NR 360 Week 3 Vital Signs Workflow Analysis Example

Who Now Carries the Vital Signs? A Workflow Analysis of Device-Integrated Documentation on a Medical Unit

Student Name

College of Nursing, Chamberlain University

NR 360: Information Systems in Healthcare

Instructor Name

Month Day, Year

Who Now Carries the Vital Signs? A Workflow Analysis of Device-Integrated Documentation on a Medical Unit

Scope

This analysis follows one routine on a composite 32-bed adult medical unit: the four-hourly vital signs round, from the moment a patient's vital signs are due to the moment the values are available in the electronic record for the nurse and the provider to act on. The routine was chosen because the unit changed it six months ago, when portable vital signs monitors were connected to the electronic record so that readings transfer automatically after the person taking them confirms the patient's identity on the device. The analysis compares the routine before and after the change over exactly the same stretch of work, from due time to available in the record, and it tags every step with who performs it, what triggers it and roughly how long it holds that person in place.

The unit's staffing did not change between the two periods: five or six patients per registered nurse, and two nursing assistants per 16 patients who take most routine vital signs. Measurements in the analysis come from direct observation of two day shifts in each period by the writer, a staff nurse on the unit, and from the record's own timestamps.

The Routine Before Integration

Step 1, trigger (nursing assistant): A paper round sheet at the station lists patients due at 0800, 1200, 1600 and 2000.

Step 2, measure (nursing assistant, about 3 minutes per patient): Temperature, pulse, respirations, blood pressure and oxygen saturation taken at the bedside and written on the round sheet or a scrap of paper.

Step 3, report abnormal values (nursing assistant to nurse, variable): Values outside set limits are supposed to be reported aloud to the registered nurse; in practice some are reported and some are written on the sheet for the nurse to find.

Step 4, transcribe (nursing assistant, about 1 minute per patient): Values typed into the electronic record at a workstation, usually in batches after the round, 30 to 90 minutes after measurement.

Step 5, review (registered nurse): The nurse sees values when she opens the patient's flowsheet, often after transcription.

The Routine After Integration

Step 1, trigger (record): The electronic record displays a task list on the device showing patients due.

Step 2, identify and measure (nursing assistant, about 3.5 minutes per patient): The assistant scans the patient's wristband with the monitor, takes the measurements, and the device displays them.

Step 3, confirm and send (nursing assistant, about 20 seconds): The assistant reviews the values on the device and presses send; they appear in the flowsheet within a minute as unvalidated.

Step 4, validate (registered nurse, about 30 seconds per patient, but deferred): The nurse must open the flowsheet and validate each set of values before they are considered final; until then, they display in gray.

Step 5, respond to abnormal values (record and nurse): Values outside limits generate a notification to the nurse's phone; the nurse decides whether to reassess.

The Steps That Changed Shape

Four changes stand out. One step was removed: transcription by the nursing assistant, which eliminated the 30- to 90-minute delay and the transcription errors that come with it. Fieler et al. (2013) reported that automated vital signs documentation eliminated transcription errors on the units they studied, and the unit's experience matches that. One step was added: validation by the registered nurse, which did not exist before. One step changed owner: reporting abnormal values moved from a spoken report by the assistant, which was inconsistent, to an automatic notification to the nurse, which is consistent but has created a new flow of notifications. And one step was split: identification of the patient is now a separate scan at the start of the measurement, which adds about thirty seconds but prevents values being entered into the wrong chart.

Where the Work Landed

The central finding is that the new routine moved work from the nursing assistant to the registered nurse, and nobody planned for it. Before integration, the assistant measured and transcribed, and the nurse reviewed values when she chose. After integration, the assistant measures and sends, and the nurse must validate every set of values for her patients and respond to every notification. For a nurse with six patients and vital signs every four hours, that is 36 validations per 12-hour day shift, plus the notifications, each requiring the nurse to open the flowsheet. At about 30 seconds each, validation alone adds roughly 18 minutes per shift, spread across the day in small interruptions. In observation, nurses validated in batches, often two to three hours after the values arrived, and a set of values that stayed gray for two hours was common.

The unit was not asked about this in advance. The implementation team consulted the nursing assistants about the device and the nurse manager about training, but the validation requirement was set by the hospital's documentation policy, which treats device values as unverified until a licensed nurse accepts them. Ash et al. (2004) describe how clinical information systems can shift work between roles and create new tasks that were not visible during design, and this is an example: the new task sits with the person who had the least say in the change.

What the Analysis Suggests

The analysis suggests two changes that would keep the benefits of integration without the unplanned load on nurses. First, the documentation policy could allow values within normal limits to be validated automatically when the assistant confirms them, reserving nurse validation for values outside set limits, which is where the nurse's judgment is needed. Second, notifications could be tiered, so that only values meeting early warning thresholds reach the nurse's phone immediately. Both are policy and configuration decisions rather than nursing practice decisions, and both would need review by the nursing informatics committee. The analysis also shows why workflow should be examined as a whole system of people, tasks, tools and environment, as the systems engineering model of Carayon et al. (2006) describes, rather than as a device installed into an unchanged routine.

Until those decisions are made, the unit can reduce the load within its own control by setting two validation times per shift, at 1000 and 1600 for day shift, when the nurse validates all pending values at once, rather than leaving validation to be squeezed between other tasks. That would make the delay predictable, which matters for providers reading the record.

References

Ash, J. S., Berg, M., & Coiera, E. (2004). Some unintended consequences of information technology in health care: The nature of patient care information system-related errors. Journal of the American Medical Informatics Association, 11(2), 104-112. https://doi.org/10.1197/jamia.M1471

Carayon, P., Schoofs Hundt, A., Karsh, B.-T., Gurses, A. P., Alvarado, C. J., Smith, M., & Flatley Brennan, P. (2006). Work system design for patient safety: The SEIPS model. Quality and Safety in Health Care, 15(Suppl. 1), i50-i58. https://doi.org/10.1136/qshc.2005.015842

Fieler, V. K., Jaglowski, T., & Richards, K. (2013). Eliminating errors in vital signs documentation. CIN: Computers, Informatics, Nursing, 31(9), 422-427. https://doi.org/10.1097/01.NCN.0000432125.61526.27