Continuity of maternity care in a community setting : a randomised controlled trial using the Zelan design

University of Technology, Sydney. Faculty of Nursing, Midwifery and Health.This research investigated a new community-based model of continuity of care provided collaboratively by a small team of midwives and obstetricians (St George Outreach Maternity Project or STOMP). The study considered whether STOMP improved maternal and neonatal clinical outcomes, resulted in a better experience for women and could be implemented within the current resources of a public teaching hospital in Sydney, Australia. A randomised controlled trial using a Zelen design was used to compare the STOMP model with standard care. One thousand and eighty-nine women were randomly allocated to either the STOMP model or standard hospital-based care. The Zelen design was used to increase the participation of women from non-English speaking backgrounds and to reduce disappointment bias in women allocated to the control group. The results suggest that the model of community-based continuity of care is associated with a lower caesarean section rate, more positive experiences for women and costs less than standard care. There were no differences in the number of medical complications experienced in either group, but more women in the control group were admitted to hospital during the antenatal period. There were four perinatal deaths in each group. Women in the STOMP group reported a higher quality of antenatal care compared with the control group. Women in the STOMP group also reported that the community-based service was accessible and convenient with reduced waiting times for appointments. Women in the STOMP group were more likely to have received adequate information about labour, birth and the postnatal period and felt more 'in control' during labour compared with the control group. Women from both groups reported problems with postnatal care, particularly when provided in the hospital. The study also examined the impact of the STOMP model on women from Chinese and Arabic-speaking backgrounds. The STOMP model appeared to reduce the rate of elective and emergency caesarean section in Chinese-speaking women compared with English-speaking women. Small numbers precluded statistical analysis on these data so the results must be interpreted with caution. Women from Chinese-speaking backgrounds reported receiving insufficient information. The STOMP model improved the provision of information, however Chinese-speaking women still reported inferior experiences. There were also differences in the method of infant feeding. The results indicate that the model provides effective, cost efficient and satisfying maternity care. New models of maternity care can be implemented within current resources when organisations have a strong commitment to change