Principal Investigator: John Lochman

Level of Intervention: Indicated

Target Population: Aggressive and disruptive children and adolescents who have difficulty with anger management.

References: Lochman (1985, 1992); Lochman, Burch, Curry, & Lampron (1984); Lochman & Lampron (1988); Lochman, Lampron, Gemmer, Harris, & Wykoff (1989); Lochman & Wells (1996)

VERSION 1
Lochman, Burch, Curry, & Lampron (1984); Lochman & Lampron (1988); Lochman & Wells (1996)

Theory (Risk & Protective Factors Targeted): Children who are aggressive often demonstrate a variety of social cognitive deficits or distortions that contribute to their lack of social competence. Their cognitive problem solving skills are also impaired when they are emotionally aroused. Aggressive behavior is often a reflection of inability to regulate emotional reactions to stimuli that are anger inducing. This behavior pattern is very stable and a risk marker for variety of poor adolescent outcomes (e.g. conduct disorder, delinquency, and substance use. Aggressive children also at risk for school failure, dropout, and poor peer relations.

It is theorized that group format is most effective because it 1) allows children to receive peer feedback, 2) provides in-vivo experiential learning, 3) increases the likelihood of generalization.

Description of Intervention:
The Anger Coping Program is a school-based group intervention designed to reduce future conduct problems, delinquency, and substance abuse. Based on a social-cognitive model of anger arousal and consistent with social information processing models of social competence the intervention focuses on the cognitive distortions and cognitive deficiencies often found in aggressive children. The lessons promote self-instruction and awareness, and builds social cognitive skills.

Sessions were held weekly for 12 weeks and last 45-60 minutes. Groups consisted of 4-6 members and were led by two co-leaders. Program contained lessons designed to improve children’s perspective-taking skills, affect recognition, self-control (through inhibatory and coping self-statements), social problem solving, and social skills strategies for managing conflict situations. Sessions included role-plays and activities that generate affective arousal. Reinforcement and feedback are used to support skill acquisition.

Goal-Setting Component
The goal-setting sessions met weekly for 8 weeks. This program was considered a minimal treatment condition. The boys were asked to establish weekly goals for themselves which were monitored by teachers and rewarded if successfully attained.

Research Subjects:
Sample consisted of 76 boys who ranged from 9 to 12 years old. Fifty-three percent of the sample was African-American and the remainder were Caucasian. Subjects were drawn from 8 schools and screened for aggressive behavior with teacher’s ratings on the Missouri Children’s Behavior Checklist (MCBC, Sines, Pauker, Sines, & Owen, 1969). The numbers of students from each school were not provided. Follow-up analyses were conducted on a sub-sample of 32 boys.

Research Design:
Subjects were assigned "on a rotating basis" to one of 4 conditions: Anger Coping (AG), Goal Setting (GS), a combination of the first two interventions (ACGS), or an untreated control group (UC).

Outcomes:
Repeated measures ANOVA, later used ANCOVA to co-vary pre-test scores.

Post-Test
Boys in the treatment conditions (AC and ACGS) exhibited lower rates of Disruptive and Aggressive Off-Task behavior on classroom observations (BOSPT, Allen, Chinsky, Larcen, Lochman, & Selinger, 1976) compared to controls (p<.03). Parents of boys in the treatment condition rated them as exhibiting significantly less aggressive behavior (p<.02) on the MCBC compared to controls. Treatment effects were greatest for the combined condition (ACGS). There were no significant differences between groups on teacher or peer ratings after pre-test scores were controlled.

Follow-up (7 month)
In independent observations, treatment boys in the AC condition were compared to controls (UC). AC boys exhibited greater on-task behavior and significantly less passive off-task behavior. Differences in disruptive off-task behavior were not maintained.

Strengths & Limitations:
The Anger Coping Program targeted the social cognitive risk factors associated with children who exhibit disruptive behavior problems by providing intervention on the individual child level. Support was found for the effectiveness of the program in making short-term improvements on children’s disruptive and aggressive behavior. The findings also suggested that behavioral strategies (e.g. monitoring and reinforcement) were useful components to combine with cognitive interventions. The fact those significant changes were found on observations and parent report counteracts the fact that teachers and peers did not report any changes. The clinical significance of the findings was not discussed and the effects were not maintained 7 months past the intervention There was also support for the reliability of the observations (i.e. high rates of inter-rater agreement). The authors did not report whether the participants received similar dosages of the interventions and no fidelity measures were included in the evaluation. Although the sample was entirely male there was an even balance of Caucasian and African-American students.

VERSION 2
Lochman (1985)

Theory (Risk & Protective Factors Targeted): See description above.

Description of Intervention:
Revised version of the Anger Coping plus Goal Setting program (ACGS) extended into an 18-session format. The 18-session program provided additional role-playing experiences and more discussion of anger arousing situations. For more detail see description above.

Research Subjects:
22 boys identified by teachers who were asked to nominate the most aggressive and disruptive students in their class. Average age 10 years 4 months. 55% African American and 45% Caucasian. The same screening procedure was used in Lochman et al. (1984).

Research Design:
In a quasi-experimental design, this sample was compared to the Lochman et al (1984) sample (N=76) who had been assigned to either the 12 session version of Anger Coping (AC-12), the Goal Setting Only condition (GS), the original combination of these two components which was 12 sessions long (ACGS-12), or an Untreated Control group (UC).

Outcomes:
Same measures used as in first program evaluation. At pretest, the groups were similar on observations of disruptive-aggressive behavior. Group differences were found for on-task, passive/off-task behavior so analyses of covariance were used. The groups were also compared on demographic variables and found to be similar except for age.

Independent observers rated the 22-treatment boys (ACGS-18) as significantly more On-Task, less Passive Off-Task, and less Disruptive-Aggressive in their behavior compared to boys in the GS and UC conditions (p levels between .05 and .001). The boys who received the most recent version of the program (ACGS-18) exhibited significantly more On Task (p<.05) and less Passive Off Task (p<.05) behavior compared to boys who received the 12 session version of the ACGS program (ACGS-12).

Strengths & Limitations:
The Anger Coping Program targeted the social cognitive risk factors associated with children who exhibit disruptive behavior problems by providing intervention on the individual child level. Support was found for the effectiveness of the program in making short-term improvements on children’s disruptive behavior in the classroom. Only observational outcomes were examined but there was support for the reliability of these measures (i.e. high rates of inter-rater agreement). The quasi-experimental design of this evaluation and the small sample size weaken the conclusions that can be drawn from the outcomes of the study. The authors did not provide information on how many referrals were received from teachers in each school and how many children were placed in each group. As was the case in Lochman et al. (1984), there was no discussion of measurement of fidelity or dosage.

FOLLOW-UP VERSION 1 & 2
Lochman (1992)

Theory (Risk & Protective Factors Targeted):
See descriptions provided in first two versions

The purpose of this study was to examine the long term effects of the Anger Coping program given that up until then, other studies of cognitive behavioral treatments had shown limited effects.

Description of Intervention:
See description provided in first version.

Research Subjects:
The sample was drawn from an original pool of 354 identified subjects. The researchers were unable to contact 121 (34%) of those families. The final sample consisted of 31 Anger Control, 52 Untreated Controls, 62 non-aggressive boys (N=145). Out of the 31 Anger Control subjects, 12 were given 6 booster sessions. The non-aggressive subjects were identified by < 7% of their peers as aggressive.

Research Design:
Three cohorts compared by combining samples from Lochman (1995), Lochman et al. (1984), and Lochman & Curry (1986). Also looked at comparison of Anger-Coping (AC), untreated aggressive (UA) and non-aggressive (NON). In addition looked at subset of AC that received 12 sessions and given 6 booster sessions (AC-BOOST) to and AC-ONLY group.

Outcomes:
Structured interviews were conducted with students 2.5-3.5 years after the end of the intervention. Two clusters of responses were identified from portions of the National Youth Survey (NYS; Elliot, Huizinga, & Ageton, 1985): substance use and general behavioral deviance (GBD). Measures of moderator variables (self-esteem & social problem solving) were included to test hypothesized process in effect of intervention.

Within each experimental group there were no significant differences between the subjects with consent and those without consent or the ones who had lost contact with the study on peer ratings of aggression or social status. The two aggressive conditions (AC and UA) and the two anger coping conditions (AC-BBOST and AC-ONLY) were comparable except there was a significant age difference between the two sets of groups that was not considered large in the absolute sense. Age was not significantly related to any dependent variables except for within the AC condition it was significantly related to general behavioral deviance. Younger subjects displayed more deviant behavior than older subjects in this condition did, but this was not considered relevant to the interpretation of intervention effects. Due to differences between the aggressive and non-aggressive conditions, age and IQ were used as covariates in subsequent analyses.

Intervention effects were tested using MANCOVA and followed by ANCOVAs to determine the source of the effect. AC subjects reported less substance use (p<.02) and higher self esteem (p<.01) compared to UA subjects but there were no group differences on disruptive-aggressive behavior indicating that effects on this dimension were not maintained from post-intervention. No significant differences between AC-ONLY and AC-BOOST except passive-off task behavior.

The differences in aggression were not maintained and no effects for delinquency were found. Secondary prevention effects were most evident on substance use outcomes. The authors used meta-analysis to contrast the aggressive sample with a normative sample and found AC subjects closer to mean of normative group on substance use, self-esteem, and problem solving compared to the UA group.

Strengths & Limitations:
The findings of this follow-up study suggest that in general, improving the cognitive-behavioral skills of at-risk youth have limited long-term effects when they are conducted in isolation. The treatment effects on aggressive behavior were not maintained over time. It is important to note, however, that these skills were found to relate to other adolescent outcomes. Although the original Anger Coping Program did not target drug resistance skills or problem solving specifically, it had a significant impact on self-reported use and two moderator variables: problem-solving and self-esteem. The authors discuss the findings in terms of their clinical significance. Although it was not mentioned in previous publications, the author noted that in all of the studies the intervention integrity was monitored through weekly meetings with group leaders.

VERSION 3
Lochman, Lampron, Gemmer, Harris, & Wykoff (1989)

Description of Intervention:
Both programs used goal setting in combination with an operant reinforcement system to reward the boys weekly for compliance with rules.

Anger Control (AC)
See description provided in first version

Anger Control Teacher Consultation (ACTC)
Teacher consultation focused on behavioral management and developing students’ problem-solving skills.

ACTC also lasted 18 sessions - It involved 6 contact hours of consultation by the group co-leaders with small groups of 2-4 teachers. Meetings varied in length but averaged 1-2 hours. The goals of the consultation were to help the teachers find ways to support the generalization of anger coping skills, support teachers’ in their efforts to solve problems with students, and help teachers’ develop a successful behavior management system within the classroom.

Research Subjects:
At 6 elementary schools in Durham Co., teachers identified the most disruptive and aggressive boys in their classroom. The average level of aggressive behavior in the sample was confirmed with teacher ratings on the MCBC (subjects were on average 1 SD > teacher norms). The sample consisted of 32 boys with an average age of 11 years. 22 boys were White and 10 were African-American.

Research Design:
Schools were alternately assigned to either the AC or ACTC treatment conditions. Three of the 6 schools referred more subjects than could be included in the group interventions so the boys in these schools were assigned on an odd-even basis to either the intervention condition or the untreated control condition. The study compared Anger Coping (AC), Anger Coping plus Teacher Consultation (ACTC), with untreated control (UC).

Outcomes:
Subjects in the 3 conditions were not significantly different at pre-test on age or intelligence level. Due to small sample size, used non-parametric Mann-Whitney U-Tests to compare conditions – change scores pre-post, one-tailed tests. The same measures were used in this evaluation as in the first two versions (i.e. MCBC, BOSPT)

Boys in combined treatment conditions (AC and ACTC) were observed as less disruptive-aggressive (p<.05) and reported higher rates of perceived self-competence (p<.05) compared to the boys in the untreated control condition. There were no significant differences between the AC and ACTC conditions.

Strengths & Limitations:
The results of this study provide some additional evidence for the effectiveness of the Anger Coping Program compared to no treatment. Boys who received both types of intervention compared to controls were described as less disruptive by observers. There was no support that the addition of a teacher consultation component improved the program. Teachers with students in the AC and the ACTC condition were compared on background variables of experience and racial status and found to be similar. As such, the lack of treatment group differences did not appear to be a function of any teacher characteristic. The results should be interpreted cautiously given the small sample size. The authors also did not include information on measures of dosage or implementation.