PP043 Topic: AS04–Emerging Sciences, Methodologies, Big Data and Technology: “POSITIVE YET PRUDENT”: NURSES’ PERCEPTIONS ABOUT DATA-DRIVEN ALGORITHMS IN PEDIATRIC HIGH DEPENDENCY CARE WITHIN LOW-RESOURCE SETTINGS, A PRE-IMPLEMENTATION AND HUMAN-CENTRED DESIGN STUDY IN MALAWI
Pediatric Critical Care Medicine, vol. 25, pp. e33–e34
Abstract
Background and Aim: In pediatric high-dependency care within low-resource settings (LRS), nurses encounter challenges in the early detection of deterioration. Data-driven algorithms show potential, but successful design and integration into clinical workflows require consideration of nurses’ perspectives. This study aims to inform the development of a data-driven algorithm and its user-interface to be implemented in a constant monitoring system for LRS. Methods: A human-centered design approach, including contextual inquiry, semi-structured interviews, and co-design sessions were carried out at the high-dependency units of Queen Elizabeth Central Hospital and Zomba Central Hospital in Malawi. Triangulating these methods, we employed inductive thematic analysis to identify what type of algorithm could assist nurses and to design a prototype for the user-interface, which could be incorporated into a vital signs monitoring system. Results: Contextual research revealed the impact of personnel shortage, the limited availability of sensors for monitoring vital signs. Interviews highlighted the benefits of predictive algorithms, especially in predicting deterioration, but stressed the importance of combining them with clinical expertise. Nurses preferred a scoring system based on familiar scales and color codes. Four prototype components were explored during the co-design sessions, with nurses favoring scores which are explainable and represented with color-codes, and a representation showing the score changes. Conclusions: Nurses in LRS perceive that data-driven algorithms, especially those that predict patient deterioration, could improve the provision of care. However, careful consideration of the context is necessary, for which a set of recommendations for design and implementation is presented. Keywords: clinical decision support, technology acceptance, patient deterioration, data-driven algorithms, Low resource setting
Authors 13
-
Margot M. Rakers Aachen
Leiden University Medical Center
Affiliation as printed
Leiden University Medical Center, Public Health and Primary Care, GZ, Netherlands
-
Kamuzu University of Health Sciences
Affiliation as printed
Kamuzu University of Health Sciences, Blantyre, Malawi
-
Affiliation as printed
GOAL3, ‘s Hertogenbosch, Netherlands
GOAL3, 's Hertogenbosch, Netherlands
-
Affiliation as printed
Training Research Unit of Excellence, Blantyre, Malawi
-
Affiliation as printed
GOAL3, ‘s Hertogenbosch, Netherlands
GOAL3, 's Hertogenbosch, Netherlands
-
Affiliation as printed
Training Research Unit of Excellence, Blantyre, Malawi
-
Kamuzu University of Health Sciences
Affiliation as printed
Kamuzu University of Health Sciences, Blantyre, Malawi
-
Affiliation as printed
Training Research Unit of Excellence, Blantyre, Malawi
-
Niels Henrik Chavannes Aachen
Leiden University Medical Center
Affiliation as printed
Leiden University Medical Center, Public Health and Primary Care, Leiden, Netherlands
-
Hendrikus J. A. van Os Aachen
Leiden University Medical Center
Affiliation as printed
Leiden University Medical Center, Public Health and Primary Care, Leiden, Netherlands
-
Affiliation as printed
Amsterdam Centre for Global Child Health, Emma Children’s Hospital, Amsterdam, Netherlands
Amsterdam Centre for Global Child Health, Emma Children's Hospital, Amsterdam, Netherlands
-
Affiliation as printed
GOAL3, ‘s Hertogenbosch, Netherlands
GOAL3, 's Hertogenbosch, Netherlands
-
Leiden University Medical Center
Affiliation as printed
Leiden University Medical Center, Public Health and Primary Care, Leiden, Netherlands
Cited by 0 stored of 0
No patents citing this paper on Lens.org (checked 2026-10-11).