Article Text

Download PDFPDF

Children who were sexually abused were more disturbed than their peers after 5 years
  1. Kate Brown, RGN, MSc
  1. Senior Lecturer School of Health, Biological and Environmental Sciences Middlesex University London, UK

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

OpenUrlAbstract/FREE Full Text

Question What are the 5 year outcomes (behaviour, depression, self esteem, anxiety, eating problems, drug use, suicide attempts, self injury, running away, criminal activity, and attributional style) in children who have been sexually abused compared with children who have not been sexually abused?


Cohort study within a case control study.


Sydney, Australia.


84 children who had been sexually abused (74% girls) were evaluated at the Child Protection Unit of 2 hospitals from 1988–90. They were 5–15 years old, lived in the metropolitan area, and had experienced some form of abusive sexual contact with either physical evidence or a clear and consistent description of events that the hospital staff deemed confirmatory. Exclusion criteria were consensual sexual activity between peers and developmental delay. Control children were matched for age and sex and chosen at random from area schools. Follow up was 81%.

Assessment of prognostic factors

Children and non-offending parents were interviewed at baseline and at 5 years. Prognostic factors were abuse characteristics, family demographic factors and socioeconomic status, family functioning, recent life events, and mental health of the mother.

Main outcome measures

Behaviour (Child Behavior Checklist, Youth Self-Report, and Teacher's Report Form); depression or unhappiness (Beck Depression Inventory or Children's Depression Inventory); self esteem (Piers-Harris Children's Self-concept Scale or the Adult Form of the Coopersmith Self-Esteem Inventory); anxiety (Revised Children's Manifest Anxiety Scale); family relationships (Parental Bonding Instrument or a project specific instrument); attributional style (Children's Attributional Style Questionnaire); and other behaviours using project specific questionnaires (self-injury, suicide attempts, running away, eating problems, drug use, and criminal behaviour). Age determined which questionnaires were used.

Main results

After 5 years of follow up, children who had been abused had completed fewer years of school (p=0.001); had lower socioeconomic status (p<0.001); were less likely to be living with a biological parent (p<0.001); experienced more parental changes (p<0.001); had mothers who had more somatic complaints, anxiety, and insomnia (p≤0.04); experienced more negative life events (p<0.001); had lower behaviour scores (2 of 3 scales); were more depressed (p<0.001) or likely to have been pregnant or had a child (p<0.04); had lower self esteem (p<0.001), more anxiety (p=0.03), and more binge eating (p=0.02); used more cigarettes (p=0.003) and hallucinogens (p=0.05); reported more parental drug use (p=0.02); thought more about (p=0.003) and attempted suicide more often since intake (p=0.03); and reported more self injury in the past 12 months (p=0.02) and since intake (p=0.009) than children who had not been sexually abused. These data were adjusted for age at time of assessment, sex of child, recent negative life events, and family functioning using regression analyses. Other outcomes did not differ between the groups.


Children who had been sexually abused were more disturbed than their peers who had not been abused, after 5 years of follow up.


The study by Swanston et al, by showing the long term consequences of child sexual abuse, provides ammunition for practitioners who have argued for short, medium, and long term services for survivors of sexual abuse. The study will be relevant to paediatric and community health nurses including health visitors and school nurses.

Research studies in the field of child abuse generally, and in child sexual abuse particularly, have often been criticised for lacking a control group or for being retrospective studies of adult populations. This study avoids both of these pitfalls. Furthermore, the use of regression analysis to control for other factors which could have led to these findings, such as changes in parent figures or lower socioeconomic groups, is very important to those who act as advocates for people who have experienced the effects of child sexual abuse. Many of the findings coincide with the reasons that children and adolescents come into contact with paediatric or community child health services (eg, eating disorders, attempted suicide, and self injury). This study should influence the assessment frameworks of practitioners in these settings. The challenge for practitioners will be to translate the findings into appropriate and sensitive questioning.

Why do research findings not influence practice? In the case of this study, 1 problem may be the cynicism of practitioners. They may feel that the study merely confirms what they have intuitively felt and have been reporting for years, and yet actual provision for mental health services for children remains uneven and lacks resources. The cost implications of providing care for children like those in this study, who may not be able to work through the original victimisation and its consequences are considerable. For the sake of the survivors of child sexual abuse, I hope that uncynical practitioners will use the findings of this study to fight for better service provision. As the authors say in their conclusion, what is needed is “long term commitment to care of these children”.


  • Source of funding: National Health and Medical Research Council of Australia.

  • For article reprint: not available.