This paper analyses the effect of ownership and system of reimbursement on mortality rates. From the statistical results we could conclude that the incentive created by fee-for-service reimbursement yields a four-point reduction in the mortality rate. However, this ranking of hospital quality is completely dependent on the characteristics and illness severity of patients. To take this difficulty into account, we use an innovative duration model applied to panel data: a duration model with both patient and hospital unobserved heterogeneity. No distributional assumptions are made regarding the latter. By this way, we control the fact that patients admitted to the private sector can be different in terms of disease severity from patients admitted to the public sector. The capacity to perform innovative procedures has more effect on the mortality than the system of reimbursement and/or ownership. As such, private sector hospitals that perform more innovative procedures provide a better quality of care, measured by the probability of dying. Nevertheless, heterogeneity within hospitals is greater in for-profit hospitals than in other types of hospital. This suggests that, by choosing a for-profit hospital, patients have on average a lower instantaneous probability of dying but are less sure about the quality of the hospital. [Ed.]