Paper 5: The Power Relationship in Sexual Abuse: An Analysis of Irish Data relating to Victims and Perpetrators with Learning Disabilities

Dr Bob McCormack, St Michael’s House, Dublin 9


Recent years have seen a growing acknowledgement of the prevalence of sexual abuse of children and adults with learning disabilities, and a corresponding effort on the part of agencies to prevent abuse happening or continuing. A variety of risk factors have been identified in the literature. These factors include gender, age, caring role, disability, communication difficulties, leisure settings and activities, need for intimate care, and number of carers. Brown & Turk (1992) have suggested that an underlying factor in abuse is the presence of a significant power imbalance between the victim and the perpetrator. This paper explores six sources of power imbalance in data derived from a large-scale incidence study of sexual abuse of and by people with learning disabilities (McCormack et al, 2004). Results indicate that power imbalance is indeed a discernable underlying factor in the majority of confirmed episodes of abuse. The implications for service users, service providers and families are discussed.


The realisation that children with disabilities were at risk of sexual abuse was slow to be acknowledged. The thought that the most vulnerable children in society were the most abused was so abhorrent as to inconceivable to many. However, there is no doubt that children with disabilities are at increased risk of sexual abuse (Crosse et al, 1993; Sobsey et al, 1997); in one large-scale Kansas study, children with disabilities were found to be more than 3 times more likely to be sexually abused than non-disabled children (Sullivan & Knutson, 2000). American estimates suggest that as many as 1-in-3 adolescents with learning disabilities have been sexually abused, compared with a UK figure of 1-in-10 (Allington-Smith et al, 2002).

What characteristics make children and adults with disabilities more at risk of sexual abuse? The NSPCC in its report It Doesn’t Happen To Disabled Children (2003) identified ...

  • a need for intimate care
  • being less able to resist or avoid abuse, and
  • communication difficulties
  • as factors which increase the vulnerability of children with disabilities.

A more extensive list of risk factors for sexual abuse in children with learning disabilities was identified by Allington-Smith et al (2002) and included:

  • The child having multiple carers
  • Having care away from the family home (respite, residential, on special transport, shared care schemes)
  • Having a continuing need for intimate care
  • Having communication problems
  • Having a sensory impairment
  • Having a physical disability – inability to escape
  • The child’s lack of sexual knowledge.

Many of these risk factors overlap. Limited communications, for example, is an obstacle to learning about risk, and to disclosure of abuse when it occurs. In Turk & Brown’s (1993) UK survey of adults with learning disabilities, 70% of adult victims had an additional handicap, most frequently communication.

A number of characteristics of perpetrators have been identified in the abuse literature. Perpetrators with learning disabilities are overwhelmingly male (Gust et al, 2003), usually 90% or more (Mansell et al, 1997); however learning disabilities studies have shown that a significant proportion of the victims are also male – up to 65% reported by Gust et al (2003). The abuse tended to be somewhat less serious (Brown & Stein, 1997) though there is little research on the impact of the abuse on the victims (O’Callaghan et al, 2003).

Perpetrators tended to be older, except where the victim was elderly (Gust et al, 2003), while victims are more likely to have less verbal ability, more severe learning disabilities and a lower age. Those who abuse others often have low self-esteem, and redirect on to the victim their frustration and anger with more powerful others (Sobsey, 1994). Their actions seek to dominate the person and create dependency and fear.

Many of the characteristics of both perpetrators and victims identified in the literature reflect a significant power imbalance – between a more powerful abuser and a more dependent victim. In the area of sexuality, a significant power imbalance is a barrier to meaningful consent (Brown & Turk, 1992) as the victim is under undue pressure to comply with the perpetrator’s wishes. Role power is one example of this imbalance. Brown & Turk identified five types of power, which staff may have over service users with learning disabilities. Staff …

  • can legitimately tell the other what to do;
  • give the other intimate care;
  • have more status or credibility than the other (e.g. money, position, looks, age, strength)
  • may have other sexual relationships
  • provide access to material goods (e.g. clothes, food, cigarettes)

McCarthy and Thompson (1997) note that some power differences exist in most sexual relationships, but where one party exploits significant power differences, any consent on the part of the other is invalid. They noted examples from their study of such exploitation:

  • Perpetrator was a family member
  • Perpetrator was in a care-giving or authority role
  • Perpetrator was significantly more intellectually able
  • Victim was afraid of perpetrator, based on past experience or perpetrator’s reputation
  • Power imbalance based on physical size
  • Power imbalance based on gender

While there may be other underlying factors in the dynamics of sexual abuse, it is clear that significant power differences are common where abuse occurs.

The Present Study

As part of a longitudinal study of sexual abuse of children and adults with learning disabilities (see McCormack et al, 2004), the authors examined 118 episodes of confirmed sexual abuse where the perpetrator or victim had a learning disability.

(An episode was defined as all the sexual abuse occurring between one particular perpetrator and one victim.)

In most of these episodes the victim had a learning disability (77%), while the perpetrator had a learning disability in more than half the episodes (56%). A notable feature was the number of perpetrators with learning disabilities who had more than one victim (almost one-third). Relatively few episodes (14%) involved children as victims. Perpetrators were predominantly male (94% where known) while victims were both male (47%) and female (53%) in almost equal proportions.

Based on a review of the literature and the available data, seven indicators of significant power imbalance were selected for analysis in the study, namely:

  • Ability level of the perpetrator and the victim, using a 5-point scale, from no learningdisability to severe learning disability;
  • Age grouping of the perpetrator and the victim: under-12, 12-18 years; 19-29 years; over-30.
  • Gender of the perpetrator and the victim: where the perpetrator was male and the victim female.
  • Role power, where the perpetrator was in a care-giving role: family member/relative, staff member or volunteer worker.
  • Communication skills, where the perpetrator was a more able communicator than the victim, based on 5-point scale, from ‘uses full sentences, easily understood’, to ‘does not use words or signs consistently and meaningfully’.
  • Physical threat, where the perpetrator was described as having Challenging Behaviour.
  • Additional vulnerability, where the victim had a physical or sensory disability in addition to their learning disability.

Each of the 118 confirmed episodes was analysed to identify the proportion where the perpetrator had significant power over the victim based on these Power Imbalance Indicators.


In every episode, sexual abuse was confirmed to have occurred. The types of abuse, based on the categories indicated in Children First (Dept of Health & Children, 1999) are set out in Table 1. The most common type of abuse was sexual touch, followed by penetration:

Table 1: Type of Sexual Abuse (N=118*)
  N %
Non-contact abuse 25 21
Sexual touch 70 59
Masturbation 24 20
Oral-Genital 9 8
Attempted penetration 8 7
Penetration 30 25

Where the perpetrator had a learning disability, the abuse was more likely to be sexual touch (80% of 54 episodes); where the perpetrator was a relative, the abuse was more likely to be attempted or actual penetration (50% of 28 episodes). Abuse by staff or volunteers were more likely to be sexual touch (54% of 13 episodes). Abuse by other familiar adults or by strangers was more likely to be actual or attempted penetration (53% of 19 episodes).

Using the seven Power Indicators already identified, the extent to which the perpetrator was in a position of power or control over the victim is set out in Table 2 below.

Table 2: Perpetrators’ Sources of Power and Control (N=118 Episodes)



Perpetrator of higher ability 59 50
Perpetrator in older age group 58 49
Perpetrator male; victim female 56 47
Perpetrator in care-giving role (family/relatives, staff, volunteer) 39 33
Perpetrator more able communicator 32 27
Perpetrator has challenging behaviour 24 20
Victim has physical or sensory disability 9 8

In many instances, perpetrators had more than one source of power, as Figure 1 shows. In only 8 of the 118 episodes was none of the seven power sources found to be present, while two or more sources were present in 76% of episodes (mean 2.3 indicators).

The most prominent power sources were ability, age and gender, at least one of which was present in all but 15 of the 118 episodes. To what extent do these Power Indicators overlap with each other? The extend of the overlap can be measured in the strength of the correlation, and is indicated in Table 3 below, using the phi coefficient test for significance.

Table 3: Significant Correlations between the seven Power Indicators (N=118; p<0.05)

Perpetrator …

…In Care giving Role…

More Able Communicator

Higher Ability





Older Age Group 198 .031 404 .000
Challenging Behaviour .390 .000 -.202 .028
Male perpetrator /Female victim -.355 .000 .204 .027

The Power Indicators were largely independent of each other, with only six significant correlations of a possible 21. Many of the overlaps make sense: Caregivers tended have higher ability, to be in an older age group, and not to have challenging behaviour. More able communicators tend to be of higher intellectual ability. The correlations between the perpetrator being a more able communicator and having a female victim suggests an element of verbal manipulation or grooming however the negative correlation between communication skills and age is not obvious – ’though this is a weak correlation.


These findings suggest that a significant power imbalance is a common feature of sexual abuse, and while abuse of power had been a theme in the literature, this is the first study to demonstrate its widespread application in relation to abuse involving children and adults with learning disabilities.

At least one of the Power Imbalance Indicators arose in 14 out of every 15 cases in this study, suggesting that some form of power imbalance arises in almost every instance. The most common indicators were ability level, age group and gender. Each of these indicates different kinds of advantage to the perpetrator: ability level representing capacity for persuasion, trickery and manipulation; age representing greater experience and authority; and gender indicating greater physical power.

Role power is evident where the abuser is in a care-giving role as a staff member, volunteer or relative. Greater communication skills suggest an ability to persuade, to groom the potential victim, as well as less risk of being reported. Where the perpetrator has challenging behaviour, there is likely to be apprehension and fear of violence on the part of the victim. Where the victim has a physical or sensory disability, there is additional vulnerability, but it didn’t feature is a strong indicator here, perhaps because relatively few service users had physical or sensory disabilities.

Of the seven Indicators examined here, only one (‘Victim has physical or sensory disability’) was found in less than 20% of episodes of sexual abuse. This would suggest that there are 6 strong indicators of power imbalance found frequently in abuse of children and adults with learning disabilities, which service providers and families should be sensitive to.

In general terms these relate to three kinds of power imbalance:

  • Physical Power: gender, age, challenging behaviour
  • Persuasive / Manipulative Power: ability level, communication skills, age
  • Role Power: carer (staff, volunteer, relative)

With any disempowered group, risk of exploitation and abuse is increased. In relation to children and adults with learning disabilities, this power imbalance arises for a great variety of reasons and in a wide range of contexts – differences in general intellectual ability and understanding of what is taking place, physical power imbalance due to lower stature and less agility/fitness than the general population, lifelong training in compliance, the number of authority figures in their lives and the wide extent of their control, a need for assistance with bodily functions, their much more limited life experience, and their lack of formal or informal sexuality education. All of these features increase their dependency on others and contribute to their inability to successfully protect themselves against abuse.


In order to minimise the risk of abuse, it is essential that service providers implement best practice in relation to screening of staff and volunteers, procedures for intimate care, and use of recreational facilities. The development of a culture of respect for the person, the elimination of all dehumanising practices, and the giving of choice and daily control back to people – all will act as a bulwark against the abuse of power.

People with learning disabilities need strategies to communicate concerns to adults they trust. This may mean learning alternative communication strategies (LAMH, PECS, Talking Mats, etc) and an understanding of their sexuality (good touch, bad touch, etc). There must also be zero tolerance of unwanted physical contract that is sometimes prevalent in centres – grabbing or being hugged, pulling of blouses and sweaters, unwanted stroking, patting and rubbing which is distressing to the recipient (Barlow, 2003).

In a wider context, there is a need for legislation, which provides similar protection to vulnerable adults as is presently provided to children (Rickard-Clarke, 2003). This includes protection for whistle blowers and the mandatory involvement of Health Boards in abuse investigations. There is also a need to extend the present screening procedures for new staff, to relief staff and volunteers and short-term workers, and to introduce a Protection of Vulnerable Adults register similar to that operating in the UK (Sale, 2004). These improvements will require moderate resourcing at national level.

Prevention of abuse requires vigilance on all our parts. Developing a better understanding of the dynamics of abuse, and the role of power differences in sexual abuse will sharpen that vigilance.


  • Allington-Smith P., Ball R. & Haytor R. (2002) Management of sexually abused children with learning disabilities. Advances in Psychiatric Treatment 8, 66-72.
  • Barlow, C. (2003) Assessing and understanding sexually abusive behaviours in men with learning disabilities. Journal of Adult Protection, 5, 15-22.
  • Brown H. & Stein J. (1997) Sexual abuse perpetrated by men with intellectual disabilities: a comparative study. Journal of Intellectual Disability Research 41, 215–224.
  • Brown H. & Turk V. (1992) Defining sexual abuse as it affects adults with learning disabilities. Mental Handicap 20, 44-55.
  • Crosse, S.B., Kaye, E. & Katnofsky, A.C. (1993) A Report on the Maltreatment of Children with Disabilities. Washington: National Centre on Child Abuse & Neglect.
  • Department of Health & Children (1999) Children First: National Guidelines for the Protection & Welfare of Children. A Summary. Stationery Office. Dublin
  • Gust D.A., Wang S.A., Grot J., Ransom R. & Levine W.C. (2003) National survey of sexual behaviour and sexual behaviour policies in facilities for individuals with mental retardation/developmental disabilities. Mental Retardation 41, 365-373.
  • Mansell S., Sobsey D., Wilgosh L. & Zawallich A. (1997) the sexual abuse of young people with disabilities: treatment considerations. International Journal for the Advancement of Counselling 19, 293-302.
  • McCarthy, M. & Thompson, D. (1997) a prevalence study of sexual abuse of adults with intellectual disabilities referred for sex education. Journal of Applied Research in Intellectual Disabilities. 10, 105-124.
  • McCormack B., Kavanagh D., Caffrey S. & Power A. (2004) investigating sexual abuse: findings of a 15-year longitudinal study. Journal of Applied Research in Intellectual Disability. Submitted.
  • National Society for the Prevention of Cruelty to Children (2003) It Doesn’t Happen To Disabled Children: Child Protection and Disabled Children. London: NSPCC.
  • O’Callaghan A.C., Murphy G. & Clare I.C.H. (2003) The impact of abuse on men and women with severe learning disabilities and their families. British Journal of Learning Disabilities, 31, 175-180.
  • Rickard-Clarke, P.T. (2003) Critical issues in disability law reform: Legal capacity – proposals for reform. Global Trends in Disability Law Reform Conference, Dublin, September 2003.
  • Sale, A.U. (2004) Barred from entry? Community Care, 22-28 July, p.30-31.
  • Sobsey D. (1994) Violence and Abuse in the Lives of People with Disabilities: The End of Silent Acceptance? Paul H. Brookes. Baltimore, MD.
  • Sobsey D., Randall W., & Parrila R.K. (1997) Gender differences in abused children with and without disabilities. Child Abuse & Neglect 21, 707-720.
  • Sullivan, P.M. & Knutson, J.F. (2000) Maltreatment and disabilities: a population based epidemiological study. Child Abuse & Neglect, 24 (10)
  • Turk V. & Browne H. (1993) The sexual abuse of adults with learning disabilities: results of a two-year incidence study. Mental Handicap Research 6, 193–216.

Page last updated: 12/08/2010