Objective: This exploratory study examined the impact of critical time intervention (CTI), a time-limited care coordination model, on community integration among formerly homeless adults with severe mental illness after discharge from inpatient psychiatric treatment. Methods: Data were from a randomized trial that enrolled 150 participants, 95 of whom completed 18-month outcome measures. Relationships between two components of community integration (social and physical) and demographic characteristics, symptom ratings, housing status, and treatment condition were examined. Results: Neither assignment to CTI nor housing stability was associated with integration outcomes. General and negative symptoms were inversely associated with both physical and social integration. Conclusions: Although CTI and similar community support models may improve housing stability and reduce rehospitalization, they may not, by themselves, improve community integration. Future studies should focus on improving measurement of community integration so that it can be effectively studied as an important outcome of mental health interventions for this population.