Despite huge improvements in prenatal care, premature births are still a big problem across the globe. Researchers from Kings College London have worked on a new predictive tool for doctors, bringing together different data sets and building an app that can be used to assess risk of preterm birth on a case-by-case basis.

Around the world, some 15 million children are born preterm – before 37 weeks gestation – every year, and more than one million of those cases have complications that prove fatal for the infant. Being able to better assess a woman's risk of giving birth prematurely allows the pregnancy to be better managed, with steps potentially being taken to prevent the early birth.

Doctors use a number of things to try and predict whether a patient is at risk of giving birth prematurely, including measuring the length of the cervix, and checking levels of a biomarker known as fetal fibronectin, which is found in vaginal fluid.

In an effort to improve predictions, the Kings College team developed an algorithm that combines the gestation period of any previous pregnancies with both fetal fibronectin levels and cervix length measurements, to provide a comprehensive estimate of risk.

The algorithm, delivered in the form of an iOS app known as QUiPP, was tested in two separate studies, one focusing on patients already deemed to be at a high risk of preterm birth, and the other working with women exhibiting signs of early labor. More than 1,600 women took part in the two studies, with data from around half being used to develop the model.

For the rest of the women, the probabilities of delivery before 30, 34 or 37 weeks were calculated, as was an estimate of risk of birth in the weeks following the fetal fibronectin test. When compared with the real-life events, with the app was found to be a more accurate tool than any one predictive method used on its own.

While further study is needed to determine whether interventions made as a result of risk assessments have a positive impact on the outcome of pregnancies, the team is positive about its invention, believing that it could have a big impact on care.

"The more accurately we can predict her risk, the better we can manage a woman's pregnancy to ensure the safest possible birth for her and her baby, only intervening when necessary to admit these 'higher risk' women to hospital, prescribe steroids or offer other treatment to try to prevent an early birth," said lead author Professor Andrew Shennan.

The researchers published their work concerning its development in the journal Ultrasound Obstetrics & Gynecology.

View gallery - 2 images