A computer-based task management system for junior medical officers during after-hours shifts

To the Editor: After-hours communication of tasks between nursing staff and junior medical officers (JMOs) traditionally uses paper lists, whiteboards, face-to-face requests and frequent paging, regardless of task urgency.1,2 This fragments workflow, which increases cognitive demand and the risk of errors.3-5 Computer-based task allocation systems, still in their infancy, have not been objectively evaluated. The aim of our study was to determine whether using a computer-based task management system to list non-urgent after-hours tasks could improve professional communication and lead to more efficient task management at a busy tertiary hospital. The system’s main webpage on the hospital intranet includes a list of all pending tasks for the selected discipline, with demographics, task details, who requested the task and when, urgency, and time due. JMOs use checkboxes to claim and then complete tasks, which then move to a completed list. Nurses and doctors enter tasks via a simple form.

Nurses and JMOs completed questionnaires using Likert scales before the system was installed at the hospital and at 1 and 6 months after. Of 19 JMOs surveyed, 12 responded before the system was installed and at 1 month after, and 19 responded at 6 months after. Of 126 nurses, 39 responded before the system was installed, 43 responded at 1 month after and 28 responded at 6 months after. Change was assessed by dividing the responses to each question into two groups (desirable and undesirable), and then compared using the Fisher exact test. Responses from JMOs indicated significant improvements in seven of 11 questions (covering paging interruptions, perceived work efficiency, patient identification and handover). Responses from nursing staff indicated significant improvements in four of eight questions (covering response times and requesting tasks). Importantly, both JMOs and nurses reported that overall communication of tasks was improved (P = 0.01). Significant changes were maintained at 6 months.

JMOs were shadowed during eight 8-hour shifts before and after installation. The number of events (such as page received, venepuncture or handover) and minutes observed for each activity were compared using the Welch t test. There were large reductions in time spent dealing with requests for tasks (Box). Time spent walking between wards was reduced (18.1 v 28.4 min, P = 0.02). There was a trend, though not reaching statistical significance, towards more minutes spent on direct patient assessment (58.0 v 47.6 min, P = 0.12). Time spent on recharting medications increased (41.8 v 18.1 min, P = 0.02) but number of charting events did not (P = 0.30). No significant difference was found in time spent on venepuncture, cannulation, documentation, handover, meals, medical emergency calls or reviewing notes and tests.

These results strongly suggest that using a computer-based task list reduces interruptions, potentially increases the efficiency of task management by JMOs and decreases errors including those in charting medications. It allows frequent overview (Box) of pending tasks across an entire hospital, and data storage facilitates monitoring of workforce requirements.