Recommendations for research

1 Relationship between midwifery staffing and outcomes

What is the relationship between midwifery staffing and outcomes in maternity settings in England, and what factors act as modifiers or confounders of the relationship between midwifery staffing and outcomes?

Why this is important

This guideline found some evidence that there is a relationship between midwifery staffing and maternal or neonatal safety outcomes, but the evidence was weak, potentially subject to bias and unclear about the direction of the effect. In particular, it is unclear if any of the following factors modify or confound the relationship between midwifery staffing and maternal or neonatal safety outcomes:

  • Maternal and neonatal factors (for example, women pregnant or in labour, maternal risk factors, neonatal needs, and stage of the maternity care pathway).

  • Environmental factors (for example, local geography and demography, birth settings and unit size, and physical layout).

  • Staffing factors (for example, midwifery skill mix, availability of and care provided by other staff, division of tasks between midwives and maternity support workers, and the need to provide additional services).

  • Management factors (for example, maternity team management and administration approaches, models of midwifery care, staff and student supervision and supernumerary arrangements).

  • Organisational factors (for example, management structures and approaches, organisational culture, organisational policies and procedures including training).

  • Cost and resource use.

Further research is needed to explore the relationships between midwifery staffing and outcomes. This research would help to establish whether staffing ratios can be identified and recommended. Current research is often limited by attempting to explain individual patient level outcomes as a function of aggregate or summary level measures of midwifery staffing resource. Such techniques may fail to adequately capture the resource input used in influencing patient-level outcomes and consequently lead to an overall biased estimate of the impact of midwifery resources on outcomes via measurement error. Future research (preferably either cluster randomised trials or prospective cohort studies) should attempt to obtain better measures of midwifery staff resource use attributable to an individual. This may also require some technique to allow for the competing demands for midwife resource on wards with several patients. In the event that observational data is used, researchers should ideally address any issues of potential endogeneity caused by non-random allocation of staff, in particular where greater numbers or higher graded midwives are allocated to address a more demanding patient case-mix.

2 Decision support methods

What is the effectiveness of Birthrate Plus compared with other decision support methods or professional judgement for identifying safe midwifery staffing requirements and midwifery skill mix for maternity services in England?

Why this is important

Birthrate Plus is widely used throughout maternity services in England, but there is a lack of evidence about what outcomes it influences. Therefore, the effectiveness and cost effectiveness of Birthrate Plus is unknown. It is also unknown whether other toolkits or methods for determining staffing requirements are better (or worse) than Birthrate Plus.

Cluster randomised controlled trials or prospective cohort studies should be designed to compare different defined approaches or decision support toolkits (including Birthrate Plus) with each other or professional judgement. These studies could be done in different maternity settings and should report outcomes relating to midwifery care, safety and satisfaction. Replicate studies should be carried out to provide evidence of reliability and validity.

These comparative studies should help to assess the value of using defined approaches and decision support aids, and to identify those that perform best.