Recipient organisationNIHR Oxford Biomedical Research Centre
NIHR supportRecorded as supported by this research centre
PeriodMar 2025 — May 2029
In plain English
AI plain-English summary
Every year, more than 8,700 women in the UK nearly die during or after childbirth, and 70 die, with women from poorer areas and certain ethnic groups at far higher risk. Current “early warning” scoring systems, which track vital signs like blood pressure and heart rate, are failing to catch deterioration early enough. They rely only on the most recent measurements and ignore other critical data, such as blood test results. This gap in detection means that many women suffer severe, sometimes lifelong harm that could have been prevented. This project will build a new electronic early warning system (eMOEWS) that incorporates a wider range of routinely collected health data. The researchers will first develop better definitions of worsening illness, then test how well existing tools perform across eight to twelve NHS maternity units. They will design a new scoring system and a clear display for maternity staff, so clinicians can quickly see which women are at risk and why. If successful, the eMOEWS could reduce preventable deaths and severe complications, and narrow the stark inequalities in maternal outcomes across the UK.
View original technical description
Every year more than 700,000 women give birth in the United Kingdom. Of these at least 8700 nearly die – called a “near-miss”, and 70 die. Many more women suffer harm, often with effects lasting for life. Women from less wealthy areas and particular ethnic groups are more likely to come to harm."Vital signs" include measurements of blood pressure, heart and breathing rates. Doctors and midwives use tools that score vital signs to spot women becoming unwell. These tools are called “Modified Obstetric Early Warning Scores” (MOEWS). Despite their use, poor outcomes still occur. This may be because MOEWS use only the most recent vital signs. Using extra data like blood tests helps spot unwell people earlier.We aim to reduce poor outcomes for women giving birth. We will find better ways of describing, spotting, and treating women becoming unwell.We plan four linked projects to develop an electronic advanced maternal obstetric early warning system (eMOEWS). Patient and Public (PPIE) collaborators have developed this work with us. We will work closely with them throughout this project.Once we have completed these four projects, we plan to carry out a trial to assess whether the new eMOEWS leads to better outcomes than the existing tools. This trial will be described in a separate protocol.Project OneWe will develop new definitions of worsening illness in women giving birth. We will work with our PPIE colleagues and other experts, reviewing published work. This will help staff use routinely collected health data to spot early illness, before a woman becomes very unwell. We will check that the new definitions reliably identify women becoming unwell.Project TwoUsing the new definitions, we will test how well current MOEWS pick up worsening illness. We will use data from eight to twelve NHS maternity units serving diverse women, and our national maternal review programme.Project ThreeWe will develop an advanced, electronic MOEWS (eMOEWS) working with our PPIE collaborators and other experts. This will use extra information known to affect the risk of poor outcomes. We will test how well the eMOEWS spots worsening illness, using our new definitions.Project FourWe will develop a way to display eMOEWS on maternity units. We will work with staff who use computers along with experts in NHS computer systems. This will allow staff to understand quickly which women are at risk, and why. We will design guidelines for how to use eMOEWS on maternity units with women and staff. This will make sure our new system helps give women the right care at the right time.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know