Extremely premature birth can leave infants facing a complicated web of medical challenges, but a new nationwide cohort study is examining whether those challenges can be measured more accurately—and used more effectively to predict children’s development years later. The research, led by Ishida, Nakanishi, Iitsuka and colleagues, focuses on neurodevelopmental outcomes at age three in infants born extremely preterm. Its central question is deceptively simple: when clinicians estimate a child’s future risk, is it enough to count the number of major illnesses an infant experienced, or should each condition be assigned a different level of importance?
The study compares two approaches to describing early-life illness. The first is an unweighted morbidity count, which treats every recorded complication as one additional medical problem. In this framework, an infant with three qualifying morbidities receives a score of three, regardless of whether those conditions have very different effects on the brain, the lungs or long-term development. The second approach is a weighted morbidity score, designed to reflect the possibility that some complications carry a greater prognostic significance than others. By placing these measures side by side, the investigators aim to determine which better predicts neurodevelopment at three years.
Extremely preterm infants are born during a period when the brain, lungs, eyes and other organ systems are still undergoing rapid development. Advances in neonatal intensive care have improved survival, but surviving the earliest weeks of life can involve respiratory support, infection, nutritional difficulties, brain injury and other complications. These conditions do not necessarily occur in isolation. Instead, they may cluster together, interact biologically and influence later development through several pathways, including inflammation, altered oxygen delivery, disrupted growth and prolonged exposure to intensive medical treatment.
That complexity creates a challenge for outcome prediction. A simple tally can be attractive because it is transparent and easy to calculate. It may also help researchers and hospitals compare groups of infants across different settings. Yet an unweighted count assumes that each morbidity contributes equally to later developmental risk. A brief complication and a severe neurological injury would both add one point, even though their potential consequences may be markedly different. A weighted score attempts to address this limitation by incorporating the relative contribution of individual conditions into a single measure of overall disease burden.
The concept resembles risk scoring systems used throughout medicine, where factors are assigned different values according to their association with an outcome. In neonatal care, however, the challenge is especially demanding because many complications are interconnected. A serious respiratory disorder may increase the risk of infection or prolonged hospitalization, while poor growth may reflect a combination of illness, feeding difficulty and inflammation. A weighting system must therefore distinguish meaningful prognostic signals from overlapping markers of the same underlying vulnerability. The value of such a score depends not only on statistical performance, but also on whether it can be applied consistently in routine clinical practice.
The researchers’ focus on three-year neurodevelopment reflects an important stage in childhood assessment. By this age, clinicians can evaluate multiple domains, including cognitive abilities, language, motor function and social or adaptive behavior. Development at three years is not a fixed prediction of adult capability, and individual children can make substantial progress over time. Nevertheless, early assessments can identify children who may benefit from closer monitoring, developmental support and timely interventions. A more accurate measure of neonatal morbidity could help healthcare teams direct those resources toward infants at greatest need without relying on a single isolated diagnosis.
A nationwide cohort provides a broad setting in which to test the competing approaches. Large, population-level data can capture variation in neonatal care, hospital practices and patient characteristics that may be missed in a single-center study. It can also make it possible to examine whether a scoring system remains useful across diverse clinical environments. The study’s design is therefore positioned to address a practical question: whether the extra complexity of weighting morbidities produces a meaningful improvement in prognostic utility compared with simply counting them.
The distinction matters beyond statistical modeling. Prognostic tools influence how clinicians communicate with families, organize follow-up and plan early developmental services. If a weighted score offers substantially better prediction, it could support more nuanced counseling by recognizing that the combination and severity of illnesses may matter more than the raw number of diagnoses. If the unweighted count performs similarly, its simplicity could make it preferable, particularly in settings where detailed data or complex calculations are difficult to obtain. The comparison may consequently help determine whether sophistication translates into real-world benefit.
The study also speaks to a broader issue in medical research: how to summarize multidimensional health histories without erasing important differences between patients. Two infants can experience the same number of complications yet follow very different developmental trajectories. Conversely, an infant with several relatively mild conditions may have a different risk profile from one with a single severe neurological complication. By testing weighted and unweighted measures directly, the investigators are evaluating whether a compact summary of neonatal illness can preserve enough biological and clinical information to improve predictions at age three.
The findings will be particularly important because neurodevelopment is shaped by more than neonatal morbidity alone. Gestational age, birth weight, sex, genetic factors, family circumstances, access to therapy and the quality of post-discharge care can all influence later outcomes. A morbidity score cannot replace comprehensive developmental surveillance, and it should not be interpreted as a destiny for any child. Its potential role is narrower but valuable: to refine risk estimation, support evidence-based follow-up and help clinicians identify which infants may require additional attention after leaving the neonatal intensive care unit. The nationwide study by Ishida and colleagues is designed to establish whether counting complications or weighting them provides the clearer signal.
Subject of Research: Prognostic utility of weighted and unweighted morbidity measures for predicting three-year neurodevelopmental outcomes in extremely preterm infants.
Article Title: Weighted and unweighted comorbidity burden for predicting three-year neurodevelopment in extremely preterm infants: A nationwide cohort study
Article References: Ishida, S., Nakanishi, H., Iitsuka, I. et al. Weighted and unweighted comorbidity burden for predicting three-year neurodevelopment in extremely preterm infants: A nationwide cohort study. J Perinatol (2026). https://doi.org/10.1038/s41372-026-02873-w
Image Credits: AI Generated
DOI: 10.1038/s41372-026-02873-w
Keywords: Extremely preterm infants, neonatal morbidity, comorbidity burden, weighted morbidity score, unweighted morbidity count, neurodevelopment, developmental outcomes, neonatal intensive care, prognostic prediction, nationwide cohort study
Tags: assessment of morbidity scoring methodsclinical implications of comorbidity burden assessmentcomorbidity burden in extremely preterm infantsimpact of early-life medical complications on childhood developmentlong-term neurodevelopment predictionmedical challenges in extremely premature infantsnationwide cohort study on preterm infant healthpredictive models for neurodevelopment at age threepreterm birth neurodevelopmental outcomesprognosis of neurodevelopmental delays in preterm childrensignificance of specific neonatal illnessesweighted versus unweighted morbidity scores
