JMIR Publications

ABSTRACT

Background:

Centralized electronic surveillance systems are widely used in intensive care settings to support continuous physiological monitoring and patient safety. Failures in health information technology (HIT) infrastructure can disrupt clinical workflows, obscure patient status, and create latent risk for serious harm. Understanding such incidents requires analytic approaches that extend beyond single classification frameworks.

Objective:

This study aimed to comprehensively classify and analyze a HIT-related incident involving loss of central patient surveillance data in a heart intensive care unit using multiple complementary patient safety and human factors frameworks.

Methods:

A qualitative case report design was used to analyze an incident in which a central surveillance system intermittently lost connection with its server, resulting in unavailable and lost monitoring data. The incident narrative was derived from an internal incident report and supporting documentation and was linguistically adapted for publication. The incident was independently classified using five established frameworks: the International Classification for Patient Safety (ICPS), the Health Information Technology Classification System (HIT-CS), the Systems Engineering Initiative for Patient Safety (SEIPS) 2.0 model, the sociotechnical model by Sittig and Singh, and the Human Factors Analysis and Classification System for Healthcare (HFACS-Healthcare). Findings were synthesized across frameworks to identify convergent themes.

Results:

Across all five frameworks, the incident was consistently characterized as a HIT-driven system failure involving loss of information availability, delayed detection, and multi-patient impact. HIT-CS identified a technical failure in system availability and recovery, resulting in data loss. ICPS classified the event as a documentation/information incident with potential for severe harm. SEIPS 2.0 and the sociotechnical model highlighted disruptions in monitoring tasks, limited system transparency, and organizational dependence on IT intervention. HFACS-Healthcare analysis indicated that the incident was primarily associated with preconditions for unsafe acts and organizational influences, with no unsafe acts by frontline staff identified. Cross-framework alignment demonstrated strong convergence on system-level and organizational contributors.

Conclusions:

This case report demonstrates that HIT-related monitoring failures in high-acuity settings are best understood as emergent sociotechnical events rather than isolated technical faults or human errors. Applying multiple classification frameworks provided complementary insights into detection delays, recovery limitations, and organizational dependencies. Integrative, multi-framework analysis can enhance learning from HIT failures and inform safer design, implementation, and governance of clinical surveillance systems.