<p>This paper considers the estimation of the change-point Cox proportional hazards model based on current status or case-I interval-censored data. Although a great deal of literature has been established for inference about the model above, all of the existing methods are for right-censored data and cannot apply to the situation considered here. To address the problem, the maximum likelihood estimation is developed, and the asymptotic properties of the resulting estimator, such as the consistency, the asymptotic independence, and the semiparametric efficiency, are established. In addition, a hypothesis test procedure is provided for testing the existence of the change point. A numerical study is conducted and suggests that the proposed methodology performs well in practical situations. Furthermore, it is applied to a set of real data arising from an AIDS clinical trial that motivated this study.</p>