Smarter follow-up and safer care: how AI could support men with diabetes
June is Men’s Health Month, an opportunity to focus on health challenges that often go unnoticed, especially among men living with chronic conditions like diabetes. At the University of Alabama at Birmingham (UAB), Forge AHEAD investigator Seung-Yup Lee, Ph.D., is leading a pilot study that combines artificial intelligence with clinical data to prevent medication mix-ups and help patients get the follow-up they need.
Why men with diabetes need a smarter safety net
Men are more likely to delay routine visits and less likely to follow up after hospital care. both of which can lead to missed medications, dangerous side effects and avoidable complications. For those living with diabetes, these risks are even higher. Managing medications correctly is critical to controlling blood sugar, avoiding hospital readmissions and staying healthy.
But for busy clinics, it’s not always clear who needs the most help. That’s where Lee’s project steps in.
Using AI to spot who needs help first
Lee’s team is developing an AI-powered risk score to help doctors and pharmacists spot patients who are most likely to have medication problems. The system analyzes a wide range of health data, including medical records, prescription refill patterns, and social factors like insurance status or housing instability, to assign a “reconciliation risk score” to each patient with diabetes.
Seung-Yup Lee, Ph.D.
Assistant Professor, University of Alabama at Birmingham
Did You Know?
Men are more likely than women to skip routine checkups, and more likely to face serious medication-related problems as a result.
The goal? Use this score to flag high-risk patients so care teams can follow up directly, either by phone or in person, to double-check medications, fix errors and offer support.
What the tool actually does, and how AI fits in
Artificial intelligence, or AI, refers to computer systems that are trained to notice patterns in large sets of information, kind of like how a person might learn from experience, but much faster. For example, just like a nurse might notice that certain symptoms usually come before a problem, AI can spot those same signs by studying thousands of patient records at once. In this project, AI is used to scan medical records, pharmacy data and provider notes to predict which patients might be most at risk for medication problems.
It’s important to know that AI doesn’t replace doctors, nurses or pharmacists. It doesn’t make decisions on its own or replace face-to-face care. Instead, it helps teams work smarter by pointing out which patients might need extra attention. That gives healthcare providers more time to focus on what people need most, support, questions answered and care that feels personal.
- Data-driven prediction: The model looks at over 5,000 patient records and uses natural-language processing (a type of AI that reads doctors’ notes) to detect warning signs.
- Real-time alerts: Risk scores are displayed inside a clinician dashboard so that care teams know who needs outreach before the next visit.
- Focused follow-up: Patients with high scores receive extra attention, including calls from pharmacists to reconcile prescriptions and catch any problems early.
Measuring what matters: fewer errors, better follow-up
The pilot study includes around 200 patients and is testing whether the tool helps reduce medication discrepancies within 30 days. It’s also tracking whether fewer patients return to the hospital with drug-related issues, and how satisfied doctors and nurses are with using the system.
This type of focused support may be especially helpful for men, who often manage more complicated medication routines and may delay reaching out for help when issues arise.
Key Terms to Know
- Medication reconciliation: The process of double-checking all a patient’s medications to make sure they match what was prescribed.
- Risk stratification: Grouping patients by risk level to focus care where it’s needed most.
- Health-related social factors: Things like income, housing or transportation that can affect someone’s ability to stay healthy.
What is AI?
Artificial intelligence (AI) refers to computer systems that can learn from large sets of data and help identify patterns.
In health care, AI is used to support, not replace, human decision-making by helping teams spot risks earlier and prioritize follow-up care.
Source: National Institutes of Health
What’s next for the project
If successful, Lee’s team plans to expand the tool across partner sites in Alabama, Mississippi and Louisiana. They’re also exploring ways to pull in data from state prescription-monitoring systems and bring the program into more community health clinics, including those that serve men with limited access to care.
How this could help you or someone you know
This project reflects a growing effort to use data to guide follow-up care and reduce preventable health problems. For men with diabetes, it could mean fewer hospital visits, safer medication use and better outcomes over time.
