NR 583NP Week 2 Workflow Analysis Example

Workflow Analysis: From a Positive Two-Question Depression Screen at Rooming to a Follow-Up That Actually Happens, Mapped Over Two Days at One Primary Care Site

Student Name

College of Nursing, Chamberlain University

NR 583NP: Informatics for Advanced Nursing Practice

Instructor Name

Month Day, Year

Workflow Analysis: From a Positive Two-Question Depression Screen at Rooming to a Follow-Up That Actually Happens, Mapped Over Two Days at One Primary Care Site

Where the Process Begins and Ends

The process starts when a medical assistant (MA) opens the rooming flowsheet for an adult patient due for annual depression screening, and ends when a patient whose PHQ-9 reaches 10 or higher has a documented contact with a clinician or the behavioral health consultant (BHC) within 14 days. I chose the 14-day end point because the practice's own protocol names it. The map covers the Linden Avenue site, observed by me on a Tuesday and a Wednesday in September 2026, across three MAs, two nurse practitioners (NPs), one BHC, two front desk staff and myself as care coordinator.

The Diagram, Lane by Lane

The diagram has five lanes. Each box is an action someone performs, with the minutes I observed beside it.

Front desk: checks the patient in (2 to 3 minutes); at checkout, books any follow-up the NP's checkout order names (1 to 2 minutes).

MA: asks the two PHQ-2 questions aloud during vital signs (under 1 minute); keys the total; if the total is 3 or more, reads the remaining PHQ-9 items aloud and keys each answer (3 to 4 minutes); writes "PHQ+" on a sticky note and puts it on the exam room door.

NP: sees the prompt after opening the chart (seconds to dismiss or act); reviews the PHQ-9 (1 to 2 minutes); either messages the BHC for a warm handoff or chooses "follow-up plan documented" and types a return interval.

BHC: receives the message; if free, comes to the room (wait 0 to 25 minutes); if in session, adds the patient to a callback list.

Care coordinator: on Thursday, reviews a list of adults with a positive screen and no visit or call in the past seven days; calls each (5 to 10 minutes per call, about half reached).

What the Two Days Counted

Over two days, 41 adults were roomed. Thirty-four were due for screening, and the PHQ-2 was asked in 29. The five not asked included two who spoke Somali, with no interpreter on the line yet, and three where the MA was running more than 15 minutes behind. Six of the 29 scored 3 or more. Five completed the PHQ-9; three scored 10 or more. One of the three met the BHC before leaving. Two left with "return in 4 weeks" typed in the plan and no appointment on the schedule, because the checkout order was never placed.

Waiting, Drawn Rather Than Implied

The longest box on the map is not an action. It is the 0 to 25 minutes a patient sits in a room while the BHC finishes another session. In two of three cases on these days, the NP decided not to make the patient wait and chose the return interval instead. That decision is reasonable for a single visit and expensive in total: it turns a same-day contact into a call I may or may not complete a week later.

The Path No Policy Mentions

The sticky note on the door is the most informative object on the map. The prompt for a positive screen fires only when the NP opens the chart, which often happens after the NP has already entered the room. The MAs started using the note so the NP knows before the door opens. No procedure document mentions it. It works, and it also means the information travels on paper that is thrown away at the end of the day. Koppel et al. (2008) found that workarounds in medication administration usually pointed to a design problem, not to careless staff, and this one points to where the prompt fires.

A second path is invisible on any screen. Two NPs told me they sometimes type the PHQ-9 answers into the note text instead of the flowsheet when the visit is running late. A score in free text never reaches my Thursday list or the quality report.

Two or Three Expensive Failures

First, the unbooked return. "Return in 4 weeks" satisfies the documentation and books nothing. Both cases on these two days ended this way.

Second, the screen not asked. The two Somali-speaking patients were not screened at all, and nothing on the map records that they were missed. The quality report counts them as not due, because no PHQ-2 row exists.

Third, the free-text score. A patient whose PHQ-9 lives only in note text disappears from every list that would prompt a follow-up.

What the Map Makes Visible

Unertl et al. (2010) describe workflow as the movement of tasks, information and people through a process over time. This map shows that the information moves well within the visit and badly after it. Inside the room, the sticky note and the prompt get a positive screen in front of an NP. After the visit, the process depends on a checkout order that is often skipped and a Thursday call list that catches only flowsheet scores. The PHQ-9 itself is a sound measure (Kroenke et al., 2001); the weakness is in the hand-offs around it.

What the Patients Said

I asked four of the six patients who screened positive, after their visits and with their permission, how the questions felt. Two said they had not expected to be asked and answered quickly to move on. One said hearing the questions read aloud with a parent in the room made it hard to answer honestly. One said nobody at the practice had ever asked before. These are four conversations, not a survey, but they suggest the rooming moment shapes the answer as much as the questions do, which the map alone cannot show.

Limits

Two days at one site is a small sample, and the busiest site may not resemble the others. My presence probably made the MAs more careful about asking. A week-long count at a second site would test whether the unbooked return is as common elsewhere.

What the Next Weeks Can Use

The map gives three baseline numbers: screens asked out of screens due (29 of 34), positive screens with a same-day BHC contact (1 of 3) and positive screens with a booked follow-up (0 of 2 of those not seen). Any change proposed later in this course can be measured against them.

References

Koppel, R., Wetterneck, T., Telles, J. L., & Karsh, B.-T. (2008). Workarounds to barcode medication administration systems: Their occurrences, causes, and threats to patient safety. Journal of the American Medical Informatics Association, 15(4), 408-423. https://doi.org/10.1197/jamia.M2616

Kroenke, K., Spitzer, R. L., & Williams, J. B. W. (2001). The PHQ-9: Validity of a brief depression severity measure. Journal of General Internal Medicine, 16(9), 606-613. https://doi.org/10.1046/j.1525-1497.2001.016009606.x

Unertl, K. M., Novak, L. L., Johnson, K. B., & Lorenzi, N. M. (2010). Traversing the many paths of workflow research: Developing a conceptual framework of workflow terminology through a systematic literature review. Journal of the American Medical Informatics Association, 17(3), 265-273. https://doi.org/10.1136/jamia.2010.004333