An investigation into the effect of implementation of the Growth Assessment Protocol on stillbirth in rates England

This research used data made available via the Office for National Statistics Secure Research Service, which is being expanded and improved with ADR UK funding. 

Authors: Oliver Hugh, Perinatal Institute

Date: November 2020

Research summary

Research by the Perinatal Institute discovered that the implementation of its Growth Assessment Protocol (GAP) improved antenatal (during pregnancy) detection of fetal growth. This resulted in significant decline in stillbirth risk in NHS maternity units that adopted the GAP.

Stillbirth rates in England have fallen steadily since 2011. However, they have been flagged as being among the highest in Western Europe. This research set out to investigate the potential causes and associations for this drop in stillbirth rates, and whether and how it related to the implementation of the GAP programme. GAP is a licensed and supported service which includes training, e-learning, customised growth charts, care pathways and tools for rolling audit, benchmarking, and reporting. It was developed by the Perinatal Institute to improve antenatal detection of small-for-gestational age fetal growth restriction, the main risk factor for stillbirth.

The authoritative analysis supports better understanding of stillbirth rates, but also stimulates further clinical and health service efforts towards prevention of stillbirths.

Data used

The project used 10-year population-based cohort study data on the GAP implementation, based on the Perinatal Institute's own core set of anonymised antenatal and postnatal records. The period spanned 2008, the date of the first regional roll-out of the GAP program within the NHS, to 2017. 

Unit-level data for live births and stillbirths was used, from the Office for National Statistics (ONS)’s Birth Registrations: England and Wales, 1982-2017 secure access dataset.  Birth data is linked by ONS to the NHS birth notification form to identify additional variables, including the postcode of mother’s residence and place of birth, as well as whether the birth took place within an NHS institution. The postcode was used to derive a deprivation score according to the 2015 Index of Multiple Deprivation. Analysis was limited to the 133 units (NHS trusts) in England.

Methods used

The 133 maternity units in England were categorised into one of three groups, according to their GAP status in 2017:

• units that had not implemented GAP
• units that had implemented GAP partially
• units with completed implementation of GAP.

Research findings

Analysis found that stillbirth rates declined across all groups from 2008 to 2017, and significantly more in units in which GAP was implemented completely than in the non-GAP units.

The steepest decline in stillbirth rate was observed in the 20 maternity units with the highest small-for-gestational age detection rates, which had a 24% lower stillbirth rate compared with the units not using GAP. Even with assessment of:

• the nine Bradford Hill causality criteria and associated characteristics
• potentially confounding effects such as social deprivation, using mixed effects logistic regression analysis

these stillbirth rate disparities were still apparent. This suggests that SGA detection rate remains a significant factor in lowering stillbirth rates when other risk factors are accounted for.