Practice shows that many users stop disclosing personal health information (PHI) in mobile health applications due to privacy concerns about potential misuse, thereby foregoing potential health benefits. We draw on privacy calculus theory to explain the reasons for this deliberate behavioral shift, which we label discontinuous disclosure. Privacy calculus theory states that users weigh the benefits of disclosure against the associated costs. We suggest that discontinuous disclosure arises from a discrepancy in users’ perceptions regarding these benefits and costs. Initially, they expect benefits from their PHI disclosure (e.g., health benefits) but subsequently do not experience these benefits or encounter unexpected costs (e.g., privacy concerns), which eventually lead them to discontinue disclosing. To account for such changing perceptions over time, we suggest that a combination of initial expectations and subsequent experiences regarding disclosure benefits and costs leads to discontinuous disclosure. We conducted a two-step fuzzy-set qualitative comparative analysis on survey data from 322 mobile health application users and reveal four configurations that explain discontinuous disclosure. We contribute to privacy research by introducing discontinuous disclosure as a novel disclosure behavior characterized by the discrepancy between expectations and experiences. We also extend privacy calculus theory by shifting from a static, one-time tradeoff to a dynamic view where (1) initial expectations form a baseline for assessing subsequent experiences and (2) experienced and expected benefits and costs work together to explain disclosure behavior. Additionally, we contribute to IS research on mobile health applications by explaining users’ disclosure behavior in this context.