Predicting and reducing “hospital‐acquired infections” using a knowledge‐based e‐surveillance system

Abstract

The use of automated computer methods when detecting hospital‐acquired infections (HAIs) enhances the validity of the surveillance in an effective manner. This is because manual infection control systems used by hospitals are time consuming and are often restricted to intensive care units. This paper proposes a new knowledge‐based electronic surveillance system to predict and reduce HAIs. The system can gather patient‐associated data from hospital databanks to automatically predict patient injury based on the standard central line‐associated bloodstream infection algorithm for HAI detection rules. The application of the proposed electronic system reduces the number of central lines associated with infection of the bloodstream and reduces the length of stay for patient treatment and thus reduces costs. The proposed system has several advantages: (a) It is a web‐based system that collects actual data from patients from several IT sources, which will help collect patient data safely and quickly, thereby predicting HAIs effectively. (b) It has an integrated simulator to generate patient records, providing the ability to train medical personnel and nurses to enhance their skills. (c) It is a multimedia‐based system to improve patient health reporting. (d) It assists policymakers in reviewing and approving control plans and policies to reduce and prevent hospital injuries. (e) The investigational results of the system showed an enhancement value equal to 87%.

How does innovative technology impact nursing in infectious diseases and infection control? A scoping review

“…To use the patients’ data more efficiently and detect HAIs sooner, one study integrated multimedia with a simulator system to process patient records automatically to diagnose and predict CLABSI. This automatic surveillance system helps to reduce the CLABSI rate, and 87% optimization was achieved (compared with traditional prevention) (Noaman et al., 2020 ). Four studies showed that using machine learning and deep learning in imaging data and vital signs data was accurate to diagnose tuberculosis and detect sepsis earlier, reducing antibiotic use and antibiotic duration and improving patient outcome.…”

Development of a knowledge-based healthcare-associated infections surveillance system in China

“…The advantage of this method is its ease of understanding and implementation, but it requires experts to develop the rule base. The work of [ 30 ] proposes a knowledge-based system that could automatically extract patient data and output possible injury based on knowledge rules on common central line-associated bloodstream infection. The article [ 31 ] discusses the development of a rule-based classification system for healthcare-associated bloodstream infections (HABSI) to improve patient safety and infection control.…”

Factors Influencing Clinicians’ Use of Hospital Information Systems for Infection Prevention and Control: Cross-Sectional Study Based on the Extended DeLone and McLean Model

“…Despite the potential benefits of information technology support in IPC practices, there are still several shortcomings that need to be addressed. First, there is a lack of timely response and user-friendly information systems at the system quality level, with system failure being the most common problem [ 8 , 9 ]. Second, at the level of information quality, the data are scattered across multiple systems, resulting in low-quality information lacking integrity and readability that cannot support analysis and decision-making [ 10 - 12 ].…”

“…However, there are few large-scale comprehensive evaluation studies on the application of IPCISs. The theoretical support of the evaluation is weak, and there is a lack of empirical evidence, especially evaluation evidence from direct user clinicians [ 9 , 26 ].…”