Characterizing nursing time with patients using computer vision.

Area of Science:

• Nursing Science
• Patient Care Technology
• Healthcare Operations

Background:

• Nurses dedicate significant time to patients, yet the precise duration and nature of bedside interactions are not well-documented.
• Understanding nurse activity patterns is crucial for enhancing patient care quality and efficiency.

Purpose of the Study:

• To characterize the quantity and nature of nurse-patient interactions at the bedside.
• To provide a comprehensive, year-long analysis of bedside nursing activities using continuous surveillance.

Main Methods:

• Employed a surveillance device (AUGi) integrating obfuscated computer vision and Bluetooth beacons to track nurse activities.
• Collected data over one year (April 2019-March 2020) across 37 patient beds in two medical/surgical units.
• Analyzed 408,588 nurse-patient interactions involving 49 nurses.

Main Results:

• Nurses spent an average of 9.39 minutes per patient per shift, with less than 1 minute for bedside shift reports.
• Hourly rounding was completed only 52.9% of the time.
• Day shifts had more interactions but shorter average interaction times compared to night shifts.

Conclusions:

• This is the first study to offer continuous, 24/7 bedside surveillance of nursing activities over a full year.
• Findings indicate significant gaps in bedside report completion and hourly rounding.
• Further research is needed to determine optimal interaction timing and duration for improving patient outcomes.