HI501 Unit 7 Workflow Redesign Proposal Example

The Page Comes First and the Order Comes Later, or Never: A Unit 7 Workflow Redesign Proposal for Inpatient Consult Requests That Changes Roles and Leaves Every Screen Alone

[Student Name]

Purdue University Global

HI501: Health Informatics

Unit 7 Workflow Redesign Proposal

[Instructor Name]

[Date]

Harlow Ridge Health, its services, its staff and every figure are composite; the sources cited are real.

Problem

An audit of 60 composite consult requests across the medicine service found that 22 had no consult order at the time the consultant was paged, 17 orders were eventually entered without a stated clinical question, and 9 requests produced no consultant note within 48 hours, discovered only when the team asked at rounds why the consultant had not come. The median time from the first page to a consultant note was 26 hours. In two of the nine lost requests, the delay mattered clinically: a patient waited an extra day for a decision about anticoagulation before a procedure, and another for a kidney biopsy decision.

Current State

The current workflow, drawn from a week of observation on two medicine units, runs across four roles.

1. Resident: decides a consult is needed, pages the consulting service's fellow or on-call physician, describes the patient by phone, and plans to enter the order later.

2. Consultant fellow: takes the call, often while doing something else, writes the patient's name on a paper list and adds the patient to the day's work.

3. Unit clerk: none; the clerk has no role in consults today.

4. Charge nurse: none; nursing learns about the consult when the consultant appears or when the order appears.

The resident's order, when it arrives, lands in a consult inbox that nobody on the consulting service checks routinely, because requests come by phone. The order screen already has a required field for the clinical question, but residents often type "please evaluate" because the conversation has already happened.

Why the Screens Are Not the Problem

It would be easy to propose a new consult screen, but the existing one already has everything needed: a field for the question, a field for urgency, a routing rule to each service's inbox and an acknowledgment status the consultant can set. The failure lies in who uses those features, in what order and whether anyone watches the result. A process that relies on a phone call and a paper list will lose requests no matter how good the screen is. The order already carries the question, the urgency and the route; the workflow simply asks people to use it after the moment it would help.

Proposed Workflow

1. Resident: enters the consult order first, writing a specific question, for example "Can apixaban be held for 48 hours before a liver biopsy given a mechanical mitral valve?" Urgent consults are followed by a page; routine consults are not.

2. Consult coordinator, an existing role on each consulting service, usually the service's nurse practitioner or the fellow holding the consult pager: checks the service inbox at 8 a.m., noon and 4 p.m., sets the order to acknowledged and assigns a consultant. If the question is unclear, the coordinator calls the resident before accepting.

3. Unit clerk: at the daily 2 p.m. bed huddle, reads aloud any consult order on the unit older than 24 hours without a consultant note, from a report that already exists.

4. Charge nurse: follows up on each open consult the clerk reads by contacting the consulting service.

5. Consultant: answers the question in the order, by name, in the first line of the note.

What Changes for Each Role

The resident gains a step at the start and loses a phone call for routine consults. The consulting service gains three inbox checks a day but loses unplanned interruptions and the paper list. The clerk gains a two-minute task at an existing huddle. The charge nurse gains visibility of consults on the unit and the follow-up of open ones. These changes have been discussed with each group's representative; residents raised a concern that urgent consults might be delayed by the inbox, which is why urgent requests keep the page. Experienced consultants have long argued that a consult works only when the question is specific and the communication is direct (Goldman et al., 1983), and a later update added that the request and the answer should both be documented where the team will see them (Salerno et al., 2007). The redesign builds both principles into the order of steps.

Steps Counted

1. Current: about 7 steps per routine consult, including two phone calls and one paper list, with no check that the consult happened.

2. Proposed: about 6 steps per routine consult, no phone call, and a daily check that catches any request older than 24 hours.

Measures

1. Share of consult requests with an order entered before or at the time of the first page: target 95 percent.

2. Share of consult orders with a specific clinical question, judged by a monthly sample of 20: target 90 percent.

3. Median time from order to consultant note: target under 18 hours for routine consults.

4. Consult requests with no note within 48 hours: target zero.

5. Balancing measure: time to consultant contact for urgent consults, which should not rise.

Risks and How They Are Handled

Three risks deserve a plan before the test begins. The first is that residents keep paging for routine consults out of habit, which would leave the old workflow running beside the new one. The chief residents will remind teams at morning report during the test, and the first measure will show within a week whether orders now come first. The second is that inbox checks slip on busy days, so a request sits from noon until the next morning. The 4 p.m. check is the safeguard, and the clerk's 2 p.m. huddle report catches anything older than a day. The third is that consultants resent being asked to answer a named question, since some prefer a broad evaluation. The note's first line can still answer the question briefly before the consultant adds whatever else they found, so nothing is taken away from the consultant's judgment.

Implementation

The change would be tested on one medicine unit with two consulting services, cardiology and nephrology, for four weeks, then spread to the other units if measures improve and the urgent consult time holds. Sociotechnical models of health IT stress that workflow, people and organizational rules must change together with technology (Sittig & Singh, 2010), and this proposal changes only the first three because the technology is already adequate.

References

Goldman, L., Lee, T., & Rudd, P. (1983). Ten commandments for effective consultations. Archives of Internal Medicine, 143(9), 1753-1755. https://doi.org/10.1001/archinte.1983.00350090131022

Salerno, S. M., Hurst, F. P., Halvorson, S., & Mercado, D. L. (2007). Principles of effective consultation: An update for the 21st-century consultant. Archives of Internal Medicine, 167(3), 271-275. https://doi.org/10.1001/archinte.167.3.271

Sittig, D. F., & Singh, H. (2010). A new sociotechnical model for studying health information technology in complex adaptive healthcare systems. Quality and Safety in Health Care, 19(Suppl 3), i68-i74. https://doi.org/10.1136/qshc.2010.042085