We studied the use of prenatal-care and pregnancy outcome in 4,148 deliveries among members of a well-established health maintenance organization (HMO) and 19,116 births among the [1973][1974] White birth cohort in the Portland, Oregon area.Mothers in the HMO were almost one year older on the average, slightly better educated, and less frequently unmarried, but had virtually identical past pregnancy histories when compared with the general population cohort.HMO members began prenatal care one month later and had three fewer visits than the general population (p<.01); 78 per cent of the general population and only 64 per cent of HMO members began prenatal care in the first trimester (p<.01). With maternal risk held constant, low birthweight, neonatal mortality, and infant mortality were 1.5 to 5 times greater with Prepaid group practices have existed in the United States since at least the 1930s, but within the past decade they have been rechristened "'Health Maintenance Organizations" (HMOs) and have been cited as one promising solution to the challenge of providing comprehensive health care to the American public at a reasonable cost and in an acceptable setting." 2 Despite the assertion that HMOs achieve savings through greater use of preventive services and more judicious use of diagnostic and therapeutic services,2'3 some groups have argued that much of the apparent cost-reduction is attributable to lower standards of care and/or to the enrollment of healthier populations whose need for medical care is less than that of the general population. Unadjusted prematurity, neonatal and infant mortality rates did not differ between the HMO and general populations. Multivariate analyses indicated that, independent of all maternal risk factors, HMO membership was associated with an increase of 30 grams in the predicted birthweight (p<.01), but had no effect on mortality. The data suggest that, in Portland, Oregon, pregnancy outcome for HMO members is comparable to that of the general population. (Am J Public Health 1981; 71:381-390.) pregnancy outcome in an HMO and general population.7 8 Based on a 100 per cent sample of Health Insurance Plan (HIP) of Greater New York births and a 10 per cent random sample of all New York City live births for the year 1955, that study demonstrated that for all ethnic groups HIP members began prenatal care earlier than the total New York City population; the differences were small when the comparison was with the population seeing private physicians. Prematurity rates, fetal death ratios, and early neonatal death rates were significantly lower for the HIP members in the years 1955-57. The study does not fully adjust for sociodemographic differences between HIP members and nonmembers. Furthermore, data drawn from the experience in New York City may be outdated and unrepresentative of current HMO performance.If the development of Health Maintenance Organizations is to continue to receive public support and if prenatal services are to be an important part of their preventiv...