Abstract Background We describe participation rates in a special interconceptional care program that addressed all commonly known barriers to care, and identify predictors of the observed levels of participation in this preventive care service. Methods A secondary analysis of data from women in the intervention arm of an interconceptional care clinical trial in Philadelphia (n = 442). Gelberg-Andersen Behavioral Model for Vulnerable Populations to Health Services (herein called Andersen model) was used as a theoretical base. We used a multinomial logit model to analyze the factors influencing women's level of participation in this enhanced interconceptional care program. Results Although common barriers were addressed, there was variable participation in the interconceptional interventions. The Andersen model did not explain the variation in interconceptional care participation (Wald ch sq = 49, p = 0.45). Enabling factors (p = 0.058), older maternal age (p = 0.03) and smoking (p = < 0.0001) were independently associated with participation. Conclusions Actively removing common barriers to care does not guarantee the long-term and consistent participation of vulnerable women in preventive care. There are unknown factors beyond known barriers that affect participation in interconceptional care. New paradigms are needed to identify the additional factors that serve as barriers to participation in preventive care for vulnerable women.