Digital tool aims to cut a country’s overuse of antibiotics : NPR

A mother and her child at a health center in Rwanda. The child’s condition was assessed by a nurse using the new ePOCT+ digital tablet (located at the corner of the table), aiming to reduce the overuse of antibiotics.

Magali Rochat

hide caption

toggle caption

Magali Rochat

On a typical day at a clinic in rural Rwanda, nurses may see 60 patients per day. Adults and children line up with injuries, coughs and fevers, often after traveling many miles.

“Nurses are very busy, they receive everything that comes from the community, complicated or easy,” explains Dr. Victor Pacifique Rwandarwacu, a Rwandan doctor. This leaves little time for diagnosis.

When faced with a patient suffering from an illness, many nurses choose to prescribe something. Often these are antibiotics. “We see them giving a lot of antibiotics, just in case,” says Rwandarwacu. “They were like, ‘OK, what if I don’t give it, and the patient comes back tonight?'”

This dynamic has led to extremely high prescription rates, according to new search by Rwandarwacu and colleagues. In 32 clinics in Rwanda, 71% of pediatric visits ended with an antibiotic prescription. This is probably much, much higher than necessary.

For a given patient, receiving antibiotics, for example, for a viral case of pneumonia, may not be a big deal, even if the antibiotics do nothing to stop the virus. But within a region or country, such high rates of unnecessary prescriptions can racial resistance.

“In sub-Saharan Africa, the rise in antimicrobial resistance is enormous,” explains Jean Claude Semuto Ngabonziza, a researcher at the Rwanda Biomedical Center who also participated in the study. “We are about to lose potential antibiotics.”

But Ngabonziza, Rwandarwacu and their colleagues have developed a new tool that could be useful.

Their tablet-based tool, called ePOCT+, guides clinicians step-by-step through diagnosing a problem and suggesting treatment. After these same 32 clinics implemented ePOCT+, antibiotic prescribing rates dropped by 71% to 25%report the researchers in PLOS Medicine. This decline has not made more children sicker.

“The most important thing is that they don’t compromise health outcomes,” says Dr. Sumanth Gandrainfectious disease researcher at Washington University in St. Louis, who was not involved in the study. “I think these tools will be really useful and can be expanded.”

A digital “second opinion”

In theory, nurses working in rural Rwanda and other parts of sub-Saharan Africa have access to clinical guidelines that provide best practices for diagnosing and treating patients. But in practice, this knowledge is often inaccessible or difficult to apply to the rural context.

“Nurses would have a lot of books on the table, so they have to go from guideline to guideline to decide each patient’s case,” says Rwandarwacu.

The researchers wanted to distill all of this information into an easy-to-use digital tool that could run on a tablet. They created ePOCT+, an algorithm that summarizes these guidelines into step-by-step instructions for assessing various symptoms. It also incorporates some simple diagnostic tests, including oxygen and hemoglobin levels, which can give clues to the underlying problem. Based on the responses, the algorithm would guide clinicians to a diagnosis and offer a treatment suggestion.

Suppose a patient arrives with suspected pneumonia. A nurse examined the patient, tablet in hand, and entered symptom information. On average, the whole process takes about 10 minutes, explains Alexandra Kulinkinaepidemiologist at the Swiss Institute of Tropical and Public Health and co-author of the study. “At the end of the day, it’s either simple pneumonia, bacterial pneumonia, or viral pneumonia,” she says. Only bacterial pneumonia would be treated with antibiotics.

Training clinic staff on how to use ePOCT+ didn’t take long, Kulinkina says. “One day of practice was enough,” she says. Researchers tracked the impact of the tool on nearly 60,000 visits. Antibiotic prescriptions have dropped significantly, to just 25% of visits.

This is probably even more than necessary, Kulinkina says. “Under ideal conditions, we would expect 10 to 15 percent of patients to walk away with a prescription,” she says. This is partly because nurses did not always follow directions. Still, it’s a “dramatic reduction,” she says.

The tool helped nurses detect other problems that often go unnoticed, Rwandarwacu says, including malnutrition and anemia. “It was a surprising effect.”

Overall, nurses and patients were generally positive about the experience, Ngabonziza says. “One mother told me, ‘It’s exceptional, I felt more evaluated and examined than during previous consultations because the clinician always checked,'” he says. The nurses also appreciated the extra support, he says, although they ended up spending more time with each patient, he says.

The Rwandan government also took note of the results.

The Ministry of Health is currently developing an electronic medical records system for the entire country. Researchers have spoken with officials about the possibility of integrating a version of ePOCT+, Kulinkina says. “The Ministry of Health is of course interested in improving the quality of care for patients.”

But she notes that it is Rwanda’s Social Security Board, which funds the public health insurance system, that has attracted the most interest. Fewer antibiotic prescriptions could save them a lot of money over time, she says.

Scaling such a tool would have challenges, including training thousands of healthcare workers. And it might not work as well as it did in this study. But trying is worth it, both for patients and to stem the rise of antibiotic resistance, says Jean Claude Semuto Ngabonziza, a researcher at the Rwanda Biomedical Center and co-author of the study.

“We are on the verge of losing potential antibiotics,” he says. “Sometimes we do research and it’s so-so, but it really has an impact.”