Conduct Disorder Research Paper

Conduct disorder is identified in childhood as a pattern of clinical dysfunction, usually during elementary school years. Yet, for many individuals, conduct disorder is a pattern of functioning over the life span. This research paper discusses characteristics of conduct disorder, continuities and discontinuities over the course of development, issues and challenge for research, and implications of selected findings for social policy.

OUTLINE

I. Conduct Disorder in Childhood and Adolescence

A. Diagnosis and Prevalence

B. Age of Onset and Subtypes of Conduct Disorder

C. Correlates and Associated Features

1. Child Characteristics

2. Parent and Family Characteristics

3. Contextual Conditions

D. Factors that Influence Onset of Conduct Disorder

1. Risk Factors

2. Protective Factors

3. General Comments

II. Conduct Disorder over the Course of Development

A. Early Development: Infancy and Preschool Years

B. Adult Outcomes

1. Antisocial Personality Disorder and Psychopathy

2. Other Outcomes

C. General Comments

III. Issues and Challenges of Developmental Perspectives

A. Continua of Dysfunction and Risk

B. Packages of Influences and Outcomes

C. Variations in Patterns of Influence and Outcome

IV. Interventions to Promote Prosocial and to Decrease Antisocial Behavior

A. Treatment

B. Prevention

C. Social Policy and Action

V. Conclusions

I. CONDUCT DISORDER IN CHILDHOODAND ADOLESCENCE

A. Diagnosis and Prevalence

Extremes of conduct problems are delineated in contemporary diagnosis, as represented by the Diagnostic and Statistical Manual of Mental Disorders. Conduct Disorder (CD) is the diagnostic category for coding antisocial behavior among children and adolescents. The essential feature is a pattern of behavior in which the child ignores the rights of others or violates ageappropriate norms and roles. Table I lists the main symptoms that conduct youths exhibit. In contemporary psychiatric diagnosis, a diagnosis of CD is provided if: (1) the individual shows at least 3 symptoms of those listed in Table I; (2) the symptoms were evident within the past 12 months; and (3) at least one of the symptoms was evident in the last 6 months.

Table I Symptoms Included in the Diagnosis of Conduct Disorder

---

1. Bullying or threatening others.
2. Fighting.
3. Using a weapon that can cause serious physical harm to others.
4. Being physically cruel to people.
5. Being physically cruel to animals.
6. Stealing and confronting a victim (e.g., mugging, purse snatching, extortion, armed robbery).
7. Forcing someone into sexual activity.
8. Firesetting.
9. Destroying property of others.
10. Breaking into someone else’s house, building, or car.
11. Frequent lying or “conning” others.
12. Stealing without confronting a victim.
13. Staying out late at night despite parental prohibitions.
14. Running away from home.
15. Being truant from school.

---

Using diagnostic criteria such as those in Table I or prior versions of the DSM, the prevalence of the disorder among community samples of school-age youth is approximately 2 to 6%. One of the most frequent findings is that boys show approximately 3 to 4 times higher rates of CD than girls. The sex differences may be explained by differences in predispositions toward responding in aggressive ways and in socialization through parent-child interactions in relation to aggression, expression of anger, experience of empathy and guilt. Differential responding on the part of parents may contribute to greater sensitivity of girls to the emotions of others, to their higher levels of empathy, and their reduced outward expression of aggression, compared to boys.

Differences in the base rates of boys and girls for a number of behaviors such as engaging in rough and tumble play, bullying others, not complying with requests, and fighting have implications for the greater prevalence of conduct disorder. The symptoms that are listed in the diagnostic criteria emphasize confrontive and violent acts that are more likely in boys than girls. Because of low rates of these behaviors in girls, even a few instances, albeit below the threshold of existing diagnostic criteria, may be clinically important. These base-rate differences have raised the possibility of a sex bias in the diagnostic criteria which would also explain, or at least contribute to, the greater prevalence of CD in boys than in girls. In general, research on normative development has revealed qualitative and quantitative differences between boys and girls in behaviors related to aggression and antisocial acts, but the information has not yet influenced diagnostic practices.

Age variations reveal interesting patterns in prevalence rates. Rates of conduct disorder tend to be higher for adolescents (approximately 7% for youths ages 12 to 16) than for children (approximately 4% for children age 4 to 11 years). The increase seems to be due to increases in onset among adolescent girls and among youths who engage in nonaggressive forms of antisocial behavior (e.g., truancy, running away). Sex differences are apparent in the age of onset of dysfunction. The median age of onset of dysfunction is 8 to 10 years of age. Most boys have an onset before age 10 (median = 7 years old). For girls, onset of antisocial behavior is concentrated in the 14-to-16 year age range (median = 13 years old). Characteristic symptom patterns are different as well. Theft and aggression are more likely to serve as a basis of referral among antisocial boys. For girls, antisocial behavior is much more likely to include sexual misbehavior.

B. Age of Onset and Subtypesof Conduct Disorder

Conduct disorder includes a heterogeneous set of problem behaviors. Research has identified subtypes in an effort to find meaningful ways of grouping various sets of symptoms and to understand processes leading to onset and course of conduct disorder. Many different ways of delineating subtypes and patterns have emerged. Recent attention has focused on age of onset as a way of accounting for prevalence differences over the course of development and sex differences in symptom patterns.

Child-onset conduct disorder consists of youths whose dysfunction is evident early in childhood, beginning with stubbornness, noncompliance (e.g., Oppositional Defiant Disorder [ODD]) and hyperactivity (e.g., Attention-Deficit/Hyperactivity Disorder [ADHD]). The symptoms may progress to those of CD, even though many of the youths retain the symptoms from these other diagnoses. Youths with child onset are more likely than those with adolescent onset to engage in aggressive and criminal behavior and are more likely to continue their dysfunction into adulthood. Thus, child-onset conduct disorder is the more severe form.

Adolescent-onset conduct disorder is more common than child onset. During adolescence, many youths engage in criminal behavior. For many of these youths, the acts are isolated; for others, the pattern meets criteria for CD. Both child- and adolescentonset conduct disorder youths engage in illegal behavior during adolescence. However, those with child onset are more likely also to engage in aggressive acts and to be represented primarily by boys. Those with adolescent onset are more equally distributed between girls and boys. Peer group influences are considered to play a central role in emergence and onset of adolescent conduct disorder.

Child onset has been particularly well studied in relation to parent-child interaction. Evidence suggests that parent child-rearing practices contribute to child onset by inadvertently promoting aversive behavior in the child. Reinforcement of deviant behavior, inattention to positive, prosocial behavior, and coercive interactions between parent and child lead to escalation of aggressive child behavior. This, in turn, leads to stable patterns of child aggression that has other consequences (e.g., poor peer relations, association with deviant peers, school failure).

Child and adolescent onset subtypes, at this point in the research, do not yet offer an explanation of the different patterns. Even so, age of onset may be a useful point of departure for connecting subtypes of conduct disorder to specific developmental processes. Perhaps influences studied in developmental research (e.g., regulation and dysregulation of emotions, bonding to parents, peers relations) and transitions over the course of development (e.g., school entry) can be readily integrated with these different patterns. Also, developmental work on understanding peer socialization may provide clues regarding early patterns and how they lead to different trajectories. In this regard, work on child popularity and rejection may be important because peer reactions predict later dysfunction.

C. Correlates and Associated Features

1. Child Characteristics

Children who meet diagnostic criteria for CD are likely to show a number of other problem behaviors than those included in the diagnosis. They are likely to argue with adults, lose their temper, actively defy and refuse to comply with requests, deliberately annoy others, and they are angry and resentful. These behaviors, as a group, are occasionally referred to as oppositional behavior and comprise their own diagnostic category (Oppositional Defiant Disorder), alluded to previously. Developmentally, oppositional behaviors are precursors to conduct disorder for many youths. Most children who evince conduct disorder probably have this early history of oppositional problems; but most children with oppositional problems are not likely to progress to conduct problems. Longitudinal research is critical in delineating the conditions leading to the continuation and escalation of behavioral problems.

In addition to oppositional behavior, many youths with severe conduct problems are considered by their teachers and parents to be “hyperactive.” There is a reasonable basis for this. A large percentage of children (e.g., 40-70%) diagnosed with CD also meet criteria for Attention-Deficit/Hyperactivity Disorder. The core symptoms of ADHD include inattention, impulsiveness, and hyperactivity. The general point to underscore here is that children and adolescents with the diagnosis of CD are likely to have many other symptoms.

There are other characteristics that affect diverse facets of functioning as well. Children with conduct disorder are also likely to show academic deficiencies. They are more likely to repeat a grade, to show lower achievement levels, and to end their schooling sooner than their peers matched in age, socioeconomic status, and other demographic variables. Such children are often seen by their teachers as uninterested in school, unenthusiastic toward academic pursuits, and careless in their work.

Poor interpersonal relations also are associated with conduct disorder. Youths with conduct disorder often are socially ineffective in their interactions with adults (e.g., parents, teachers, community members) and engage in behaviors that promote deleterious interpersonal consequences such as peer rejection. Conduct disorder youths are often deficient in attributional processes and cognitive problem-solving skills that underlie social behavior. For example, such youths are more likely than their peers to interpret gestures of others as hostile and are less able to identify solutions to interpersonal problem situations and to take the perspective of others.

2. Parent and Family Characteristics

Several characteristics of the parents and families of conduct disorder children are relevant to conceptualization of the dysfunction. Among the salient characteristics are parent psychopathology and maladadjustment, criminal behavior, and alcoholism. Parent disciplinary practices and attitudes also are associated with conduct disorder. Parents are likely to show especially harsh, lax, erratic, and inconsistent discipline practices. Dysfunctional relations are also evident, as reflected in less acceptance of their children, and in less warmth, affection, and emotional support, compared to parents of nonreferred youths. At the level of family relations, less supportive and more defensive communications among family members, less participation in activities as a family, and more clear dominance of one family member are also evident. In addition, unhappy marital relations, interpersonal conflict, and aggression characterize the parental relations of antisocial children. These characteristics are correlated with, and often antecedent to, conduct problems, but do not, of course, necessarily cause or inevitably lead to those problems.

3. Contextual Conditions

Conduct disorder youths are likely to live in conditions of overcrowding, poor housing, and high crime neighborhoods, and to attend schools that are in disadvantaged neighborhoods. Many of the untoward conditions in which families live place stress on the parent or diminish the threshold for coping with everyday stressors. The net effect can be evident in adverse parent-child interaction in which parents inadvertently engage in patterns that sustain or accelerate antisocial and aggressive interactions. Also, contextual factors (e.g., poor living conditions) are associated with other influences (e.g., deviant and aggressive peer group, poor supervision of the child) that can further affect the child.

D. Factors that Influence Onset of Conduct Disorder

1. Risk Factors

Risk factors refer to characteristics, events, or processes that increase the likelihood (risk) for the onset of a problem or dysfunction (e.g., conduct disorder). Risk factors, as antecedents to the dysfunction, may provide clues as to development and progression of conduct problems, possible mechanisms and processes through which the dysfunction comes about, and periods during development that might be used to identify cases at risk and to intervene. The factors that predispose children and adolescents to conduct disorder have been studied extensively in the context of clinical referrals and adjudicated delinquents. Numerous factors have been implicated. Table II highlights several risk factors that have been studied along with a general statement of the relation that has been found.

Table II Factors that Place Youths at Risk for the Onset of Conduct Disorder

---

Child factors

---

Child Temperament. A more difficult child temperament (on a dimension of “easy-to-difficult”), as characterized by more negative mood, lower levels of approach toward new stimuli, and less adaptability to change.
Neuropsychological Deficits and Difficulties. Deficits in diverse functions related to language (e.g., verbal learning, verbal fluency, verbal IQ), memory, motor coordination, integration of auditory and visual cues, and “executive” functions of the brain (e.g., abstract reasoning, concept formation, planning, control of attention).
Subclinical Levels of Conduct Disorder. Early signs (e.g., elementary school) of mild (“subclinical”) levels of unmanageability and aggression, especially with early age of onset, multiple types of antisocial behaviors, and multiple situations in which they are evident (e.g., at home, school, the community).
Academic and Intellectual Performance. Academic deficiencies and lower levels of intellectual functioning.

---

Parent and family factors

---

Prenatal and Perinatal Complications. Pregnancy and birth-related complications including maternal infection, prematurity and low birth weight, impaired respiration at birth, and minor birth injury.
Psychopathology and Criminal Behavior in the Family. Criminal behavior, antisocial personality disorder, and alcoholism of the parent.
Parent-Child Punishment. Harsh (e.g., severe corporal punishment) and inconsistent punishment increase risk.
Monitoring of the Child. Poor supervision, lack of monitoring of whereabouts, and few rules about where youths can go and when they can return.
Quality of the Family Relationships. Less parental acceptance of their children, less warmth, affection, and emotional support, and less attachment.
Marital Discord. Unhappy marital relationships, interpersonal conflict, and aggression of the parents.
Family Size. Larger family size, i.e., more children in the family.
Sibling With Antisocial Behavior. Presence of a sibling, especially an older brother, with antisocial behavior.
Socioeconomic Disadvantage. Poverty, overcrowding, unemployment, receipt of social assistance (“welfare”), and poor housing.

---

School-related factors

---

Characteristics of the Setting. Attending schools where there is little emphasis on academic work, little teacher time spent on lessons, infrequent teacher use of praise and appreciation for school work, little emphasis on individual responsibility of the students, poor working conditions for pupils (e.g., furniture in poor repair), unavailability of the teacher to deal with children’s problems, and low teacher expectancies.

---

Note. The list of risk factors highlights major influences. The number of factors and the relations of specific factors to risk are more complex than the summary statements noted here.

Merely enumerating risk factors is misleading without conveying some of the complexities in how they operate. These complexities have direct implications for interpreting the findings, for understanding the disorder, and for identifying at-risk children for preventive interventions. First, risk factors tend to come in “packages.” Thus, at a given point in time several factors may be present such as low income, large family size, overcrowding, poor housing, poor parental supervision, parent criminality, and marital discord, to mention a few. Second, over time, several risk factors become interrelated, because the presence of one factor can augment the accumulation of other risk factors. For example, early academic dysfunction can lead to truancy and dropping out of school that further increases the risk for conduct disorder. Third, risk factors may interact with (i.e., be moderated or influenced by) each other and with other variables. As one example, large family size has been repeatedly shown to be a risk factor for conduct disorder. However, the importance of family size as a predictor is influenced by income. If family income and living accommodations are adequate, family size is less likely to be a risk factor. As another example, risk factors often interact with age of the child (e.g., infancy, early or middle childhood). For example, marital discord or separation appear to serve as risk factors primarily when they occur early in the child’s life (e.g., within the first 4 or 5 years). How risk factors exert impact in childhood and why some periods of development are sensitive to particular influences underscore the importance of understanding “normal” developmental processes.

2. Protective Factors

Research on risk factors leads naturally to the study of positive outcomes. The reason is that even under very adverse conditions with multiple risk factors present, many individuals will adapt and will not experience adverse outcomes. A conceptually interesting and potentially critical set of influences that may affect onset are referred to as protective factors. These are characteristics, events, or processes that decrease the impact of a risk factor and likelihood of an adverse outcome. Although protective factors have been less well studied than have risk factors, significant progress has been made.

Researchers have identified protective factors by studying individuals known to be at risk (i.e., show several risk factors) and by delineating subgroups of those who do, versus those who do not, later show conduct disorder. Youths can be identified who are at risk for delinquency based on a number of factors. Yet, not all at-risk youths become delinquent. Those who do not evince delinquency by adolescence are more likely to be first born, to be perceived by their mothers as affectionate, to show higher self-esteem and locus of control, and to have alternative caretakers in the family (than the parents) and a supportive same-sex model who played an important role in their development. Other factors that reduce or attenuate risk include above average intelligence, competence in various skill areas, getting along with peers, and having friends. In many cases, these protective factors seem to be the absence or inverse of a risk factor. For example, easy temperament, academic success, and good relations with parents reduce risk, as does a good relationship with an emotionally responsive, caregiving adult, whether a parent or nonparent figure.

Among the many protective factors, three general categories help to organize current findings. The first is personal attributes of the child. Beginning in infancy and unfolding throughout development, these include such factors as easy temperament, sociability, and competencies at school. The second category is family factors and includes such characteristics as caretaking style, education of the parents, and parent social competence. The third category consists of external supports and includes friendships, peer relations, and support from another significant adult. The categories are useful ways to describe protective factors, but it is important to bear in mind that they tend to be interdependent and reciprocal. For example, child attachment to the parent is important as a protective factor and probably reflects personal attributes of the child in combination with characteristics of the parent. In general, it is useful to conceptualize many of the protective factors as part of transactions between the child and the environment.

3. General Comments

Risk and protective factors refer to variables that influence the probability of onset of an outcome in a population. Although many risk and protective factors have been identified, we do not understand how most of the factors operate. In some cases, there are clues as to the processes and mechanisms that have direct influences on the outcome. For example, harsh punishment practices serve as a risk factor for conduct problems. Punishment is part of a broader set of inept child-rearing practices that have been shown to escalate coercive and aggressive behavior directly. How the parent responds (e.g., coercively or passively) in response to the demands of the child has been shown to increase systematically the level of aggressive child behavior. Moreover, intervening with special training programs that alter how the parents respond to their children decreases child aggression and antisocial behavior. Research on parent discipline practices has made significant gains in moving from identification of a descriptor (risk factor) to the process (means of operation). Understanding the processes leading to dysfunction provides an excellent basis for preventive interventions. Also, understanding discipline practices and their relation to conduct problems draws attention to broader developmental issues. For example, inept discipline practices do not invariably lead to behavior problems. Understanding influences that may attenuate the role of these practices in development could be important. Thus, the study of conduct problems draws attention to discipline practices more generally in development, as well as to the search for protective factors among youths who are subjected to those practices that promote antisocial behavior.

II. CONDUCT DISORDER OVER THE COURSE OF DEVELOPMENT

The manifestations of conduct disorder are likely to change over the course of development. Even so, there may be a continuity in the inferred trait or characteristic that underlies these manifestations. For example, young children (3 to 4 years of age) with conduct problems may be mildly stubborn, break other children’s toys, and “borrow” (take) things that belong to their friends. These behaviors may not predict these same behaviors 10 years later. Yet, these early behaviors may predict other behaviors, such as stealing from stores and confronting strangers with a weapon, that are conceptually related or that belong to the same general class of behaviors. A life-span perspective emphasizes continuities and discontinuities over time and paths and progressions. Behavioral and other manifestations may be discontinuous but still reflect continuity at a broader level of conceptualization. Charting the course over the life span begins with descriptive characterization of conduct problems at different points in development. The period of schoolage years through early adolescence has been especially well studied. The present discussion of conduct disorder focuses on early development and adult outcomes, to fill out the life course of the problems.

A. Early Development: Infancy and Preschool Years

Risk-factor research suggests that a number of signs may be evident in the child, parent, and family context beginning in infancy. Child characteristics (e.g., difficult temperament, neuropsychological deficits, high activity), parent characteristics (e.g., prenatal and perinatal birth complications, parental punishment of the child), contextual characteristics (e.g., stress, marital conflict), and other factors, noted earlier, are likely to be present. In addition, diverse psychological processes and experiences (e.g., development of affect, attachment, and cognition) are likely to be implicated. It is likely that a set of general factors may emerge in early development that increase vulnerability to dysfunction and some set of more specific factors that move the child more specifically to conduct disorder.

Charting the influence of any single factor is difficult because characteristics of child, parent, and contexts are dynamic rather than static. The dynamic feature emphasizes complex interrelations such as the reciprocal and mutual influence of the child on the parent and the parent on the child. As the child interacts with others (e.g., peers), reciprocal and dynamic influences continue and have their own consequences (e.g., early aggression may lead to peer rejection).

In addition to dynamic influences at a given point, there is a developmental progression over time. The influences can place children on a trajectory or path that refers to a course leading to a particular outcome such as conduct disorder and criminality. The trajectory or path is not necessarily a fixed or determined course, but rather a matter of increased likelihood (probability) that specific behaviors will unfold in the short run and lead to other outcomes in the long run. Some outcomes become more probable (e.g., being arrested, bonding with delinquent peers), and other outcomes become less probable (e.g., graduating high school, sustaining employment). A variety of influences can converge to alter the probabilities.

The progression of characteristics in early development toward conduct disorder has been examined in longitudinal studies from birth through adolescence and young adulthood. In such research, the same individuals are studied at multiple points in time (e.g., every few years) and then early predictors of later behavior can be identified. Through longitudinal studies, one can chart the course over short and long periods and identify transitions from one time to another and the relations among proximal and distal manifestations. Recent research has characterized progressions and different paths and how conduct disorder symptoms and their associated features emerge. Among the salient findings is a progression of severity of conduct problems over time. Trivial antisocial acts precede more severe acts in the child’s repertoire. Youths who show the more serious behaviors (e.g., assault, firesetting) are likely to have progressed through the less severe behaviors (e.g., temper tantrums, noncompliance) but, of course, not all youths who engage in less severe antisocial behaviors progress to more severe antisocial behaviors.

B. Adult Outcomes

1. Antisocial Personality Disorder and Psychopathy

Longitudinal studies show that conduct disorder in childhood predicts conduct disorder up to 10, 20, and 30 years later. Antisocial behavior when continued into adulthood falls into another diagnostic category, namely, Antisocial Personality Disorder (APD). The essential features include a pervasive pattern of disregard of others, violation of the rights of others. The main symptoms of APD include repeatedly engaging in unlawful behavior, deceitfulness (e.g., repeated lying, conning others), impulsivity, irritability, aggressiveness (repeated fighting), disregard for the safety of others, consistent irresponsibility (e.g., repeated failure to retain a job), and lack of remorse. The presence of CD in one’s youth is a prerequisite for the diagnosis of APD. The criteria include many concrete behavioral acts of CD but also encompass more pervasive personality patterns, as reflected in deceit, manipulation, impulsivity, and irresponsibility.

Large-scale epidemiological research has revealed a life-time prevalence rate of APD of 2.1 to 3.3%. Males are approximately 4 to 8 times more likely to be diagnosed with the disorder. The greater prevalence of APD among males compared to females is in keeping with the sex-difference pattern evident in childhood. Follow-up of child conduct disorder has elaborated this sex difference. Boys are much more likely to continue conduct disorder into adulthood and show APD. In contrast, girls are likely to shift into more internalizing types of disorders (e.g., depression, anxiety) in adulthood. This pattern is especially interesting in light of research showing different reactions of boys and girls who are exposed to factors that might increase risk for conduct disorder. For example, exposure to family violence in childhood (ages 6 to 11) is associated with externalizing and internalizing symptoms in boys but primarily internalizing symptoms among girls. The process leading to symptom pattern differences have yet to be elaborated.
The symptoms required for a diagnosis of APD, noted previously, emphasize overt behavioral signs. Over the history of the study of antisocial behavior in adulthood, emphasis has also been accorded internal experience such as lack of guilt or remorse, lack of empathy, and egocentricity. A distinction has been drawn between APD, which emphasizes the behavioral components, and psychopathy, which focuses more on the motivational and interpersonal processes. APD has as its characteristics adverse family background (e.g., low socioeconomic status) and lower IQ. Psychopathy is correlated negatively with anxiety and positively with narcissism. Interestingly, individuals with both APD and psychopathy are those who exhibit the most severe and enduring patterns of antisocial behavior in adulthood. The distinction in the adult literature between behavioral and motivational/interpersonal components is important from a developmental perspective because it identifies different end points of earlier developmental trajectories. Unfortunately, to date there have been few efforts to connect the different outcomes of adulthood with characteristics of early development.

2. Other Outcomes

Among youths who are severely antisocial during childhood, slightly less than 50% continue their conduct disorder into adulthood. What happens to the remainder of youths? If all diagnoses are considered, rather than continuation of conduct disorder alone, 84% of the full sample received a diagnosis of psychiatric disorder as adults. Moreover, diagnosis of dysfunction does not adequately characterize the scope of adjustment difficulties in adulthood. There are many other outcomes identified by following conduct disorder children. As adults, multiple domains may show continued dysfunction, as reflected in psychiatric symptoms, criminal behavior, physical health, and social maladjustment. The characteristics that conduct disorder youths are likely to show when they become adults are presented in Table III. As the table indicates, individuals with a history of conduct disorder evince a broad range of untoward outcomes.

Table III Long-Term Prognosis of Youths Identified as Conduct Disorder:

Overview of Major Characteristics Likely to Be Evident in Adulthood

---

Characteristics in adulthood

---

Psychiatric Status. Greater psychiatric impairment including anatisocial personality, alcohol and drug abuse, and isolated symptoms (e.g., anxiety, somatic complaints); also, greater history of psychiatric hospitalization.
Criminal Behavior. Higher rates of driving while intoxicated, criminal behavior, arrest records, and conviction, and period of time spent in jail.
Occupational Adjustment. Less likely to be employed; shorter history of employment, lower status jobs, more frequent change of jobs, lower wages, and depend more frequently on financial assistance (welfare). Served less frequently and performed less well in the armed services.
Educational Attainment. Higher rates of dropping out of school, lower attainment among those who remain in school.
Marital Status. Higher rates of divorce, remarriage and separation.
Social Participation. Less contact with relatives, friends, and neighbors; little participation in organizations such as church.
Physical Health. Higher mortality rate; higher rate of hospitalization for physical (as well as psychiatric) problems.

---

Note. These characteristics are based on comparisons of clinically referred children identified for conduct disorder relative to control clinical referrals or normal controls or from comparisons of delinquent and nondelinquent youths.

C. General Comments

From a developmental standpoint, it is important to understand the continuities and discontinuities of conduct disorder over the life span. Longitudinal studies have identified intriguing patterns, yet to be explained. For example, the continuity of conduct disorder among boys (ages 7 to 12) is influenced by APD of a parent or child intelligence. With either an APD parent and lower level of intelligence, boys are likely to continue conduct disorder symptoms. How these characteristics operate and combine and the other variables with which each is associated has yet to be studied. The continuity of conduct disorder over the life span warrants mention in another light. The continuity extends beyond the life of the individual, because conduct disorder extends across generations. For example, children are more likely to show antisocial behaviors if their grandparents have a history of these behaviors. Similarly, one of the best predictors of how aggressive a boy will be in childhood is how aggressive his father was when he was about the same age. Thus, the life-span perspective requires consideration of how the dysfunction is extended to one’s offspring and whether there are different modes of transmission.

III. ISSUES AND CHALLENGES OF DEVELOPMENTAL PERSPECTIVES

A. Continua of Dysfunction and Risk

Research often focuses on youths who meet diagnostic criteria for CD. In principle, it is quite useful to specify criteria in this fashion so that diagnoses can be made reliably and that research on these samples can be replicated. Yet, the criteria themselves are difficult to defend. Where one draws the cutoff point to decide dysfunction (e.g., 3 symptoms rather than 4 or 8; duration of 12 months rather than 18, 24, or more) is likely to lead to different findings with regard to risk and protective factors, developmental trajectories, responsiveness to treatment, and prognosis.

Clearly, youths who meet the criteria are likely to be significantly impaired. Yet to understand the nature of conduct disorder more generally, it would be important to extend research to the full spectrum of severity of impairment and dysfunction. For example, youths who show symptoms of CD but who are below, at, and above threshold (e.g., fewer than 2 symptoms, 3 symptoms, or more than 4 symptoms, respectively, as only one way to operationalize threshold) for meeting the diagnosis would be important to study. This type of analysis would permit evaluation of factors that predict functioning across the spectrum of severity and frequency, as well as those that are only predictive of more severe levels or types of dysfunction. In general, conduct disorder is a “fuzzy” insofar as some individuals are at each extreme (clearly conduct disorder, and clearly not) with many shades in the middle. Presumably, there are points on the continuum at which there is a particularly poor prognosis, failure to respond to treatment, and so on. The full spectrum warrants much more attention to understand where the points are warranted to be delineated for intervention and for policy decisions as well.

In a similar vein, many of the factors that contribute to conduct disorder can be conceived along continua. In much of the research that focus on risk factors, groups are selected and compared based on their exposure to and experience of an event. For example, the effects of abusive child-rearing practices on children and adolescents are often studied in this way. Typically in research, one selects abused and nonabused children and then identifies the other characteristics they might show at some later point in time (e.g., symptoms of psychopathology, poor school performance, dysfunctional peer relations). Identification of extreme groups is an excellent point of departure, but we wish to understand the continuum of the risk characteristic. Evaluation of the continua of discipline practices is required to understand the impact of various levels and types of punishment and the point at which these practices become risk factors for various outcomes.

Studying multiple levels of a proposed risk factor is important to reveal the function (or relation) in a more fine-grained fashion than the study of two groups or the presence or absence of a particular characteristic. Many influences are likely to bear curvilinear relations to an outcome of interest, and assessment of different levels of the risk characteristic can reveal this. For example, parental efforts to control their adolescents is related to externalizing symptoms and drug use. However, the relation between degree of parental control and symptoms is not linear. Extremely high or low parental control, but not intermediate control, is associated with adolescent dysfunction. Similarly, adolescent substance abuse is correlated with current dysfunction and predicts lack of academic pursuits, job instability, and disorganized thought processes years later. Yet, the relation of substance use and untoward consequences is not linear. Heavy alcohol or drug use predicts later problems; no alcohol or drug use or consumption whatsoever is associated with undesirable personal and social characteristics as well. Use of a small amount of alcohol or drugs (primarily marijuana) is associated with positive outcomes such as decreased loneliness, reduced self-derogation, improved relationships with family, and increased social support.

The point of these examples is to convey the need to study multiple levels of factors presumed (or indeed known) to increase risk for dysfunction. There may be points at which a given factor has one effect (risk), another at which it has no effect, and another level at which it has an opposite (protective) effect for an outcome of interest. Developmental research examines continuities and discontinuities over time for individuals and groups. Research is needed that examines continuities and discontinuities over dimensions of behavior (e.g., conduct disorder), risk factors (e.g., childrearing practices), and contextual influences (e.g., socioeconomic status). How the dimensions influence development and the points at which risk and impairment are especially likely are not well known.

B. Packages of Influences and Outcomes

A significant challenge for research is the finding that many influences and outcomes come in “packages.” It is difficult to identify simple profile of risk factors that are associated with and are unique to conduct disorder. The reason is that many influences and outcomes come in “packages,” that is, sets of factors that go together and have multiple deleterious outcomes. For example, socioeconomic disadvantage, adverse childrearing practices, parental neglect, and low parental interest in the child’s academic accomplishments are interrelated. The presence of one or two of these risk factors increases the likelihood of a child accumulating more of them. Thus, early child aggression or academic retardation is often associated with peer rejection, association with deviant peers, and placement in a class designed for socially and emotionally disturbed youths. These qualities in turn can lead to a “snowbailing” of additional risk factors.

“Packages” are also evident in the outcomes (e.g., problem behaviors, disorders). Although we are interested in understanding the development and course of specific emotional and behavioral patterns, many of these are embedded in, or are part of, larger packages. For example, antisocial acts often are part of a larger cluster involving multiple problem behaviors (e.g., substance abuse, early sexual activity, and academic dysfunction). A challenge is to explain how these behaviors come together developmentally. A prominent view, referred to as problem behavior theory, suggests that problem behaviors serve similar functions in relation to development. Autonomy from parents and bonding with peers are two of the functions that may be served by such behaviors. Another view is that there is a trait or pervasive tendency to engage in deviant, delinquent, and criminal behavior. The tendency, referred to as low self-control, reflects a propensity to seek pleasures of the moment and short-term solutions to problems.

What has been well established is that multiple deviant behaviors co-occur. Evidence points to complex interrelations among behaviors and patterns that are idiosyncratic and that vary reliably across individuals, situations, and contexts. Understanding the organization of affect, cognition, and behavior and how they emerge and evolve developmentally are central to understanding problem behavioral patterns. No doubt specific factors (e.g., risk and protective) relate to specific outcomes (e.g., conduct disorder), and these are obviously important to identify. Yet from what we know so far, two general conclusions can be reached: (1) multiple paths (e.g., different packages of risk factors) can lead to a specific outcome (e.g., conduct disorder); and (2) a single path (e.g., single or seemingly identical packages of risk factors) can lead to multiple outcomes (e.g., diverse types of dysfunctions or other outcomes). Elaborating specific lines of development and exploring the bases for variation are rich in opportunities for both theory and research.

C. Variations in Patterns of Influence and Outcome

A challenge for research stems from the prospect that some influences and relations may vary systematically as a function of other variables. For example, the relation between characteristics of early development and outcome in relation to conduct problems varies as a function of child sex. It is not merely the case that boys and girls differ on a particular characteristic (e.g., degree of aggressiveness), but rather how the relations among other variables differ as a function of sex. We know, for example, that early signs of aggression in the school is a risk factor for conduct disorder, delinquency, and crime in adulthood for boys but not for girls. The issue in relation to the present discussion is not merely the fact that there are sex differences, but rather the relation between antecedents and outcomes and how and why they are influenced by sex.

Race and ethnicity are also likely to influence relations among factors related to conduct disorder. Differences are known to exist among European American, African American, Asian American, and Hispanic American children in relation to prevalence, age of onset, and course of dysfunction. For example, among youths with substance abuse, one of many behaviors associated with conduct problems, ethnic variation exists in the specific substances used, degree of family monitoring, and amount of exposure to substance use. In addition, whether a particular influence emerges as a risk factor varies as a function of ethnicity. That is, a relation between a particular antecedent and outcome is moderated by ethnicity.

Sex and ethnic differences are not the only factors that influence the relations among other variables. Yet, these two are important and serve as a basis for articulating the challenge for research. Investigators are often interested in developing theories of dysfunction with implied widespread generality of explaining conduct disorder. It is likely that key variables such as sex and ethnicity, but no doubt others as well, influence onset and course of conduct disorder and the suitability of various interventions.

IV. INTERVENTIONS TO PROMOTE PROSOCIAL AND TO DECREASE ANTISOCIAL BEHAVIOR

Conduct disorder represents a serious clinical problem for individuals and their families, as well as a major mental health problem for society at large. From a social perspective, conduct disorder is considered to be the most costly mental disorder, at least in the United States. The costs stem from the fact that for many youths conduct disorder is a life-long problem. Over the course of childhood and adolescence, youths are likely to enter into many systems and programs, including special education classes, mental health services (inpatient or outpatient treatment), and the juvenile justice system. As adults, entry into mental health services and the criminal justice system may continue. Thus, the costs that accrue to care for conduct problem individuals is exorbitant. These costs extend beyond the individual’s lifetime, insofar as conduct disorder tends to be transmitted from one generation to the next. With these considerations in mind, identifying interventions to combat the problem is obviously important. Three broad levels of intervention can be delineated, namely, treatment, prevention, and social policy. An overview of each is presented next.

A. Treatment

Treatment refers to a broad range of interventions that are applied to youths who have been identified as showing conduct disorder symptoms and who experience impairment in their everyday lives. Conduct problem symptoms constitute the most frequent basis for which children and adolescents are referred for treatment. Consequently, there is a need for effective interventions. Many different treatments have been applied to youths with conduct disorder, including psychotherapy, pharmacotherapy (medications), psychosurgery, home, school, and community-based programs, residential and hospital treatment, and assorted social services. Few treatments have been carefully evaluated in controlled studies and shown to reduce conduct disorder problems and to improve functioning of the child in everyday life (at home and at school). A few treatment approaches have been studied and show considerable promise in treating children and adolescents who are referred for treatment. Three of the more well studied treatments are highlighted here.

Parent management training refers to a treatment in which parents are trained to interact with the child in ways that promote prosocial behavior. The treatment is based on learning research from psychology and focuses on the use of reinforcement (e.g., use of contingent consequences), mild punishment (e.g., very brief time out, response cost), and a variety of related techniques (prompting, shaping) to develop child behavior. Extensive research has shown that many parent-child interaction patterns in the home unwittingly foster and escalate child aggressive behavior. Parent management training teaches skills to the parent, develops interactions between parents and the child that promote positive parent and child behavior, and in the process decreases aggressive and antisocial behavior. Several controlled studies have shown that parent management training reduces oppositional, aggressive, and antisocial behavior at home, at school, and in the community. The effects have been maintained in many studies up to 1 to 2 years. Few studies have evaluated the longer term impact of treatment, but evidence has been favorable for these studies.

Another technique is cognitive problem-solving skills training. This technique is based on research showing that conduct problem youths often show distortions in various cognitive processes. Cognitive processes reflect how individuals perceive, code, and experience the world, as reflected in beliefs, attributions, and expectations. A variety of cognitive processes pertain to interactions with others, including the ability to generate solutions to interpersonal problems, to identify the means to obtain particular ends (e.g., making friends) or consequences of action (e.g., what would happen after a particular behavior), and to make attributions to others of the motivation of their actions. Distortions and deficits in these and related cognitive processes relate to conduct problems at home and at school. Problem-solving skills therapy develops skills in approaching interpersonal situations. Youths learn a series of steps or self-instruction statements that help identify prosocial or adaptive solutions and alternative consequences of actions. Children practice using the approach in treatment sessions and at home with their parents. Within the sessions, several techniques are used, including modeling by the therapist, practice, role playing, and reinforcement to shape appropriate behavior of the child. Several controlled studies have shown that problem-solving skills training can reduce aggressive and antisocial behavior in children and adolescents. The effects have been maintained up to one year after treatment.

Multisystemic therapy is a family systems approach to antisocial behavior. The treatment focuses on the child behavior within the context of various systems (e.g., the family, peer group, schools) which may contribute to the child’s problem behavior or could be used to help alter that behavior. Many different techniques are applied within the context of treatment to modify specific behaviors of the child and others whose behaviors may be affected by the child. Parent management training and problem-solving skills training, mentioned previously, often are incorporated into treatment. A focus on the family as a system is designed to build better communication, to reduce negative interactions, and to improve the ability of the parents to function. Factors that can affect these interactions and the child’s problems, such as stress that the parent experiences, marital conflict, association of the child with a deviant peer group, are focused on as well. Several controlled studies have shown that multisystemic therapy reduces delinquency and antisocial behavior. The effects of treatment surpass the effects achieved with other types of treatment routinely provided to antisocial youths (e.g., counseling, probation) and have been maintained up to 5 years after treatment.

There are other promising treatments for conduct disorder, but only a small number of techniques have been carefully evaluated. Although the promising treatments have evidence from controlled studies in their behalf, they still leave many questions unanswered. For example, we do not yet know the longterm effects of even the best available treatments, whether they influence adaptation into adulthood, and for whom they are likely to be most effective. Considerable attention in the field focuses on these questions with the goal of improving treatment.

B. Prevention

Ideally we would like to prevent the onset of conduct disorder so there would be no need for treatment. Actually, prevention includes a number of goals. Primary, secondary, and tertiary prevention have been delineated to note whether the intervention is designed to prevent the onset of dysfunction (incidence), to reduce the severity, duration, or manifestations among cases with early signs (prevalence), and to delimit the disability or dysfunction and its complications among persons who have early signs or the dysfunction itself, respectively. A more recent classification distinguishes types of interventions and includes: universal interventions, which are designed for the general population, are low cost, and deemed beneficial for persons in general; selective interventions, which are targeted to subgroups that have elevated risk for the disorder; and indicated interventions, which are targeted to high-risk individuals who already show detectable signs or symptoms of developing the problem.
Research on risk factors has been very helpful in guiding preventive efforts. Because many risk factors for conduct disorder are known, we can identify youths who are at high risk and provide preventive interventions. In addition, the risk factors may suggest processes through which antisocial behavior may emerge. For example, we know that harsh punishment practices can contribute to antisocial behavior and that altering these practices reduces antisocial behavior. Parenting is one of the foci of many early intervention programs.

Prevention programs come in many different forms. Early intervention programs with the family have been effective in altering conduct problems. High-risk families are identified, usually by such factors as low socioeconomic status, low educational attainment, and high-stress living conditions. Intervention programs sometimes begin before the child is born to provide counseling regarding maternal care, to provide support in the home to reduce stress, and to prepare the parents for child-rearing demands. After the infant is born, the program may continue for a few years to help support parents, to develop cognitive skills of the child, and to enroll the child in a preschool program. Programs of this type can have broad impact beyond reducing the incidence of conduct problems. Adolescents who have received such programs when they were young, compared to those who did not, show lower arrest rates, higher educational attainment, less substance use and abuse.

Prevention programs are often conducted in the schools because there are opportunities to provide programs to youths in larger numbers, in the context of peers, and on a regular basis for protracted periods. Programs often focus on developing positive skills and success experiences at home and at school. The reason for this focus is that bonding to deviant peers and poor connections with the family are risk factors for such behaviors as delinquency and substance abuse. Developing such success experiences in the schools among elementary school children has increased bonding to families and decreased rates of antisocial behavior and substance abuse.

Prevention is an obviously critical focus. To date, the evidence shows that preventive interventions can have impact on child functioning and onset of conduct problem behaviors. At the same time, the very best and most effective programs show that the incidence of conduct problems can be reduced, but by no means eliminated. The long-term effects of prevention are not well studied. However, the value of preventive efforts is that many outcomes may be improved by early intervention.

C. Social Policy and Action

Social policy refers to governmental and legislative efforts to implement changes to benefit society or a particular segment of society and, in this sense, is a social intervention. In principle, the interventions rely on practices that have emanated from research on the nature of the problem (e.g., risk and protective factors) and intervention practices that are or appear to be promising. Thus, policy interventions are not necessarily different from those discussed in the context of treatment or prevention. As an example, Head Start has been implemented as an early preschool program to have broad impact on child development and families. Many of the practices are designed to improve educational and social goals of the children and to improve conditions that in the long term will have impact on children.

Efforts to influence policy are reflected in recommendations to alter or modify practices with the goal of decreasing aggression, violence, and other conduct problems on a large scale. As a recent example, a Commission on Violence and Youth of the American Psychological Association (1993) completed a 2-year study and concluded that, “society can intervene effectively in the lives of children and youth to reduce or prevent their involvement in violence” (p. 5). Several specific suggestions were elaborated to convey how this can be accomplished. Table IV summarizes the categories of actions that can be taken. Each of these was developed in detail to convey their connection to what is known from current research on risk factors, onset of dysfunction, and interventions.

Table IV Overview of Recommendations to Curb Violence

---

• Early childhood interventions directed toward parents, child-care providers, and health-care providers to help build the critical foundation of attitudes, knowledge, and behavior related to aggression.
• School-based interventions to help schools provide a safe environment and effective programs to prevent violence.
• Heightened awareness of cultural diversity and involvement of members of the community in planning, implementing, and evaluating intervention efforts.
• Development of the mass media’s potential to be part of the solution to violence, not just a contributor to the problem.
• Limiting access to firearms by children and youth and teaching them how to prevent firearm violence.
• Reduction of youth involvement with alcohol and other drugs, known to be contributing factors to violence by youth and to family violence directed at youth.
• Psychological health services for young perpetrators, victims, and witnesses of violence to avert the trajectory toward later involvement in more serious violence.
• Education programs to reduce prejudice and hostility, which are factors that lead to hate crimes and violence against social groups.
• Efforts to strengthen the ability of police and community leaders to prevent mob violence by early and appropriate intervention.
• Efforts by psychologists acting as individuals and through professional organizations to reduce violence among youth.

---

From the Executive Summary of the Report of the American Psychological Association Commission on Violence and Youth (1993). Violence and youth: Psycbology’s response (Vol. 1). Washington, DC: American Psychological Association.

There are multiple opportunities within society to reduce influences that can contribute to conduct problems and aggression more generally. For example, the use of corporal punishment in child-rearing and school discipline, violence in the media, especially television and films, and social practices that permit, facilitate, or tacitly condone violence and aggression (e.g., availability of weapons), to mention salient issues, are some of the practices that are relevant to the issue of aggression and antisocial behavior in society. We take as givens a backdrop of factors and practices that contribute in significant ways to aggression and antisocial behavior in society. The factors need to be scrutinized in relation to policy regarding child management and care.

As an illustration, the use of corporal punishment (e.g., physical aggression against children) is already implicated as a contributor to child aggression. The extensive use of corporal punishment is one of the givens in our society – a right that accompanies parenting and often teaching – that might be challenged if there is broad interest in delimiting aggression and antisocial behavior. Corporal punishment in child discipline at home and at school has been banned in a number of countries (e.g., Austria, Denmark, Finland, Norway, and Sweden). Large-scale efforts to reduce risk factors in this fashion are critically important in addition to the more common prevention and treatment efforts.

It might be useful to conceptualize the full range of influences in terms of a risk-factor model in which there are multiple influences that contribute to the outcome. Yet, in a risk-factor model, multiple influences add and combine to increase the likelihood of the outcome (e.g., aggression). Small influences can combine (additively and synergistically) and have significant impact, even if their individual contribution would be nugatory. We want to reduce risk factors not because individually they are the cause or because they will eliminate the problem, but because they are likely to have a palpable impact.

Limiting violence in the media can be seen as one influence likely to affect the level of violence and aggression in society. Efforts to quell gross displays of violence in the media are countered with arguments noting the benefits of television (e.g., education) and the responsibilities of others (e.g., parents) in policing what children watch. Yet, the significant impact of the media on antisocial and at-risk behaviors already has been well documented. Reducing aggression in the media is likely to have impact, even though media violence is not “the cause” of violence in society.

A commitment at the policy level and at the level that can mobilize social forces that influence, express, or model aggression could have significant impact on the problem. Social influences involving the matrix of societal displays, encouragement, and implicit endorsement of aggression including the media at all levels ought to be mobilized more systematically for a broad effort to ameliorate aggression and antisocial behavior. Again, this is not the solution nor a reflection on the cause of aggression in society, but rather a way to have impact in one more incremental way.

V. CONCLUSIONS

Conduct disorder represents a special challenge given the multiple domains of functioning that are affected. For many individuals, severe antisocial behavior and associated dysfunction in multiple spheres represents a lifelong pattern. Advances have been made in understanding the characteristics and patterns evident in school-age children and adolescents. Also, efforts have been made to chart the life course longitudinally, different paths leading to conduct disorder in childhood and adulthood, and the role that specific influences play (e.g., parent child-rearing practices, peers) at different points in development. Research has identified characteristics of the child, parent, family, and contexts that contribute to the emergence and maintenance of conduct disorder.

The lifelong pattern of conduct disorder and the transmission of the problems within families from one generation to the next underscore the importance of a developmental and life-span perspective. It will be important to identify the course and various paths and to examine developmentally opportune points of intervention. Over the course of development, influences vary in their contribution to conduct disorder. For example, during adolescence, the influence of peers on the appearance of conduct problem behavior is marked. Peer influences have been implicated in the onset, maintenance, and therapeutic change of antisocial behaviors. Identifying how such influences operate and precursors to such influences has obviously important implications for intervening.

Although many fundamental questions remain about conduct disorder over the course of development, sufficient information is available to advance policy recommendations, a few of which were noted previously. A broad range of social interventions are required to have impact on such conduct problems. Specific programs and interventions developed by mental health professionals play a major role, but these programs do not exhaust the options. Broader social practices warrant scrutiny in ways that balance individual freedoms and responsibilities. Recommendations from research on ways of reducing conduct problems require addressing these broader issues.

BIBLIOGRAPHY:

1. American Psychiatric Association (1994). Diagnostic and statistical manual of mental disorders. (4th ed.). Washington, DC: Author.
2. American Psychological Association, Commission on Violence and Youth (1993). Violence and youth: Psychology’s response (Vol. 1). Washington, DC: American Psychological Association.
3. Kazdin, A. E. (1995). Conduct disorder in childhood and adolescence (2nd ed.). Thousand Oaks, CA: Sage.
4. Ketterlinus, R. D., & Lamb, M. E. (Eds.) (1994). Adolescent problem behaviors: Issues and research. HiUsdale, NJ: Erlbaum.
5. McCord, J., & Tremblay, R. E. (Eds.) (1992). Preventing antisocial behavior. New York: Guilford.
6. Mrazek, P. J., & Haggerty, R. J. (Eds.) (1994). Reducing risks for mental disorders: Frontiers of preventive intervention research. Washington, DC: National Academy Press.
7. Patterson, G. R., Reid, J. B., & Dishion, T. J. (1992). Antisocial boys. Eugene, OR: Castalia.
8. Pepler, D. J., & Rubin, K. H. (Eds.) (1991). The development and treatment of childhood aggression. Hillsdale, NJ: Erlbaum.
9. Peters, R. D., McMahon, R. J., & Qinsey, V. L. (Eds.) (1992). Aggression and violence throughout the life span. Newbury Park, CA: Sage.
10. Robins, L. N. (1991). Conduct disorder. Journal of Child Psychology and Psychiatry, 32, 193-212.
11. Robins, L., & Rutter, M. (Eds.) (1990). Straight and devious pathways from childhood to adulthood. Cambridge: Cambridge University Press.