Objective: To model typical trajectories for improvement among patients treated in an integrated primary care behavioral health service, multilevel models were used to explore the relationship between baseline mental health impairment level and eventual mental health functioning across follow-up appointments. Method: Data from 495 primary care patients (61.1% female, 60.7% Caucasian, 37.141 plus or minus 12.21 years of age) who completed the Behavioral Health Measure (Kopta & Lowry, 2002) at each primary care appointment were used for the analysis. Three separate models were constructed to identify clinical improvement in terms of number of appointments attended, baseline impairment severity level, and the interaction of these 2 variables. Results: The data showed that 71.5% of patients improved across appointments, 56.8% of which (40.5% of the entire sample) was clinically meaningful and reliable. Number of appointments and baseline severity of impairment significantly accounted for variability in clinical outcome, with trajectories of change varying across appointments as a function of baseline severity. Patients with more severe impairment at baseline improved faster than patients with less severe baseline impairment. Conclusions: Patients treated within an integrated primary care behavioral health service demonstrate significant improvements in clinical status, even those with the most severe levels of distress at baseline. (Contains 2 tables and 1 footnote.)