Objective: The purpose of this study was to determine sociodemographic, psychosocial, and health belief factors that explain the association between maternal drug use and little or no prenatal care.
Study design: A cohort of 812 low-income women who delivered at Johns Hopkins Hospital were administered a postpartum survey. Drug use was determined by self-report, medical record, and toxicologic screens. Medical records were abstracted to determine little or no prenatal care, as defined by </= 1 visit.
Results: Adjustments for sociodemographic characteristics and cocaine and opiate use were predictive of little or no prenatal care. The effect of cocaine was explained by psychosocial and health belief factors: external locus of control, fear of being reported to police, and disbelief in the efficacy of care. Opiate use remained strongly related to little or no care in fully adjusted models (odds ratio, 3.16; P < .001).
Conclusion: Different outreach and education strategies may be necessary to enroll cocaine- vs opiate-using women into prenatal care.