4 - (SYM 22) Development and Implementation of Modified-positive Behavior Interventions and Supports on a Child and Adolescent Inpatient Psychiatry Unit

Background: Positive Behavior Interventions and Supports (PBIS) has been found to reduce use of discipline practices in schools, residential settings, and inpatient psychiatry units. Over the course of a 4-month period, a new model of care using Modified-PBIS (M-PBIS) was created for an inpatient psychiatry unit (IPU) serving children and adolescents. This new model aimed to decrease safety issues on the IPU, decrease the use of restraint, and improve culture and morale on the unit.

Methods: M-PBIS program development was led by psychologists with input and collaboration from multi-disciplinary IPU staff. Superusers from nurses and floor staff, for example, were actively involved in developing several aspects of the M-PBIS framework and related training. Primary drivers of this quality improvement initiative included the development of a multi-tiered system of support and enhanced staff resources/training. M-PBIS was introduced across a variety of forums to ensure that all IPU patient facing staff had exposure to the model. The M-PBIS framework introduced several changes and interventions to the IPU including universal unit values and expectations, statement of purpose, community meetings, a token economy system, the use of Collaborative and Proactive Solutions, and enhanced proactive behavior management strategies. To implement the M-PBIS model, IPU staff went through two 8-hour in-person training days which included didactic, embedded anti-bias training, and simulations. Post training, M-PBIS was launched on the unit with additional psychologist, leader, and superuser presence to give in the moment coaching to staff. Ongoing training occurs for all new staff oriented to the unit. Fidelity was monitored via observational periods conducted by psychologist and trained superuser staff.

Results: While total number of restraints increased, minutes in restraint decreased, with total hours in restraint for the program reduced by approximately 200 hours. Mean minutes in restraint per event reduced from 29 minutes pre intervention to 11 minutes post intervention. 30-day readmission rates and length of stay remained stable across the study period.

Conclusions: Staff engagement in creation and implementation of the M-PBIS model were key to successful implementation. Over the course of the first year of implementation, we had variable fidelity to the model and noticed drift to previous practices. Continued and ongoing training is recommended for increased fidelity to the model. Confounding variables will be discussed.