Visiting a doctor's office after a patient with flu-like illness appears to be a risk for getting sick, according to University of Minnesota research that could have implications for infection control and the use of online visits for primary care.

Researchers from the U, Harvard University and athenahealth, reviewed electronic medical records for 6,709 U.S. primary care clinics in 2016 and 2017 and found that people who visited their clinics the same day after patients with flu-like illnesses were more likely to return within two weeks with their own respiratory ailments.

The risk is only slight, as the rate of post-visit illnesses was 2.9 per 1,000 in people who visited clinics the same day after flu patients, compared with 2.2 per 1,000 in patients who visited clinics before flu patients. However, lead author Hannah Neprash said the nearly 32% difference in the rates was significant and provided some of the first documented evidence that clinics can be infection sources for patients — not just hospitals.

"There's a lot of attention that gets paid to the fact you can go into a hospital and get something infectious you hadn't bargained on," said Neprash, a U assistant professor of health policy and management. "There is just no research on whether a similar thing happens in physician offices. And of course, to the extent that any of us consume medical care, the most common thing to do is go to a doctor's office."

Doctor visits in the study occurred well before the declaration in early 2020 of the COVID-19 pandemic, but Neprash said lessons learned in the pandemic could likely address the causes of infections in clinics. Mask-wearing in clinics could be one solution along with the continued use of telehealth services for basic primary care when appropriate.

"Maybe some of it should stick around even after we emerge from the pandemic," Neprash said.

The study didn't offer evidence of how flu-like illnesses might have spread from one patient to another.

"It's hard to say exactly what the mechanism of transmission is — that patients are all touching the same surface or someone sneezes in the waiting room," Neprash said.

The study presented challenges, including making sure in the medical records that they were identifying patients with flu-like illnesses as well as patients who arrived before and after them.

Patients with flu-like illnesses were identified if doctors ordered influenza medications or flu tests for them. Patients with appointments at least 90 minutes before flu patients were labeled as unexposed, while patients who arrived at the same time or later on the same day were considered at risk for exposure.

The researchers conducted the same analysis with back pain and urinary tract infections and found no similar patterns, lending credence to their findings regarding flu-like illnesses.

Jeremy Olson • 612-673-7744