An engineering leader’s insights on optimizing healthcare and preventing burnout
In this Q&A, Engineering Professor and Michigan Medicine Chief Transformational Officer Amy Cohn discusses problem-solving and workplace well-being
Amy Cohn became Michigan Medicine’s first chief transformation officer in 2021, bringing an engineer’s mindset and a passion for problem-solving to benefit patients and practitioners alike. She leads a team charged with redesigning and enhancing processes to improve health system operations and outcomes.
Healthcare
Michigan Engineering is advancing
critical technologies for a healthier world
Cohn was a guest on the podcast Well-Being at Michigan Medicine, hosted by Chief Wellness Officer Elizabeth Harry, M.D. In this edited excerpt of their conversation, Cohn shares her unconventional career path and approach to systems-level change.
Cohn is director of the Center for Healthcare Engineering and Patient Safety (CHEPS) and an Arthur F. Thurnau Professor in the Department of Industrial and Operations Engineering. In addition to her role at Michigan Medicine, Harry is clinical associate professor of learning health sciences and internal medicine at the U-M Medical School.
Elizabeth Harry: How did you get into this very unique niche at the intersection of engineering and medicine?
Amy Cohn: I joke that whatever room I’m in, I don’t quite fit in, but that that’s kind of a secret superpower. My first job was actually in the trucking industry. And somehow, that matches up with solving complex problems in health care. I’m really interested in complex systems and how what we think is the problem is often not really the problem, but just a symptom. The engineer in me just really wants to fix problems. And health care, unfortunately, has lots of problems.
I’m incredibly grateful to have a position at a university like this where we have so many different talents and depth and breadth in so many fields, so that someone in engineering can come together with someone in medicine and improve patient care and provider well-being and make sure that our staff have what they need to do their jobs.
What are some operational issues your team has tackled?
There are so many big problems, but most of them need to be fixed by solving many, many little problems. I’ll give an example from our call system.
There were a lot of messages going from the centralized call center and triage nurses into the providers… and they didn’t feel like they were conveying important information. We talked to the providers getting the largest number of them and many nearly begged us to make them stop. But a couple appreciated the information about their patients and said, “Don’t take away that connection that I feel.” So we changed the system that was kind of a push to a pull. We created some reports that you could pull whenever you wanted.
Your team brings an evidence-based framework around how systems and people work best, and you involve students in unique ways. Tell us about that.
In CHEPS, we hire students as paid members of professional teams working to solve real problems. These are engineering students, public health students, nursing students, pre-med, data analytics. It’s a great experiential learning opportunity. This year we had 500 applications to fill 12 slots.
We then bring in a clinician with a problem. And they start off by talking about the challenge of finding an inpatient psychiatric bed for a child in crisis, or the difficulty in ensuring that maternal care access is available to women independent of their socioeconomic and medical needs. They set the stage not only for operationally what’s happening, but also why it’s important, how many women die in childbirth, or how many days a child can wait for an inpatient bed.
And then—I call it flying monkeys—we throw anything out there that might help solve this in a way that people coming from a traditional clinical background wouldn’t think of. And usually the flying monkey ideas are terrible, but they trigger two things. One, “Oh, that won’t work because I forgot to tell you about this other piece of the puzzle.” So it’s a way of eliciting information. And the other is better ideas, like we can’t actually use flying monkeys to deliver pharmaceuticals to rural communities, but we could use drones, right?
Those kinds of connections, I find so valuable.
Where in healthcare do we need more systems thinking?
I want to use a medicine analogy. When you’re trained as a physician, you learn how to treat diseases and you also learn how to diagnose. And one of the things that’s so challenging and why you need to spend time learning to diagnose is because the human body is a system and a human being also lives in an environmental ecosystem. So I can’t just understand how your heart works, I have to understand how your heart interacts with your lungs and your kidneys and your liver, and how treating your heart is going to impact on your liver. Whether you are in medicine, public health, engineering, we’re taught to solve problems. We’re taught to treat the disease, but we’re not taught to diagnose.
Where I view the biggest challenge and opportunity right now is in diagnosing our health care system. Take the example of overcrowding in emergency departments, which is a problem across the country. You might jump in and try to solve it by making your emergency departments bigger, hiring more ER docs. But in many cases what we really need are more inpatient beds or better discharge processes because many ER patients are waiting to be admitted.
It starts with how we scope problems and thinking about this sort of input, throughput, output. What happens before the problem? What happens during the problem? What happens after it? …(I)t’s helpful to begin not by trying to fix things, but by understanding the domain.
What’s your vision for workplace well-being and how can operational engineering contribute?
In the more concrete realm, there are a couple of things that are cornerstones for me. One is to standardize absolutely everything that you should—and not one bit more.
I see opportunities to take the low risk, low variability, low importance stuff off of people’s plates. I don’t know how many times I’ve come out of a full day of meetings and I’m just going delete, delete, delete, delete, delete to all the junk emails and I accidentally delete something really important. A [better] mechanism to prioritize those messages might occasionally fail, but that already happens. On the other hand, if you could focus only on the 10 messages that only you can do, that makes a big difference.
We have to find this balance of allowing people autonomy and control over their work, but clearing out the noise so that the work that I do is work that really only I can and should be doing.
To read the full transcript and listen to the 40-minute podcast, visit Operational improvement through problem solving and efficiency from Michigan Medicine, published in February 2025.