Prediction models for determining the success of labour induction: A systematic review and critical analysis

The purpose of this study was to systematically identify and critically appraise models developed in the past few years, using universally accessible clinical and demographic factors, that have been derived and validated to predict the success of labour induction. Our search identified 26 studies describing 24 prediction models derived between 1966 and 2021. Models varied with regard to methodology, scope and performance. Before any prediction model can be recommended for use in clinical practice, there is a need to determine thresholds of risk at which IoL should not be offered, subgroups that are most likely to benefit from the use of prediction models, and the clinical impact of prediction models on shared decision-making, parental satisfaction, caesarean rates, clinical outcomes and costs. A list of recommendations for improving model performance and utilization, as well as measures for encouraging appropriate use of prediction models and directions for future research, is provided.