Few studies have examined predictors of long-term sustainability of evidence-based practices in mental health. This study used assessments of five evidence-based practices implemented in 49 sites in eight states at baseline and years 2, 4, and 8. Program characteristics, implementation characteristics, reinforcement activities, and sustainability factors were used to predict program survival status. The majority of predictors were not significant. Supervisor turnover in year 4 predicted survival status in year 8, but site characteristics, fidelity at implementation, quality improvement activities, and post-implementation activities had little impact on long-term program survival. This study extends previous sustainability research by examining the long-term impact of internal program factors over a substantial period of time using longitudinal prediction. Future research should also consider the influence of external factors such as financial policies.