Clinical Provider Testing Behavior at Michigan Medicine

This project is the result of a partnership between the Division of Quality Health Improvement (DQHI) in the Michigan Medicine Pathology Department and the University of Michigan Stamps School of Art & Design, Masters of Design (MDes) in Integrative Design program. I co-designed alongside a cohort of five Master's students over the course of a four month semester during
the winter of 2018.

Problem

Approach

Observations
& Interviews

Synthesis

Outcome

Conclusion

Our objective was to investigate unnecessary test utilization (inappropriate testing) in acute care settings at Michigan Medicine. Research shows inappropriate testing occurs due to a variety of factors including “defensive” medical practice as providers fear making mistakes, lack of knowledge about tests protocols and lack of communication regarding previous tests results.

Test Utilization Communication Channels

The working model for current state communication channels with Attendings and Residents communicating regularly regarding test-ordering.

We utilized a human-centered design approach to explore our problem space including on-site observations, contextual inquiry, interviewing experts, clinical providers, and patients, and leveraging design methods. For our design process we utilized the Double Diamond method focusing primarily
on future strategy recommendations.

1

2

3

4

5

Expert
Interviews

On-site
Observations

Primary
Interviews

Data
Synthesis

Final
Report

We conducted a combined total of 40 hours of observations within Michigan Medicine including the Trauma Burn Intensive Care Unit, Internal Medicine Intensive Care Unit, and the Gastroenterology department. Our observations focused on gaining a deeper understanding of clinical provider workflow, how tests are ordered and by who, and how decisions are made to determine test appropriate test in real time.  We had a total of 11 interviews with Pathologists (6), Pathologist Residents (2), and Specialist Physicians (3).

Through our observations and interviews we identified two distinct ordering patterns in comparing the ICU and Non-ICU setting. We found that in the ICU context, ordering occurs in real time during rounds in the hallway with a central person ordering on a single computer. In the Non-ICU context, ordering occurs after rounds in a huddle room through a decentralized process in which various Residents order on their own computer based patient case load.

FORCE FIELD ANALYSIS

Force Field Analysis is used to analyze the pressures or 'forces' working for and against an intended change. We used this method to aid in our decision-making regarding proposed recommendations for the Pathology department. We determined this method to be a valuable tool as it clearly identified for our partner the impediments and opportunities for behavior change.

Forces for Change

Available educational resources:
Pathology Handbook, 'Uptodate' software

Position hierarchy influencing individual behavior i.e. Attendings influence Residents

Modeling informal work environment
of Pathology across other services

Huddle room as communal decision-making location for communication

Current behavior change interventions:
Pre-test Probability decision-making process

Decision-making support tools

Pathology dashboard for behavior change

Forces Against Change

Position hierarchy influencing individual behavior i.e. Attendings influence Residents

Lack of extensive knowledge regarding which test to choose and when

Varied perceptions of test appropriateness across all positions

Perception of lack of access to specialist knowledge for ordering system

Providers perception of patient expectations for what patients describe as 'good care'

Lack of formal and regular evaluation system to provide constructive feedback regarding test utilization

Michigan Medicine academic culture leads to:

Transient Residents

Inability to appear wrong

Less questions are asked as a result

Using learning as an excuse for over-testing

MiChart issues:

Processing time

Functional and usability problems

Redundant system features

Lack of notification when test name is changed

PROBLEM TREE

A problem tree is a tool to clarify the hierarchy of problems addressed by the team within a design project; it represents high level problems or related sublevel problems. We used this method as both a data synthesis tool as well as a prioritization too to aid in our decision-making regarding which problem areas would be the first to approach for behavior change.

RECOMMENDATION ONE: Alternative Rounding Protocol to Cultivate Collaborative Open Dialogue

Instead of having only pre-rounds and rounds as a team which includes the Attending, we suggest the addition of post-rounds as a team within the huddle room in order to cultivate a sense of shared decision-making.

​

While Residents transition in and out of hospital units throughout their residency term, the huddle room remains a consistent location for conversation regardless of the unit. Residents grow accustom to learning from Senior Residents and Attendings within this environment and we see the huddle rooms as an opportunity to cultivate behavior change. 

Typical Huddle Room Layout

Proposed Huddle Room Layout

RECOMMENDATION TWO: Co-design Workshop with Pathology and Attendings

To strengthen communication and connect specialist knowledge (Pathologists) to the point of ordering (Attendings and Residents), we suggest hosting a co-design workshop as an intervention to open more enriching lines of communication within the hospital. 

​

Pathologists have key knowledge and expertise which can aid in decision-making to ensure appropriate use of resources and best care for patients. The goal of the workshop will be to not only open lines of communication, but develop shared goals and a better communication system leveraging the current tools at use within the hospital.

We presented our final recommendations to our partners with the Pathology Department DQHI and stakeholders from Michigan Medicine. One MDes student continued this project in the following year for their thesis project including prototyping and testing new tools with Internal Medicine and the Pathology department DQHI to improve behavior change.