Abstract
INTRODUCTION: Women with pre-existing morbidity arising from medical conditions or previous caesarean section are at higher risk of adverse pregnancy outcomes compared to women without such morbidity. Women often face complex pregnancy-related decision-making that may be characterized by conflicting maternal and perinatal priorities. The aim of this systematic review and meta-analysis was to assess randomised controlled trials of decision aids to evaluate whether they are effective at reducing decisional conflict scores and to evaluate what type of decision aids are most effective for women with pre-existing morbidity in pregnancy. METHODS: We searched Medline (via Ovid), Embase (via Ovid), CINAHL (via EBSCO) from the earliest entries until September 2021. We selected randomised controlled trials comparing patient decision aids for women with pre-existing morbidity with usual clinical practice or a control intervention. Study characteristics and Jadad risk of bias was recorded. Meta-analysis by pre-existing morbidity type was performed using Stata 17 and the data was presented with a Forest Plot. Random effects models were used to calculate summary estimates if there was substantial clinical or statistical heterogeneity and post mean DCS scores were described in a sensitivity analysis and presented as a line graph, to improve clinical interpretation of results.. A narrative synthesis of the selected studies evaluated what type of decision aid works and for in what circumstances. RESULTS: Ten randomised controlled trials, which reported data from 4028 women, were included. Patient decision aids were evaluated in women with pre-existing morbidity who were undertaking pregnancy-related decision-making. Patient decision aids reduced decisional conflict scale scores by an additional - 3.7, 95% Confidence Interval - 5.9% to - 1.6%) compared to the control group. Women with pre-existing medical conditions were more conflicted at baseline and had greater reductions in decisional conflict scale score (mean difference vs. control group: - 6.6%; 95% CI - 9.8% to - 3.3%), in contrast to those with previous caesarean section (mean difference - 2.4%; 95% CI - 4.8% to - 0.1%). There was limited evidence on the effect of decision aids on health outcomes. Decision aids reduced unwanted variation in decision-making support across maternity settings. CONCLUSION: Patient decision aids are effective tools to support personalised care planning and informed decision-making in women with pre-existing morbidity. Women with pre-existing medical morbidity were more conflicted at baseline and were more likely to benefit from decision aids. Adoption of aids in this population may lead to improve adherence and health outcomes, warranting further research.