Strengths and limitations
This research has generated detailed data on utility weights and costs relating to IOL that can be used in other economic evaluations. It adds to the scarce literature in this context. Conducting a cost-utility analysis (CUA) with generic health outcome measure (QALY) rather than other types of cost-effectiveness analyses (CEA) with outcomes measured in clinical terms (e.g. reduction in stillbirths), enables comparison to other cost-effectiveness outcomes across different fields, and aids decision-makers in selecting healthcare interventions with best value for money. In addition, we used real-life data from a large multi-centre RCT including actual costs and clinical events observed. The results of this trial have been reported in three phases, including clinical outcomes (16), women’s preferences (15) and now cost-effectiveness outcomes. Our findings from different angles strengthen the confidence in our conclusions with the economic benefit, safety and preference of nulliparous women for outpatient-balloon IOL.
Although this was the largest RCT of outpatient IOL undertaken, more research with larger patient cohorts and in different settings is warranted to confirm our conclusions and explore generalisability of results. The costing approach also had some limitations due to the nature of administrative records as hourly costs per length of stay in each clinical unit were calculated based on patient cohort averages (bottom-down approach) rather than a sum of each individual cost component per patient (bottom-up approach), which is regarded as the gold standard in costing. We are nonetheless confident that our costing methodology did not significantly influence overall costs since all relevant components were detailed, including midwifery, nursing, medical and allied health staff costs, pathology, radiology, equipment, consumables, hotel services and business overheads. Furthermore, the time horizon in this evaluation only included immediate adverse events until discharge from hospital for both mother and baby. It is possible that some women may have experienced long-term physical or psychological impact associated with their birth experience, but this was beyond the scope of this study. Finally, all model-based analyses have inherent limitations and might not capture all real-world events.