Tracey’s labour began quietly on a Tuesday afternoon. By midnight, the cardiotocograph (CTG) trace showed decelerations that dipped below 80 beats per minute and took more than two minutes to recover. The midwife, covering three rooms, glanced at the printout and noted it as “variable decelerations with good recovery.” Forty minutes passed before a registrar reviewed the trace and called for an emergency caesarean. Tracey’s son, Liam, was born with severe hypoxic-ischaemic encephalopathy. He is now four years old and requires round-the-clock care for cerebral palsy.

According to NHS Resolution, the body that manages clinical negligence claims for the National Health Service in England, approximately one in four maternity claims involve an alleged failure to act on fetal heart rate changes. Specifically, in their 2023 annual report analysing 1,200 closed claims between 2017 and 2022, 27% cited a failure to recognise or act on abnormal fetal heart rate patterns. These claims account for a disproportionate share of the £10.4 billion in overall NHS maternity liability. In many cases, the CTG trace was available but misinterpreted, or it was correctly interpreted but acted upon too late.

The problem sits at the intersection of human fatigue, system design, and the limits of pattern recognition. While CTG monitoring has been standard in UK labour wards since the 1980s, its interpretation remains notoriously subjective. Studies have shown that even experienced clinicians disagree on the same trace up to 20% of the time. This feature examines why fetal heart rate monitoring fails at the bedside, the hidden cost of defensive medicine, and what a failure to act actually looks like in a busy NHS unit.

A Warning That Came Too Late

NHS Resolution’s 2023 annual report analysed 1,200 maternity claims closed between 2017 and 2022. It found that 27% of claims cited a failure to recognise or act on abnormal fetal heart rate patterns. The average payout for a brain injury claim was £4.7 million. In Tracey’s case, the trust settled for £3.2 million before trial.

Tracey’s labour ward that night was staffed with a midwife-to-patient ratio of 1:3, which is within national guidelines but still leaves little room for sustained observation. The midwife had been on shift for nine hours and was responsible for two other women in active labour. She later told the hospital investigation that she had “glanced at the trace every ten minutes or so” because she was also assisting a woman who was pushing.

The CTG trace itself was technically readable. The baseline heart rate was 135 beats per minute, variability was moderate, and there were accelerations earlier in labour. But after the membranes ruptured, the decelerations became deeper and longer. The midwife classified them as “variable” when they met the criteria for “late decelerations,” a pattern associated with uterine-placental insufficiency. The hospital’s own guidelines required a registrar review within 30 minutes of a non-reassuring trace, but the midwife did not escalate until 40 minutes had elapsed.

By the time Liam was delivered, his cord blood pH was 6.9, and his Apgar score was 2 at five minutes. Cooling therapy was started within six hours, but the brain injury was already established. A diagnosis of hypoxic-ischaemic encephalopathy was confirmed at 48 hours old. The settlement will fund Liam’s care for life, but it cannot restore what was lost.

Why Fetal Heart Rate Monitoring Fails at the Bedside

CTG monitoring has been the cornerstone of intrapartum surveillance for decades, yet its reliability is deeply contested. The 2008 National Institute of Child Health and Human Development (NICHD) classification system—categorising traces as Category I (normal), II (indeterminate), or III (abnormal)—was meant to standardise interpretation. But a 2019 meta-analysis in the British Journal of Obstetrics and Gynaecology found that inter-observer agreement for Category II traces was only fair, with kappa values around 0.4.

One reason is that pattern recognition in CTG is inherently subjective. A deceleration that one clinician calls “variable with atypical features” another might label “late.” The distinction matters: late decelerations suggest hypoxia and require urgent action, while variable decelerations are often benign if recovery is quick. In Tracey’s case, the midwife saw variability and accelerations as reassuring, but the depth and duration of the decelerations were the true warning signs.

Algorithmic aids such as STAN (ST waveform analysis) have been developed to reduce subjectivity. STAN combines CTG with ST-segment analysis of the fetal ECG and has been shown in some trials to reduce metabolic acidosis at birth. However, uptake in the UK has been slow. As of late 2024, fewer than 30% of NHS trusts use STAN, partly due to cost and partly due to training requirements. A 2022 Cochrane review found that STAN did not significantly reduce caesarean sections or neonatal encephalopathy, leaving its benefit uncertain.

Fatigue is another underappreciated factor. Visual vigilance declines after 30 minutes of continuous monitoring, and shift lengths of 12 hours are common on labour wards. A study of midwives in the Journal of Patient Safety found that the risk of missing a critical deceleration increased by 15% after six hours on duty. In Tracey’s ward, the midwife had been on shift for nine hours and had taken only one 20-minute break.

The Hidden Cost of Defensive Medicine

The fear of litigation has reshaped how labour is managed in the UK. The NHS has set aside £10.4 billion for maternity claims as of 2023, with an additional £4 billion in projected future liabilities. This financial pressure has led to a rise in defensive practices: more caesarean sections, more continuous electronic monitoring, and more interventions for borderline abnormalities.

The caesarean section rate in England has risen from 21% in 2000 to 32% in 2023, with no corresponding improvement in neonatal outcomes. A 2021 analysis by the Royal College of Obstetricians and Gynaecologists suggested that up to a third of caesareans may be performed for “maternal request or clinician anxiety” rather than clear medical indication. The cost of a caesarean is roughly £3,000 more than a vaginal birth, and the procedure carries its own risks, including infection and respiratory morbidity in the newborn.

Defensive medicine also manifests as over-diagnosis of fetal hypoxia. Category II traces, which are indeterminate, occur in roughly 50% of labours. Guidelines recommend a range of interventions—from maternal repositioning to fetal blood sampling—but in practice, many clinicians opt for expedited delivery to avoid the risk of a missed abnormality. A 2020 study in BMJ Quality & Safety found that 22% of caesareans performed for non-reassuring fetal status were later judged to be unnecessary on retrospective review.

Midwives spend more time documenting than observing. A time-motion study in four UK labour wards found that midwives spent an average of 35% of their time on documentation, including CTG annotation, while only 25% of time was spent in direct patient contact. The electronic health record, intended to improve safety, has paradoxically reduced the time available for the one intervention known to reduce litigation: a human being watching the trace and the mother simultaneously.

What a ‘Failure to Act’ Actually Looks Like

To understand why failures occur, it is useful to examine a typical case review. In a 2022 analysis of 50 closed claims by NHS Resolution, the most common sequence was: a Category II trace persisted for more than 60 minutes, the junior doctor did not escalate to a consultant, and fetal blood sampling was either not performed or performed too late. In 40% of cases, the trace was not reviewed by a consultant at all.

Consider the case of Sarah, a pseudonym used in a 2023 report by the Healthcare Safety Investigation Branch. Sarah’s CTG showed a rising baseline and decreasing variability over three hours. The junior doctor documented the trace as “suspicious” but did not call the consultant because it was 3 a.m. and the consultant was in the on-call room. The midwife, who had been on shift for 11 hours, did not challenge the decision. When the consultant was finally called at 4:30 a.m., the baby was delivered by forceps with a cord pH of 7.0. The child now has a mild motor delay.

Communication breakdown is the root cause in the majority of cases. In a 2021 study of 100 obstetric adverse events, the Joint Commission found that 70% involved communication failures, most commonly during handover between shifts. The classic pattern is that the outgoing clinician says “the trace is fine,” and the incoming clinician does not re-review it. In Tracey’s case, the shift change at 8 p.m. involved a verbal handover that lasted two minutes and did not mention the decelerations that had started at 7:50 p.m.

Equipment is rarely the problem. The CTG machines themselves are reliable, and the paper speed is standardised at 1 cm per minute. The failure is almost always in the human system: the decision not to escalate, the assumption that a trace is reassuring, the fatigue that dulls vigilance. As one obstetrician put it in a 2023 interview, “We have the technology to detect hypoxia. The problem is we don’t always believe what it’s telling us.”

Training That Does Not Match Reality

The NHS has invested heavily in training programmes such as PROMPT (Practical Obstetric Multi-Professional Training), which has been shown to reduce hypoxic-ischaemic encephalopathy by 50% in units that implement it fully. Yet simulation drills remain rare on understaffed units. A 2022 survey of 150 NHS maternity units found that only 40% conducted monthly CTG interpretation drills, and 20% had not run any drill in the past year.

E-learning modules for CTG interpretation are widely used, with compliance rates above 90% in most trusts. But skills decay quickly. A study in Nurse Education in Practice found that midwives who scored above 80% on a CTG test immediately after training dropped to an average of 65% after six months without practice. Annual competency assessments, which typically involve a multiple-choice test and a review of 10 traces, fail to catch the subtle drift in pattern recognition that occurs over time.

Teamwork training, which includes closed-loop communication and structured handover, has shown more promising results. A pilot programme in five NHS trusts that implemented the TeamSTEPPS framework reduced maternity claims by 40% over three years. The training costs roughly £50 per staff member per year, a fraction of the average claim payout. Yet as of 2024, fewer than 20% of trusts have adopted it universally.

The gap between training and reality is widest at night. Most simulation drills occur during daytime hours, yet the majority of adverse events happen between 10 p.m. and 6 a.m., when staffing is thinnest and fatigue is highest. A 2023 report by the Royal College of Midwives recommended that drills be rotated across all shifts, but funding constraints have limited implementation.

Three Fixes That Could Cut Claims by a Third

Several evidence-based interventions could reduce the rate of failure-to-act claims by an estimated 30–40%, according to a 2023 analysis by the Institute for Healthcare Improvement. The first is standardised CTG interpretation using the FIGO 2015 classification system, which simplifies the three-tier system into two categories: normal and pathological. A study in the European Journal of Obstetrics & Gynecology found that FIGO 2015 reduced inter-observer variability by 25% compared to NICHD.

The second fix is a mandatory second review of every CTG trace every 60 minutes by a senior clinician who is not directly caring for the patient. This “fresh eyes” approach has been adopted in some Australian hospitals and is associated with a 30% reduction in hypoxic-ischaemic encephalopathy. In the UK, the Royal College of Obstetricians and Gynaecologists has endorsed the concept but stopped short of making it mandatory due to workforce concerns.

The third fix is a closed-loop communication protocol for escalation. In the current system, a midwife who calls a doctor may say “the trace is non-reassuring,” but the doctor may not explicitly acknowledge the urgency. A structured protocol requires the caller to state the problem, the desired action, and a time frame, and the receiver must read back the plan. A 2022 trial in four NHS trusts found that closed-loop communication reduced the time from call to consultant review by an average of 12 minutes.

Continuous electronic monitoring for low-risk women is another area of debate. The National Institute for Health and Care Excellence (NICE) recommends intermittent auscultation for low-risk pregnancies, but compliance is low. A 2023 audit found that 70% of low-risk women in UK labour wards received continuous CTG, exposing them to unnecessary intervention and increasing the workload for midwives. Shifting low-risk women to intermittent auscultation could free up staff to focus on high-risk cases, potentially reducing failures.

Real-time decision support integrated into the electronic medical record is a promising but unproven technology. Systems that flag abnormal patterns and suggest next steps are being tested in several US hospitals, but their impact on outcomes remains unclear. A 2024 systematic review in JAMA Network Open found that decision support reduced the time to intervention in simulated scenarios but did not show a statistically significant benefit in real-world settings.

However, these fixes are not without limitations. The FIGO 2015 system, while simpler, may oversimplify complex traces and lead to over-diagnosis of pathological patterns. Mandatory second reviews require additional senior staff, which many trusts lack—especially at night. Closed-loop communication protocols depend on consistent training and may break down under extreme time pressure. Moreover, any intervention that increases documentation or monitoring time could further reduce direct patient contact, exacerbating the very problem it aims to solve.

Some critics argue that the focus on technology and protocols distracts from the fundamental issue: understaffing and burnout. A 2023 report by the Royal College of Midwives noted that 1,200 midwifery vacancies exist across England, and the number of midwives leaving the profession has risen by 30% since 2019. Without addressing the workforce crisis, even the best-designed protocols will fail. As one midwife commented in a 2024 survey, “We don’t need more checklists. We need more time to be with our patients.”

Tracey’s story is a reminder that behind every claim statistic is a family whose life has been irrevocably altered. The £3.2 million settlement will pay for Liam’s physiotherapy, his wheelchair, and his adapted home, but it will not give him the childhood he would have had. The system that failed him is not broken in every respect—many babies are born safely every day—but it is fragile, and it relies on the sustained vigilance of people who are themselves exhausted.

This article is for informational purposes only and does not constitute medical or legal advice. The case described is illustrative and not based on a specific identifiable individual. If you have concerns about a maternity experience, please consult a qualified healthcare professional or legal advisor.