Effectiveness of Behavioral Preschool Teacher Training for Externalizing

Study:

A cluster randomized controlled pre-post effectiveness trial of behavioral preschool teacher training (BPTT) delivered in a practitioner assisted group format for children with externalizing behavior problems. Preschools were randomized to either intervention in 25 preschools or to 22 waiting list control preschools, where preschool classes with the target child or children were program receivers. Informed preschools could apply for participation in the study, then, target children were screened for eligibility. Participants were 100 target children 3-5 years old together with 72 enrolled preschool teachers and 83 parents as informants of behavioral outcomes after a five months period of implementation (at six months).

Intervention:

The intervention was part of the Swedish evidence-based parent and teacher training programs (COmmunication METhod - Comet) for children and youths aged 3-12 and 12-18 years with moderate or elevated externalizing behavior, usually delivered in practitioner assisted group formats but previously evaluated as delivered via internet, single workshops followed by self-administered training, and universal prevention as well. Program implementations and evaluations are executed as collaboration projects between university researchers and the social services administration at place. Intervention content is influenced by operant conditioning, social learning theories, applied behavior analysis, and coercion theory. The focus is to establish a positive and effective interaction and communication style primarily through different reinforcement techniques (e.g., selective attention, more to positive behaviors and less to negative behaviors) and modeling. Parents or teachers meet in psychoeducational group sessions (often 9 to 11 sessions à 2.5 to 3 hours) led by one or two practitioner supervisors. They each follow a comprehensive manual and a highly structured curriculum. Training occurs at sessions (role-plays) and between sessions together with the children, followed-up with feedback in the next session.

Here an adapted version was tried in the preschool setting for the first time. The program corresponded in much to the parent training supplemented with techniques from the school teacher training and a group level administrated glove-puppet play technique to foster children's prosocial skills. (Results from the universal part of the program are not reported.) The nine sessions curriculum consisted of 2.5-hour biweekly meetings and two optional visits from supervisors with coaching on the spot. Practitioner supervisors (n = 27) were educated by a cognitive-behavior oriented psychologist during five days term one and two days term two.

Investigation issues:

The primary aim was to investigate behavioral outcome effects of the preschool teacher program. Would effects (Cohen's d) be in the medium-large range as found for Comet parent training (i.e., d = .50-.90) or in the small-medium range as often found for preventive developmental preschool programs (i.e., d = .20-.40)? With effects of about .40, a sample size restricted to 100 subjects, and an alpha at .05, power would be close to sufficient (i.e., .70). Of interest was to compare the effects of this program with effects found for other preschool program investigations of externalizing behavior problems.

Also investigated was eventual generalized effects to the children's homes. Of what magnitude would such effects be, if any, and would the parents' ratings validate the teachers' ratings, and thus, support the intervention? Such effects may have implications for future implementations, for example if the program is sufficiently efficient as a stand-alone intervention in reducing externalizing behavior problems or not sufficiently efficient. There was also a question of degree of informants agreement/discrepancy. Furthermore, would improved prosocial and regulatory skills function as a predictive and an intermediate mechanism for reduced problem behavior, and/or would there be room for other intervention features to contribute as well? In addition, as a trial in the real world, effects of implementation fidelity as well as social acceptability and relevance, such as proportions of children with reliable change, were investigated.

Inclusion criteria

• Elevated level of externalizing behavior
• Exceeding cutoff value 11 on the Strengths and Difficulties Questionnaire, total scale

Exclusion criteria

-