In England and Wales, some investigation frameworks use a gestational threshold of 37 weeks when determining which cases fall within scope.
A baby born at 36 weeks and 6 days falls outside those frameworks.
But a baby born just one day later, at 37 weeks, may then qualify for investigation, subject to additional criteria such as whether the death occurred during labour.
This creates a stark boundary in which two babies, potentially separated by only hours, minutes or even seconds and within the normal uncertainty of clinical dating, are treated differently when it comes to investigation and accountability.
Yet the 37-week threshold did not originate in law, and it was not designed as a rule for determining which deaths should receive independent scrutiny. Over time it has simply been carried forward from earlier uses into later policies, consultation and policy proposals without clear justification
Gestational age
Gestational age is not a fixed measurement. It is a clinical estimate based primarily on ultrasound dating early in pregnancy.
Even with early ultrasound scans, recognised medical guidance explains that gestational dating carries a margin of error of several days and potentially a week. Later assessments may have wider margins.
In practice, the difference between 36 weeks + 6 days and 37 weeks falls entirely within the normal uncertainty of clinical measurement.
Despite this, policy frameworks increasingly treat gestational age as if it were a precise legal boundary.
Where did the 37-week threshold come from?
Many readers assume that the 37-week boundary reflects a clear medical or legal rule.
While 37 weeks is widely recognised as the beginning of “term” pregnancy, this clinical classification was never designed to determine which deaths should receive independent investigation.
In reality, its origins lie in research methodology rather than law or principles of justice.
Each Baby Counts (2015)
The Royal College of Obstetricians and Gynaecologists’ Each Baby Counts programme analysed cases of stillbirth and severe brain injury in babies born at or after 37 weeks gestation.
The purpose of this boundary was methodological. Restricting analysis to babies born at term helped avoid confounding factors linked to prematurity.
It was a research classification, not a boundary intended to determine which deaths should receive independent investigation.
HSIB (2017) Now MNSI
When the Healthcare Safety Investigation Branch established its maternity investigation programme, it adopted the same 37-week threshold from the Each Baby Counts criteria.
However, the function of the boundary had changed.
A threshold originally used to define a research dataset had become a gateway determining which cases qualified for investigation.
From this point onward, the 37-week threshold began to appear repeatedly in later frameworks and proposals.
The 2019 Act
The Civil Partnerships, Marriages and Deaths (Registration etc.) Act 2019 required the Government to publish a report considering whether coroners should investigate stillbirths.
Importantly, the legislation itself contains no reference to gestation.
Section 4 simply requires the Government to consider whether, and if so how, coronial jurisdiction could be extended.
The 2019 consultation
When the Government consulted on possible reforms, again the 37-week threshold appeared in relation to which stillbirths might be investigated.
However, the consultation provided limited explanation as to why a research boundary should determine eligibility for coronial oversight.
The MNI (2025-2026?)
The Maternity and Neonatal Investigation also uses a 37-week threshold when defining cases within its scope in its Terms of Reference.
When families queried why this threshold had been included when they, earlier, raised concerns about the issues of its use in non-coronial investigations, Baroness Amos wrote and confirmed that this was due to its presence in the 2019 consultation.
The Chair also confirmed that the workstream would not only consider whether oversight would apply after 37 weeks, yet did not amend the Terms of Reference when requested, despite concerns that these would become an artifact for the next desktop review.
Its original research context is largely lost with each iteration showing less examination of the basis for the threshold than the previous.
Moving the threshold does not solve the problem
The difficulty is not limited to the specific boundary of 37 weeks.
If the threshold were moved earlier, the same problem would simply shift. A baby born at 36 weeks + 5 days would then be one day outside the line. If the threshold moved again, the boundary would continue to move with it.
Any fixed gestational threshold will always create cases that fall just outside it, even though the underlying concerns about care may be identical.
Some discussions suggest aligning investigation with the legal definition of stillbirth at 24 weeks gestation. While this would align with existing registration law and better reflect the full spirit of the Act, it would not remove the underlying difficulty.
Gestational age remains an estimate, and cases such as 23 weeks + 5 days would still fall just outside any fixed boundary. The same uncertainties about dating would still apply.
Negligence, missed opportunities, or failures in care can occur at any gestation.
The central issue therefore is not where a line is drawn, but whether the presence of serious questions about care should determine whether an independent investigation can take place.
The Result
A boundary originally created for research convenience has gradually become embedded within investigation frameworks and policy discussions.
Yet gestational age remains a clinical estimate rather than a precise measurement.
The Central Question
If serious concerns exist about the care provided to a baby, why should the difference of one day in estimated gestation determine whether those concerns are independently examined?
A boundary created for research has gradually become a boundary for accountability.
Investigation frameworks currently draw a sharp line based on gestational age, even though that age is only an estimate.