Mapping the Path: Why Doctors Overuse High-Flow Oxygen in Infant Bronchiolitis

Bronchiolitis is one of the most common respiratory infections of infancy, and it sends a remarkable number of babies to hospital wards and pediatric intensive care units every winter. For decades, clinicians have debated how best to support these infants’ breathing, with options ranging from simple observation to low-flow nasal cannula oxygen, high-flow nasal cannula therapy, continuous positive airway pressure, biphasic positive airway pressure, and, in the most severe cases, mechanical ventilation. Yet a curious paradox has emerged in emergency departments around the world: although the disease itself has not become pathologically more severe, and intensive care admission rates have not risen, the use of high-flow nasal cannula therapy, known as HFNC, has climbed steadily. A new study published in Pediatric Research asks why this is happening, and its answer has less to do with the biology of the virus than with the psychology and structure of the emergency department itself.

The rise of HFNC is not without justification. Research over the past fifteen years has demonstrated that the technique, which delivers warmed, humidified oxygen at high flow rates through small nasal prongs, improves patient comfort, produces better respiratory scores, and appears to reduce the rate of intubation in infants with severe respiratory distress. A 2024 Cochrane review pooling sixteen randomized controlled trials with 2,813 pediatric patients found that HFNC outperformed low-flow nasal cannula therapy in reducing respiratory and heart rates, shortening hospital stays, cutting the total duration of oxygen therapy, and lowering the need for treatment escalation. A more recent meta-analysis echoed these findings, reporting significantly reduced treatment failure and shorter pediatric intensive care stays, although it found no improvement in the need for intubation, a result consistent with earlier network meta-analyses. Despite this evidence, clinicians still cannot accurately predict which individual patients will benefit most from HFNC, raising the possibility that the therapy is being used more widely than its genuine advantages warrant, particularly given its considerable resource implications.

Clinical guidelines have responded to this tension. The 2025 Australasian Bronchiolitis Guidelines Update recommends low-flow nasal cannula as the initial treatment for hypoxemic infants with bronchiolitis, reserving HFNC for patients who fail initial oxygen therapy with low-flow support. The World Health Organization offers similar guidance, suggesting that patients responding well to low-flow therapy should not be switched to high-flow systems. These recommendations exist precisely because HFNC use has expanded to the point where quality improvement programs are now actively trying to reduce its use in emergency departments. But guidelines, however well crafted, only matter if clinicians follow them, and ensuring consistent, evidence-based practice in pediatric emergency medicine has proven remarkably difficult.

The barriers to guideline adherence are numerous and deeply embedded. Organizational obstacles include inadequate staffing, limited pediatric resources, low levels of pediatric readiness training, and poor integration of guidelines into clinical workflows. Personal barriers include variable clinical experience and confidence in treating children, along with honest concerns about whether guidelines developed elsewhere apply to local settings. Layered on top of all this is the fundamental character of the pediatric emergency department itself: a fast-paced, high-pressure environment where decisions must be made constantly, often with incomplete information and under time pressure that leaves little room for reflection or consultation of written guidance.

Into this gap steps a study by Shaw and colleagues, which took a fundamentally different approach to the problem of guideline adherence. Rather than simply publishing another guideline and hoping for the best, the research team used a technique called process mapping, combined with clinician co-design, to understand exactly how and why decisions about HFNC are made in real emergency departments. Twelve senior clinicians, both nurses and doctors, from Australian metropolitan, regional, and rural emergency departments participated in a mixed-methods process designed to identify the drivers of inappropriate HFNC use in pediatric bronchiolitis care and to co-design implementation strategies that might actually work on the shop floor.

The findings are illuminating. The behavioral barriers identified included fear of clinical deterioration, limited confidence in pediatric respiratory assessment, and a tendency toward early escalation of care. Contextual barriers included limited exposure to pediatric emergency presentations, high staff turnover, and a scarcity of data on de-escalating patients already started on HFNC. In other words, clinicians were not ignoring the guidelines out of negligence; they were responding to genuine anxieties and systemic pressures that push toward escalation. Perhaps most importantly, the study identified five key decision points, or nodes, in the clinical pathway where the choice to use HFNC is actually made, providing focused targets for intervention rather than vague exhortations to do better.

For each of these five decision points, the researchers mapped the barriers using two established implementation science frameworks: the Theoretical Domains Framework combined with Behavior Change Techniques, and the Consolidated Framework for Implementation Research combined with Expert Recommendations for Implementing Change. These mappings led to the development of suggested implementation strategies, along with co-design concerns and suggestions from the participating clinicians. Combined with results from the Knowledge Translation-Bronchiolitis studies, the authors concluded that a theory-based approach married to a co-design strategy can be feasibly applied to HFNC use in bronchiolitis. The authors do acknowledge a limitation: with only twelve clinicians, including just two representatives from regional and rural emergency departments, the identified barriers may not fully capture the perspectives of those settings, although the results align well with prior Australian bronchiolitis research.

The practical recommendations that emerged are refreshingly concrete. The authors call for standardized oxygen threshold messaging so that all staff interpret saturation targets the same way, micro-learning modules on oxygen saturation interpretation, explicit reassessment timeframes so that escalation decisions are revisited rather than simply inherited, and senior-endorsed guidance on weaning and stopping HFNC. Each of these targets a specific decision node rather than the pathway as a whole, reflecting a central lesson of the study: interventions must be tailored to the clinical setting and to the specific point of decision-making, rather than delivered as one-size-fits-all mandates. This emphasis on co-production and grounded implementation theory marks a shift in how quality improvement in pediatrics is conceived, moving from what works in principle to what can actually be implemented in practice.

The process mapping approach, the commentary accompanying the study suggests, could be readily applied to other pediatric conditions where clinical practice has drifted from the evidence over time, with asthma and gastroenteritis named as prime candidates. But the authors are careful to note that the recommendations are education-focused, and that deeper cultural, resource, and systemic challenges remain. A referring clinical team in one area may correctly recognize that a patient can continue on low-flow oxygen, only for the receiving clinical area, including its senior staff, to refuse the patient as being too ill. Alternatively, insufficient clinical capacity may mean that patients on HFNC are prioritized over others, creating perverse incentives to escalate care simply to secure a bed or attention. No amount of education alone can resolve such structural tensions.

What makes this study resonate beyond pediatric respiratory medicine is its underlying message about the nature of clinical improvement. Process mapping provides an unusually clear window into the challenges of changing care in emergency settings, and it reminds the medical community that understanding the behavioral and contextual determinants of clinical decision-making has become just as important as generating new evidence in the first place. The future of pediatric quality improvement, the authors argue, lies not only in discovering what works best, but in understanding how to implement it in the messy, pressured, human reality of the emergency department. For a therapy as resource-intensive and widely adopted as high-flow nasal cannula oxygen, that insight may prove as consequential as any randomized trial.

Subject of Research: Process mapping to reduce inappropriate high-flow nasal cannula use in pediatric emergency bronchiolitis care

Article Title: Process mapping: finding new routes to guideline adherence

Article References: Karageorgos, S., & Roland, D. (2026). Process mapping: finding new routes to guideline adherence. Pediatric Research. https://doi.org/10.1038/s41390-026-05550-y

Image Credits: AI Generated

DOI: 10.1038/s41390-026-05550-y

Keywords: bronchiolitis, high-flow nasal cannula, pediatric emergency medicine, guideline adherence, process mapping, implementation science, quality improvement, clinical decision-making, infant respiratory care, co-design, Theoretical Domains Framework, de-implementation