To evaluate patterns of prenatal care utilization stratified by medical and psychosocial risk. A retrospective cohort of 786 pregnant women who subsequently delivered live births from 1999 to 2003 at the University of Michigan were classified into high medical, high psychosocial, high medical and high psychosocial (dual high risk) and low-risk pregnancies. Chi-square and logistic regression analyses assessed the association between risk and prenatal care utilization using the Kotelchuck Index. Of 786 pregnancies, 202 (25.7%) were high medical risk, 178 (22.7%) were high psychosocial risk, 227 (28.9%) were dual high risk and 179 (22.8%) were low-risk. Over 31% of dual high risk and 25% of high medical risk pregnancies received "adequate plus" prenatal care versus 10% of high psychosocial risk pregnancies. In multivariate analyses, adjusted for risk, race and insurance, high psychosocial risk pregnancies (OR = 1.69; 95% CI 1.06-2.72) were significantly more likely to receive inadequate prenatal care than care of greater intensity. Many high psychosocial risk pregnancies do not receive adequate prenatal care.