THE WHO ANTENATAL CARE RANDOMIZED CONTROLLED TRIAL - RATIONALE AND STUDY DESIGN

The World Health Organisation and collaborating institutions in develo ping countries al-e conducting a multicentre randomised controlled tri al to evaluate a new antenatal care (ANC) programme, consisting of tes ts, clinical procedures and follow-up actions scientifically demonstra ted to be effective in improving maternal and newborn outcomes. These activities are distributed, for practical reasons, over four visits du ring the course of pregnancy and are aimed at achieving predetermined goals. The study is taking place in four countries, Argentina, Cuba, S audi Arabia and Thailand. Recruitment of study subjects started on 1 M ay 1996. All 53 ANC clinical units had been enrolled by December 1996. Clinics in each country were randomly allocated (cluster randomisatio n) to provide either the new programme or the traditional programme cu rrently in use. Approximately 24000 women presenting for ANC at these clinics over an average period of 18 months will have been recruited. As women attending the control clinics receive the 'best standard trea tment' as currently offered in these clinics, individual informed cons ent is requested only from nwomen attending the intervention clinics. Authorities of the corresponding health districts and all participatin g clinics have provided written institutional informed consent before randomisation. The primary outcome of the trial in relation to materna l conditions is the rate of a morbidity indicator index, defined as th e presence of at least one of the following conditions for which ANC i s relevant: (a) preeclampsia or eclampsia during pregnancy or within 2 4h of delivery; (b) postpartum anaemia (haemoglobin < 90 g/L); or (c) severe urinary tract infection/pyelonephritis, defined as an episode r equiring antibiotic treatment and/or hospitalisation. The primary feta l outcome is the rate of low birthweight (< 2500 g). Adverse maternal and fetal outcomes are expected for approximate to 10% of the control group. Several maternal and perinatal secondary outcomes are also cons idered. A comprehensive cost-effectiveness analysis and women's and pr oviders' satisfaction evaluation are performed concurrently with the t rial. Health-care programmes should be rigorously evaluated by randomi sed controlled trials, which are feasible in developing countries and should be? conducted before introducing new treatments or health inter ventions.