Each week, Daniels is featuring a researcher who conducts meaningful research that impacts their field and the wider community. Learn more about their work in Q&As with Melissa Archpru Akaka, associate dean for faculty research. Email Melissa to nominate yourself or a colleague for a future Q&A. 

Zlatana Nenova

Zlatana Nenova is an assistant professor in the Department of Business Information and Analytics at the Daniels College of Business. Her primary interests are in the areas of operations management, data mining and health care analytics. In her work, she leverages big data to develop empirical models, which can be used to improve the management for patients with chronic diseases. She was a Lee B. Lusted Student Poster Competition finalist at the 2016 Medical Decision Making Annual North American Meeting, and her work has been published in the Journal of Operations Management, Production and Operations Management, and Palliative Medicine. 

What are you interested in studying? Have your interests evolved over time?

I’ve always been interested in health care analytics. I use available data to answer questions that practitioners and health care professionals want to tackle, especially at the patient level. I want to answer the question of how to improve care for individual patients rather than, for example, how do I optimize how a hospital is run?

Over time, I’ve become more focused on chronic conditions. I am currently really interested in the substance-use disorder field. What I have noticed from working in this field is that most facilities do not collect data on their patients past discharge. Thus, you can’t know how they’re faring outside of the facility. So that’s what I’m working on now, is trying to find a way to collect data on clients after they have been discharged.

What methods for data collection do you use in your research?

We’re currently using a platform called Prolific which allows you to filter which individuals receive a survey based on specific treatments that they have or haven’t received. We will split our survey into parts so that we can first collect background information, and then follow up with individuals immediately to ask questions about their experiences with different treatments. Finally, we will follow up with a third survey to ensure we’re not tiring anyone out with too many questions at once. That last survey is geared more toward building a narrative of the client’s overall experience with substance use disorder treatment facilities. We will use text analytics tools to gain even better insights from those written narratives.

What do you see as the impact of your research? How do you measure that impact?

Doctors of course have recommendations on how a patient should be treated. As a statistician, I want to see what’s behind their choices. Is there some correlation between doing X type of therapy and people having better outcomes? I see the most impactful outcome of my work as being able to quantify what actually works for patients with substance use disorder—treatment modalities, environmental exposures, etc.

In addition to all of the background and treatment questions, we also want to ask people, did anyone follow up with you after discharge? The answer is going to be very important for policymakers. If, for instance, funding is going to facilities that do not follow up with their patients at all, and we can see that follow-up is important for patient outcomes, maybe facilities should be properly incentivized to reach out to recently discharged patients.

We hope this work will show at least a handful of things that do make a difference in patient outcomes.