In Malaysia, private health insurance coverage is usually limited to inpatient treatment or hospitalisation. With private health insurance, there is a possibility that individuals will use health care services more frequently or spend more on health care (known as moral hazard effects) because they know they are protected. This study estimates the importance of factors affecting the demand for private health insurance and how it affects health care utilisation. This paper also provides an empirical test for the existence of moral hazard effects in health care utilisation. The analysis uses the second and third National Health and Morbidity Surveys (NHMS), which were conducted in 1996 and 2006. The analysis applies a bivariate probit model to estimate the demand for private insurance and its effect on the utilisation of health care. Utilisation of health care is defined as being admitted to either a public or private hospital. The results show that taking up private health insurance is lower among disadvantaged individuals, such as those with lower income, a lower level of education, those living in less developed regions and the unemployed. The findings also show that health conditions have a very strong effect on the hospitalisation decision. Furthermore, the results demonstrate that evidence of moral hazard existed in the 1996 but not in the 2006 data. From the results, policy makers can target an appropriate population for providing health subsidies if the National Health Insurance Scheme is implemented.