Albert Burgess-Hull, PhD

Data Scientist

Albert Burgess-Hull, PhD is an addiction researcher and data scientist who serves as MATClinics data scientist. He earned a PhD from the University of Wisconsin-Madison and completed his postdoctoral fellowship training at the intramural research branch of the National Institute on Drug Abuse (NIDA) at the National Institutes of Health (NIH). Dr. Burgess-Hull has been recognized nationally for his research and leadership and has held training fellowships at both Duke University and Yale University.

At MATClinics, Dr. Burgess-Hull conducts research focused on the development, validation, and implementation of analytic frameworks designed to improve medical decision-making, clinic/staff efficiency, and the delivery of addiction treatments. He actively collaborates with addiction researchers at NIDA, the NIH, and various universities and health departments around the U.S. to disseminate this research to other addiction scientists.  

Q: How would you describe MATClinics’ proprietary analytics?

A: We have developed algorithms that our clinicians use to identify patients who may need additional help during their treatment, including identifying potential roadblocks they may encounter on their path to recovery. Many of our proprietary analytics use machine learning to construct models that can make predictions on future data points – models that can often outperform traditional statistical methods.

Q: How does the use of the proprietary analytics benefit patients?

A: Essentially, our models enable us to place patients into different categories of stability based on how far along they are on their path to recovery. This allows our clinical staff to allocate additional resources to patients who may need more help. And because we continually update MATClinics’ adherence model (which predicts whether a patient is taking their medication and abstaining from drugs) we’re able to identify important changes in a patient’s health status and take steps to prevent negative outcomes from occurring.

Q: If you were a MATClinics patient, would you be more confident in your recovery knowing how they use analytics?

A: Absolutely, yes. If I was struggling with opioid or alcohol addiction, I would want to receive treatment at a clinic that was using all available resources to make my recovery more attainable. I would also want exceptionally smart and creative people who were thinking outside the box and trying to harness the huge amounts of data we now have at our disposal to improve my care. At MATClinics, we have put together a great team that is really pushing the limits of how predictive analytics can be used in real-world addiction treatment settings.  

Q: How new is the use of analytics in addiction treatment?

A: The use of analytics is not new in addiction treatment. As one example, the composite international diagnostic interview, or CIDI, was developed by the World Health Organization years ago to assess substance-use dependence using statistical methods. A variation of this diagnostic tool is still used in practice today. However, a lot of the addiction research using advanced statistical methods never find their way to real-world settings, which is why MATClinics’ use of proprietary analytics is so important. We are deploying real-time predictive and explanatory analytic methods in a real-world addiction clinic. In doing so, we’re focused on translating cutting-edge research results to real-world tools that can help patients.

Q: Looking forward, what are other things you hope MATClinics’ proprietary analytics will do for patients?

A: I believe we can develop even better predictive tools that can be used to match the right treatment to the right patient at the right time. To that end, we’re testing an algorithm for optimizing the buprenorphine dosage that a patient receives, which will make life a lot easier for our patients, especially those who are struggling with craving. Also, I’m hoping we will use data to identify ways we can deliver the same standard of care with minimal disruption to our patient’s everyday lives.

Q: You and MATClinics Medical Director Enrique Oviedo, MD are amongst the co-authors of an article recently published by the Journal of Addiction Medicine about predictive modeling. How important is it that your findings are shared throughout the industry?

A: The article describes the development and validation of MATClinics’ prediction model, which forecasts opioid use and medication adherence at a patient’s next appointment. Each day in our clinics, we use this model to generate an adherence score for each patient, which we believe has improved the quality of care that we are delivering. Because we believe that our proprietary analytics have the potential to help not just our patients, but all patients, we strongly believe that our work should be disseminated to the broader addiction community.

Q: Finally, as a data scientist, what draws you to addiction treatment?

A: In graduate school and during my postdoctoral training at the National Institutes of Health, I became very interested in how artificial intelligence could be used to identify the groups of individuals who might make the transition to more problematic patterns of drug use. I really believe that the use of AI, along with novel data collection methods like smartphones and smartwatches, can revolutionize how we treat and identify problematic patterns of drug use in real-world settings, which is why I joined MATClinics.  

Begin Your Recovery Today

MATClinics currently accepts all Maryland Medicaid programs as well as Medicare Part B and all CareFirst (or BlueCross Blue Shield) plans. Learn more about what to expect and a list of the insurance programs accepted.