Background: The objectives of this study are to identify the individual- and neighbourhood-level determinants of utilization of prenatal care, and to identify self-reported reasons for not receiving prenatal care in Turkey. Methods: A household-based cluster sample of 1249 women who had a child less than two years old were interviewed in five Turkish cities. Multilevel regression analysis was run to predict the influences of individual- and neighbourhood-level characteristics on utilization of prenatal care. Results: Utilization of prenatal care and the quality of the care received were found to be significantly lower in poorer neighbourhoods. Using multilevel regression analysis (two levels), educational level, income, parity and having health insurance were found to be individual-level determinants, while quality of care offered and stability of the local population were found to be neighbourhood-level determinants of utilization of prenatal care. The most frequent self-reported reason for receiving no prenatal care was 'not having any complaint', and the second was 'insufficient financial resources'. Conclusion: There was a big difference between poor and wealthy neighbourhoods in utilization of prenatal care. This difference was partly due to a contextual effect of neighbourhood status; but mostly due to individual-level variables. Improving the quality of prenatal care may increase not only the benefits of prenatal care, but also its utilization, especially in the public sector. Health and social policies have to take into account diversity among individuals and neighbourhoods in the course of efforts to improve service quality.