Longitudinal exploration of the delivery of care following a successful antenatal practice change intervention

Alix Hall ^1* Emma Doherty ^1,2 Nicole Nathan ^1,2 John H Wiggers ^1,2 John Attia ¹ Belinda Tully ² Elizabeth J Elliott ^3,4 Christopher Oldmeadow ⁵ Simon Chiu ⁵ Melanie Kingsland ^1,2

• ¹ The University of Newcastle, Callaghan, Australia
• ² Hunter New England Health, New Lambton, New South Wales, Australia
• ³ The University of Sydney, Darlington, New South Wales, Australia
• ⁴ Sydney Children’s Hospitals Network, Sydney, New South Wales, Australia
• ⁵ Hunter Medical Research Institute, The University of Newcastle, New Lambton, New South Wales, Australia

Sustainment of evidence-based care is required for ongoing population benefits. In a randomised stepped-wedge controlled trial of public maternity services in three health sectors in New South Wales Australia, we demonstrated positive practice change related to addressing alcohol use in pregnancy after delivering a seven-month implementation strategy between February 2018 and November 2019. However, evidence suggests the impact of implementation strategies may reduce over time. It is important to document if and when recommended care reduces so timely support for sustainment can be provided. Interrupted time series analysis of outcomes from the largest sector of the randomised stepped-wedge controlled trial was conducted as a secondary analysis. The analysis explored the rate, time points and extent of change in women’s reported receipt of recommended antenatal care for alcohol consumption, following delivery of an effective implementation strategy. 4,909 (82% consented) women were surveyed. The proportion of pregnant women receiving recommended care reduced significantly per week following the withdrawal of implementation support, for three of the four outcomes: assessment of alcohol consumption (% change per week: -0.66, 95% CI: -1.1, -0.26); advice not to consume alcohol during pregnancy and of potential risks (% change per week: -0.63, 95% CI: -1.1, -0.22); and complete care relevant to alcohol risk level (advice and referral) (% change per week: -0.64, 95% CI: -1.1,-0.22). Similar results were observed regardless of timing of antenatal visit. A more rapid decline occurred for most outcomes from the end of implementation until approximately 30 weeks post-implementation. Despite a reduction in the receipt of recommended care, rates were still higher post-implementation than pre-implementation. Receipt of recommended antenatal care for alcohol consumption declined after active implementation support was withdrawn. The findings suggest the need for ongoing monitoring of care delivery and introduction of additional sustainability strategies at key time points post withdrawal of implementation support.