Frustrated and desperate, Genesee County health official Jim Henry did not mince words as he demanded information from the city of Flint on a 2014 outbreak of Legionnaires’ disease that had sickened 45 people in Michigan, killing five of them.

In his March 2015 email, Henry noted his previous efforts, which included a Freedom of Information Act request, and warned that another outbreak could be coming as the warm summer months approached.

“The increase of the illnesses closely corresponds with the time frame of the switch to Flint River water,” he wrote. “The majority of the cases reside or have an association with the city. . . . This is rather glaring information and it needs to be looked into now.”

Yet a year later, despite a second outbreak and a total of 87 illnesses and nine deaths, no government agency has tested the water supply for the legionella bacteria that cause the infection, which flourished as the beleaguered city’s tap water was being poisoned by lead.

Without a scientifically proven match between the bacteria in the water and strains cultured from victims, it is impossible to determine whether the tainted water supply caused the deadly infections, officials have told the public.

That could complicate a special counsel’s efforts to assess criminal culpability for the fatalities and at least one lawsuit seeking damages — as well as efforts to protect the public from future outbreaks, experts said.

The county health department, two state agencies, the Environmental Protection Agency and the Centers for Disease Control and Prevention have offered a variety of reasons why Flint’s water has not been tested, at times pointing fingers at one another. A state health department spokeswoman noted that chlorine is being added to the water to kill bacteria, including legionella.

But three experts in the control of Legionnaires’ disease expressed varying degrees of surprise and dismay that testing still has not been done.

“It seems likely that the public water supply contributed to the outbreak of Legionnaires’ disease in Flint,” said Ruth L. Berkelman, a professor of epidemiology at the Rollins School of Public Health at Emory University. “The public water supply was at risk for multiplication of legionella, and the timing of the outbreak is unlikely to be coincidental.”

Brian Shelton, owner of PathCon Industries, a Georgia-based company that specializes in preventing Legionnaires’ disease, said sites throughout the city — including private homes — should be tested for the bacteria.

“This is a preventable disease. It’s a disease that’s transmitted from the environment. It’s not transmitted person to person. So the only way to prevent transmission is to know what’s going on in the environment,” Shelton said.

Take a look at the key moments that led up to Flint, a city of 90,000, getting stuck with contaminated water. (Claritza Jimenez/The Washington Post)

Flint’s water was contaminated with toxic lead soon after the city switched to the Flint River as its source in April 2014 and a Michigan state department failed to ensure that anti-corrosive chemicals were added to the supply. That allowed lead to leach from aging pipes into the tap water of the city’s 95,000 people, many of whom are still living on bottled water.

The magnitude of the crisis has overshadowed the outbreaks of Legionnaires’ disease, which have resulted in the only deaths so far. The same corrosion encouraged the growth of legionella bacteria, which flourish in warm water that contains flakes of iron — a nutrient for the bacteria — from aging pipes.

The bacteria pose no threat in drinking water. But when water vapor is inhaled or contaminated water is used during invasive medical procedures, bacteria enter the lungs, where they can cause a virulent pneumonia. Cooling towers, showers, fountains and hot tubs are common sources for the spread of bacteria to people.

Genesee County’s 87 cases came in two waves that crested as the weather warmed between June 2014 and November 2015.

As they interviewed the victims, public health authorities quickly realized that more than half had been treated on an inpatient or outpatient basis at one of the city’s two main hospitals. Some people also received Flint water at home, complicating efforts to determine where they might have been exposed. The bacteria’s incubation period is about 10 days.

Tim Monahan, 58, a carpenter who has lived in Flint for a decade, said he began to feel sick toward the end of June 2014, about two months after the city switched to the Flint River for its water. At the Hurley Medical Center emergency room, his fever registered 104.6 degrees on July 5. He also was having problems with his lungs.

Doctors were initially puzzled by his symptoms. But soon enough, they gave him a surprising diagnosis: Legionnaires’ disease.

He lost 20 pounds but recovered after nine days in the hospital. Like others, he suspects that he was sickened by the contaminated water supply and has been frustrated by officials’ explanation of why it is difficult to link the water with his illness.

“ ‘You can’t prove it.’ That’s what they keep saying,” Monahan said. “You can’t prove it because you didn’t do your jobs, people.”

One group of victims, which includes the family of a woman who died, has sued another hospital, McLaren Flint, over the Legionnaires’ cases, but Monahan is not part of any lawsuit. He thinks he contracted the infection before he went to Hurley and sees no point in suing the government.

“The truth is, I survived,” he said. “There are nine people who didn’t.”

McLaren Flint has since disinfected its water system. Hospital spokeswoman Laurie Prochazka declined to answer questions but said in a statement that “we want to assure our patients, visitors, employees, volunteers and community that the hospital’s water is safe.”

Special counsel Todd Flood, appointed in January by Gov. Rick Snyder (R), told reporters last month that he would consider charges — including involuntary manslaughter — as he conducts his investigation. But two criminal law experts said that assigning responsibility for a death would be impossible without the ability to prove the water caused it.

“If the prosecutor cannot prove causation beyond a reasonable doubt, charges are not appropriate,” said David Uhlmann, a law professor at the University of Michigan and former chief of the Justice Department’s environmental crimes section.

Flood’s spokeswoman, Andrea Bitely, said in an email that she could not comment during the probe, but “every lead will be investigated.”

Janet Stout, president of the Special Pathogens Laboratory, a consulting firm in Pittsburgh, thinks it might still be possible to connect legionella bacteria taken from victims to legionella in the water. A state lab cultured bacteria from eight victims — too few to show that city water was generally responsible for the outbreaks. But each one could be matched with bacteria found in a home, even after nearly two years, Stout said. Once it takes hold in a water heater, Stout said, legionella can remain indefinitely unless the system is disinfected.

For now, it appears that the only water testing was conducted by Stout at McLaren Flint, and by the Virginia Tech University laboratory that revealed high levels of lead in the water last year.

The team found no alarming levels of legionella in private homes last August, said Amy Pruden, a professor of civil and environmental engineering at the university. But when they took more samples from large buildings, including two hospitals, in October, the researchers found as much as 1,000 times the normal level of legionella bacteria, she said.

Pruden agreed that it’s “very difficult” for someone sickened by legionella to prove definitively where and when he acquired the disease. “Ideally, the CDC should have been called in to do culturing of the patients and the water; then you match the DNA and have a possible source,” she said. That didn’t happen, she added, “which is unfortunate.”

Alice Crites contributed to this report.