Assessing electronic health record potential for adaptive learning in multimorbidity care in Sub-Saharan Africa: a mixed-methods study of Zimbabwe's Impilo system

Imagine a hospital as a bustling city where every patient is a traveler with a unique map. In the past, doctors kept these maps on paper, but as technology advanced, many cities switched to digital GPS systems called Electronic Health Records (EHRs). The idea is simple: instead of carrying a stack of crumpled papers, a doctor can click a button and see a traveler's entire history instantly. This is especially important for people living with "multimorbidity," a fancy word for having two or more long-term health conditions at once, like having both HIV and high blood pressure. Think of it like a traveler who needs to visit both the bakery and the mechanic on the same day; if the GPS only shows the route to the bakery and forgets the mechanic, the traveler gets lost.

The big question scientists are asking is: Do these digital maps actually help doctors learn how to take better care of these complex travelers? Or do the maps just show the wrong roads? This paper dives into that question by looking at a specific digital system in Zimbabwe called "Impilo." The researchers wanted to see if this system helps doctors figure out how to treat patients with multiple diseases at the same time, or if it just adds more confusion. They aren't just checking if the computer turns on; they are checking if the computer helps the whole city learn and adapt to help its people better.

---

The Story of the Digital Map That Forgot the Whole Person

In Zimbabwe, the government built a massive digital health record system called Impilo. Think of Impilo as a giant, national library where every patient's medical story is supposed to be stored. The dream was that this library would be a "Learning Health System"—a place where the library doesn't just sit there, but actually reads the stories, figures out patterns, and tells the doctors, "Hey, we noticed that patients with both HIV and high blood pressure need to be treated together, not separately!"

The researchers, a team of curious detectives, went to two clinics in Zimbabwe to see if this dream was real. They watched patients, talked to doctors, and looked at the computer screens. They were looking for a specific kind of magic: adaptive learning. This means the system should help doctors adjust their care based on what they learn from previous visits, creating a smoother, smarter path for the patient.

The Big Discovery: The System is a "Single-Story" Library

The team found that while Impilo is a powerful tool for tracking one specific disease (like HIV), it struggles to tell the whole story of a patient with multiple conditions.

Imagine a patient, let's call her Sarah. Sarah has HIV and high blood pressure. When she walks into the clinic:

1. She goes to the HIV desk. The doctor opens the computer and sees Sarah's HIV history. It's perfect!
2. Then, Sarah goes to the blood pressure desk. The doctor opens a different part of the computer (or sometimes a paper notebook) to see her blood pressure history.
3. The Problem: The computer doesn't automatically show the doctor that Sarah is at the blood pressure desk because she also has HIV. The two stories are in separate rooms. The system doesn't say, "Hey, look at Sarah's whole life!"

The researchers found that doctors are incredibly smart and hardworking. They use their own brains, memory, and paper notes to connect the dots. They act like human bridges, carrying information from the HIV room to the blood pressure room. But the computer itself? It's not helping much. It's like having a GPS that gives you perfect directions to the bakery but forgets to tell you that you also need to go to the mechanic, leaving you to figure out the second part of the trip on your own.

What the System Gets Wrong (and Right)

The paper suggests that the main issue isn't that the doctors are bad or that there isn't enough data. The problem is how the data is organized.

• The "Vertical" Trap: The system was built to track diseases one by one (vertically), like a stack of separate folders. It's great for counting how many people have HIV, but it's bad at showing how HIV and blood pressure interact in one person.
• The "Single-Loop" Learning: The doctors are learning, but only in a small way. They learn how to fix a problem today (Single-Loop). For example, "Sarah's blood pressure is high, let's give her more medicine." But the system doesn't help them learn how to change the system for tomorrow (Double-Loop). It doesn't help them realize, "Wait, maybe we should stop treating these diseases in separate rooms and start treating them together."
• The Paper Crutch: Even though they have a fancy computer, the doctors still rely heavily on paper notebooks and patient-held cards. Why? Because the computer doesn't show the full picture. It's like having a smartphone that can't connect to the internet, so you have to use a paper map instead.

The "Workaround" Reality

The study found that the doctors are constantly doing "workarounds." This is a fancy way of saying they are inventing their own solutions because the official tools aren't working. They might ask a patient, "Do you have high blood pressure too?" even though the computer should already know. They might write notes on a sticky pad and stick it to the screen. They are doing the job of the computer, manually stitching the broken pieces of the patient's story together.

What the Paper Says About the Future

The researchers suggest that the solution isn't to throw away the computer or build a new one from scratch. Instead, they propose co-production. This means the people who actually use the system—the doctors, nurses, and patients—should be the ones helping to design the next version.

Think of it like designing a video game. If you ask the gamers what they need, they will tell you, "I need a map that shows all my quests at once, not just one!" The paper suggests that if the designers of Impilo sit down with the doctors and ask, "How do you actually take care of a patient with two diseases?", they can build a system that actually helps.

The Bottom Line

The paper concludes that Zimbabwe's Impilo system is a great start, but it's not yet the "Learning Health System" it was promised to be. It collects data, but it doesn't yet turn that data into wisdom for the whole system. The doctors are doing the heavy lifting of learning and adapting, but the computer is just sitting there, waiting to be told what to do.

The authors suggest that the path forward is to stop treating the computer as just a digital filing cabinet and start treating it as a partner in learning. By involving the frontline workers in the design process, they hope to build a system that doesn't just record history, but helps write a better future for patients living with multiple health challenges. It's a reminder that technology is only as good as the way it fits into the messy, complex, and beautiful reality of human care.