Montreal Prevention Experiment

OVERVIEW

The Montreal Prevention Experiment was designed to reduce antisocial behavior among elementary school boys exhibiting disruptive behaviors. The program consisted of both a parent and a child component. In an effort to reduce antisocial behaviors and promote school adjustment, parents and their children participating in the Montreal Prevention Experiment (MPE) received a series of skills training sessions over the course of two years. The training sessions largely emphasized management and monitoring skills for parents, as well as prosocial and self-control skills for boys. Upon completion of the treatment, results indicate that boys who participated in the MPE performed better on measures of school adjustment and delinquency in comparison with peers who did not receive the treatment. Some impacts lasted through age 15.

DESCRIPTION OF PROGRAM

Target population: Boys ages 7-9 exhibiting disruptive behaviors and their parents

The Montreal Prevention Experiment is a manualized program designed to target boys in elementary school who had been identified in kindergarten by teachers as exhibiting disruptive behavior. The treatment was intended to prevent the onset of subsequent antisocial behaviors and offset the effects such behavior may have on school achievement. The Montreal Prevention Experiment began when children were approximately seven years old and was two years in duration. During that time, parents and children met in a series of sessions with caseworkers. Caseworkers assessed each family individually and determined the number of sessions based on the severity of behavior.

The parent sessions each lasted less than an hour and taught parents skills to manage family crises, monitor behavior, discipline effectively, and reinforce positive behaviors. Boys were invited, but not required, to attend parent sessions. On average, families participated in a total of 20 sessions over the course of the two-year period.

The boys in MPE received separate skills training sessions in small groups that included some of their more prosocial peers at school. Sessions incorporated role-playing, peer modeling, reinforcement techniques and coaching. In the first year of the program, the boys participated in nine sessions that emphasized prosocial skills such as offering help and asking for reasons. In the second year, ten sessions were dedicated to self-control skills like following rules and anger management. Although caseworkers also conducted the boys’ sessions, boys and parents were not treated by the same caseworker.

EVALUATION(S) OF PROGRAM

McCord, J., Tremblay, R.E., Vitaro, F., & Desmarais-Gervais, L. (1994). Boys’ disruptive behaviour, school adjustment, and delinquency: The Montreal Prevention Experiment. International Journal of Behavioral Development, 17(4), 739-752.

Evaluated population: The study sample included 166 Canadian boys who were part of a larger study of youth called the Montreal Longitudinal-Experimental Study. Children were primarily White, came from families with below-average household incomes, and lived in a large city. The sample was selected because they had been rated above the top 70^th percentile on the disruptive behavior scale administered to teachers. Boys were French-speaking and came from families in which parents had less than 14 years of schooling.

Approach: In the spring of 1984, teachers from 53 schools submitted evaluations on 1,161 kindergarten boys on measures of disruptive behavior, anxiety, prosocial behavior, and inattentiveness. Children who scored above the 70^th percentile on teacher evaluations of disruptive behavior were asked to participate in the study (n=319). Of those asked to participate, 172 agreed. Upon verification of original sample demographic characteristics, six cases were dropped from the current analyses. The resulting sample of 166 participants includes n=43 treatment cases, n=82 comparison cases, and n=41 control cases.

The Montreal Prevention Experiment began in 1985 when the participants, on average, were seven years old. The treatment was completed in 1987, and follow-up evaluations were conducted annually beginning one year after program completion, when the boys were approximately 10 years old. Upon entry into the study, subjects were randomly assigned to one of three conditions. The treatment group (n=46) received the MPE training sessions outlined previously. The attention-control group (n=84) did not participate in sessions but was similar to the treatment group in the amount of general attention they received from the study and the control group (n=42) was only contacted at the time of evaluation.

Six follow-ups were obtained for this randomized controlled study. Program outcomes were assessed using teacher and child reports of disruptive and delinquent behavior, administrative records of school adjustment and delinquency, and self-reported perceptions of parenting behavior.

Results: With the exception of maternal occupational status, no differences were found on basic demographic or academic measures among the three groups at baseline. Mothers in the treatment group were found to have slightly lower occupational status than those in the other groups. Comparisons of teacher evaluations at kindergarten revealed that there were slight group differences on measures of disruptive behavior. The treatment group was generally midway between the comparison and control groups with the exception of fighting, which was highest among treatment boys.

Outcomes assessed in the study included school adjustment and delinquency. School adjustment measures were based on school records, in addition to annual teacher and peer evaluations. Delinquency was based on self-reported measures. Over 80 percent of each condition group was available for follow-up post treatment. The control and comparison groups were systematically compared on the outcome variables and found to have no significant differences. For purposes of analyses, these groups were combined into a single comparison group.

School adjustment was determined according to age-appropriate classroom placement and disruptive behavior. Upon program completion, 64 percent of the control group and 76 percent of the treatment group were in regular third-grade classrooms. These numbers continued to decline through age 12 for both groups; however the study found a significant group effect where treatment boys were more likely than control group boys to be in age-appropriate classrooms.

Boys were considered to have disruptive behavior if they scored above the 70^th percentile on a rating based on either the teacher and peer-rated questionnaires. At post-treatment in 1987, 68.6 percent of treatment boys and 74.3 percent of control boys were classified as disruptive. In 1990, those numbers dropped to 55 percent and 71.2 percent, respectively. Differences between the two conditions were not statistically significant.

Boys were considered to have achieved satisfactory school adjustment if they were in age-appropriate class placements and scored at or below the 70^th percentile on reports of disruptive behavior. At each year of follow-up, a greater proportion of treatment boys achieved satisfactory school adjustment than control boys, however results were only significant in 1989 and 1990 (p<.05). After controlling for group differences reported by teachers in the kindergarten year, differences between groups were significant in 1990 only.

Finally, boys were assessed on measures of delinquency which included theft, trespassing with the intention of stealing, fighting, and engaging in mischievous behavior with peers/gangs. Because sample sizes were often too small to analyze separately, measures were combined into theft and misbehavior and aggregated across years. Data revealed that 31.8 percent of treatment boys, compared with 54.9 percent of control boys committed at least one theft between the ages of 10 and 12. Differences were significant at the p<.01 level. In 1990, 2 percent of treatment boys and 19 percent of control boys reported engaging in misbehavior with gangs. This difference was significant at the p<.03 level.

Although impacts from the evaluation of the Montreal Prevention Experiment are promising, there are several limitations to the study. First, although the control and comparison groups were examined consistently on the outcome measures before collapsing the two, the two groups did have different experiences in the program. Second, mothers had originally completed evaluations on the boys’ disruptive behaviors that were not used in analyses. The study designated mother-report questionnaires as unreliable due to varying levels of program participation and therefore, awareness, of boys’ skills and behaviors.

Tremblay, R. E., Pagani-Kurtz, L., Masse, L. C., Vitaro, F., & Pihl, R. O. (1995). A Bimodal Preventive Intervention for Disruptive Kindergarten Boys: Its Impact through Mid-Adolescence. Journal of Consulting and Clinical Psychology, 63(4), 560-568.

Evaluated Population: This is a follow-up study to the 1994 study, therefore the population is the same.

Approach: Unless otherwise noted, sampling design, research design, and study measures do not differ from the previous evaluation.

Results: Findings indicate that the program has long-term effects on self-reported delinquency, however impacts on teacher ratings and administrative records of disruptive behavior were not found. Patterns of effects on self-reported delinquency were not influenced by time. Interestingly, compared with untreated boys, treated boys were more likely to be in age-appropriate classrooms from ages 10 to 12, but not from ages 13 to 15.

Note: Analysis procedures did not adjust the standard errors of the estimates to account for the nesting of observations within individuals. Therefore, findings should be interpreted with caution.

Tremblay, R.E., Masse, L., Pagani, L., & Vitaro, F. (1996). From childhood physical aggression to adolescent maladjustment: The Montreal Prevention Experiment. In R.D. Peters and R.J. McMahon (Eds.) Preventing Childhood Disorders, Substance Abuse, and Delinquency. Thousand Oaks, CA: Sage Publications.

Evaluated Population: This is a follow-up study to the 1994 study, therefore the population is the same.

Approach: Unless otherwise noted, sampling design, research design, and study measures do not differ from the previous evaluation.

Results: In the current analyses, one additional demographic characteristic, parental age at child’s birth, was found to differ among conditions. Mothers in the comparison group were younger than their control group counterparts, while fathers in the treatment group were older than those in the comparison group.

Because these analyses follow boys through early adolescence, additional self-report measures were included which consist of age of sexual initiation, substance abuse, delinquent behavior and school motivation.

Repeated comparisons of gang membership at ages 11, 13 and 15 revealed that treatment boys were significantly less likely than the control group to be involved in gangs (p <.01). During roughly the same time period, the percentage of all sample children who reported having been drunk or taking drugs in the past 12 months increased linearly until age 14. At age 15, however, when compared to boys in the control group, significantly fewer boys in the treatment group reported having been drunk (p<.02) or having taken drugs (p<.05).

Boys’ reports of delinquent behavior through early adolescence continue to mirror earlier trends. Boys who participated in the treatment group were significantly less likely to engage in delinquent acts than those in the control group and this trend held constant from age 10 through ages 15. Boys also reported on their own police arrests and those of their friends in the past 12 months. From ages 11 through 13, no differences were found between groups on the frequency of subjects’ arrests. Although the control group reported more arrests at 13 to 15 years of age, differences from the treatment group were not statistically significant. Friends’ police arrests were less frequent among treatment boys but were only significantly different from the control group between the ages of 13 and 15 (p<.05).

On measures of sexual initiation, no significant differences were found between conditions from ages 13 to 15. School motivation was found to decline from ages 10 through 15, but differences between groups were not statistically significant. Finally, researchers expected that the better school adjustment experienced by treatment boys in elementary school would continue into high school. No differences were found, however, between groups on age-appropriate classroom placements. In fact, only 33.3 percent of treatment boys and 32.2 percent of comparison boys were in age-appropriate classrooms at age fifteen.

SOURCES FOR MORE INFORMATION

References

Tremblay, R. E., Pagani-Kurtz, L., Masse, L. C., Vitaro, F., & Pihl, R. O. (1995). A Bimodal Preventive Intervention for Disruptive Kindergarten Boys: Its Impact through Mid-Adolescence. Journal of Consulting and Clinical Psychology, 63(4), 560-568.

Program categorized in this guide according to the following:

Evaluated participant ages: 7-9 years

Program age ranges in the Guide: 6-11

Program components: clinic-based, provider-based, or miscellaneous; counseling/therapy; parent or family component; school-based;

Measured outcomes: education and cognitive development; social and emotional health and development; behavioral problems; teen pregnancy and reproductive health

KEYWORDS: Elementary School, Gender Specific (Males Only), Behavioral Problems, Social/ Emotional Health and Development, Education, Academic Achievement, Life Skills Training, Skills Training, Children, Middle Childhood (6-11), Urban, Disruptive Behavior Disorders, Delinquency, Violence, Educational Expectations, School Engagement, Clinic-Based, Provider-Based, Cognitive Development, Teen Pregnancy, Reproductive Health, Sexual Initiation, Substance Abuse.

Program information last updated on 12/15/08.