As many as 20% of adults with intellectual disabilities (ID) display challenging behaviours such as self-injury or aggression, which may negatively impact quality-of-life, learning, employment, and social integration (Bowring et al., 2019; Emerson et al., 2001; Sheehan & Hassiotis, 2017). Without adequate intervention, disruptive challenging behaviours may lead to a behavioural crisis (McCombe et al., 2022; Weiss et al., 2009). A behavioural crisis occurs when an individual with ID engages in aggressive behaviour that results in imminent and significant threats to safety (their own, or safety of others) (McCombe et al., 2022; Rubin et al., 2007). In the context of behavioural crises, behavioural stabilization services can provide assistance through behavioural consultation and assessment to create a path for successful community integration (Rubin et al., 2007; Wieseler et al., 2002). Previous literature is lacking in guidance on the best way to structure a behaviour stabilization service (McCombe et al., 2022). A logic model framework to evaluate a behaviour stabilization program provides a consistent and systematic way to understand program components and compare programs. Prior evaluations of behaviour stabilization programs that performed outreach and capacity building did not use a logic model framework and lacked social validity information from multiple perspectives. The purposes of this research were to comprehensively evaluate St.Amant's Community Stabilization Service (CSS) through the use and co-construction of a logic model and social validity measures, to identify the strengths and weaknesses of CSS at St.Amant, and provide a list of recommendations to strengthen the program and improve service. Two studies were conducted that used anonymized retrospective data and prospective data. Retrospective data from CSS and two outside organizations were collected and analyzed from April 1, 2017-March 31, 2021, to characterize the program’s inputs, activities, outputs, and outcomes. The nine prospective survey respondents were generally satisfied with the program. This research contributes to the current, limited research in this area by providing a methodological improvement over prior work by using a logic model framework to evaluate a behaviour stabilization program.