A review of risk scoring for preterm birth.

Risk scoring is a quantitative method used to screen populations to identify persons at increased risk of developing a specific adverse health outcome. Risk scoring for preterm birth has taken the form of a simple additive score with risk factors based on clinical experience to sophisticated multivariable risk models using a large number of factors from several domains. It appears that our attempts at defining an effective risk scoring method for preterm birth have been unsuccessful. The positive predictive values (the percent of women defined as high risk that actually go on to have a preterm birth) have been relatively low (20% to 30%) and vary depending on the population studied. Several difficulties with risk scoring for preterm birth are identified. The biggest limitation of current efforts at screening for preterm birth is our ignorance of the causes of preterm birth. The role of recently defined risk factors, of the potential for synergistic relationship between risk factors, and of physician or hospital factors needs to be evaluated. Epidemiologic evidence is suggestive that preterm birth can be subclassified into preterm labor, preterm rupture of the membranes, and medically induced preterm births; however, substantial problems with misclassification, detection bias, and questionable biologic feasibility argue that these subgroups need to be validated before they can be used with confidence. Careful attention should also be paid to the statistical and methodologic limitations of risk scoring. Unless a single risk factor for a disease outcome has a relative risk over 5 and is very prevalent in the population (more than 25%), most of the disease cases will not have the risk factor. It may be unrealistic, however, to think that a single risk factor will adequately define women at high risk for preterm birth. Therefore in the future, if several lesser risk factors can be identified, multivariate modeling may prove useful in defining a valid high-risk group. In addition, attention should be paid to ensure that the risk factor assessment tools obtain accurate information and are easily reproducible among different health care providers. The reduction of preterm birth in France that occurred over a decade was accompanied by radical social changes in the ways in which the population and government perceived pregnancy and childbirth. In addition to universal access to prenatal care, population-based social changes to liberalize pregnancy leave policies and a commitment to supporting women during pregnancy were the focus of the government's commitment to reducing preterm birth.(ABSTRACT TRUNCATED AT 400 WORDS)