Introduction

Child abuse is a serious and common risk affecting the long-term mental health of individuals. A growing body of evidence indicates that the mental health of a substantial minority of abused individuals appears relatively unaffected. Gaining a fuller understanding of the factors and processes involved in positive adaptation is important for many reasons:

• Theoretically, models of resilience have the potential to enhance the understanding of the mechanisms by which abuse affects psychosocial development.
• Clinically, some protective factors may be amenable to external manipulation and could present a potential focus for future treatments and interventions.

This study uses longitudinal data from a general population sample studied first in adolescence and again at mid-life to examine correlates and outcomes of child abuse, the extent of resilience for adult psychopathology, and the factors that best predict such resilience.

The implications for children’s psychological development and long-term mental health have been well documented. Consequences include cognitive delays and lowered IQ, neurobiological abnormalities, dysfunctional behaviours like conduct problems, aggression, and substance abuse, and an increased risk of adolescent and adult psychiatric disorders including depression, suicide, anxiety disorder, PTSD, and somatization disorders.

Current evidence also makes clear, however, that not all abused children go on to experience mental health problems later in life. Researchers have realized that understanding positive adaptation in the face of adversity is important, but there has been a debate on how to best define and study the concept of resilience. It’s agreed that a working definition should consider two points:

1. The experience to which individuals have been exposed should present a sufficient “risk” to which individuals can be considered to have shown “resilience”.
2. Markers of resilience should encompass a variety of domains and be evident across an extended time period.

This study defines resilience by identifying individuals who (1) had experienced repeated, ongoing or severe sexual and/or physical abuse, and (2) who reported no psychiatric disorders or suicidality over a 30-year adult follow-up period.

The first aim of this paper was to examine adolescent and adult psychopathology in individuals who reported being abused in childhood and to establish the extent of resilience in this group. The second aim was to identify factors that distinguished resilient and non-resilient individuals with experiences of abuse.

Method

Data are drawn form a follow-up of the Isle of Wight study, an epidemiological sample assessed in adolescence and at midlife. Ratings of psychiatric disorder, peer relationships and family functioning were made in adolescence; adult assessment included a lifetime psychiatric history, personality, and social functioning assessments, and retrospective reports of childhood sexual and physical abuse.

Measures

• Psychopathology: psychiatric disorder was assessed through interviews with parents and children through reports form teachers. Interviews assessed the frequency, severity, and duration of specific behaviours and symptoms over the past year.
• Peer relationships: accounts by adolescents and parents were used to rate the adequacy of peer relationships over the past year on a 3-point scale (normal, moderate abnormality, marked abnormality).
• Family functioning and demographics: interviews with parents provided information on parental separation and divorce, parental discord, family size, repeated long-term separations from parents, family social class, and housing tenure.
• Childhood abuse: childhood abuse was defined using retrospective reports collected during the adult interviews with the intensively studied sample. The sexual and physical abuse modules were based on the Childhood Experience of Care and Abusive Interview (CECA).
• Adult psychopathology: adult psychopathology was assessed using the Schedule for Affective Disorders and Schizophrenia-Lifetime version (SADS-L), revised as appropriate to cover DSM-IV diagnostic criteria.
• Parental care: study members completed mother and father versions of the shortened seven-item Parental Bonding Instrument (PBI), to measure their perceptions of the parenting they experienced as children.
• The Adult Personality Functioning Assessment (APFA): the APFA is an investigator-based interview designed to assess patterns of specific and general social dysfunction. It covers six domains of functioning: work, marriage/cohabitation, friendships, non-specific social contacts, day-to-day coping, and negotiations.
• Relationship history: informants’ descriptions of their first long-term relationship were used to rate how supportive their first partner was (both emotionally and practically).
• Personality: a 48-item Eysenck Personality Questionnaire (EPQ-R) was administered, focusing on the Neuroticism sub-scale.
• Crime: study members completed a questionnaire about involvement in illegal activities since age 18.
• Self-rated health: individuals rated their current health on a five-point scale (poor, fail, good, very good, excellent).
• “Quality of relationships” index (adolescent/adult composite): a summary scale of positive relationships across the life span was derived by counting across the following indicators: either parent rated as very caring; adolescent peer relationships rated as normal; adult relationships rated as positive; first adult partner rated as supportive; stable relationship history.

Results

10% of individuals reported repeated or sever physical or sexual abuse in childhood. Prospective measures revealed increased rates of adolescent psychiatric disorders in this group. Rates of adult psychopathology were also high. A substantial minority of abused individuals reported on mental health problems in adult life. Resilience of this kind was related to perceived parental care, adolescent peer relationships, the quality of adult love relationships, and personality style.

Discussion & conclusion

This study provided further evidence that child abuse is relatively common, and that it constitutes a serious risk for adult psychopathology. Risks for adult recurrent depression, suicidal behaviour, PTSD, and substance abuse were elevated several-fold among abused individuals, even controlling for prospective indicators of other types of adolescent family adversity. However, not all individuals with abusive experiences showed such difficulties. A large portion reported no psychiatric problems over the 30-year follow-up period. Further tests also showed positive adaptation in other domains such as health, inter-personal relationships of non-criminality in this non-disordered group, supporting the view that these individuals can be described as “resilient” in the face of abuse.

Findings also highlighted two domains of particular relevance for understanding the risk of psychopathology in the context of abuse:

1. Variations in the characteristics and severity of abuse were strongly related to better or worse outcomes in adulthood.
2. Prospective and retrospective assessment of individuals’ relationships with parents, friends, and partners were potent predictors of adult resilience.

This design did not allow to test for the direction of these effects. It’s possible that psychiatric problems will undermine individuals’ relationship competence, just as chronic problems in interpersonal relationships may elevate the risk of future mental illness. These findings suggest that understanding the processes whereby relationship competencies are developed and maintained constitutes an important goal for future research on resilience in individuals exposed to abusive experiences and may be a core target for clinical interventions.