Economic evaluation assesses whether healthcare interventions represent good value for money (1). Well-established methods estimate the incremental costs and benefits of an intervention compared to an alternative, often “treatment as usual” (1). Health benefits can be measured using quality-adjusted life-years (QALYs) in cost-utility analysis, as recommended by the UK National Institute for Health and Care Excellence (NICE) (2). A broader perspective can also be taken, in which non-health outcomes (such as education and employment) are included. Regardless of the perspective taken, the judgment about whether incremental costs are justified by incremental benefits depends on context. For example, high-income countries may be willing to pay more for health gains. In some jurisdictions, this trade-off is built into policy through the adoption of an explicit cost per QALY gained willingness-to-pay (WTP) threshold (1, 3). New interventions for which the cost per QALY exceeds the threshold are deemed not cost-effective and are not implemented. Complementary to this incremental cost per QALY ratio, the incremental net monetary benefit (INMB) statistic can be used to inform decision-making. The INMB combines incremental costs, benefits (i.e., QALYs) and the WTP threshold into a summary statistic (1) to express the value of an intervention in monetary terms. For this purpose, the WTP for QALY gains is established, and the INMB is calculated as [(incremental QALY x WTP)—incremental cost]. Therefore, a positive INMB indicates that the intervention is cost-effective at the relevant cost per QALY threshold.

In addition to informing decision-making on specific healthcare interventions, economic evaluation can inform decisions about implementation strategies (4, 5). This is important because the sub-optimal implementation of cost-effective treatments leads to forgone health and other economic benefits, known as opportunity costs (1). While there is wide recognition of the importance of promoting implementation initiatives, comparatively little attention has been paid to how economic analyses can be used to inform the development of quality improvement (QI) programmes to optimise the implementation of cost-effective treatments (5, 6). For this purpose, health economics methods in implementation science can calculate the benefit lost over time due to sub-optimal implementation through the cumulative INMB forgone. For example, failing to optimally implement an intervention that offers good value for money—such as a smoking cessation treatment—results in a cumulative loss. This loss can be estimated as the total INMB foregone by all individuals who did not receive the intervention (e.g., those who go on to develop avoidable chronic conditions due to continued smoking). This measure can inform the scope for potential benefits to be gained from undertaking implementation initiatives, such as QI programmes. Moreover, the value of a future implementation initiative can be estimated as the cumulative INMB resulting from forecast increases in treatment uptake (7–13).

A practical illustration of this approach can be seen in the case of antenatal magnesium sulphate (MgSO₄) therapy given to women at risk of preterm birth. MgSO₄ is an effective and cost-effective treatment for the prevention of cerebral palsy (CP) in preterm delivery (14–17). Since 2015, the World Health Organisation (18) and NICE (19) have recommended the administration of MgSO₄ in preterm deliveries as a core part of maternity care. Despite these guidelines, the adoption of MgSO₄ has been irregular in the UK and other high-income countries (20–23). For example, in infants below 30 weeks' gestation, the UK National Neonatal Audit Programme (NNAP) reported that in 2017, only 64% of eligible women received it (24). There was also high variation in uptake between different regional networks (range 49%–78%) (24). Following the launch of the National PReCePT Programme in England in 2018 (25), the Preterm Perinatal Wellbeing Package in Scotland in 2017, and locally led QI initiatives in Wales (26) uptake increased to 86% by 2022. The variation in uptake between and within units persisted over time and suggested that the adoption of MgSO₄ remained sub-optimal (27). There is room for improvement, especially in the case of MgSO₄, when small gaps in implementation represent large economic and health opportunity costs. The National PReCePT Programme (25) established 95% as a “stretch” target for MgSO₄ uptake. However, the feasible optimal target may also be informed by actual unit and network performance, which is shaped by contextual factors. Implementing an intervention, even when there is evidence-based and clinical practice guidelines, is complex and dependent on implementation contexts (26). Previous studies (26, 28) suggest that variation in practice may stem from how QI initiatives interact with enabling factors—such as team structure or leadership—to influence the uptake of MgSO₄.

This study has two complementary aims: first, to contribute methodologically to the use of health economics methods, including INMB, to inform resource allocation decisions for implementation initiatives; and second, to illustrate these methods by evaluating the adoption of MgSO₄ for fetal neuroprotection in preterm births in the United Kingdom. For this purpose, using the uptake of MgSO₄ in England, Scotland, and Wales over time, we estimate the optimal implementation level and the cost-effectiveness of a hypothetical future QI programme to promote further adoption of MgSO₄ based on different scenarios of implementation effectiveness and implementation costs (i.e., assuming different levels of increased uptake of MgSO₄ and different costing of the implementation strategies to achieve it).

Our study demonstrates that health economic methods can effectively quantify the value of implementing evidence-based treatments and support the prioritisation of future initiatives within a budget-constrained health and care system. In the case of MgSO₄ for preterm neuroprotection, we estimate that its implementation generated over £900 million in societal benefits between 2014 and 2022 for babies born under 32 weeks' gestation. However, sub-optimal uptake during this period resulted in forgone benefits exceeding £555 million, highlighting a significant missed opportunity in preventing CP among preterm infants. These findings underscore the need for consideration for further investment in implementation initiatives. We estimated that an additional QI programme costing £1 m to further increase adoption of MgSO₄ would be a good use of resources (i.e., cost-effective) if it achieved at least a 1% absolute increase in uptake, and would be highly cost-effective if it achieved 5% or more additional uptake.

In England and Wales, NICE considers two key aspects of health technology implementation. First, its Health Technology Assessment Manual (2) emphasises the importance of evaluating the resource use and costs associated with implementation. It also advocates using economic evaluation evidence to estimate appropriate levels of implementation and the expected impact on the population. Second, NICE acknowledges that implementing its guidelines may require additional resources for clinicians and commissioners (34). However, it does not provide detailed methods for evaluating implementation. In the wider literature, methods including value of implementation (6, 7, 35) and policy cost-effectiveness (35, 36) have been proposed, though their application remains limited (5, 35). Our approach reconciles these methods and assesses value for money of implementation initiatives using the INMB measure (6).

We applied this approach to the case MgSO₄, recommended in NICE guidelines since 2015 for preterm neuroprotection (19). Using clinical judgement and unit-level performance data, we estimated an optimal level of implementation and combined evidence on the treatment's cost-effectiveness with the cost-effectiveness of strategies to increase uptake. Despite strong clinical consensus and policy support for perinatal optimisation, research has identified several barriers to optimal MgSO₄ uptake. These include context-specific and clinician-related factors, such as ambiguity in clinical guidance and variation in team capacity, resources, and safety culture across organisations (26, 28). Effective implementation strategies must address these challenges. Our previous work showed that a national QI programme, focused on building perinatal team capacity and capability, enabled improved uptake through a co-developed toolkit and implementation blueprint (28). However, different settings require tailored levels of QI input depending on local context (26). Therefore, we recommend that future implementation funding prioritise lower-performing units at the ODN level, where targeted support could enhance service integration and uptake. Such initiatives are likely to represent a good use of limited resources, even if they achieve only modest improvements in adoption.

Our results reveal notable differences in MgSO₄ uptake between babies born under 30 weeks' gestation and those between 30 and just under 32 weeks. In England and Wales, this may reflect how NICE guidelines are interpreted, which emphasise treatment for babies under 30 weeks, with less attention to those at higher gestational ages (19). Scotland, where these guidelines do not directly apply, showed even lower uptake for the 30–32 week group. Economic evidence supports MgSO₄ as a highly cost-effective intervention for babies under 32 weeks (14, 31). While recent reviews confirm its effectiveness up to 34 weeks (15, 37), emerging studies have introduced uncertainty about its protective effect between 30 and 34 weeks (38). This has led to the consideration of 30 weeks as a safety cut-off (39). However, given the large number of births between 30 and 32 weeks, our data suggest that the opportunity costs of non-treatment in this group are substantial. Balancing treatment costs, opportunity costs, and potential adverse effects needs to be considered in determining the most appropriate gestational age threshold. In light of recent evidence, our findings, and WHO's recommendation to treat babies under 32 weeks (18), future guidelines and QI programmes should explicitly aim to increase uptake in the 30–32 week subgroup. These differences also point to the potential value of targeted QI interventions.

Despite the important implications of our study in applying health economic methods to assess the impact of implementing evidence-based treatments, several limitations and areas for future research remain. There is scope for further methodological development, particularly for interventions that require sustained adherence (e.g., weight loss medications or psychological therapies), assess greater uncertainty between implementation and treatment outcomes, or raise equity considerations. While these methods are transferable to other settings, results are likely to be context-specific. In the case of MgSO₄, our analysis is constrained by limited evidence on its lifetime cost-effectiveness (40) and does not account for substantial clinical negligence litigation costs associated with perinatally acquired CP. More robust data on the long-term impact of CP are needed to fully understand its societal and individual burden. Additionally, our estimates rely on assumptions about optimal uptake. Although informed by clinical judgment and unit-level performance, further work—such as engagement with key informants—could refine these assumptions. Achieving higher levels of uptake may also require more intensive efforts and greater resource investment. Therefore, results from scenario analyses involving high-performance implementation (i.e., a 10% increase in uptake) should be interpreted with caution, as such levels may be difficult to achieve in practice. Finally, our analysis is limited to the lower WTP threshold of £20,000 per QALY. It is important to highlight that if higher WTP thresholds (e.g., £30,000 per QALY) are used, our conclusions would be even more strongly supported.

Health economic methods offer a valuable framework for informing decision-makers about the potential impact and value for money of future implementation initiatives. In our case study of antenatal magnesium sulphate for the prevention of cerebral palsy in preterm babies, we found that its sub-optimal implementation results in substantial societal benefits lost, highlighting the importance of investing in strategies to improve uptake. This approach enables the quantification of the potential value for money of future quality improvement programmes, and offers a compelling rationale for targeted investment in the implementation of magnesium sulphate.