Introduction: Management of critically ill patients with respiratory failure demands immediate and precise interpretation of physiological data. Conventional intensive care unit (ICU) workflows are often challenged by latency and fidelity issues arising from manual transcription or complex middleware. This study evaluated the longitudinal impact of a novel Optical Character Recognition (OCR) tool for direct bedside vital sign acquisition on respiratory support escalation times and patient outcomes. Methods: A multi-center longitudinal study was conducted in 85 ICUs across 18 Indian states, analyzing data from 23,306 intubated patients from December 2019 to June 2025. The median age was 57, with 81.27% ARDS admissions. Data were captured directly from monitors using a proprietary OCR model (98% accuracy), bypassing middleware. The study cohort included 9,331 patients escalated to invasive mechanical ventilation (IMV) from a non-IMV baseline (64.7% Room Air, 21.6% Conventional O2, 12.2% NIV, 0.9% HFNC). Outcomes, including time to IMV escalation, mortality, total IMV hours, and ventilator-free days (VFD), were analyzed using linear, multivariable and segmented regression. Results: Median time to IMV escalation decreased by a net 17.80 hours (0.27 hours/month, p< 0.001). Total monthly IMV hours increased by 1296.33 hours/month (p\< 0.001), and mean VFD increased by 0.088 days/month (p< 0.001). Mortality in the IMV escalation cohort significantly declined by a net 27.75 percentage points (0.42 percentage points/month, p\< 0.001), independent of age, length of stay, APACHE II score, or city tier. Segmented analyses showed no significant trend acceleration post-OCR deployment, suggesting improvements were a result of OCR integration and broader critical care advancements. Conclusions: These findings underscore the critical role of streamlined, real-time data acquisition in enhancing timely respiratory interventions and improving patient outcomes in high-volume settings. The progressive integration of the OCR tool was associated with a significant decrease in IMV escalation time and a substantial reduction in mortality, highlighting the value of technological advancements in critical care data management.
Sekar et al. (Sun,) studied this question.