Real-Time Streaming Analytics for Healthcare: An Architecture for Continuous Clinical Data Processing

Abstract

This study investigates delayed, fragmented, and weakly integrated data processing in environments where electronic health records, monitoring devices, laboratory platforms, pharmacy systems, wearable sensors, cloud services, and enterprise analytics applications generate continuous patient data not always transformed into timely clinical insight. The purpose of the study was to examine how real-time streaming analytics architecture supports continuous clinical data processing, improves clinical decision-making efficiency, and strengthens healthcare operational performance. A quantitative, cross-sectional, case-based design was adopted using structured five-point Likert-scale survey data from cloud-enabled and enterprise healthcare cases involving healthcare IT professionals, clinicians, hospital administrators, clinical informatics specialists, data analysts, and decision-support staff. Out of 200 distributed questionnaires, 176 valid responses were analyzed, producing an 88.0% valid response rate. The key variables included real-time data ingestion, streaming analytics capability, system interoperability, data security and privacy controls, continuous clinical data processing, clinical decision-making efficiency, healthcare operational performance, and clinical event latency perception. Data were analyzed using descriptive statistics, Cronbach’s alpha reliability testing, Pearson correlation, and multiple linear regression. The findings showed high agreement across major constructs, with mean scores ranging from 4.09 to 4.26 and reliability values ranging from 0.781 to 0.892. Real-time data ingestion was strongly associated with continuous clinical data processing (r = 0.681, p < 0.001), while streaming analytics capability was strongly associated with clinical decision-making efficiency (r = 0.704, p < 0.001). Regression results showed that architecture variables explained 63.4% of operational performance and 59.8% of decision-making efficiency. Continuous clinical data processing was the strongest predictor of operational performance (β = 0.301, p < 0.001), while latency negatively affected decision efficiency (β = -0.263, p < 0.001). The study implies that healthcare organizations should prioritize real-time ingestion, interoperable cloud-enterprise architecture, privacy governance, and low-latency analytics to improve patient monitoring, workflow coordination, decision visibility, operational responsiveness, safety, and outcomes.