In the Zelen design randomization precedes consent. The potential participants were randomly allocated to test and control groups, then approached and informed about the aims of the study and their group allocation. They had the opportunity to accept or refuse the group to which they were randomly allocated. Lack of blinding (14) and potential loss of statistical power (if many participants refuse the allocated group) are the main disadvantages of the design. The main advantage is that the intervention could begin immediately after the consent form was signed (15). With a 9.9% perinatal and neonatal
mortality rate (per 1000 births) in year 2002 and 2.4% of all live births notified with congenital abnormalities in South Australia (10), some negative pregnancy outcomes were expected. For this reason medical records were checked for
pregnancy outcomes before mothers were contacted
again. This check resulted in some further
exclusions (Fig. 1; Table 1).