OBJECTIVE: To estimate the proportion of interfacility transfers (IFTs) transported by '108' ambulances and to compare the characteristics of the IFTs and non-IFTs to understand the pattern of use of '108' services for pregnant women in India. DESIGN: A cross-sectional analysis of '108' ambulance records from five states for the period April 2013 to March 2014. Data were obtained from the call centre database for the pregnant women, who called '108'. MAIN OUTCOMES: Proportion of all pregnancies and institutional deliveries in the population who were transported by '108', both overall and for IFT. Characteristics of the women transported; obstetric emergencies, the distances travelled and the time taken for both IFT and non-IFT. RESULTS: The '108' ambulances transported 6 08 559 pregnant women, of whom 34 993 were IFTs (5.8%) in the five states. We estimated that '108' transferred 16.5% of all pregnancies and 20.8% of institutional deliveries. Only 1.2% of all institutional deliveries in the population were transported by '108' for IFTs-lowest 0.6% in Gujarat and highest 3.0% in Himachal Pradesh. Of all '108' IFTs, only 8.4% had any pregnancy complication. For all states combined, on adjusted analysis, IFTs were more likely than non-IFTs to be for older and younger women or from urban areas, and less likely to be for women from high-priority districts, from backward or scheduled castes, or women below the poverty line. Obstetric emergencies were more than twice as likely to be IFTs as pregnant women without obstetric emergencies (OR=2.18, 95% CI 2.09 to 2.27). There was considerable variation across states. CONCLUSION: Only 6% institutional deliveries made use of the '108' ambulance for IFTs in India. The vast majority did not have any complication or emergency. The '108' service may need to consider strategies to prioritise the transfer of women with obstetric emergency and those requiring IFT, over uncomplicated non-IFT.