The price of saving a newborn’s life can extend far beyond the hospital walls. A new analysis published in the Journal of Perinatology is drawing attention to a largely invisible consequence of neonatal intensive care: the direct medical out-of-pocket costs paid by families. The study, led by researchers Gorka, Profit and Phibbs, examines who carries the greatest financial burden during a neonatal intensive care unit admission and investigates how structural conditions may shape those costs.

Neonatal intensive care is among the most technologically advanced forms of medicine. Infants in NICUs may require mechanical ventilation, incubators that regulate temperature and humidity, continuous heart and oxygen monitoring, intravenous nutrition, specialized medications and repeated laboratory testing. These interventions can be lifesaving, particularly for babies born extremely prematurely or with serious medical complications. Yet the financial consequences of receiving this care are not experienced equally by every household, even when families receive treatment in the same health-care system.

The researchers’ focus is direct medical out-of-pocket spending: the money families must pay themselves for medical services connected to a NICU admission. This can include deductibles, copayments, coinsurance and other forms of cost sharing that remain after insurance has paid its portion. Unlike the total hospital charge or the amount reimbursed by an insurer, out-of-pocket cost reflects the portion that reaches a family’s own budget. For households already coping with an unexpected high-risk birth, even relatively small individual charges can accumulate rapidly.

The study is designed not only to document these expenses, but also to explore the structural forces behind differences in cost burden. In health economics, structural contributions refer to the way institutions and systems distribute risk and resources. Insurance design, employment-linked coverage, household income, geographic location, hospital characteristics and access to public assistance can all affect how much a family ultimately pays. Two infants with similar clinical needs may therefore generate very different financial experiences for their parents, depending on the architecture of coverage surrounding them.

That distinction matters because the effects of a NICU admission are not limited to the newborn’s medical condition. Parents may need to travel long distances to reach a specialized hospital, arrange accommodation near the facility, pay for transportation or meals, and take unpaid leave from work. These expenses may not always appear in estimates of direct medical out-of-pocket costs, but they can intensify the economic shock associated with a prolonged admission. The medical bill is only one component of the broader financial strain that can accompany neonatal illness.

A central question raised by the analysis is whether the families with the fewest financial resources are also the ones most exposed to high cost sharing. A payment that is manageable for a high-income household may consume a substantial share of a lower-income family’s monthly budget. Economists often describe this pattern as a disproportionate cost burden: the absolute amount paid may not be the highest, yet the payment represents a greater fraction of household resources. This distinction can reveal inequities that remain hidden when researchers look only at average dollar amounts.

The findings are expected to be relevant to health-policy debates about medical debt, insurance protection and the financial consequences of premature birth. If structural factors account for meaningful differences in family payments, reducing inequality may require more than improving clinical care. Potential responses could include limits on neonatal cost sharing, stronger protections against unexpected bills, expanded eligibility for public coverage and assistance with transportation or family support services. The analysis may also help hospitals and policymakers identify families at risk of financial hardship while an infant is still receiving care.

The issue has significance beyond individual hospitalizations because NICU care can influence a family’s financial stability long after discharge. Parents may face follow-up appointments, medications, developmental assessments and specialist care for children who experienced complications of prematurity or serious illness. Medical expenses can overlap with reduced work hours, job loss or the need for one parent to become a full-time caregiver. A short period of intensive treatment can therefore produce a long economic aftershock, particularly when families lack savings or paid leave.

By placing family payments at the center of neonatal research, Gorka, Profit and Phibbs emphasize a dimension of NICU care that is often absent from clinical success measures. Survival, complications and length of stay remain essential indicators, but they do not fully describe the experience of families navigating intensive care. Understanding who pays the most, why they pay it and how those payments relate to broader social structures could help redefine quality in neonatal medicine. The study’s message is potentially viral because it connects cutting-edge life-saving technology with a simple but consequential question: when medicine saves a baby, how much of the cost is left for the family to survive?

Subject of Research: Direct medical out-of-pocket costs of neonatal intensive care admissions and structural factors contributing to differences in financial burden among NICU families.

Article Title: Who pays the most for neonatal intensive care? An analysis of the out-of-pocket cost burden borne by NICU families

Article References: Gorka, N., Profit, J. & Phibbs, C.S. “Who pays the most for neonatal intensive care? An analysis of the out-of-pocket cost burden borne by NICU families.” Journal of Perinatology (2026). https://doi.org/10.1038/s41372-026-02869-6

Image Credits: AI Generated

DOI: 10.1038/s41372-026-02869-6

Keywords: neonatal intensive care, NICU, out-of-pocket costs, medical expenses, health-care inequality, premature birth, financial burden, insurance, neonatal health, health policy

Foundation models are promising to become the “universal translators” of biomedical data—but a new perspective argues that medicine should resist treating them as all-knowing digital doctors. In a review published in Nature Biomedical Engineering, researchers describe how large, adaptable artificial intelligence systems are reshaping biomedical imaging while warning that impressive benchmark scores can conceal serious weaknesses in clinical environments. Their central message is direct: foundation models are most likely to improve healthcare by augmenting specialists, not replacing them.

Foundation models are designed to learn broad representations from enormous and diverse datasets before being adapted to specific tasks. In biomedical imaging, those data may include magnetic resonance imaging, computed tomography, X-rays, ultrasound, digital pathology slides and ophthalmic images. The same underlying model could theoretically identify tumors, segment organs, estimate disease risk, retrieve similar cases and generate clinical reports. The vision becomes even broader when imaging is combined with pathology, electronic health records and genomic information, creating a composite system intended to analyze a patient across multiple biological scales.

That ambition reflects a major change from traditional medical AI. Earlier systems were generally built for one narrowly defined task, such as detecting pneumonia on chest radiographs or outlining a brain lesion on MRI scans. Foundation models instead attempt to create a reusable backbone that can be transferred across diseases, hospitals and imaging technologies. Technically, they learn statistical patterns in high-dimensional data, often through self-supervised training in which the system predicts missing, transformed or associated information. Afterward, developers fine-tune or prompt the model for particular clinical applications.

But biomedical imaging is not a single, uniform world. Images differ according to scanner manufacturer, acquisition protocol, patient population, disease prevalence and local clinical practice. A model trained primarily on data from large academic hospitals may perform very differently in community clinics, rural settings or countries with limited equipment. This problem, known as domain shift, can occur even when images appear visually similar. Small changes in image quality, contrast, hardware or patient demographics may alter the statistical distribution learned by the model and undermine its predictions.

The authors introduce a framework called real-world evaluation and assessment of foundation models, or REAL-FM, to examine whether these systems are ready for practical use. Rather than focusing only on accuracy scores, REAL-FM considers several dimensions at once, including the quality and representativeness of training data, technical readiness, clinical value, integration into existing workflows and responsible artificial intelligence. The framework is intended to help clinicians interpret claims about new models and to push developers toward evaluations that resemble actual medical practice.

One of the sharpest distinctions in the perspective is between pattern recognition and causal reasoning. Foundation models can become extraordinarily skilled at recognizing visual associations—for example, linking a particular texture or anatomical feature with a diagnosis present in their training data. Yet association is not the same as understanding why a disease occurs, how it will progress or whether a treatment will benefit an individual patient. A model may identify a correlation that is valid in one hospital but reflects a hidden confounder, such as a scanner type, reporting convention or patient-selection pattern, rather than a biological signal.

This limitation becomes especially dangerous when models are moved beyond simplified benchmarks. Many benchmark datasets provide carefully curated images, clear labels and narrowly defined tasks. Clinical care is messier: scans can be incomplete, diagnoses can be uncertain, records may contain contradictory information and several conditions may coexist. The perspective highlights a lack of verified generalization across such conditions, along with a shortage of prospective, outcome-based validation. In a prospective study, an AI system would be evaluated while care is actually being delivered, with researchers measuring whether it improves diagnostic accuracy, treatment decisions, patient outcomes or workflow efficiency.

Data scarcity is another obstacle. The largest foundation models require vast quantities of data, but medical information is difficult to collect, standardize and share. Patient records are fragmented across institutions, imaging data may lack reliable annotations and rare diseases are inherently underrepresented. Privacy regulations and governance requirements further complicate the creation of centralized datasets. As a result, a model may appear broadly capable while remaining poorly tested in the populations and clinical circumstances where errors could have the greatest consequences.

The researchers argue that human oversight therefore remains indispensable. In practical terms, foundation models may be most useful as clinical assistants that prioritize images for review, highlight suspicious regions, summarize longitudinal records or offer a second opinion that specialists can interrogate. Safe deployment will require transparent reporting of uncertainty, monitoring for performance drift, mechanisms for correcting errors and clear accountability when recommendations influence care. The future envisioned by the authors is not a single monolithic medical oracle, but a coordinated ecosystem of specialized AI systems, each evaluated for a defined clinical role and connected to expert-led workflows.

The perspective arrives as the biomedical AI field races toward increasingly general-purpose systems. Its warning is not that foundation models lack value, but that technical scale alone cannot establish clinical reliability. Before these models can be trusted with consequential decisions, they must demonstrate robustness across domains, usefulness in real workflows, safety under unexpected conditions and measurable benefits for patients. The proposed REAL-FM framework offers a way to separate viral demonstrations from durable medical progress—by asking not only what a model can recognize, but where it works, why it works and whether it makes care better.

Subject of Research: Foundation models and their real-world evaluation in biomedical imaging and clinical medicine.

Article Title: Foundation models in biomedical imaging: turning hype into reality

Article References: Muneer, A., Zhang, K., Hamdi, I. et al. Foundation models in biomedical imaging: turning hype into reality. Nature Biomedical Engineering 10, 1557–1575 (2026). https://doi.org/10.1038/s41551-026-01762-z

Image Credits: AI Generated

DOI: 10.1038/s41551-026-01762-z

Keywords: foundation models, biomedical imaging, medical artificial intelligence, clinical AI, multimodal AI, domain shift, causal reasoning, clinical validation, responsible AI, REAL-FM