Integration of Big Data in Patient Safety Predictive Systems: A Systematic Literature Review

Abstract

Big data plays a transformative role in enhancing patient safety through the development of predictive systems based on real-time analytics and artificial intelligence (AI). This study aims to map the integration of big data technology into patient safety systems, identify thematic trends, and highlight innovations in clinical and nursing practices. A total of 50 articles published between 2014–2024 were analyzed using the PRISMA approach. The findings were classified into three main categories: Big Data in Patient Safety, Patient Risk Prediction, and AI Implementation in Healthcare. The findings indicate that the successful integration of big data in patient safety is heavily dependent on data validity and technological readiness at the point of care. To address the challenge of global disparities, future research should focus on developing predictive systems that are adaptive to local conditions in developing countries as well as strengthening interprofessional collaboration to ensure that the resulting solutions are practical, contextual, and sustainable. Several significant implementations include machine learning models to predict acute kidney injury (AKI), fall risk, ICU needs, and drug-drug interactions, with Python being the most dominant software. AI integration is also evident in nursing tasks such as risk assessment, care planning, and perioperative monitoring via wearable devices.