Digging Deeper: Legionnaires' timeline at Kent Hospital

With cases of hospital-acquired Legionnaires' disease, the Rhode Island Department of Health said Wednesday it would notify the public of an outbreak, but doesn't consider what happened at Kent Hospital as an outbreak.

In May, an elderly woman with multiple health issues was admitted to Kent Hospital in Warwick and tested negative for legionella, the bacterium that causes Legionnaires' disease.

Ten days later, while in the hospital, she then tested positive. The Health Department was notified immediately of the woman's exposure.

She died four days later of an undetermined cause. The woman's name hasn't been released.

"There's a fine line between alerting the public when there's a threat and then saying there's a threat when there isn't one," said Dr. James McDonald of the Health Department.

Hospital spokesman Jim Beardsworth said the bacteria were found in four of the hospital's 24 ice machines, but there's no evidence that tainted ice caused the problem.

"We know where the positive results were, in four ice machines. But we also know we cannot directly link that to the patient," he said.

Legionnaires' disease is a rare form of pneumonia that's contracted when people breathe in tiny droplets of contaminated water.

McDonald said there was another possible case at Kent Hospital.

"Someone had been in the hospital, was discharged, was home for two days, and came in the hospital. So it's considered a possible case, not a confirmed one," he said.

The Health Department said the patient did not meet the Centers for Disease Control definition for hospital-acquired infection as the patient was not in the hospital for 10 continuous nights prior to developing legionella.

The Health Department also said the CDC only recognizes positive cases and not potential cases. It said that Kent Hospital can be commended for identifying the confirmed case right away and working proactively on corrective actions.

McDonald said the Health Department doesn't have a hard number to call a case or cases an outbreak.

"It depends on the situation when we would label it an outbreak," he told NBC 10. "I'd like to give you a more specific answer. It's something that, we look at every case and make our best judgment as to when to alert the public."

There have been 32 cases in Rhode Island this year.