Objectives.\p=m-\Although there has been substantial policy interest in interventions to improve the neonatal outcomes of disadvantaged women, little attention has been paid to the health status of pregnant women themselves. We therefore ex- amined whether the provision of health coverage to uninsured low-income pregnant women affects maternal health status or the use of cesarean section. Design.\p=m-\Naturalexperiment in Massachusetts. Patients.\p=m-\Allin-hospital, single-gestation births in 1984 (N=57257) and 1987 (N=64346). Intervention.\p=m-\HealthyStart is a statewide health coverage program for unin- sured pregnant women. In 1985, it covered women with incomes below 185% of the federal poverty level. Main Outcome Measures.\p=m-\Ratesof adverse maternal outcome (severe pregnancy-related hypertension, placental abruption, and a length of stay at least 1 day longer than infants' stay) and cesarean section for uninsured women, and for two concurrent control groups, women with Medicaid and women with private in- surance. We calculated the difference in rates between the uninsured and each concurrent control. We then examined the change in these interpayer differences in rates between 1984 and 1987 to measure the effect of Healthy Start. Main Results.\p=m-\In1984, uninsured women had higher rates of adverse mater- nal health outcome than privately insured women (5.5% vs 5.1%, respectively; in- terpayer difference, 0.4%) and received fewer cesarean sections (17.2% vs 23.0%; interpayer difference, -5.8%). Between 1984 and 1987 there was no statistically significant change in the interpayer difference in adverse outcome relative to wom- en with private insurance. However, the interpayer difference in cesarean sections between the uninsured and the privately insured was reduced by 2.3% (95% con- fidence interval (CI), +0.4% to +4.2%), although the uninsured continued to undergo fewer cesarean sections (22.4% vs 25.9%). Similar results were observed when the uninsured women were compared with women with Medicaid. Conclusions.\p=m-\Theprovision of health insurance alone to low-income pregnant women may not be associated with an improvement in maternal health. Expanded coverage was associated, however, with an increase in the rate of cesarean section. (JAMA. 1993;270:61-64)