It seems like a chicken-or-egg scenario, but the question of which comes first—diabetes or depression—may finally have an answer: Both.

In a new study, people who displayed depressive symptoms were 42 percent more prone to developing type 2 diabetes over five years than less depressed people. And 54 percent of people being treated for type 2 diabetes had depressive symptoms over that time.

The researchers followed a large multiethnic group of people aged between 45 and 84. At each of three successive meetings over the five-year study, the participants’ fasting glucose was taken to determine if they had diabetes. At the first and last meetings, they filled out questionnaires to gauge their behavior and mood.

“People who had symptoms of depression were less physically active, ate more, and were more likely to smoke … clearly risk factors for developing type 2 diabetes,” says the study’s lead author, Sherita Hill Golden, MD, MHS, an associate professor of medicine and epidemiology at Johns Hopkins University.

Previous studies had established the “depression-to-diabetes” link, but this study clarifies that the converse is also true: Diabetes can lead to depression. “We think it is the illness burden associated with treating diabetes” that increases the risk of depression, says Golden.

One of the surprising findings was that those with untreated diabetes—people who may not have known they had diabetes until their fasting glucose was measured as part of the study—were 20 percent less likely to experience depressive symptoms.

Golden emphasizes that this isn’t a reason to forgo treatment for diabetes, but thinks it provides further evidence that dealing with the condition can bring people down. “If you have treated diabetes, you know you have the diagnosis,” she says. “That could contribute to symptoms of stress, which could lead to symptoms of depression.” She hopes to explore this association in future studies on the causes and alleviation of the illness burden.

This study was published in the June 18, 2008, issue of the Journal of the American Medical Association.