offenders is almost non-existent. Only one study was mentioned by Hanson and Morton-Bourgon (2004). Williams and Nicholaichuk (2001) followed up sixty one of seventy two women who sexually offended between 1972 and 1998. Recidivism was 32.8% (most non-violent such as theft, drugs and prostitution), 11.5% (n=7) were violent and 3.3% (n=2) were sexual. In a sample of ten female sex offenders, 80% offended with a co-offender and 87% of the victims were female. Both sexual recidivists were extra-familial child molesters, committed the offence on their own and had prior sex offences on record. One reoffended against both genders (younger than 2 years old), the other against a 15 year old female. "No woman who committed her crime in conjunction with a male confederate re-offended. Every woman who assaulted a stranger and who committed the offence on her own re-offended. It was a non-overlapping distribution. … The women who recidivated were so clearly criminalised and disordered, it took no effort to identify them" (T. Nicholaichuk, personal communication, August 2, 2005).

Nathan and Ward (2001) identify possible indicators for female sexual re-offending because "some of the risk predictors used in relation to male offenders may not have the same utility" (p. 52):

• existence of self-harm prior to or after the offence or the potential for self harm in the future
• whether the woman was unable to express her rage, rejection at the time of the offencewhether the previous factor was a chronic condition or aggravated by situational factors
• whether the victim was extra-familial or intra-familial
• the degree of emotional attachment shown to the victim.

In addition other risk predictors may include: homosexual orientation, intellectual deficits, deviant arousal and fantasies, unaccompanied offending, sexual dysfunction, use of force in previous sexual offending, anti-social tendencies (Nathan & Ward, 2001) and psychological dysfunction (Hunter & Mathews, 1997).

The different types of female sexual offenders invite tailored treatment (Hunter & Mathews, 1997; Nathan & Ward, 2001). Treatment needs are oriented toward intimacy and relationship issues, self-esteem, victimization experiences, what problems the sexual abuse solved for the offender and what specific needs are met by the offending, such as power and control, affiliation, esteem, social and physiological needs (Nathan & Ward, 2001).

For assessment and treatment Grayston & De Luca (1999) recommend to include the offender's own history of physical, sexual and psychological abuse as a child, adolescent or adult; any mental health problems that may be interfering with her capacity to interact appropriately with children and youth (depression, substance use, personality disorder), any interpersonal or marital problems that may contribute to abusive episodes, any other types of child maltreatment, specific stressors impairing the offender's capacity to adequately cope with daily demands and the offender's perceptions of and responses to various child stimuli.

Some research has been conducted on juvenile female sexual offenders (Hunter & Mathews, 1997; Mathews, Hunter, & Vuz, 1997; Vick, McRoy, & Matthews, 2002). Results show that juvenile female sexual offending "typically occurs in the context of more pervasive emotional and behavioral disturbances" (Hunter & Mathews, 1997, p. 468). Compared to their male counterparts juvenile females have a more extensive and severe history of sexual and physical abuse, were much younger at the time of first victimization and had a greater likelihood of having had multiple perpetrators (Mathews, Hunger, & Vuz, 1997). Further, they had been exposed to high levels of trauma, interpersonal violence and aggression. "A surprising number … reported having themselves been molested by a female" (Mathews, Hunger, & Vuz, 1997, p. 194) and about the same percentage of juvenile females and males used force in the commission of their acts. Both males and females were most likely to choose victims of the opposite gender and younger children. Females typically abuse children known to them.

Vick, McRoy and Matthews (2002) state that "there is no solid estimation of the likelihood of recidivism among young female sex offenders" (p. 19). Definitions of inappropriate sexual behaviour for males may not be suitable for females, "because of the sexual double standards and societal beliefs about young women, sexuality and violence" (p. 20) and sexual abuse is often captured by child abuse (Vandiver & Walker, 2002). Others claim little difference in the dynamics and characteristics of female and male youth sexual offenders hence "the treatment interventions are basically the same" (Ryan & Lane, 1997, p. 389) but observe that motives for the abuse may be different (Kaufman, Wallace, Johnson, & Reeder, 1995). It is suggested however that treatment with a developmental perspective and victimization processing would be more appropriate than the more confrontational treatment programmes for male sexual offenders (Mathews, Hunger, & Vuz, 1997).

The Static-99 (Harris, Phenix, Hanson, & Thornton, 2003) assesses only static factors "that have been seen in the literature to correlate with sexual reconviction in adult males" (p. 3). The authors are clear this instrument is not recommended for females. It is probable that the same principle applies to the Static-AS. The Static-AS is a brief actuarial instrument designed to estimate the probability of sexual recidivism among New Zealand adult males who have already been convicted of at least one sexual offence against a child or non-consenting adult. The scale contains 7 items that assess static factors relating to risk. The minimum information required for scoring Static-AS is the offender's official criminal record, information about victim gender, and current age of the offender. The Static-AS has been found to accurately classify male offenders into four risk categories from low to high risk of sexual recidivism.

In summary, empirical data on recidivism of female sexual offenders is virtually non-existent. One study by Williams and Nicholaichuk (2001) identified stranger victims and unaccompanied offenders as particular risk factors for re-offending. When the sexual offence was committed with a male accomplice, it is recommended to assess the woman's ability to resist pressure from male co-offenders, her role as caretaker of children, access and attachment to (mainly) female victims, perceptions and responses to various child stimuli, history or potential to self-harm, chronicity of inability to express rage and rejection, other types of child abuse, and ability to cope with daily demands. In addition, psychological dysfunction (mental health issues, substance abuse, personality disorder) needs to be assessed with any female sex offender. Some important profile differences exist for juvenile female sexual offenders compared to adult female sex offenders, but little with male youth sex offenders, except for elements of personal victimization history. There are no actuarial risk assessment tools available for female sex offenders.