What is known so far about the human cases of bird flu in the U.S. is enough to worry those who spend their professional days tracking and monitoring the spread of viruses, particularly strains with the ability to spread widely.
But at a time when public dissemination of data and details about new cases is critical, another worrying development has become trackable: the erratic drips of information coming from the federal agency responsible for such matters.
Late last Friday, the Centers for Disease Control and Prevention(CDC) made public that a second healthcare worker in Missouri who had cared for a patient hospitalized with H5N1developed respiratory symptoms of their own, a fact that had not been presented during the most recent H5N1 press briefing. The CDC said that Missouri health officials didn’t know about the case until it was too late to test the worker for bird flu—and it was unclear whether or how long the agency had known about it before the briefing.
It was the second time in two weeks that a significant development in the case was announced at the tail end of the week, near the close of business. On Sept. 13, the agency belatedly disclosed that a household contact of the patient with bird flu had become sick on the same day with similar symptoms as the individual, but was not tested for the virus. In addition, a third contact–a healthcare worker with exposure to the patient–showed mild symptoms but tested negative for flu.
“The household contact’s simultaneous development of symptoms should have been mentioned in the press briefing, along with the additional context, to fully highlight all available information about the case and to further demonstrate why CDC has not changed its risk assessment,” a CDC spokesperson told me when asked about the initial delay in providing the information. The agency said the immediate risk of bird flu to the general public remains low.
Since the federal agency began issuing weekly and “noteworthy” bird flu updates in late July, nearly every release has come on a Friday, often later in the day.
‘Not looking hard enough for the virus’
Despite these revelations, the CDC said that there is “no epidemiological evidence at this time to support person-to-person transmission of H5N1, though public health authorities continue to explore how the H5N1-positive individual in Missouri contracted the virus.” But to Rick Bright, the eminent American immunologist and vaccine researcher, the problem with the CDC’s statements lies in what’s missing.
“It’s very difficult to support the CDC’s statement—or even a risk assessment—without critically missing data that can only be analyzed from ongoing serological surveillance(antibody testing) of those in close or direct contact with infected animals and environmental exposure, and their close contacts,” says Bright.
Bright adds that there is no epidemiological evidence to support a claim about human-to-human transmission either way. “The CDC and state health departments are simply not looking hard enough for the virus or its impact,” the former public health official says.
Without a doubt, the agency’s perspective has varied. On the Sep. 12 press call, Nirav Shah, principal deputy director of the CDC, said, “None of the individuals that this individual came into contact with have developed any signs and symptoms. So we haven’t seen any evidence of it at this time…We are right now of the view that this is a one-off.” The following day, the CDC disclosed two other symptomatic contacts.
The Missouri individual was hospitalized on Aug. 22, but the CDC and the Missouri Health Department did not disclose the bird flu case until Sep. 6, after the patient had been hospitalized, discharged, and recovered. Shah explained the delay in part by saying that this case of H5 was detected through the country’s national flu surveillance system, which is different than the systems used in clinical medicine. Here, positive flu samples are batched up by the hospital and sent to the state labs to scan for anything in the sample set that might be novel. This is performed on an occasional basis, not as frequently in the summer as in the winter, he said, “so we don’t view that as a delay.”
Certainly, none of this equals a pandemic. The H5N1-positive patient in Missouri was officially only the 14th confirmed human case in the U.S. this year, and the previous 13 were all traced to close contact with infected poultry or cows.
But for a nearly limitless number of reasons, quick and full disclosure by the CDC—not to mention coordination and cooperation with state and local agencies—looms large in this equation. And so far, experts say, the returns on those fronts are not encouraging.
“Obviously, that type of information release pattern raises questions and is not ideal,” says Amesh Adalja, a senior scholar at the Johns Hopkins Center for Health Security, speaking of the situation in Missouri. “Without prompt and complete information, no risk assessment can be made. States need to be proactive with bird flu in cattle and humans, not reactive and evasive.”
Adalja, who is also an associate editor of the journal Health Security, was referring to a key component of this equation. Though the CDC has been belatedly adding facts to the H5N1 story, it’s not clear whether the agency is receiving timely communication from state or local administrators—in this case, officials in Missouri. I have reached out to the Missouri Department of Health and Senior Services and the CDC with several questions but did not receive an immediate reply.
This all matters tremendously. The Missouri patient represents the first known case of a human bird flu infection that doesn’t connect directly or indirectly to exposure to sick farm animals, wild birds, or other wildlife prior to the illness. The individual also reported no exposure to unpasteurized milk or dairy products.
To date, no H5N1 infection has been reported in dairy cows in Missouri—but testing in that state is not required. (The Missouri Department of Agriculture wrote in an email that just 84 out of a total of about 60,000 dairy cattle have been tested for H5N1. Testing on farms, they state, is completely up to dairy owners.) The origin of the patient’s infection is unknown, at least to the public, and the incidence of other people in close contact showing symptoms of their own cries out for more information and background.
The CDC has said that Missouri health officials, who are leading the investigation, collected blood samples from the H5N1-positive individual and the household close contact for serological testing, which could reveal antibodies that confirm a previous bird flu infection. The federal agency will test the samples. Serologic testing will also be offered to the second health worker.
But the CDC lacks the authority to go much further. As with other states and local agencies, only Missouri officials can ask for more widescale testing of workers, or for testing of the dairy or poultry farms themselves at which H5N1 infection has been detected.
Agency statistics show that since late March the CDC has monitored roughly 5,000 people as a result of their exposure to H5N1-infected or potentially infected animals, and tested fewer than 250 people who developed flu-like symptoms. Testing occurs at the state and local level and the CDC does confirmatory testing.
At the more than 230 dairy farms with infected herds, it’s not clear how many exposed workers have undergone testing, according to an article published this week in Nature. “Veterinarians visiting H5N1-infected dairy farms anecdotally reported suspected human cases that never received testing, including workers with and without direct contact with cattle,” the authors wrote. “Picking up rare transmission chains requires intensive contact tracing among workers, family members, and other contacts.” So far, that isn’t happening at any scale.
Nationally, public health labs have run fewer than 50,000 specimens that could pick up influenza A (H5) since late February. Without more extensive testing, experts say it is likely that we are missing human cases.
We don’t know what we don’t know
“Overall, I think that there are likely more response efforts going on than are publicly communicated,” says Stanford infectious disease physician and post-doctoral researcher Abraar Karan, who has worked on both state- and county-level responses in the past. “This is fine, but the issue is that from what has been communicated, it seems that the CDC cannot go in and assist in testing or response and tracing without invitation. I think we are missing some cases for sure.”
As Scott Hensley, a viral immunologist at the University of Pennsylvania Perelman School of Medicine in Philadelphia says, “The fear is that the virus is spreading within the community at low levels, and this is the first time we’re detecting it. There’s no data to suggest that to be the case, but that’s the fear.”
One way to know more, albeit belatedly, is via wastewater testing. Between March and mid-July, for example, researchers at the University of Texas Health Science Center detected H5N1 in the wastewater of all 10 cities they tested, a result described as “troubling” by the authors of a letter to the New England Journal of Medicine. (The CDC also monitors H5 virus in wastewater at more than 250 sites in over 40 states.)
“We are seriously alarmed for the possibility of human-to-human transmission,” Anthony Maresso, a co-author of that letter and a professor of molecular virology and microbiology at Baylor College of Medicine, told me.
Maresso says the researchers’ sequencing of the Texas wastewater samples suggests that possible H5 sources are dairy cows, cats, and birds. However, the researchers cannot yet determine whether human infection or transmission is occurring. They’re picking up mutations that they can’t explain. “Humans represent a massive natural biomass to take advantage of and infect,” says Maresso. “If this virus can learn to infect and transmit between us, it will. And when it does, the world must be prepared.”
The CDC’s Shah told reporters that “nothing in the area or the region (in Missouri) is suggestive of increased rates of influenza,” whether measured by emergency or urgent care visits, lab tests, or respiratory tests. Rick Bright counters that “these types of passive surveillance approaches will miss many cases in the community until there is sufficient ongoing transmission and increased severity in infections to start sending people to the hospital, or worse.”
In the confusing labyrinth of local, state, and federal responses to a bird flu outbreak that has already infected more than 100 million poultry, over 10,000 wild birds, and 238 dairy herds (in 14 states) across the U.S., Shah’s agency stands at the top. Its leadership likely will determine H5N1’s course, especially if, as many researchers fear, new strains will make human infection more likely.
“The very best thing we can do is to not downplay its seriousness,” says Maresso. Among other things, the researcher advocates health agencies building up antiviral reserves, getting an updated vaccine prototype in the pipeline (for both humans and animals), and using next-generation testing and sequencing of environmental and clinical samples “to get ahead of (the virus’s) secret activities before it’s too late.”
In the end, perhaps only the CDC, through forceful persuasion, can kickstart those vital processes in states that may otherwise want to avoid the bad news. “There needs to be more aggressive testing of contacts, both acutely and in convalescence,” Amesh Adalja says. “This dripping out of information from Missouri is not acceptable, and the state should ask for official CDC assistance now.”
More must-read commentary published by Fortune:
The opinions expressed in Fortune.com commentary pieces are solely the views of their authors and do not necessarily reflect the opinions and beliefs of Fortune.
