Prenatal care is all of the care that a pregnant woman receives from organized health care services, as well as from family, relatives, and friends. It begins with planning for pregnancy, and should be seen by those who give it as part of a process that continues through labor and delivery and into the neonatal period. The primary aim of prenatal care is promotion of maternal and fetal health, viewed as a unit until the pregnancy reaches full term (3). This encompasses the goals of reducing maternal and infant mortality and morbidity, detecting early factors that heighten the perinatal risk in both individual pregnancies and vulnerable groups, intervening to improve outcomes, educating all who provide or receive prenatal care, and helping women make their pregnancies and birth a positive life experience.

In 1979, maternal and child health issues were discussed during the annual meeting of the 33 Member States of the European Region of the World Health Organization. During this discussion many countries expressed concern about the services offered for pregnancy, birth and the period following birth. The countries recognized that, as yet, unsolved problems remain which must be examined and they asked the European Regional Office of WHO to mount activities to study and report on these problems surrounding birth and birth care. In response to this request, the maternal and child health unit of the European Regional Office organized a Perinatal Study Group. The Group's 15 members came from 10 countries and spanned 10 professional disciplines: economics, epidemiology, health administration, midwifery, nursing, obstetrics, pediatrics, psychology, sociology, and statistics. For five years the Group conducted surveys, reviewed the literature and brought its own personal and professional experience to discussions of the health services for women and their babies, during pregnancy and birth, and after birth. The entire group met together at least once a year, at which time findings from the previous year's work were presented, followed by lengthy, sometimes heated, open and free discussions.

Latin America and the Caribbean had about 370 million inhabitants in 1980 (8), and about 10 million births occurred in this population that year. Even if, as expected, the birth rate drops 14% between 1980 and the year 2000, the population exposed to reproductive risks would still increase by 26%, because Latin America's population is expected to rise to 600 million by the year 2000 (17). Thirteen million births are expected in the year 2000.

Improvements in nutrition, sanitation, housing, and medical care have been associated with reductions in infectious diseases of infancy, such as diarrhea, and, concomitantly, with a decline in infant mortality. Although deaths from congenital abnormalities have also decreased, the rate of their decline has not kept pace with the overall rate. Thus in the United States they have become the most frequent cause of infant death (58). For abnormalities detected in the newborn period, which comprise 3% of all births, about 8 per cent are due to single gene disorders, 6 to 12% to chromosome abnormalities, and less than 5% to viral infections of the mother (39,69). Little is known about underlying causes in the remainder; polygenic factors, environmental agents, and, probably the most common, genetic-environmental interactions, account for them. Congenital abnormalities also contribute substantially to childhood hospitalizations in several developed countries (17). One chromosome disorder, Down's syndrome, accounts for about 16 to 30% of all severe mental retardation, with other genetic and chromosome disorders contributing another 15 to 20% (58). Although the incidence of congenital abnormalities differs to some extent geographically (39), it seems likely that as any country develops, the proportion of infant mortality and childhood disability due to congenital abnormalities will increase.

When a National Health System was established in Cuba and the health services regionalized, three levels of increasing complexity were established for prenatal care. This led to the use throughout the country of a system for the classification of pregnant women according to risk.

The first system used in 1969 was a qualitative system which divided pregnant women into 4 risk groups. This made it possible to establish levels and standards of care, to supervise and control the quality of prenatal care, and to use nurse-midwives for providing prenatal care efficiently. This very expeditious and practical system seemed somewhat inadequate because it gave little attention to biological and social factors, but stressed the complications of pregnancy and concomitant diseases.

Beginning in 1975, another system was used throughout the country to classify pregnant women during prenatal care. In this quantitative risk classification, an arbitrary value was given to each risk factor, and when these values were added up, the final score made it possible to divide pregnant women into 4 risk categories. This new system was also used for the administration of maternal and child health, but its use resulted in too many pregnant women being put into the high-risk category and many protective measures being taken for pregnant women who possibly did not need them. Validation tests such as sensitivity, specificity, and predictive power were used in this system, and it was found that it suffered from problems in discriminating the risk groups.

The need for a system of classifying pregnant women according to risk led to an epidemiological study whose purpose was to give a more appropriate numerical value to each risk factor. It is also emphasized that the objectives of prenatal care include not only the preventive treatment of diseases of women and children, but also the preparation of women for delivery, encouragement of breast feeding, and preparation of couples for assuming their responsibility as parents.

The field of perinatal medicine has been experiencing dramatic changes recently, mostly due to technological breakthroughs in antepartum and intrapartum care. Concomitantly, a substantial decline in perinatal morbidity and mortality has been observed. Whether these two phenomena are causally related has yet to be demonstrated. In fact, a recent report has suggested that this may not be the case (18).

As critics such as Thomas McKeown (12) have pointed out, Western medicine is concerned firstly with the promotion of medical care and only secondarily with the promotion of health care. Medical care and health care are not necessarily the same thing. There are ways of caring for people's health which do not involve medical examinations, procedures, and drugs—they involve such activities as health education, preventive health care, self-help, and community support. However, since these alternative routes to health mostly lie outside the theory and practice of professional medical care provision, their impact on health and illness is largely unexamined.

Pregnancy care is undergoing an evolution. The trend toward professional control and technological intervention begun decades ago is under challenge by some women's groups and health professionals, who seek to make the pregnant woman herself the locus of control. In the same period, a developing literature concerning the relation of health to psychological and social factors has demonstrated their importance (3) and raised the issue of the place of social support and social action in health care (24).

Ceará occupies an area of 146,817 km2 in northeastern Brazil, where the climate is tropical and there are only two seasons, rainy and dry. Frequent droughts have serious consequences for subsistence farming, resulting in precarious living conditions for the rural inhabitants. In the absence of any industry to provide employment for labor, low incomes prevail, with 90% of the population receiving incomes below the minimum wage for farm workers (9).

Drs. Weinstein and Wulff (in issue 1:3) have raised legitimate and important concerns about the ethical acceptability and feasibility of randomized control trials (RCTs) of diagnostic technologies. Their criticisms are all the more compelling in light of their strong advocacy of rigorous evaluation, and their insights into the advantages of RCTs and potential problems with alternative approaches as outlined with great clarity in Dr. Weinstein's paper. We shall not attempt a point by point rebuttal (partly because we are in full or partial agreement with most of their comments) but instead will highlight several issues which we believe merit further discussion.