A breast cancer diagnosis typically results in dramatic and negative effects on an individual's quality of life. Web-based interactive support systems such as the Comprehensive Health Enhancement Support System (CHESS) offer one avenue for mitigating these negative effects. While evidence supports the efficacy of such systems, evaluations typically fail to provide a true test of the theorized model of effects, treating self-determination theory's constructs of competence, relatedness, and autonomy as outcomes rather than mediators. Using path analysis, this study tests the nature of the proposed mediated relationship between system engagement and quality-of-life indicators utilizing data collected from women (N = 90) who participated in the treatment condition of a CHESS randomized controlled trial. Findings support a latent model, indicating that system effects are mediated through an intertwined measure of autonomy, competence, and relatedness.