1. Introduction
2. Methods
   • Study design
   • Variables
   • Group-base Trajectory Models (GBTM)
3. Results
4. Conclusions
5. Limitations

• Definition Oppositional Defiant Disorder:

"Child or teenager has a frequent and persistent pattern of anger, irritability, arguing, defiance or vindictiveness toward you and other authority figures" (DSM V, 2010).

• Angry and irritable mood:
  • Often and easily loses temper
  • Is frequently touchy and easily annoyed by others
  • Is often angry and resentful

• Argumentative and defiant behavior:
  • Argues with adults or people in authority
  • Actively defies or refuses to comply with adults' requests or rules
  • Deliberately annoys or upsets people
  • Blames others for his or her mistakes or misbehavior -Vindictiveness:

• Why is it important?
  • Children who have this behavior in childhood are more likely to have criminal behavior in adulthood (Tramblay R, 1999)
  • It is also related with substance use and sexual risk behavior.
• Why in Colombia?
  • Colombia have history of violence related with guerillas, paramilitary, and drug trafficking (Garcia H, 2001)
  • Medellin had highest homicide rate.
    • In 1993, homicide rate was 383 per 100.000 inhabitants.
    • Therefore, it is important create and evaluate programs to prevent violence.

• Cluster Randomized control trial.
  • The study aim was to evaluate the effectiveness of a program to early childhood aggression (Restrepo A, 2015).
  • 16 schools were assigned to the program and 16 schools were assigned to the control group.
  • We random selected children to being follow-up within (n=862 intervened group and n=822 in control groups).
• Intervention: we called! PILAS! that mean pay attention.
  • 28 hour of training with teacher and 10 months of advisor for the multidisciplinary.
  • 309 teachers and 1,200 parents were trained

• Socidemoghrapic variables:
  • Child: sex, and age.
  • Caregiver: sex, age, and SES.
• Outcome:
  • We used a questionnaire to evaluate symptoms of aggressive and oppositional behavior (parent report) we conduct the Confirmatory Factorial Analysis.

• Description demographic variables accoring study group.
• Description outcome variable.
• Trajectories over time using GBTM.
  • Number of groups: 2-4
  • Quadratic trajectory for the three groups.

• Classification of outcomes: continuous to categorical groups
  • Based on experience and clinical guidelines = “classification rules”
  • Arbitrary
  • Example: mild, moderate, severe depression rating
  • Groups may reflect random variation
  • Miss important and unusual developmental patterns

Goal of GBTM (Nagin, 1995)

• Identify and classify these distinctive life trajectories as they evolve over time or age
• Test if hypothesized trajectories emerge from the data
• Identify factors that change life trajectories

1. Outcome variable: children’s oppositional behavior reported by the parent
2. Time: # measures (2009, 2012, 2016).
3. Number of groups: 2-4 clusters
4. Trajectories:
   • Trajectories: quadratic (2),
5. Run models
6. Compare fit indicator
7. Calculate probabilities and coefficients
8. Graphic models

• haven
• crimCV
• splines
• reshape2
• Hmisc

## Loading required package: splines

## Loading required package: lattice

## Loading required package: survival

## Loading required package: Formula

## Loading required package: ggplot2

## 
## Attaching package: 'Hmisc'

## The following objects are masked from 'package:base':
## 
##     format.pval, units

## Warning in min(abs(tFrtr$llike - bllike)): no non-missing arguments to min;
## returning Inf

• We found fourth trajectories model that explain the oppositional behavior on this study
  • Group 1. High level of oppositional behaviors with decreasing trajectory
  • Group 2. Middle level of oppositional behaviors with constant trajectory
  • Group 3. Low level of oppositional behaviors with constant
  • Group 4. Children without oppositional behavior cross time

• Only can test one type of trajectory for all cluster.
• Usually other software allows to test different trajectories for each group.
• We cannot test factor that may explain the trajectories.
• Time consuming.