In a survey of 52 studies, Roberts et al. (1998) found that the average prevalence of psychiatric disorder in children and adolescents was 15.8%, with a range of 11–22%.

Sawyer et al. (2000) surveyed a representative sample of 4500 Australian school children aged 4–17 years, using the Child Behavior Checklist (CBCL) (Achenbach, 1991), the Diagnostic Interview Schedule for Children (DISC-IV) (Shaffer et al., 2000), and the Child Health Questionnaire (CHQ) (Landgraf, et al., 1996). Adolescents aged 13–17 years also completed the Centre for Epidemiologic Studies Depression Scale (CES-D) (Radloff, 1977) and the Youth Risk Behavior System Questionnaire (YRBS) (Berener et al., 1995). Sawyer et al. found that 14.7% of children and 13.1% of adolescents were in the clinical range. Children and adolescents with mental health problems had a poorer quality of life, lower self-esteem, and worse school performance than those who did not have, and were a greater burden to their families. Adolescents with mental health problems reported high rates of suicidal behavior, smoking, drinking, and drug use. Few psychiatrically disturbed children were receiving any professional treatment, and those who did receive help usually obtained it from general practitioners, school counselors, or pediatricians. Very few had attended a specialized mental health service or clinician.

These sobering statistics indicate that, even if primary diagnostic and therapeutic services were effective, and even if the links between primary and specialized professional services were efficient, there are far too many seriously disturbed families for existing facilities to serve. Furthermore, there is little empirical support for different kinds of mental health treatment outside of laboratory studies (Weisz et al., 1995), and the dropout rate from community services is alarmingly high. For these reasons, the idea of prevention has been promoted. Prevention aims to avert or divert unfavorable developmental trajectories in such a way as to reduce the incidence or severity of psychopathology and promote mental health.

The risk factors known to be associated with later psychiatric disorders vary from biological (e.g., genetic or chromosomal abnormality; exposure to intrauterine toxins such as alcohol or nicotine; premature birth; exposure to toxins such as lead during early development; and chronic physical disability such as epilepsy or brain injury) and temperamental (e.g., behavioral inhibition or difficult temperament) to familial (e.g., parental depression, alcoholism and antisocial personality; disorganized infant–parent attachment; coercive child rearing; single-parent or blended families; marital discord and domestic violence; physical abuse, sexual abuse and neglect), socioeconomic (e.g., poverty; membership of a disadvantaged minority group) and catastrophic (e.g., civilian disaster or war).

Protective factors counterbalance risk. It is known, for example, that an easy, likeable temperament, above-average intelligence, good support from at least one parent, a cohesive family environment, and social capital in the form of good schools, adequate community resources for sport and skill-building, and good employment prospects protect otherwise vulnerable individuals from psychiatric disorders. Protective factors act by moderating the effect of risk factors or by promoting alternative, compensatory processes that enhance personal effectiveness and self-esteem.

Hypothetically, prevention might work by eliminating risk factors (e.g., the cessation of smoking or drinking during pregnancy), or decreasing their impact (e.g., reducing the incidence of premature birth by providing good antenatal care and nutrition for socially disadvantaged women) or by enhancing protective factors (e.g., providing good schools and social opportunities).

In reality, most outcomes have multiple determinants. Psychopathology following child sexual abuse, for example, is associated with three groups of moderating factors: antecedent factors such as the quality of child attachment prior to the abuse; the nature of the abuse experience—repeated, coercive, intrafamilial, genital penetration being the most adverse; and the quality of parental support after the child discloses the abuse. Moderating factors operate through mediating factors, for example: whether the child has sustained posttraumatic stress disorder; the child’s attitude to self and others; and the child’s coping methods (denial, dissociation, distraction, escape, and repetition–compulsion being the most pathogenic).

Preventive intervention that aims to reduce the incidence of psychopathology after child sexual abuse might therefore focus on improving the quality of family support to the child, treating parental psychopathology, counteracting the child’s adverse attitudes to herself and others, helping the child to assimilate and cognitively reconstruct memories of abuse, treating posttraumatic stress disorder, and promoting healthy self-assertiveness and self-protection. Prevention which aims to stop sexual abuse from occurring in the first place relies upon promoting community awareness, teaching children to avoid or report perpetrators, and treating as soon as possible sexually abused boys, a proportion of whom will otherwise grow up themselves to be sexual perpetrators as adults.

Repucci et al. (1999) have reviewed the debate between those who advocate mental health promotion and competence building as the cornerstone of prevention and those who reject competence building and sociopolitical change in favor of reducing the risk of developing a DSM-IV-TR disorder. In other words, universal approaches to prevention are pitted against programs that target at-risk individuals or groups for intervention. This debate has not been resolved.

Prevention refers to intervention that aims to eliminate, or reduce the incidence of, or ameliorate the severity of, general or specified psychopathology in the population as a whole or in particular groups that are at risk of developing psychiatric disorder or impairment.

Preventive intervention can be classified according to when in the course of the development of a psychiatric disorder the intervention is applied (Caplan, 1964). Primary prevention refers to intervention in normal populations to avert future mental ill health (e.g., school-based alcohol and drug education). Secondary prevention focuses on special at-risk groups in order to stave off the development of psychopathology (e.g., the treatment of sexually abused children). The term tertiary prevention (the early treatment of patients with established psychiatric disorders) has fallen into disuse.

An alternative classification system has been proposed by the Institute of Medicine (Mrazek & Haggerty, 1994). Universal prevention is offered to the entire population of a particular area (e.g., good antenatal care). Targeted prevention is offered to particular groups. Targeted indicated prevention is directed at groups who are identified as being at risk by virtue of biological markers or symptom patterns (e.g., children with epilepsy, or highly aggressive preschool children). Targeted selective prevention is aimed at children who are at increased risk by virtue of their membership of a vulnerable subgroup (e.g., the children of highly stressed, economically disadvantaged, single mothers), or because they are experiencing or about to experience a life transition or stressful event (e.g., change of schools or divorce).

Prevention can also be classified in accordance with the level and timing of intervention. The level of intervention refers to whether the intervention is delivered to the individual, the family, the peer group, the school, the workplace, or the community as a whole. Timing refers to the developmental period when the intervention occurs: antenatal, infancy, preschool, middle childhood, adolescence, or adult.

Take, for example, the following program. Olds et al. (1986) successfully reduced the incidence of child maltreatment by delivering a nurse-home-visiting program to mothers who were at risk of abusing their children by virtue of adolescent pregnancy, poverty, and single parenthood. The program began antenatally and continued until the child was 2 years old. This was a primary, targeted, selective prevention of family-level type, delivered in the antenatal/infancy period.

Discontinuity refers to breaks or changes in development, which lead to a different outcome. For example, a number of studies have shown that a proportion of children who have been seriously sexually abused have no discernible psychopathology at the time of ascertainment. However, Gomez-Schwarz et al. (1990) found that, when assessed 18 months later, many previously asymptomatic children had serious psychopathology.

Divergent development refers to the way that a single stressor can lead to a variety of later outcomes (e.g., child sexual abuse is linked to disturbed self-concept, chronic emotional distress, self-harm, substance abuse, dissociative disorder, somatoform disorder, sexual problems, and revictimization or sexual perpetration in adulthood).

Equifinality or convergent development refers to the way that a variety of risk factors affecting the child at different periods of development can lead to the same outcome. For example, antisocial behavior in male adolescents is predicted by genetic background, disorganized attachment, coercive child rearing, aggressive behavior in early childhood, impaired verbal intelligence, learning problems, gravitation toward delinquent companions in late childhood, and early initiation into alcohol and drug use. Some of these risk factors may be causative whereas others represent points on the longitudinal development toward an undesirable end point.

It is important to determine whether a risk factor is causal (e.g., genetic factors in attention-deficit/hyperactivity disorder) or a noncausal correlate (e.g., abnormal saccadic eye movements in schizophrenia). Prevention should be aimed at those causal factors that are accessible for intervention. Furthermore, a single risk factor can play different parts in the causation of psychiatric disorder. Family dysfunction, for example, can precede and foreshadow sexual abuse, increase the likelihood that it will occur, be precipitated by the disclosure of abuse, and aggravate its effect (Spaccarelli, 1994).

The possibility of adverse and favorable reciprocal interactions should be considered. For example, premature children raised by disadvantaged parents are more likely to develop depressed intelligence and learning problems than are those raised in families that provide good social and language stimulation to the child during infancy and the preschool period. Preventive intervention might, for example, intervene to generate progressively more and more favorable interactions between premature child and mother.

To exemplify the reasoning behind prevention programs, the problem of conduct disorder and antisocial behavior will be discussed.

Conduct disorder is associated with four kinds of behavior: aggressiveness (intimidation, use of a weapon, cruelty, fire-setting, coercive sexual behavior); deceitfulness (lying, stealing); rule violation (refusal to follow the rules at home or school, running away from home, truancy, vandalism); and impulsiveness (explosive anger, thoughtless destructiveness). Early-onset conduct disorder is more common in boys and has a worse prognosis than the adolescent-onset variety. Conduct disorder is often preceded by oppositional-defiant disorder and, in 25% of cases, evolves into adult antisocial personality disorder, which is likely to be associated with serious criminal behavior. Since the prevalence of conduct disorder is high (about 5.5% in both sexes; 8.1% in boys; 2.8% in girls), any reduction in the number of those with the disorder would be likely to result in huge savings for the community by virtue of increased personal productivity, less crime, and sparing of the police, criminal justice, and correctional systems.

The following risk factors have been identified: genetic factors (e.g., the inheritance of callous, remorseless personality traits); disorganized parenting; parental depression, personality disorder, antisocial behavior, and substance abuse; disrupted attachment experiences; poor language stimulation and impaired verbal intelligence; child maltreatment; failed fostering; academic failure; school dropout; gravitation toward antisocial peers; and early introduction to alcohol and substance abuse. The influence of criminogenic environment is probably through the stress experienced by vulnerable parents in such circumstances and the malign influence of antisocial peers. These risk factors interact in a transactional, developmental cascade.

Hypothetically, intervention could begin in the antenatal period with the identification of vulnerable parents and the provision of supportive intervention; in the preschool era with the identification and treatment of depressed or disorganized parents with oppositional children and the provision of language enrichment education; and in primary school with identification of those at risk of dropping out of school, the availability of remedial education, and the promotion of therapeutic intervention for children with disruptive behavior. It is apparent that intervention has much less chance of success if the adolescent has dropped out of school and begun to associate preferentially with antisocial peers.

Infancy

The University of Rochester Nurse Visitation Program (Olds et al., 1986) compared the effect of four programs for antenatal women in a random controlled assignment design: level 1, information and support; level 2, free transport to antenatal and well-baby clinics; level 3, nurse home visiting; and level 4, nurse home visiting continuing regularly for 2 years. Adolescent antisocial behavior was averted, particularly in the level 4 program. For example, level 4 versus level 1 groups had 50% fewer arrests, 30% fewer convictions, and 10% fewer juvenile supervision orders. In earlier years, in level 4 families there were less child abuse, neglect, substance abuse, and parental crime, and fewer pregnancies.

The Pre-School Period