JCCAP FDF/2017/Conduct disorder

This is the landing page created at the First JCCAP Future Directions Forum to help organize information about publicly available data sets as well as some suggestions for best practices in designing and reporting research looking at these types of variables. There were four keynote addresses: Dr. Eric Youngstrom discussing future directions in assessment, Dr. Matthew Nock discussing suicidal and self injurious behaviorDr. Mary Fristad discussing bipolar disorder, and Dr. Daniel Shaw discussing trajectories and treatment for conduct problems. Each of these was the focus for 2-3 smaller breakout discussion sessions led by content experts. There are a set of four pages that gather the ideas and resources related to these sessions.

Dr. Daniel ShawEdit

Dr. Shaw was the keynote speaker who focused on the development and prevention of early conduct problems.

Developmental course of conduct problemsEdit

Basic development work of Campbell et al. and Richman et al. in the 1980s

Early intervention efforts of Eyberg et al. and Webster-Stratton et al. in 1990s

Why do we start in early childhood? The earlier the onset, the more serious the outcome. Moffitt and Patterson's work looked at onset at under age 10, now it's as young as age 2. Some evidence suggest that conduct problems (CP) are more malleable before age 5.

Stability of CP in early childhoodEdit

From sample of 310 low-income toddler males recruited from WIC in early 1990s. Among those who scored in the 90th percentile on externalizing and aggressive factors on the CBCL and were identified at age 2 by their mothers, 60% remained in clinical range at age 6 and 100% remained above the median. Only 16% of those below the 50th percentile at age 3 moved into clinical range at 6. Patterson (1982) saw similar findings.

Longitudinal studies initiated prior to age 2 find modest evidence of stability across time and informant/context, but this changes by age 3, especially using observation of CP. If you change informants, then this stability drops.

Risk factorsEdit

Negative emotionality, inhibitory control, fearlessness, deceitful-callous/unemotional behavior. There are still relatively few genetically informed studies to tease apart biological and contextual factors (Leve et al., 2009). Parent attributes show a stronger onset prior to age 2 than child behavior (there is depression, anxiety, inter-parental conflict, etc...). A lot of different parenting factors during their 2's have strong influences on the children's behaviors. Attachment theory emphasizes lack of responsibility during first 2 years, unwittingly training child to use oppositional/aggressive means to reliably elicit attention from parent. (Gard et al., in press; Quevedo et al., 2017, Sitnick et al., 2017).

Caspi & Moffitt's (2001) work looked at the genetic and environment interaction in the risk factors of child onset of CP.


Nonoffenders vs. non-violent offenders:

  • Family income

Violent offenders vs. nonoffenders:

  • Family income
  • Oppositional behavior
  • Emotional regulation (ER) problems
  • Minority status

Violent vs. nonviolent offenders:

  • Rejecting parenting
  • Oppositional beahvior
  • ER

Stinick, Shaw et al., 2017, Child Development

Future directions in development of early starting CPEdit

Child sexEdit

There is very few research done that includes girls. Recent research, though still not extensive, have included girls and have found that the magnitude of the risk factors do not differ between girls and boys. Early steps multisite sample


Broad issues here is formulating models that better reflect the pervasiveness of risk factors associated with living in poverty. The family stress model (Conger, Mistry) have traditionally said that poverty leads to increased stress, which adds to CP. Poverty limits parental involvement, decreases access to resources, creates social isolation due to stigma, leads to differences in parenting styles (ex: corporal punishment is seen as a better trade off than having child misbehave and get themselves killed), and sees differences in environment (ex: lead exposure).

New InterventionsEdit

Most new interventions focused on promoting positive parent-child relationships and general child outcomes rather than CP per se, using the attachment theory as a basis.

Challenges in prevention and treatmentEdit

Emphasizing similar types of parenting would not be relevant for al kinds of parents of children with CP, despite the strong evidence that a diverse number of children benefit from this.

Dr. Deborah Drabick: AssessmentEdit

Developmental approachEdit

I. Change, timing, and multiple determinants 

II. Factors that initiate and maintain disordered trajectories may differ

III. Hierarchically integrated or cumulative

IV. Identify early patterns of maladaptation related to later disorder

V. Developmental psychopathology

  1. —Typical vs. atypical development
  2. Development of skills (e.g., verbal, cognitive) and neural systems

VI. Preschool: ↑ aggression, noncompliance

VII. Middle childhood: decreases


  1. —Normative increases in risk-taking and sensation-seeking

IX. Early adulthood: typically, desistance

Developmental pathwaysEdit

I. Equifinality: a variety of pathways can lead to the same outcome

II. Multifinality: —same pathway or process can lead to multiple outcomes

III. Transactional and reciprocal processes (—e.g., conduct & academic problems; —e.g., coercive interchanges)

IV. Hierarchical motility

  1. —Past processes are carried forward

Future directions of conduct problems: Shaw (2013)Edit

I. Understanding conduct problems among girls

  1. —Child vs. family factors
  2. —Different risk factors based on sex?

II. Contribution of poverty & urban settings

  1. —Proxy for other contextual risk?
  2. Differences based on urbanicity

III. Cascade models that include prenatal risk

IV. Prevention and intervention

V. Engaging parents (e.g., community settings)

Future directions of assessments: Youngstrom (2013)Edit

I. Combine evidence-based and psychological assessment procedures for determining prognosis and treatment

  1. —Prediction: ability to predict to criterion
  2. —Prescription: informing treatment
  3. —Process: progress over time

12 assessment steps (Youngstrom, 2013)Edit

I. Identify most common diagnoses based on setting

  1. —Availability of assessment instruments

II. Know base rates

  1. —Prioritize order of assessment and strategies

III. Evaluate relevant risk and moderating factors

  1. —Differential diagnosis & treatment targets

IV. Synthesize broad instruments into revised probability estimates

  1. —Combining scores may change probability of conduct problems vs. other descriptions

V. Add narrow and incremental assessments to clarify diagnoses

  1. —May have better validity, improve differential dx

VI. Interpret cross-informant data patterns

  1. —Understand common patterns & relevance to context/settings

VII. Finalize diagnoses

VIII. Treatment planning & goal setting

IX. Measure processes

  1. —Repeated measurement of conduct problems, mechanisms/mediators, etc.

X. Chart progress and outcome

  1. —Assess at expected points in tx (e.g., midpoint & end)

XI. Monitor maintenance and relapse

  1. —Continued assessment for stressors & challenges

XII. Solicit and integrate client preferences

Co-occurring conditionsEdit

I. Explanations

  1. —1 disorder confers risk for another
  2. —Shared risk processes account for co-occurrence (e.g., parenting, temperament)

II. Relations among ADHD, ODD, and CD

  1. —ADHD -> ODD -> CD
  2. ADHD + ODD
  3. —ODD + CD

III. Anxiety

  1. —↓ conduct problems in childhood (buffer)
  2. ↑ conduct problems in adolescence (exacerbate)

IV. Depression

  1. —Follows conduct problems (“failure” model)

V. Substance use/abuse

  1. —Follows conduct problems
  2. —Earlier onset of substance use
  3. —Linked to deviant peer affiliations

VI. Language difficulties

  1. —Confer risk for behavior problems
  2. —Exacerbated by behavior problems

VII. Academic problems

  1. —Difficulty attending in classroom and to homework
  2. —Conflict with teachers and peers in school
  3. —Cognitive/neuropsychological difficulties

Implications for assessmentsEdit

I. Continuity linked to assessment approach

  1. —Different informants lead to different estimates (higher rates with parent reports)
  2. —Observations best for seeing stability

II. Developmentally sensitive approaches

  1. —Cascade models

III. Multiple domains (youth & context)

  1. —Child factors (e.g., temperament, neurodevelopmental variables, CU traits)
  2. —Parent-child factors (e.g., harsh discipline, monitoring)
  3. —Family factors (e.g., disorganization, social support, parental psychopathology)
  4. —Peer factors (e.g., rejection, victimization)

Based on Shaw (2013)Edit

I. Assess early (before age 5)

  1. —Early starters have more severe course and outcomes
  2. —More continuity from ages 4-5 to adolescence
  3. Later starters have less stability in course

II. Comorbid conditions linked to continuity

Resilience or protective factorsEdit


II. Verbal abilities

III. Parenting factors

  1. —Routine, consistent discipline, supervision

IV. Low levels of family adversity

V. Prosocial peer relationships

VI.Family social support

VII. Interaction of factors leads to risk/resilience



I. Use multiple methods and informants

II. “Packages” or groups of risk factors

  1. —Multiple domains

III. Developmental history

IV. Normed instruments

V. Course and developmental pathways

VI. Maintenance and resilience

VII. Context & transactional relations

Types of problemsEdit

I. Persistence linked to

  1. —Number of behavior problems
  2. —Types of behavior problems
  3. —Severity
  4. —Level of impairment
  5. —Co-occurring conditions

II. Possible strategies

  1. —Interview with parent and youth
  2. —Behavior rating scales
  3. —Behavioral observations
  4. —Include teachers for cross-setting evidence


I. Achenbach Child Behavior Checklist

  1. —Aggressive, attention problems, delinquent, DSM
  2. —Parent, teacher, and youth (ages 11-18)

II. Behavioral Assessment System for Children

  1. —Aggression, conduct problems
  2. —Parent, teacher, youth (ages 8-21)

III. Conners Rating Scales

  1. —Oppositional, hyperactivity, anger control problems
  2. —Parent, teacher, and youth (ages 12-18)

IV. Child Symptom Inventory

  1. —Diagnostic categories (ADHD, ODD, CD)
  2. —Parent, teacher, and youth (age 11-18)


I. Clinical Interview

  1. —Type, frequency, severity
  2. Parent-child interactions
  3. —Antecedents and consequences

II. Structured Interview

  1. —Assess range of psychological disorders
  2. —Impairment
  3. —Age of onset
  4. —More structured, formal format
  5. Examples:
    1. Diagnostic Interview Schedule for Children (DISC)
    2. Diagnostic Interview for Children and Adolescents (DICA)
    3. Schedule for Affective Disorders and Schizophrenia for School-Aged Children (K-SADS)
    4. Anxiety Disorders Interview Schedule for Children (ADIS-C/P)


I. Identify contributing and maintaining contextual factors

II. Home or office

  1. —Free play, directed tasks for children
  2. —Problem-solving, communication for adolescents

III. Classroom

  1. —Academic and peer issues

IV. Playground, lunch room

  1. —Higher levels of aggression and victimization

Co-occurring conditionsEdit

I. Behavioral checklists

  1. —CBCL, BASC, Conners, CSI

II. Clinical interview

III. Structured interviews

IV. Psychoeducational testing

  1. —Language and academic performance

Sample testing domainsEdit


II. Receptive and expressive language

III. Memory

  1. —Short- and long-term, working

IV. Executive functioning

  1. —Sustained attention, planning, set-shifting

V. Decision-making

VI. Academic performance

  1. —Reading, math, writing

Risk factorsEdit

I. Physical history

II. Social cognitive biases

III. Parenting practices

  1. —Interviews, checklists, observations

IV. Screen for parental psychological difficulties

  1. —Symptom and marital conflict checklists

V. Parenting stress

  1. —Interview, checklists

VI. Peer processes

  1. —Interview, checklists, observations

Importance of contextEdit

I. School

  1. —Classroom, teachers, peers, resources

II. Neighborhood

  1. Cohesion, danger/decay

III. Home

  1. —Parent-child, siblings, family

IV. Psychosocial stressors

V. Assessment

  1. Opportunity to excel (or at least show their best)!


I. Relative to strengths

II. Grounded in knowledge of typical development

III. Considered across contexts & domains of functioning

IV. Opportunities for support & practical remediation

—Individual & contexts

  1. —Need “buy in”
  2. —Strengths to buffer areas of challenge
  3. —Developmentally appropriate
  4. —Implementation across settings

Potential challenges to assessing youthsEdit

I. Establish who is client

II. Clarify confidentiality issues

  1. —Consent/assent
  2. Who gets report
  3. —What are goals (of child, family, school, etc.)?

III. Identify strengths & areas of interest

IV. Attend to youth’s performance

  1. —Maintain motivation & take breaks as needed

V. Be consistent with praise

VI. Feedback session

  1. —Motivational interviewing model

Prevention/intervention overviewEdit

I. Behavioral principles

  1. —Positive reinforcement
  2. —Negative reinforcement
  3. —Ignoring
  4. —Punishment

II. Address conduct problems AND risk factors or mechanisms

III. Developmentally appropriate

IV. Consider multiple domains and settings

V. Varying effectiveness

Future directionsEdit

I. Multiple levels of analysis

II. Initial assessments may predict treatment outcomes

  1. —Parenting behaviors
  2. —Family functioning
  3. Child-specific factors

III. Moderators and mediators

IV. Augmenting interventions

V. Prevention

VI. Assessments (and interventions) for low-resource settings

VII. Applying for grant funding

  1. —Psychometrics, gold standards, incremental validity

VIII. RDoC Domains

  1. —Negative & positive valence, arousal & regulatory, and cognitive systems; social processes

IX. Person-centered approaches

  1. —Growth mixture modeling, latent growth curve modeling, latent profile/class analysis

X. External validation of profiles/classes

  1. —Risk factors, correlates, outcomes, course, tx response

XI. Differential susceptibility

  1. More sensitive to negative AND positive contexts

Dr. Arielle Baskin-Sommers: NeuroscienceEdit

Useful linksEdit

General population neuroimaging data: http://www.humanconnectomeproject.org/data/https://openfmri.org/;  

Data-Sharing and Open-Source Initiatives through Child Mind: https://childmind.org/data-sharing-initiatives/ 

Open source imaging informatics platform: https://www.xnat.org/about/xnat-implementations.php 

NIH Data Sharing: https://www.nlm.nih.gov/NIHbmic/nih_data_sharing_repositories.html


Data with justice involved individuals: http://www.pathwaysstudy.pitt.edu/