Integrating artificial intelligence into nursing practice: Development and clinical evaluation of the NURSYA care model based on Levine's Conservation Principles

Although artificial intelligence (AI) is increasingly used in healthcare, its integration with established nursing theories remains limited. Levine's Conservation Model provides a holistic framework for nursing care, yet its routine application can be challenging without structured decision support. This study aimed to develop and evaluate NURSYA, a theory-based AI-supported nursing care system grounded in Levine's Conservation Model, and to examine changes in nursing outcomes following its implementation. A single-group pre-post intervention study was conducted with 300 nurses working in diverse clinical settings in Türkiye. Nurses used NURSYA for three months. Data were collected using a nursing model evaluation scale assessing planning, implementation, evaluation, and usability. Paired t -tests and effect size analyses were performed. Perceived effectiveness scores increased significantly after implementation ( p < 0.001), with improvements observed across all domains. The evaluation scale showed high internal consistency (Cronbach's α = 0.92). NURSYA was associated with significant improvements in nurses' perceived effectiveness in care planning and clinical decision-making. Theory-based AI systems may support evidence-based and patient-centered nursing practice, although patient-level outcomes require further investigation. • A theory-based AI nursing care model grounded in Levine's Conservation Model was developed. • NURSYA significantly improved nurses' care planning and clinical decision-making. • Overall effectiveness of nursing care improved significantly after NURSYA implementation. • High usability and reliability supported the clinical applicability of NURSYA.