For primary prevention, an effective, affordable and safe intervention applied to all mothers without preceding testing is likely to be the most cost-effective approach in asymptomatic women in early pregnancy. For secondary prevention among women at risk of preterm labour in later pregnancy, a management strategy based on the results of testing is likely to be more cost-effective. Implementation of a treat-all strategy with simple interventions, such as fish oils, would be premature for asymptomatic women. Universal provision of high-quality ultrasound machines in labour wards is more strongly indicated for predicting spontaneous preterm birth among symptomatic women than direct management, although staffing issues and the feasibility and acceptability to mothers and health providers of such strategies need to be explored. Further research should include investigations of low-cost and effective tests and treatments to reduce and delay spontaneous preterm birth and reduce the risk of perinatal mortality arising from preterm birth.

6.29 (95% CI,

This review aims to determine the accuracy with which published risk scores predict spontaneous preterm birth in pregnant women. Studies were identified without language restrictions through nine different databases (up to June 2002), and manual searching of bibliographies of known primary and review articles. Two reviewers selected studies independently and extracted data on their characteristics, quality and accuracy. Accuracy data were used to form 2 x 2 contingency tables of the results of risk scoring with spontaneous preterm birth as the reference standard. Heterogeneity was assessed and its reasons were explored. Summary estimates of accuracy were produced within clinically appropriate subgroups. There were 19 primary accuracy articles that met the selection criteria, including a total of 67390 women. There are 12 different risk-scoring systems, the one developed by Creasy being the most commonly evaluated. Quality features of an ideal study, such as blinding and consecutive enrolment, were frequently missing from the included studies, no study fulfilled all criteria for high quality study, and there was heterogeneity between their accuracy estimates. The reference standard most often used was birth before 37 weeks' gestation. The point estimates for the likelihood ratios (LRs) varied widely among the studies. LRs for an abnormal score (LR+) ranged from 1.0 (95% confidence interval (CI) 0.6-1.4) to 38.8 (95% CI 23.5-63.9) while that for a normal score (LR-) ranged from 0.1 (95% CI 0.02-0.6) to 1.2 (95% CI 0.9-1.6). In otherwise asymptomatic women, risk scoring in early pregnancy has a wide range of accuracy in predicting spontaneous preterm birth before 37 weeks' gestation. The evidence is of a relatively poor quality and lacks clinically important reference standards.

Objective To determine the accuracy with which various types of tests for bacterial vaginosis predict spontaneous preterm birth in pregnant women. Data sources Studies were identified without language restrictions through nine different databases, and manual searching of bibliographies of known primary and review articles. Study selection and data extraction There are four different bacterial vaginosis testing methods: Gram staining tests using either Nugent's or Spiegel's criteria, and gas liquid chromatography are laboratory based, and the fourth method uses clinical (Amsel's) criteria to diagnose bacterial vaginosis. Two reviewers independently selected studies and extracted data on their characteristics, quality and accuracy. Accuracy data were used to form 2 Â 2 contingency tables of the bacterial vaginosis test results with spontaneous preterm birth as the reference standard. Data synthesis Data on asymptomatic women and women with symptoms of threatened preterm labour were analysed separately. Data were pooled to produce summary estimates of likelihood ratios for positive (LRþ) and negative (LRÀ) test results for the various types of tests. Results There were 18 primary articles, involving altogether 17,868 women. There was unexplained heterogeneity in the meta-analyses of the accuracy results, which requires caution in their interpretation. Meta-analyses of studies testing asymptomatic women in the second trimester showed that clinical criteria had a LRþ of 0.98 (95% confidence interval 0.59 to 1.6) and a LRÀ of 1.00 (0.93 to 1.1), Gram staining (Nugent's criteria) had a LRþ of 1.6 (1.4 to 1.9) and a LRÀ of 0.9 (0.8 to 0.9), and Gram staining (Spiegel's criteria) had a LRþ of 2.4 (1.4 to 4.9) and a LRÀ of 0.81 (0.64 to 1.0). Among symptomatic women, Gram staining (Spiegel's criteria) had a LRþ of 1.3 (1.0 to 1.6) and LRÀ of 0.9 (0.7 to 1.0). Conclusion There was a lack of difference in the accuracy of the various bacterial vaginosis tests for predicting preterm birth in both asymptomatic and symptomatic women of threatened preterm labour.