People who developed the most common form of diabetes tended to take more antibiotics in the years before their diagnosis than counterparts who didn’t have the condition, Danish researchers have found.

Using records from three national health registries in Denmark, researchers tracked antibiotic prescriptions for 170,504 people who had Type 2 diabetes and compared them with 1.3 million people who did not.

The people with Type 2 diabetes used more antibiotics, going as far back as 15 years before their diagnosis. What’s more, that association existed across all 16 groups of antibiotics the researchers tracked. (They looked at drugs most commonly prescribed by Danish doctors.) People with diabetes filled 0.8 prescriptions a year, on average, compared to 0.5 prescriptions a year among the healthy counterparts, according to a team of researchers, including those at Gentofte Hospital and the University of Southern Denmark.

The study was published Thursday in the Journal of Clinical Endocrinology & Metabolism.

Past studies have shown that antibiotics can change the bacteria in a person’s gut, and other research has suggested that antibiotics could affect the body’s ability to metabolize sugar and fat.

In the United States, diabetes is at epidemic proportions in the United States because of sugary diets and lack of exercise. If current disease rates continue, one in three American adults could have diabetes by 2050.

So do these results show that increased use of antibiotics increases the risk of developing diabetes? Or do the results show a greater demand for antibiotics from people with increased risk of infection because of yet-undiagnosed diabetes?

Turns out, it’s both.

“The study does not tell us which interpretation is the right one,” said Kristian Hallundbaek Mikkelson, a doctor at the Center for Diabetes Research at Gentofte Hospital, and one of the study’s authors. “Both interpretations are possible and both are supported by other research.”

More research is needed, he said, with randomized studies to determine whether there is a direct cause and effect between increased antibiotic use and the development of diabetes.

Read more: