After a quarter-century of experience as a pediatric infectious diseases specialist, Asunción Mejías has gained extensive knowledge about the deadly unpredictability of respiratory syncytial virus (RSV). This infection leads to the hospitalization of up to 80,000 children under the age of five annually in the United States.
Mejías, who is part of St. Jude Children’s Research Hospital in Memphis, Tennessee, describes RSV as a disease with rapidly changing conditions. She has observed that for every two children admitted, one might require intensive care within three hours, while the other may be discharged the following day, highlighting its unpredictable nature.
RSV infections are exceedingly common, with nearly every child contracting the virus by the age of two. While most children exhibit symptoms akin to a cold, some may develop severe lung disease. Globally, RSV is responsible for over 100,000 infant deaths each year, with nearly half occurring in babies under six months of age.
Determining which children will suffer the most from RSV remains challenging, except for a few known risk factors such as premature birth and preexisting lung conditions. Mejías notes that 80% of children hospitalized with RSV appear to be perfectly healthy, having been born full term without any risk factors for severe disease.
In response, various research groups globally are seeking to develop machine learning algorithms and statistical models to identify children most vulnerable to RSV. Utilizing extensive databases of electronic health records, these tools aim to pinpoint risk factors that can predict the likelihood of hospitalization due to RSV. Health care providers can leverage this information to prioritize vaccinations and other preventative measures for those most at risk.
Respiratory epidemiologist Tina Hartert and her team at Vanderbilt University have developed a tool based on a statistical model trained on data from over 400,000 infants in the Tennessee Medicaid program. This tool identifies 19 risk factors for RSV and enables the calculation of an individual infant’s risk at birth, according to Hartert.
Some risk factors used in the model are familiar. For instance, prenatal smoking, which is known to impair lung development in the fetus, makes a baby more susceptible to viral pneumonia. Babies with low birth weight also struggle to breathe normally. However, Hartert emphasizes that often, it is a combination of various risk factors that renders a child vulnerable, and focusing on individual factors alone fails to identify many at-risk infants.
In 2023, U.S. regulators approved a vaccine known as Abrysvo, designed for administration to mothers during weeks 32 to 36 of pregnancy, to ensure babies are born with protective antibodies against RSV. Additionally, they approved a drug called Beyfortus, a monoclonal antibody, which can be given through a single injection to provide protection prior to the winter RSV season.
