The Numbers That Should Have Broken the Internet
By mid-2025, the CDC had confirmed 70 human cases of H5N1 in the United States. Let that sink in. A single case in 2022. Seventy in one year. I started making calls the afternoon the agency updated its case count, the way I used to when the police scanner picked up a code I didn’t immediately recognize. My contacts at the health department seemed tired. Not worried-tired. Defeated-tired. The kind of tired you hear in someone’s voice when they know they’ve been shouting into a room nobody’s paying attention to.
The math itself was explosive. Thirty-five of those confirmed infections came from dairy farm workers in California alone. California. The state with the most robust agricultural surveillance system in the nation. The state where infection clusters should have been caught early and contained. Instead, they multiplied. I kept thinking about the workers I’d never interviewed, dairy employees whose names would never appear in the evening broadcast, who went to work because they needed the paycheck and went home because they had no choice.
The Louisiana Case That Changed Everything, Quietly
In January 2025, the CDC confirmed the first human death from H5N1 in the United States. A Louisiana patient. Someone with exposure to both backyard poultry and wild birds. Read that twice. Not a healthcare worker. Not someone in a lab. Someone who moved through a landscape we all recognize: a backyard, birds, the ordinary intersection of human life and animals.
I called three epidemiologists before I published anything about this. I needed to understand what the data was actually telling us versus what we were projecting onto it. That’s the job: source discipline. The death happened. The CDC confirmed it. But what hadn’t been confirmed yet was equally important. Was this an outlier? A harbinger? The question wasn’t being asked loudly enough in the places where policy gets made.
The Dairy Herd Picture Nobody Connected
Here’s where source criticism becomes essential reporting. The USDA APHIS H5N1 Dairy Herd Detection Map showed the scale clearly: seventeen states. Over 1,200 individual dairy herds since the outbreak began in March 2024. This wasn’t a California problem. It wasn’t even a West Coast problem. It was a national agricultural infrastructure problem that had been quietly spreading for eighteen months.
What stuck with me, the way certain details do when you’ve been reporting long enough to recognize a pattern, was how the information traveled. The USDA released the data. The CDC absorbed it into its situational reports. Agricultural trade publications covered it. General assignment reporters? Mostly silent. The story didn’t live in any single newsroom’s beat anymore. It fell through the cracks between agriculture coverage and public health coverage, like something nobody quite owned.
The Science That Should Have Triggered Alarms
In October 2025, researchers at Scripps Research Institute published findings in Nature that made the containment picture much darker. They’d identified evidence of limited human-to-human transmission in two California family clusters. Not theoretical evidence. Not lab models. Actual documented transmission between people. The WHO elevated its pandemic risk assessment following this publication, which should have been a front-page story everywhere. Should have been.
I spoke with three virologists about what this meant in practical terms. The answer was consistent: it meant the virus was learning. It meant we were watching a real-time shift from an occupational hazard to something with pandemic potential. The data suggested we hadn’t crossed into a pandemic yet, but we were walking toward it through a familiar epidemiological landscape. We’d been here before with other flu strains. The warning signs looked exactly like this.
The Policy Response That Raises Questions
Congress allocated $306 million in emergency supplemental funding for H5N1 vaccine stockpiling in December 2025. That’s real money. That’s action. But here’s where the source story gets complicated. HHS Secretary Robert F. Kennedy Jr. simultaneously declined to endorse the existing CSL Seqirus mRNA vaccine candidate in trials. Call it what it is: a mixed signal from the highest level of health policy.
I made calls about this too. What I found was institutional confusion masquerading as debate. You had funding allocated for a vaccine stockpile strategy without agreement on which vaccine should fill it. You had emergency appropriations moving through Congress on the same timeline as regulatory hesitation from the agency designed to oversee the response. The data was being created and collected. The policy response looked less like a symphony and more like three different bands playing in the same room.
What Reporting Requires Now
The H5N1 story of 2025 wasn’t ignored because people didn’t care. It was fragmented because it lived in too many places at once. Dairy farmers saw it through their industry publications. Public health people saw it through CDC reports. Families in affected areas saw it through local news. But nobody was stitching it together into a narrative that made clear what the stakes actually were: seventy cases, one death, hundreds of herds, limited human transmission documented, and a vaccine policy that looked uncertain even as money was being allocated.
That’s what source criticism means to me now, after thirty years of making phone calls and checking facts before publication. It’s not just about verifying numbers. It’s about understanding how information moves, where it gets stuck, what stories get told and which ones vanish into the gap between agencies and beat structures. The CDC data was there. The USDA data was there. CDC H5N1 Bird Flu Current Situation Summary kept getting updated. But the story as a whole? That got lost.
If you’ve been following H5N1 tracking your own way, through whatever combination of government reports and medical journals and agricultural updates you can access, I want to hear about it. What story are you seeing that the mainstream coverage missed? Where do you think the next transmission cluster is most likely to emerge? Send me your observations. That’s how this kind of reporting works now. Not top-down from official sources alone, but built together from people who are paying attention and willing to connect the dots.
