Risk-based computational model can predict spontaneous preterm birth

Newly developed predictive models help predict the imminent risk of preterm birth and avoid the negative and costly consequences of overtreatment and misdiagnosis.

Premature baby. Image Credit: M.Moira / Shutterstock.com

Improve predictions of imminent preterm birth

Preterm birth is a difficult case to diagnose due to the nonspecific nature of the condition. Current diagnostic methods are inadequate because overtreatment of symptoms is common and costly to the patient. However, the health consequences of preterm birth not being treated effectively raise concerns not only about the long-term effects on the mother, but also about the success of the birth.

A new study published by Sarah Stock and a colleague in the journal at the University of Edinburgh, UK PLOS medicineProvides a new way to address the challenge of diagnosing the symptoms of preterm birth. This was achieved by developing a risk prediction calculation model that improves the prediction of imminent preterm birth.

First, researchers identified clinical risk factors for preterm birth by analyzing individual participant data from five European prospective cohort studies, including 1,783 pregnant European women, and used them naturally. We have developed a model that predicts the risk of preterm birth.

The model was then validated in a prospective cohort study of 2,924 women with signs and symptoms of early labor from 26 obstetric units led by a British consultant. This allowed researchers to verify the accuracy of the model by comparing the predicted and observed results across the dataset.

The findings show that the best predictive model includes a combination of key symptoms. On the surface, the inclusion of vaginal fluid fetal fibronectin concentration analysis with clinical risk factors exceeded expectations of imminent preterm birth and was more cost effective than fetal fibronectin alone.

Risk prediction models show promising performance in predicting spontaneous preterm birth within 7 days of testing and can be used as part of a decision support tool to guide administrative decisions for women at risk of preterm birth. It is easily feasible and can immediately benefit women and babies and medical services by avoiding unnecessary hospitalizations and treatments. “

Unpredictable Prediction-Models Show Possibility to Limit Health Impacts During Spontaneous Preterm Birth

Implementation of such a predictive model may help identify the most important factors that lead to the clinical diagnosis of adverse events such as spontaneous preterm birth.

Nonetheless, this study pointed out limitations in data collection and subsequent analysis that could limit the representative meaning of this study.

The data collected focused primarily on white European and British patients, with non-white participants providing limited insight into ethnic or regional differences. Therefore, including other populations can improve the accuracy of the model. In addition, risk prediction models improve actual clinical outcomes and should be further investigated if data from the Risk Predictor Development Cohort are missing.

By developing more accurate and applicable predictive models, we can provide important guidelines for successful clinical treatment and avoid misdiagnosis and overtreatment that can have adverse health consequences.

The majority of women with signs or symptoms of preterm birth do not give birth pretermly, but many accept unnecessary hospitalization in case of preterm birth. Risk predictors developed by our research team can help women understand their potential for early childbirth and decide whether to be hospitalized and treated.

We are currently working on linking predictors to birth records, so they are easy to use as part of women’s care and can be continuously improved as more women use them. “

Dr. Stock

Extending the prediction model beyond preterm birth and using risk prediction in a computational database may also provide support for other clinical treatments. This may include the diagnosis of neurodegenerative and cardiovascular disease. These disorders can have a variety of symptoms and often require specific treatment to be successfully treated.