Editor’s Note: November’s theme at the International Nursing Knowledge Association (INKA) is Innovation in Nursing Practice. Throughout the month, we’ll explore how nursing research, knowledge, and new approaches to care can improve clinical practice and patient outcomes.
From Nursing Knowledge to Clinical Innovation
Innovation in nursing practice doesn’t necessarily begin with a new technology. It begins with a clinical question: how can we use the knowledge generated through nursing practice to provide safer, more effective, and more person-centred care?
Research is essential to answering this question. It helps nurses understand patients’ needs, evaluate interventions, identify risks, and translate evidence into improvements in clinical practice. Today, advances in artificial intelligence (AI), data science, and digital health offer new opportunities to transform routinely collected clinical information into actionable knowledge.
One particularly promising area is the use of nursing data to predict clinical deterioration. An example is the CONCERN Early Warning System (Communicating Narrative Concerns Entered by Registered Nurses), developed by Sarah Rossetti and colleagues, which demonstrates how research can turn patterns in nursing documentation into information that supports clinical decision-making.
Making nursing surveillance visible through data
Nurses continuously observe patients, assess their conditions, monitor changes, and adapt care accordingly. Sometimes, subtle changes in a patient’s condition become apparent through the frequency, timing, and nature of nursing assessments and documentation before deterioration is clearly reflected in conventional physiological indicators.
The CONCERN Early Warning System uses machine learning to identify patterns in nursing documentation that reflect increased nursing surveillance. Rather than relying exclusively on physiological measurements, it analyses how nursing documentation patterns change over time as nurses respond to patients’ evolving clinical conditions.
The system translates these patterns into an early warning score that is displayed within the electronic health record, helping the clinical team recognise patients who may be at increased risk of deterioration.
This approach illustrates an important principle: nursing data do not simply describe care that has already been delivered; they can also contain signals that help anticipate what may happen next.
The clinical relevance of this approach has been evaluated in a multisite, pragmatic cluster-randomised trial involving 60,893 hospital encounters across 74 clinical units. Compared with usual care, exposure to CONCERN was associated with a 35.6% lower instantaneous risk of in-hospital death and a 4% reduction in length of stay. The study also found a 7.5% lower instantaneous risk of sepsis and a 24.9% higher instantaneous risk of unanticipated transfer to intensive care. These findings demonstrate the potential of integrating nursing surveillance data into clinical decision support, while also highlighting the importance of evaluating individual outcomes carefully.
Importantly, the innovation was not simply the development of an algorithm. It was the integration of research findings into an operational system that made relevant information available to the clinical team when it could inform decisions and actions.
Why the quality of nursing data matters
The CONCERN example also raises a broader question: what kinds of innovation become possible when nursing information is systematically recorded and made available for research?
Nursing assessments, diagnoses, interventions, and outcomes provide information about patients’ functional abilities, symptoms, self-care, risks, responses to treatment, and evolving care needs. When this information is incomplete, inconsistently documented, or difficult to interpret across information systems, its potential value for research and clinical innovation is limited.
Standardized nursing terminologies can help address this challenge by providing a consistent representation of nursing phenomena across clinical settings and information systems. Structured nursing data can then support the investigation of relationships between patient characteristics, nursing assessments, care processes, and outcomes.
CONCERN demonstrates that useful predictive signals can emerge from patterns in nursing documentation, even without depending exclusively on the clinical values recorded. Other research opportunities involve combining standardized nursing diagnoses, interventions, and outcomes with physiological measurements and medical information to develop and evaluate predictive models.
These approaches are complementary. Together, they can help generate a more comprehensive representation of the patient’s condition and the nursing care provided.
For nursing science, this is particularly important. Better data can enable researchers to investigate nursing-sensitive outcomes, evaluate the contribution of nursing interventions, and generate evidence that informs both clinical decisions and the organization of healthcare services.
From predictive performance to meaningful clinical impact
Predictive models alone do not improve patient outcomes. Their value depends on whether they are reliable, appropriately validated, integrated into clinical workflows, and used to support timely and effective action.
Implementation also requires attention to data quality, potential algorithmic bias, usability, alert burden, and the consequences of incorrect predictions. Nurses must be involved throughout this process, from identifying relevant clinical questions to designing, evaluating, and refining digital tools.
The objective is not to replace clinical judgement with automated predictions. It is to provide nurses and the wider clinical team with additional information that supports situational awareness, prioritization, and decision-making.
Research must therefore extend beyond measuring model accuracy. It should establish whether an innovation changes clinical practice, improves the timeliness of interventions, reduces avoidable harm, and produces outcomes that matter to patients and healthcare professionals.
Research as the foundation of nursing innovation
The CONCERN experience illustrates how nursing research can challenge conventional approaches to clinical monitoring and demonstrate the value of information generated through everyday nursing practice.
It also reminds us that the digital transformation of healthcare depends not only on technological development, but on recognizing the knowledge embedded in nursing care and making it visible, measurable, and usable.
As electronic health records and interoperable health information systems continue to evolve, nursing data can become an increasingly important resource for clinical research, predictive analytics, and quality improvement. Achieving this potential requires investment in structured documentation, standardized terminologies, nursing informatics, and interdisciplinary collaboration.
Ultimately, innovation should not be measured by the sophistication of an algorithm or the volume of data collected. It should be measured by the difference it makes to clinical decisions, professional practice, and patient outcomes.
When nursing research transforms routinely collected data into actionable evidence, innovation becomes more than a technological achievement: it becomes an opportunity to improve care and demonstrate the contribution of nursing to better health outcomes.
References
- Rossetti SC, Dykes PC, Knaplund C, et al. Real-time surveillance system for patient deterioration: a pragmatic cluster-randomized controlled trial. Nature Medicine. 2025;31:1895–1902. https://doi.org/10.1038/s41591-025-03609-7
- Rossetti SC, Dykes PC, Knaplund C, et al. Author Correction: Real-time surveillance system for patient deterioration: a pragmatic cluster-randomized controlled trial. Nature Medicine. 2026;32:1556. https://doi.org/10.1038/s41591-026-04382-x
About the Author
Dr. Fabio D’Agostino, RN, PhD, is an Associate Professor and Dean of the BSc Nursing Program at UniCamillus University in Rome, Italy. He serves as Treasurer/Secretary of the International Nursing Knowledge Association (INKA) and has extensive experience in nursing research, education, and clinical practice. His research interests include nursing informatics, digital health, and the use of standardized nursing data to improve clinical practice and patient outcomes. He has contributed to several INKA committees and international research initiatives aimed at advancing nursing knowledge.
Comments (0)
Leave a comment