Background Proper utilization of antenatal and postnatal care services plays an important role in reducing the maternal mortality ratio and infant mortality rate. This paper assesses the utilization of health care services during pregnancy, delivery and post-delivery among rural women in Uttar Pradesh (UP) and examines its determinants. Methods Data from a baseline survey of UP Community Mobilization (UPCM) project (2013) was utilized. A cross-sectional sample of currently married women (15 to 49 years) who delivered a baby 15 months prior to the survey was included. Information was collected from 2208 women spread over five districts of UP. Information on socio-demography characteristics, utilization of antenatal care (ANC), delivery and postnatal care (PNC) services was collected. To examine the determinants of utilization of maternal health services, the variables included were three ANC visits, institutional delivery and PNC within 42 days of delivery. Separate multilevel random intercept logistic regressions were used to account for clustering at a block and gram panchayat level after adjusting for covariates. Results Eighty-three percent of women had any ANC. Of them, 61% reported three or more ANC visits. Although 68% of women delivered in a health facility, 29% stayed for at least 48 h. Any PNC within 42 days after delivery was reported by 26% of women. In the adjusted analysis, women with increasing number of contacts with the health worker during the antenatal period, women exposed to mass-media and non-marginalized women were more likely to have at least three ANC visits during pregnancy. Non-marginalized women and women with at least three ANC visits were more likely than their counterparts to deliver in an institution. Contacts with health worker during pregnancy, marginalization, at least three ANC visits and institutional delivery were the strong determinants for utilization of PNC services. Self-help group (SHG) membership had no association with the utilization of maternal health services. Conclusions Utilization of maternal health services was low. Contact with the health worker and marginalization emerged as important factors for utilization of services. Although not associated with the utilization, SHGs can be used for delivering health care messages within and beyond the group.

The limitations of individual level interventions in changing behaviors to improve global maternal, newborn and child health have generated more interest in the patterns of social influence and decision making embedded in families, friends and communities. The purpose of this study is to expand the understanding of village dynamics in India and how first degree social and advice networks and cognitive perceptions of 185 recently delivered women (RDW) in areas with and without women's Self-Help Groups (SHGs) affect immediate breastfeeding. Data was collected in 6 blocks and 36 villages in Uttar Pradesh, India. The expansion of RDW's social worlds and creation of social capital through the organization of Self-Help Groups in their villages allowed us to examine basic relationships and advice formation as well as perceptions of interconnectedness of known groups. RDW living in SHG villages and blocks had consistently higher numbers of relationship ties, health advice ties and higher density of health advice networks than RDW living in the non-SHG areas. RDW's perceived knowing ties were also significantly higher between family and health workers in the SHG areas with related higher immediate breastfeeding rates. These results suggest that SHGs can accelerate community social capital and promote more accountability in the health system to engage with families and support the change from traditional to more evidence-based health practices.