Menu Bar

Home           Calendar           Topics          Just Charlestown          About Us

Wednesday, October 21, 2020

Inside the Fall of the CDC

Inside the Fall of the CDC
How the world’s greatest public health organization was brought to its knees by a virus, the president and the capitulation of its own leaders, causing damage that could last much longer than the coronavirus.
By James Bandler, Patricia Callahan, Sebastian Rotella and Kirsten Berg for ProPublica

SHORT ATTENTION-SPAN WARNING: This is a long and detailed article. For those only able to absorb Tweet-length content, this may be too much for you.  - Will Collette

At 7:47 a.m. on the Sunday of Memorial Day weekend, Dr. Jay Butler pounded out a grim email to colleagues at the Centers for Disease Control and Prevention in Atlanta.

Butler, then the head of the agency’s coronavirus response, and his team had been trying to craft guidance to help Americans return safely to worship amid worries that two of its greatest comforts — the chanting of prayers and singing of hymns — could launch a deadly virus into the air with each breath.

The week before, the CDC had published its investigation of an outbreak at an Arkansas church that had resulted in four deaths. The agency’s scientific journal recently had detailed a superspreader event in which 52 of the 61 singers at a 2½-hour choir practice developed COVID-19. Two died.

Butler, an infectious disease specialist with more than three decades of experience, seemed the ideal person to lead the effort. Trained as one of the CDC’s elite disease detectives, he’d helped the FBI investigate the anthrax attacks, and he’d led the distribution of vaccines during the H1N1 flu pandemic when demand far outstripped supply.

But days earlier, Butler and his team had suddenly found themselves on President Donald Trump’s front burner when the president began publicly agitating for churches to reopen. That Thursday, Trump had announced that the CDC would release safety guidelines for them “very soon.” He accused Democratic governors of disrespecting churches, and deemed houses of worship “essential services.”

Butler’s team rushed to finalize the guidance for churches, synagogues and mosques that Trump’s aides had shelved in April after battling the CDC over the language. In reviewing a raft of last-minute edits from the White House, Butler’s team rejected those that conflicted with CDC research, including a worrisome suggestion to delete a line that urged congregations to “consider suspending or at least decreasing” the use of choirs.

On Friday, Trump’s aides called the CDC repeatedly about the guidance, according to emails. “Why is it not up?” they demanded until it was posted on the CDC website that afternoon.

The next day, a furious call came from the office of the vice president: The White House suggestions were not optional. The CDC’s failure to use them was insubordinate, according to emails at the time.

Fifteen minutes later, one of Butler’s deputies had the agency’s text replaced with the White House version, the emails show. The danger of singing wasn’t mentioned.

Early that Sunday morning, as Americans across the country prepared excitedly to return to houses of worship, Butler, a churchgoer himself, poured his anguish and anger into an email to a few colleagues.

“I am very troubled on this Sunday morning that there will be people who will get sick and perhaps die because of what we were forced to do,” he wrote.

When the next history of the CDC is written, 2020 will emerge as perhaps the darkest chapter in its 74 years, rivaled only by its involvement in the infamous Tuskegee experiment, in which federal doctors withheld medicine from poor Black men with syphilis, then tracked their descent into blindness, insanity and death.

With more than 216,000 people dead this year, most Americans know the low points of the current chapter already. A vaunted agency that was once the global gold standard of public health has, with breathtaking speed, become a target of anger, scorn and even pity.

How could an agency that eradicated smallpox globally and wiped out polio in the United States have fallen so far?

ProPublica obtained hundreds of emails and other internal government documents and interviewed more than 30 CDC employees, contractors and Trump administration officials who witnessed or were involved in key moments of the crisis. Although news organizations around the world have chronicled the CDC’s stumbles in real time, ProPublica’s reporting affords the most comprehensive inside look at the escalating tensions, paranoia and pained discussions that unfolded behind the walls of CDC’s Atlanta headquarters. And it sheds new light on the botched COVID-19 tests, the unprecedented political interference in public health policy, and the capitulations of some of the world’s top public health leaders.

Senior CDC staff describe waging battles that are as much about protecting science from the White House as protecting the public from COVID-19. It is a war that they have, more often than not, lost.

Employees spoke openly about their “hill to die on” — the political interference that would prompt them to leave. Yet again and again, they surrendered and did as they were told. It wasn’t just worries over paying mortgages or forfeiting the prestige of the job. Many feared that if they left and spoke out, the White House would stop consulting the CDC at all, and would push through even more dangerous policies.

To some veteran scientists, this acquiescence was the real sign that the CDC had lost its way. One scientist swore repeatedly in an interview and said, “The cowardice and the caving are disgusting to me.”

Collectively, the interviews and documents show an insular, rigorous agency colliding head-on with an administration desperate to preserve the impression that it had the pandemic under control.

Some of the key wounds were self-inflicted. Records obtained by ProPublica detail for the first time the cataclysmic chain of mistakes and disputes inside the CDC labs making the first U.S. test for COVID-19. A respected lab scientist made a fateful decision to use a process that risked contamination, saw signs of trouble, but sent the tests to public health labs anyway. Many of those tests didn’t work, and the scramble to fix them had serious consequences.

Even when the CDC was not to blame, the Trump administration exploited events to take control of the agency’s messaging. As a historically lethal pandemic raged, the White House turned the CDC into a political bludgeon to advance Trump’s agenda, alternately blocking the agency’s leaders from using their quarantine powers or forcing them to assert those powers over the objections of CDC scientists.

Once seen as an apolitical bulwark, the CDC endured meddling on multiple fronts by officials with little or no public health experience, from Trump’s daughter Ivanka to Stephen Miller, the architect of the president’s immigration crackdown. A shifting and mysterious cast of political aides and private contractors — what one scientist described as young protégés of Trump’s son-in-law, Jared Kushner, “wearing blue suits with red ties and beards” — crowded into important meetings about key policy decisions.

Agency insiders lost faith that CDC director Dr. Robert Redfield, a Trump appointee who’d been at the agency only two years, would, or could, hold the line on science. One division leader refused to sign what he viewed as an ill-conceived and xenophobic Trump administration order. Redfield ultimately signed it himself.

Veteran CDC specialists with global reputations were marginalized, silenced or reassigned — often for simply doing what had always been their job. Some of the agency’s most revered scientists vanished from public view after speaking candidly about the virus.

The Trump administration is “appropriating a public enterprise and making it into an agent of propaganda for a political regime,” one CDC scientist said in an interview as events unfolded. “It’s mind-boggling in the totality of ambition to so deeply undermine what’s so vitally important to the public.”

The CDC repeatedly declined to make Butler, Redfield or any other employees mentioned in this story available for questions, and a CDC spokesperson declined to comment on behalf of the agency. The White House did not respond to an email seeking comment.

A spokesperson for the Department of Health and Human Services, which oversees the CDC, rejected accusations of political interference.

“Under President Trump, HHS has always provided public health information based on sound science,” the HHS spokesperson said. “Throughout the COVID-19 response, science and data have driven the decisions at HHS.”

People interviewed for this story asked to remain anonymous because they feared retaliation against themselves or their agency.

In interviews and internal correspondence, CDC employees recounted the stunning fall of the agency many of them had spent their careers building. Some had served on the front lines of the CDC’s most storied battles and had an earned confidence that they could swoop in and save the world from the latest plague, whether it was E. coli on a fast-food burger or Ebola in a distant land. Theirs was the model other nations copied. Their leaders were the public faces Americans turned to for the unvarnished truth. They’d served happily under Democrats and Republicans.

Now, 10 months into the crisis, many fear the CDC has lost the most important currency of public health: trust, the confidence in experts that persuades people to wear masks for the public good, to refrain from close-packed gatherings, to take a vaccine.

Dr. Martin Cetron, the agency’s veteran director of global migration and quarantine, coined a phrase years ago for what can happen when people lose confidence in the government and denial and falsehoods spread faster than disease. He called it the “bankruptcy of trust.” He’d seen it during the Ebola outbreak in Liberia in 2014, when soldiers cordoned off the frightened and angry residents of the West Point neighborhood in Monrovia, the capital. Control of a pandemic depended not just on technical expertise, he told colleagues then, but on faith in public institutions.

Today, some CDC veterans worry that it could take a generation or longer to regain that trust.

“Most of us who saw this could be retired or dead by the time that’s fully fixed,” one CDC official said.

Dr. Anne Schuchat, the CDC’s top career scientist, was one of the first to notice a brief report about four cases of “unexplained pneumonia” in Wuhan, China, in an emerging diseases bulletin. It followed a warning about a “red blotch disease” in the grape industry.

As a disease detective in 2003, Schuchat had been dispatched to China to investigate the outbreak of SARS, a respiratory disease that killed about 800 people and shut down parts of Asia. Her role in that outbreak and in later epidemics inspired the virus hunter played by Kate Winslet in the movie “Contagion.” Unflappable and regarded as brilliant, Schuchat eases the tension at meetings by singing ditties about the latest outbreak set to Broadway tunes. Nobody wants to disappoint her.

At 8:25 a.m. on Dec. 31, Schuchat emailed Butler and other colleagues asking if “any of your folks know more about the ‘unknown pneumonia’” in Wuhan.

Emails and calls bounced among the agency’s leaders, a handful of veterans with more than a century of experience among them. Dr. Dan Jernigan, the flu chief, and his boss, Dr. Nancy Messonnier, met at headquarters to plan. Within hours, they learned there were 27 cases — seven of them severe — with fever, difficulty breathing and a buildup of abnormal substances in the lungs. All the cases were believed to be connected to an outdoor seafood market. “Raises concern about SARS,” Messonnier wrote in an email.

The news reached Cetron in New Hampshire. While celebrating the holidays at a beer-and-tacos pub across the river in Vermont, he told family and friends about a new virus in China that he worried could affect the whole world. “We should be bracing ourselves,” he said.

If the outbreak had been a movie, this would have been the scene where the heroine mobilizes an all-star squad of specialists to save the planet. Schuchat’s team is seen as among the top infectious disease experts in the world. All of them had started out in the CDC’s Epidemic Intelligence Service, an elite corps of globetrotting disease fighters. They were a brain trust forged by decades of defending the country from outbreaks.

But in the 11 years since the H1N1 flu pandemic, the terrain had shifted. Politics and budget cuts had weakened the agency at home and abroad. Meanwhile, the regime in Beijing had grown increasingly aggressive and authoritarian. The Trump administration’s trade war had worsened tensions. And after a series of tough-minded leaders who were adept at protecting the agency and its mission, Trump’s first choice as director quit after Politico reported that she had purchased tobacco stocks while leading the CDC, which fights lung diseases.

Trump appointed Redfield in 2018. He was an HIV researcher who had treated AIDS patients since the earliest days of the disease. He’d wanted the CDC job for decades, and had been passed over for it twice. During his first all-hands meeting at the Atlanta campus, he’d choked up describing the honor of leading the agency.

In the fierce chaos of Trump’s Washington, the CDC needed a streetfighter. Instead, it got “the nicest grandfather you can imagine,” a senior health official said. A former colleague described how Redfield, a devout Catholic, prayed with the ailing Elijah Cummings, a Democratic congressman from Baltimore, during a visit to the Capitol.

Redfield took over an agency that, despite its $8.3 billion budget, was feeling the chronic funding woes of the American public health system, which has been quietly gutted since the Great Recession. As the coronavirus began its march through the United States, years of federal and state cuts had left about 26,000 fewer employees at state, county and municipal health agencies since 2009, according to the nonpartisan Trust for America’s Health.

With a mission of protecting America from diseases, the CDC was stretched thin. Over the decades, its portfolio had expanded to include almost every malady, chronic or acute.

The CDC’s global presence was suffering too. An infusion of hundreds of millions of dollars at the time of the Ebola epidemic in 2014 allowed the agency to increase its presence to as many as 65 countries, but a large chunk of those funds ran out in 2019. As funding expanded and contracted in recent years, the CDC had to cut over 300 posts overseas, including both Americans and foreigners. By the time Schuchat noticed the blurb about an outbreak in Wuhan, her agency no longer had an office inside the Chinese Center for Disease Control and Prevention, its counterpart in Beijing. While the U.S. agency once had more than a dozen Americans in China, by January only three remained.

On Jan. 3, Redfield phoned his agency’s closest ally in Beijing, George Gao, the director of China’s CDC, a microbiologist trained at Oxford and Harvard. Gao said his agency had sent a field investigation team to Wuhan. But during conversations in the next few days, many of Redfield’s questions about the mystery disease went unanswered. Gao, who was usually open and talkative, sounded guarded, according to several officials familiar with the conversations.

Nevertheless, Redfield assured federal health and national security officials that information was flowing from China thanks to his rapport with Gao, knowledgeable people said.

On Jan. 6, Redfield sent Gao a carefully worded letter offering the help of CDC experts. Expecting the Chinese to accept “very soon,” CDC leaders began preparing a team to go to China, emails show.

To Redfield’s chagrin, however, the conversations with Gao came to a sudden halt. Ominous news accumulated: the first recorded death, Jan. 9, the first case outside China, Jan. 13. In the secure, high-tech room where the CDC brain trust met, the mood turned dark as the scientists began to fear they were confronting a pandemic.

“We were slowly convincing folks: It doesn’t matter if you believe it or not, but this is the circumstantial evidence,” a senior lab official said. “And you have to prepare.”

Amid the scramble to find out what was happening in China, CDC officials began telling the public not to panic. But they conveyed the serious nature of the threat.

On Jan. 17, for example, Messonnier said that the CDC was “especially concerned about a novel coronavirus” because related viruses — SARS and Middle East Respiratory Syndrome — were “difficult outbreaks with many people getting ill and deaths.”

It appeared that the illness had been spreading since at least early December, but data on cases provided by Chinese authorities was woefully incomplete, listing only the dates patients were hospitalized, not what symptoms they had or for how long, the senior lab official said.

“We knew they were good enough epidemiologists to get that data,” the official said. “Why aren’t they announcing the results?”

The lab official tried to contact a chief virologist at the China CDC who was usually helpful, but got no response. Neither did colleagues who reached out to Chinese scientists with whom they had collaborated for years. The Americans concluded that the regime in Beijing was telling them to keep quiet.

Gao had also run up against a cover-up by authorities in Wuhan, health and national security officials said. Gao’s field investigators were “told there was no evidence of human-to-human transmission,” said Dr. Ray Yip, a former country director for the CDC in China. “They didn’t show them all the cases. They had a couple of cases of hospital workers infected by then, and that’s obviously human-to-human, how else did they get it?”

During the SARS epidemic in 2003, Time magazine reported that Chinese authorities had hidden 31 infected health workers from the world by pulling them from their hospital, loading them into ambulances, and driving them around Beijing until a visiting delegation from the World Health Organization left the hospital.

In January of 2020, the bond between the U.S. and Chinese health agencies became a double-edged sword. Chinese leaders were wary about Gao’s relationship with the Americans, who heard rumblings that he would be made the scapegoat for the outbreak. Meanwhile, Redfield’s reputation suffered in Washington because he didn’t deliver.

“The China CDC and the U.S. CDC were almost seen as one,” a senior U.S. health official said. “Dr. Redfield contributed to this by talking about how much he talked to Dr. Gao, the information exchange they had going. There was a sentiment blaming Dr. Redfield for the inability to get more information.”

In reality, the blame went beyond Redfield and his agency. China was a hard target. Even U.S. spy agencies struggled to gather intelligence on the evolution of the disease. Still, at the moment of truth, the CDC’s decades of investment in building a network in China did not pay off. That failure created an early and significant schism between the agency and the Trump administration.

“What the fuck are we paying for people to be in China if they can’t go where there’s an outbreak when there’s an outbreak,” Joe Grogan, then the head of the White House’s Domestic Policy Council, recalled saying repeatedly at the time.

Deputy National Security Advisor Matthew Pottinger was another influential critic of the CDC and one of the first senior White House officials to realize the magnitude of the coronavirus threat. Pottinger had served as a Marine intelligence officer and worked in China as a correspondent for The Wall Street Journal. His coverage of the SARS pandemic had helped shape his view of China as what he called “an expansionist totalitarian empire.”

Pottinger clashed with CDC officials when he pushed to limit travel from China. Many of the agency’s scientists held the traditional public health view that border closures interfere with the movement of medical personnel and goods. On Jan. 31, Trump issued an order restricting most foreigners from entering the United States if they had been in China within the 14 days before their arrival.

The CDC deployed personnel to airports to screen incoming passengers for symptoms, a measure that leaders now admit was futile, given the high number of asymptomatic cases. (Of the 754,124 travelers screened at U.S. airports by mid-September, only 24 cases of COVID were confirmed, according to CDC records.)

The CDC had gone from being the world’s finest disease SWAT team to batting back claims from the administration that it was doing a lousy job.

Another blow came on Feb. 25, after an ill-fated press conference about the steps Americans might need to take to protect themselves. Leading that briefing was Messonnier, the no-nonsense director of the CDC’s powerful immunization and respiratory diseases center, who’d come to prominence during the 2001 anthrax attacks.

Asked by the media team to add a personal touch, Messonnier said she’d told her children they needed to prepare for a significant disruption of their lives and had called their school to ask about plans for online learning. Afterward, she left to take her children to the dentist.

But her words had rocked Wall Street and the White House. Soon the staff in the Atlanta Emergency Operations Center saw a news alert with a photo of Messonnier pop up on their phones. A CDC veteran remembers thinking: “Oh, crap, the stock market dropped!”

The market’s fall infuriated the president. Trump had privately confessed to author Bob Woodward that he was publicly downplaying the virus to prevent panic. The CDC would pay the price for undercutting that narrative.

The next day, Trump put Vice President Mike Pence in charge of his coronavirus task force and assumed the role of communicator-in-chief. The CDC, which had been the public face of the government during every health crisis in memory, soon became nearly invisible. After a few more briefings, a Pence aide told the agency’s media staff that this was the president’s stage, not theirs.

Even when Redfield was allowed to speak publicly, his sleepy eyes and soft, droning tone anesthetized listeners. The agency had been effectively muzzled.

“When it mattered the most, they shut us up,” a senior CDC official said. “The threat is clear. If we want to ever be able to talk tomorrow or next week or next month — or whatever is being dangled in front of us, you stay inside the lines.”

A friend of one CDC scientist ribbed him: “We keep waiting for the CDC to show up on a milk carton as a missing child.”

In the months that followed, CDC scientists watching the president’s news conferences on a wall of screens in the agency’s Emergency Operations Center were dumbfounded as Trump countermanded science in a flurry of inaccuracies and dangerous advice, saying the virus would soon go away, theorizing about injecting disinfectant as a treatment, and dismissing recommendations about wearing a mask.

As the agency stumbled in China and at home, a group of lab scientists was assigned a high-stakes mission: developing a test for the coronavirus.

Inside a small lab on the CDC’s Atlanta campus, microbiologist Stephen Lindstrom was put in charge. A Saskatchewan native who speaks at a breakneck clip, Lindstrom had studied in Tokyo and defended his Ph.D. dissertation in Japanese. During the H1N1 flu pandemic, his team had invented a test, jumped through regulatory hurdles and shipped it around the world in just two weeks’ time.

“Frankly, he kind of lives for the pressure,” said one of his colleagues.

But this time around, just about everything that could go wrong did. Calculated decisions went sideways, and Lindstrom couldn’t find a quick way to right them. Mystifying contamination appeared at every turn, relegating tests to the trash heap. Precious weeks were lost.

The CDC declined to make Lindstrom available for questions. But lab records obtained by ProPublica and interviews reveal for the first time the mounting pressure and the cascading troubles inside the lab.

As soon as Lindstrom’s team received the genetic sequence from scientists in China in January, they got to work. By the time German researchers on Jan. 13 announced the recipe for the test that would be adopted by the World Health Organization, Lindstrom’s team was almost done building its own.

Lindstrom had turned to the lab’s expert on coronaviruses to design the U.S. test. They chose one that looked for three targets on the same coronavirus gene. While the first and second targets were unique to the new virus spreading in China, the third would identify a broader family of coronaviruses, useful if the virus circulating in China mutated as it infected Americans.

Such a test works like this: Imagine three different pieces of velcro, each custom-made to stick to one of those three genetic targets. If any of them finds a perfect match in a patient’s sample, the test will cause that snippet of genetic material to duplicate over and over until there’s enough to light up a signal, alerting a technician that there is a positive test result.

To make sure the tests work properly, microbiologists prefer to validate a test using actual virus samples taken from people. Lindstrom didn’t have that, but he could use lab-made pieces of the virus to do the same thing. He also needed to make the velcro-like testing ingredients that find matches in patient samples.

Making both the testing ingredients and the snippets of the virus in the same location, though, goes against best practices. Even in world-class labs, manufacturing pieces of a virus can leave microscopic traces in the environment and on equipment for months. Those can later contaminate tests so that even water would give a positive result. That kind of false positive renders the tests useless.

Lindstrom’s lab didn’t have the equipment or expertise needed to make the raw materials for the test. But an underground corridor led to another CDC lab — the “core facility” — in a gleaming glass tower. Lindstrom had used it many times to quickly make testing materials. The facility could make what Lindstrom needed, but it was risky.

Hiring a private company to take on one of those tasks would add at least 10 days to production times, an eternity during an outbreak. So Lindstrom hedged his bets. He placed an order with a contractor for the genetic pieces he needed, but also asked the core facility to make those snippets along with the velcro-like ingredients.

“It’s a pretty dangerous procedure to make that in the same facility” due to contamination, said one CDC scientist. “Trying to fast-track it this way was risky.”

Years ago, low-level contamination ruined some CDC tests for Middle East Respiratory Syndrome, even though the core facility made the viral pieces on a different floor from the velcro-like ingredients, according to a person familiar with the matter.

Initially, it looked as if Lindstrom had made a good call. The core facility cranked out the parts needed for the tests and they passed quality checks, suggesting that making all of them in house wasn’t a problem. On Jan. 20, his lab was able to identify the first positive U.S. case. Still, Lindstrom showed a rare flash of anxiety, telling colleagues: “This is going to either make me or break me.”

Soon specimens were pouring in. At that point, Lindstrom’s lab was the only one in the country able to test samples to confirm whether patients had COVID-19. At the same time, his team was racing to get authorization from the Food and Drug Administration for test kits that could be distributed to state and local public health labs. Exhausted CDC scientists arrived at 7 a.m. and left after 11 p.m.

With that authorization in the works, Lindstrom asked the core facility to begin mass-producing the ingredients that stick to the three genetic targets in a human sample. Then Lindstrom made a second risky decision. He had his team produce the stand-in for the virus that labs would use to check that a positive sample would trigger a positive result, lab records show.

The ingredients made by Lindstrom’s lab and the core facility passed the quality checks, records show, so Lindstrom sent them to another CDC lab to process and put in vials for the test kits.

The first sign of trouble appeared on Feb. 3. Lindstrom’s team performed quality checks on two lots of tests. In one lot, the third target was showing up as present when testers were using only water — a false positive result. The other lot was fine, records show. Though the flawed lot was set aside, this was a red flag. Contamination can be difficult to eliminate once it occurs, and the batch that failed had gone through the same lab spaces as the one that passed. Nevertheless, Lindstrom released the good lot of tests to be sent to public health labs.

While those tests were in transit, his team performed one last round of quality checks. This time, one of the test kits that they believed was fine also came back with a false positive, records show. Confoundingly, the next day that same kit performed as it should when Lindstrom’s lab checked it, according to a lab record.

Complaints poured in as soon as the tests arrived at the public health labs. Before screening any samples from patients, scientists checked to ensure the tests worked, using water for a negative and the stand-in for the virus for a positive. They found the same problem with the third target: It registered as positive when just testing water.

“There is likely a widespread issue that will need to be addressed immediately,” a California public health official said in an email to the CDC on Feb. 8.

“Aw Shit!” Lindstrom muttered to his staff. His team rechecked bulk testing ingredients from that lot, and found no issues. Then they pulled a portion from the freezer that hadn’t been opened since they received it from the core facility. A few false positives turned up, records show. So Lindstrom’s lab ordered from the core facility a replacement for the ingredient that is supposed to stick to the third target. But he also had contractors make some too.

At first, it looked like the problem could be solved quickly. The core facility delivered test ingredients that passed quality checks on Feb. 11. But subsequent checks — after they had been put in vials again — showed problems, records show.

Lindstrom told colleagues he was convinced there was contamination, but some CDC leaders insisted that the problem was actually a faulty design akin to a software bug — that Lindstrom had chosen genetic sequences that could cause a glitch and show a false positive, according to emails and interviews. While they debated, public health labs with the faulty kits couldn’t process samples, and the FDA still hadn’t authorized any tests made by commercial labs. Instead of a network of labs around the country testing sick people, Lindstrom’s team remained one of the few that could do it, using kits they’d made before the problem arose.

The air was filled with tension. At one point, a manager on the CDC coronavirus response team banged on the door to Lindstrom’s lab and demanded test results from his staff rather than waiting for them to be entered in the agency’s database, according to a scientist who was present. During a meeting, Lindstrom yelled at his colleagues for going around him and browbeating his people, according to an official who was present.

When it seemed things couldn’t get any worse, they did. Public health labs began reporting on Feb. 12 that they also were having problems with the part of the test that was supposed to stick to the first target. Subsequent checks by Lindstrom’s lab found the same problem, records show. Lindstrom now had an issue with the ingredients that were supposed to match two of the three targets. And it wasn’t clear whether there was contamination in his lab, the core facility or the separate facility that put the material into vials. Two weeks after the first complaint, the CDC still didn’t have a solution.

The FDA’s head of lab diagnostics showed up to troubleshoot and found Lindstrom’s lab in disarray. The Wall Street Journal later reported that the FDA official’s boss told CDC leaders that if it had been any other lab, they would have shut it down.

Public health labs were clamoring for tests, and Lindstrom was running out of options. The replacement material that was supposed to stick to the third target was made incorrectly and had to be scrapped, records show. The test kits he had ordered from contractors hadn’t arrived yet.

It seemed like the virus’ fingerprints were everywhere. So when the core facility sent some test ingredients that passed quality checks, Lindstrom hired a contractor to put them in vials. Even those tests came back with problems, a lab record shows.

With the FDA’s blessing, Lindstrom cobbled together test ingredients from different batches that had all passed quality checks, and they dropped the troublesome third target.

By the end of February, three weeks after public health labs first reported problems, the CDC started to send new test kits.

In the aftermath, an investigation by HHS lawyers pointed to Lindstrom’s lab as a likely source of contamination and praised the core facility for following “extreme precautionary measures” that minimized risk. Lindstrom fumed to colleagues that the HHS report was inaccurate. He was adamant that evidence showed the contamination originated in the core facility, not his own lab, records show.

The CDC did its own review but never released it. Separately, the HHS inspector general has been investigating. And some CDC scientists remain convinced that the problem wasn’t contamination but faulty design.

Anger and mistrust caused by the shortage of tests fell on the CDC — even if the FDA shared the blame for sticking to a cumbersome regulatory process that delayed the rollout of more tests. The combination of delays and missteps by the nation’s two top health agencies put the United States dangerously behind in assessing the spread of the virus. In contrast, South Korean officials gave near instantaneous approval to commercial labs, and they quickly began testing 10,000 people a day.

In a written statement, FDA spokeswoman Lauren-Jei McCarthy said her agency “has demonstrated unprecedented regulatory flexibility in order to speed development and quickly authorize tests.” The FDA, she said, streamlined its process to allow “diagnostic tests to be developed, validated, and deployed within weeks rather than several months to over a year, as traditionally required.”

In July, the acting director of Lindstrom’s division summoned him. He was reassigned to a new job with no official title and few responsibilities.

The following month, a CDC journal published a study that showed that Lindstrom had not been the only one struggling with faulty tests. Commercial labs in Europe had similar problems that delayed testing in at least nine countries.

By then, though, the damage had been done. To the public and within the federal government, the CDC had failed catastrophically at a critical juncture.

As the virus hopscotched across the globe, cruise ships became early symbols of the pandemic. Overnight, they morphed from bastions of leisure into pariahs of the sea, floating hotspots crammed with tourists, sick and well.

The Diamond Princess quickly became the most infamous. During excruciating weeks in February, the disease ripped through the massive ship, infecting hundreds of passengers off the port of Yokohama, Japan. Relatives of those stranded on board pleaded with the U.S. government to evacuate them, likening the recirculated air to a gas chamber.

At the CDC, the dilemma of what to do with the ships and their passengers, many of them Americans, fell to Cetron, who had led the agency’s quarantine division for more than two decades.

Cetron, 61, bore his responsibilities with a grim knowledge of the past. The CDC doesn’t have much statutory authority. Its influence lies in the ability to coax the public into acting in the nation’s collective interest. But the agency has one formidable power: the ability to control border movement during an outbreak and deprive people of their freedom to protect the public’s health.

Cetron had talked openly about how that power had been used in the past as a weapon to stigmatize. His academic research partner, the medical historian Howard Markel, had written a book about the mistreatment of Jewish immigrants in New York during cholera and typhoid outbreaks in 1892. Even a group sent to help called them “human maggots.” Authorities shunted them off to a quarantine island where they endured squalor and isolation. Some died.

But with the coronavirus, the agency’s singular authority would be undercut, abused and politicized — and Cetron would be unable to stop it.

As the Diamond Princess languished, U.S. diplomats assured passengers that nobody with the virus would board the evacuation flights. However, after packing the American passengers on buses headed for chartered planes, officials learned that 14 had tested positive. The State Department pushed for all of the passengers — uninfected and infected — to fly out together, according to CDC officials who were involved in the discussions.

Schuchat and Butler objected. Dr. Robert Kadlec, the HHS official in charge and a former Air Force colonel, sided with the State Department. Kadlec told colleagues the priority was bringing Americans home. On one of the planes, the only thing separating the infected from the non-infected was a flimsy plastic sheet.

The Washington Post reported that Schuchat demanded the removal of all references to the CDC from the State Department press release about the repatriation.

CDC officials involved told ProPublica that they were appalled by both the decision and its sloppy execution. “There’s a four-foot gap at the top of the shower curtain that you bought from Home Depot — and you’re calling this a quarantine area?” one said. “If I were to write a book, it would be called Operation Clusterfuck, and it would start with this chapter.”

Spokespeople for the State Department and HHS said diplomats and federal health experts took stringent precautions on the evacuation flights.

“Individuals who tested positive were moved in the most expeditious and safe manner to a specialized containment area on the evacuation aircraft,” a State Department spokesperson said in a written response. He added, “All passengers were closely monitored by medical professionals throughout the flight and were provided masks for additional protection.”

Despite that very public ordeal, cruise lines kept packing more passengers on board and heading out to sea. Days after the Diamond Princess evacuation, a ship from the same company, the Grand Princess, set sail from San Francisco on another ill-fated voyage. On March 5, a military helicopter had to fly to the ship to deliver tests after passengers got sick.

The next day, with the Grand Princess floating off the coast of San Francisco, Trump flew to Atlanta for an impromptu tour of the CDC laboratories. Wearing a red “KEEP AMERICA GREAT” cap, Trump briefly praised the CDC’s tests as “perfect” and talked about the record high ratings for his recent appearance on Fox News. Asked by a reporter about cruise ships, the president said he preferred that the Grand Princess passengers remain on board because their arrival — even at a federal quarantine site — would cause a spike in U.S. case numbers.

“I don’t need to have the numbers double because of one ship,” Trump told reporters.

Cetron and his team mapped every cruise ship at sea with COVID patients, working feverishly to build support in the government for a no-sail order that could prevent more outbreaks. “These cruise ships are the equivalent of mass gatherings of hundreds if not thousands of the most vulnerable populations” at risk for severe illness or death from COVID, and any of these passengers could seed the virus in their communities when they returned home, he said in an email to Redfield.

The cruise industry resisted and put forth a plan that would allow companies to keep sailing with extra safety precautions. The day after Trump’s appearance in Atlanta, Pence and Redfield met in Florida with cruise executives. After Pence praised the industry’s “spirit of collaboration,” the chairman of the industry’s largest trade group said, “Given the significance of travel and tourism, it is critical that Americans keep traveling.”

Employees watching in the CDC’s command center in Atlanta let out an audible groan.

Cetron told colleagues in an email that the industry’s plan was inadequate, given the “sardine can density” of these ships, records show. Every day the federal government delayed shutting down this industry meant more illness and death. At a meeting in March, Cetron railed against the industry’s recalcitrance and his own government’s unwillingness to act, according to people who attended.

“This is unconscionable,” he told Schuchat and more than a dozen others around the conference table, his voice so anguished it alarmed some who were there.

Colleagues could see the toll the battle was taking on him. Raccoon-like rings deepened around his eyes. He looked like an unmade bed, often wearing the same shirt, pants and rumpled tweed jacket with elbow pads as the day before. At one point, the CDC’s chief of staff became so worried about Cetron’s health that he ordered him to surrender his phone to Butler, who answered the late-night calls. “Go home and get some sleep,” the chief of staff commanded, according to people who overheard the conversation.

When the CDC finally issued a 30-day no-sail order on March 14, it excluded the majority of cruise operators since their trade group, Cruise Lines International Association, voluntarily agreed the previous day to stop launching any new ships from U.S. ports during that time. The order praised the trade group’s actions, “and the commitment it demonstrates to protecting the health of both cruise ship passengers and the public at large.”

Outbreaks continued on ships that were already at sea. The trade group had drafted a plan to hire a global rescue team staffed by special-operations veterans who would extract infected passengers and take them to medical facilities contracted to care for them “without burden on the U.S. government,” records show. Yet by April 6, the group still hadn’t hired the rescue company, and public health authorities had to scramble to help evacuate critically ill people from ships, records show.

Cetron worked on a new no-sail order that exposed the industry’s failures and required cruise operators to care for the 79,800 crew members on ships in or near U.S. ports without further strain on public health workers, records show.

“Poor planning by the industry, failure to adhere to recommendations and unsafe transport operations used by ships to get passengers and crew home has posed significant risks to local, state, national and international spread of the virus,” Cetron told Redfield in an email. “Dozens of vessels are still at sea with active COVID infections on board,” he added, “heading toward US waters requesting arrival in our ports.”

Cetron told Redfield this tougher order was “urgently needed.” Yet, the Department of Homeland Security refused to sign off. Officials wrote that they disagreed with CDC’s “narrative describing the actions of the cruise line industry.”

After four days of wrangling, DHS agreed to keep the force of the order, but Cetron’s criticisms of the cruise industry were censored or softened. A section titled “Failure of Cruise Ship Industry to Develop and Implement a Response Plan” became “Critical Need for Further Cooperation and Response Planning.”

Representatives from Cruise Lines International Association did not return emails or a phone call.

In September, the CDC proposed extending the no-sail order into February 2021, but the White House Coronavirus Task Force instead sided with the cruise industry and picked an end date of Oct. 31.

At the same time as they were watering down Cetron’s criticism of the cruise industry, the White House and DHS were pushing him to invoke quarantine powers to stop a problem that barely existed: the spread of coronavirus by migrants trying to cross the U.S.-Mexico border.

Two days after the no-sail order in March, Trump’s senior adviser Stephen Miller scheduled a meeting to discuss “Emergency Border Planning.” Like Cetron’s ancestors, Miller’s great-great-grandfather escaped anti-Semitism in Eastern Europe and found refuge in the United States. But Miller was a driving force behind Trump’s so-called Muslim ban, as well as the family-separation policy and efforts to build a wall spanning the 1,954-mile U.S.-Mexico border.

In a call on March 17, Miller urged the administration to use the CDC’s powers to close the border immediately because “the Southern Border is in crisis and will get worse as COVID-19 spreads in Mexico,” according to an email from a deputy general counsel at HHS.

Shortly after 7 a.m. the next day, an HHS lawyer sent Cetron’s team a proposed CDC order that largely closed the borders with Mexico and Canada. A deputy of Cetron’s lamented to the agency’s chief of staff that the order cited a “misrepresented and incomplete piece of data” to overstate the threat.

“I’m also not a fan of trying to make the case that Canada and Mexico represent a big risk on the land border based on what we ‘believe’ is occurring vs. what we know about the # of cases (which are far fewer than the # of cases in the US now due to community spread),” she wrote.

Cetron refused to sign off on the order, according to people who worked with him. “I will not be a part of this,” a furious Cetron told a colleague. “It’s just morally wrong to use a public authority that has never, ever, ever been used this way. It’s to keep Hispanics out of the country. And it’s wrong.”

With Cetron engaged in a personal act of civil disobedience, Redfield signed the order.

For the first time since the enactment of the Refugee Act of 1980, people who came to the border saying they feared persecution or torture in their home countries were turned away with no chance to plead their case for asylum.

Ken Cuccinelli, a senior Homeland Security official, later boasted to a congressional committee that border agents had expelled “90 percent of aliens crossing the Southern Border within two hours of encountering them — an incredible feat and of critical importance to the public health and the protection of our workforce in response to COVID.”

The order signed by Redfield said the CDC had invoked its powers “to protect the public health from an increase in the serious danger of the introduction of Coronavirus Disease 2019.” Nevertheless, border officials tested unaccompanied children seeking asylum — and expelled them even if their results were negative.

An HHS spokesperson said the department does not discuss internal deliberations. A CDC spokesman declined to make Cetron available for interviews.

During an online talk in August hosted by Dartmouth College, he said that one of the lessons of this pandemic was the importance of “a full bank account of trust” in institutions.

“And if there’s a bankruptcy of trust,” he said, “it can be really tough.”

By April, the numbers were brutal. There were 608,000 cases of COVID nationwide. More than 26,000 people had died, about 10,000 of them in New York City, where the per capita death rate had surpassed Italy’s. Morgue trucks appeared outside hospitals.

Inside the CDC, scientists scrambled to gather and analyze data that could alert them to emerging hotspots. The data was their fuel, driving almost every decision they made. Early in the outbreak, the lack of widespread testing had caused a shortage of data, obscuring the agency’s vision as the virus spread in Washington state, New York and New Jersey. The CDC updated its well-regarded hospital tracking system to collect information about COVID.

But in a startling power play this spring, the Trump administration stripped the CDC of its lead role in handling this vital hospital data, bringing in a private contractor that would struggle to gather reliable information. The unprecedented move, CDC scientists and public health specialists said, struck at the heart of the agency’s mission.

Now, with fall pushing people indoors and threatening a new wave of infections, CDC scientists worry they will again have trouble tracking outbreaks and directing doctors, nurses, medicine and equipment to hotspots.

“When you don’t have quality data that is accurate and reliable, you miss out on signals,’’ a CDC data scientist said. “It can have a devastating impact.”

Like many of the agency’s travails, politics played a role in the battle over data. Powerful critics worked backstage to sideline an agency that they saw as unresponsive and ineffective.

In February, Pottinger, the deputy national security advisor, had lobbied hard for Dr. Deborah Birx, his wife’s friend and former boss, to be named the White House coordinator of the federal response. Pottinger was exasperated by the CDC’s testing debacle and its failures in China. But it was also personal. As a former CDC scientist, Pottinger’s wife had helped invent an HIV test, which was adopted overseas, but not in the United States due to what Pottinger believed was bureaucratic dispute within the CDC. Pottinger told White House colleagues that the agency had a “culture where petty rivalries between egos tend to subordinate the public good.”

Birx, too, was no fan of the agency, even though she’d once run its global AIDS program, according to officials who know her. Since 2014, she’d overseen the State Department’s international AIDS-fighting initiative, which is seen as one of the most effective federal health programs in U.S. history. Birx was a leader who sent emails at 3:45 a.m. A former CDC colleague praised her as brilliant and “data-driven.”

Others were less impressed. Senior officials claimed she amassed power by undermining colleagues, stoking upheaval and presenting herself as the lone savior in a crisis. In February, an audit of her AIDS program by the State Department’s inspector general found that 49 of 68 respondents were critical of the leadership, with some describing it as “dictatorial” or “autocratic.” Several employees complained about intense pressure to meet performance targets, with one saying, “You’re incentivizing data cooking.”

With the CDC now under her ambit, Birx made similar demands. During contentious meetings, she clashed with Schuchat and others over the coronavirus data the CDC collected from hospitals, according to people who were present. She wanted many more details, and she wanted them faster.

Birx expected “every hospital to report every piece of data every day, which is in complete defiance of statistics,” a CDC data scientist said. “We have 60% [of hospitals] reporting, which was certainly good enough for us to have reliable estimates. If we got to 80%, even better. A hundred percent is unnecessary, unrealistic, but that’s part of Birx’s dogma.”

In April, HHS hired TeleTracking Technologies Inc. to collect COVID data along with the CDC. But the Pittsburgh company had trouble getting accurate information, records and interviews show. A CDC analysis in May discovered that data about ventilator use was missing from 57% of hospitals that reported to TeleTracking, compared with 6% of hospitals reporting to the CDC system during the same week. Rather than acknowledge that data was missing, the company reported zeroes instead, according to the CDC analysis.

“It would be like reporting on race and assuming that everybody for whom that variable is missing is white,” a senior CDC official said.

Still, TeleTracking agreed to add many data fields to the forms that hospitals had to fill out every day. CDC data experts refused to do that, warning that hospitals confronted with a form with 91 categories would leave them blank or provide unreliable numbers.

At an impasse, the government in July told hospitals to stop reporting coronavirus data to the CDC.

“That’s really almost like the final blow to show CDC you are out of the game,” said Yip, the agency’s former country director in China. “We don’t even trust you to handle the basic data.”

A TeleTracking spokesperson defended the company’s performance.

“TeleTracking, under HHS’s direction, has developed a data collection system that has enabled more hospitals to report their data more quickly and reliably than ever before,” the spokesperson said. “Since the switchover in July, compliance has improved more than 25%.”

Spokespeople for TeleTracking and HHS also pointed out that Redfield has publicly praised the new system and said his agency’s experts still have access to the data.

The pandemic has required a different and more flexible approach, an HHS spokesperson said. “Rather than reject incorrect data outright, HHS allows it to flow into our system,” the spokesperson said. “The error is flagged and then resolved directly with the hospital.”

Birx did not respond to requests for comment. During a press briefing on Oct. 6, she said she had worked with hospitals to pare back some daily requests to weekly. But at the same briefing, she and other health officials announced that hospitals now would have to provide information about flu patients as well as COVID. If they didn’t, the officials said, they could lose their Medicare and Medicaid funding — a fatal blow for a hospital.

CDC experts fear hospitals may cut corners as they try to comply. A scientist predicted that the tough new policy would “convert a problem of incomplete data to a problem of invalid data.”

By the summer, communities were wracked with anxiety about how to safely reopen schools that had been shuttered since the spring. They looked to the CDC for advice.

From past experience, the CDC’s career scientists knew that schools were tricky political terrain. The last time a pandemic hit the U.S., in 2009, the CDC caused a political backlash when it suggested one- to two-week school closures. But the outcome of the inevitable tug of war between politics and science was much different.

Just months after President Barack Obama took office, a novel flu jumped from pigs to people, then spread across the nation. CDC scientists identified it as an H1N1 virus and initially feared it might be as deadly as the 1918 flu pandemic that had infected a third of the world’s population, killing more than 50 million people.

While Schuchat warned the public, Acting CDC Director Rich Besser flew to Washington. A telegenic pediatrician, Besser told the president and his cabinet that the CDC would be recommending brief school closures in areas where Besser’s disease detectives had identified cases. Obama was clear: All decisions had to be made quickly and grounded in the best available science.

Besser, who recalled the events in an oral history in 2010, said he was then called to another meeting by Rahm Emanuel, Obama’s intimidating chief of staff. Obama’s top political advisor, David Axelrod, and several cabinet secretaries told Besser that his school closure plan wasn’t “going to fly.” Among the many problems: kids who counted on schools for meals would go hungry.

“Let me take a stab at rewriting this,” Besser recalled Emanuel saying as he began scribbling on a pad.

Besser was flabbergasted. Hadn’t the president just said that science was going to drive policy? He looked around, thinking, “I’m the only scientist at this table.”

He turned to his new boss, HHS Secretary Kathleen Sebelius. “Madam Secretary, I’m not real comfortable with this,” he recalled saying. Sebelius hushed him, urging him to wait. Emanuel read his new version aloud. Then Axelrod spoke. “You know, Rahm,” Besser recalled him saying. “I don’t think it’s a good idea for you to be writing scientific guidance.”

Cursing, Emanuel crumpled the paper in his fist, threw it aside and began eating his lunch. At a crucial moment, science prevailed.

In 2020, time and again, the crumpled paper hurled into the corner was the work of the scientists.

In late June, the CDC posted a checklist for reopening schools, which included advice on social distancing and masks. Trump raged on Twitter that he disagreed with the CDC’s “very tough & expensive guidelines for opening schools.”

One CDC official recalls seeing the July 8 tweet and sighing in defeat. “Come on, man, this is your team! You don’t have to tweet it like that! You can just pick up the phone and call Redfield!”

That checklist was supposed to be just the beginning of the agency’s advice on school reopening. Everyone nitpicked the CDC’s subsequent proposals, records show — even Trump’s daughter Ivanka, who suggested granting paid sick leave to teachers and administrators at high risk for COVID-19 complications. In a section that described the higher proportion of cases among Hispanic children, the White House counsel’s office wanted the CDC to add a reference to one of the president’s favorite bugaboos, the Mexican border.

But the most heated disputes involved an HHS mental health office that emphasized the role of schools as integral to the psychological well-being of children. It chastised the CDC team for writing an overly negative “tome” that was a “recipe for schools to stay closed.” The HHS unit was even critical of the suggestion that schools might need to close in areas where the virus was raging uncontrolled. The mental health office scolded the CDC for its “lengthy list of cautions” and said it had written its own guide for parents that had the “opposite tone,” records show.

The White House insisted that the mental health office’s missive lead the CDC’s schools page when it was unveiled in late July. To the outside world, it looked as if the president had snapped his fingers and the CDC caved. Those who bothered to drill down into the real CDC guidance posted beneath were confused by the conflicting messages.

“We didn’t know at CDC that it was going to be forced upon us to post it on our website,” said an agency staffer involved in the discussions.

Scientists at the agency commiserated, calling it “propaganda.”

The HHS mental health office “strongly supports the reopening of schools with appropriate safety measures,” a spokesman said in a written statement. “Parents should be equipped with all perspectives to make an informed decision about the whole health of their child.”

In August, the White House crafted new guidance from Trump. Titled “SCHOOLS SHOULD SAFELY REOPEN,” it contradicted the CDC recommendations on social distancing and masks, and minimized the risks to teachers and students.

The CDC objected, but the White House published it anyway.

The months of defeats were taking a toll. Redfield looked beaten. When his boss, HHS Secretary Alex Azar, upbraided him, he could only mumble, “Yes sir” or “I understand, sir” or “I agree, sir,” according to people who heard these exchanges. (Asked about these exchanges, an HHS spokesperson said: “The American people are fortunate to have Dr. Redfield leading the CDC.”)

Even Kyle McGowan, Redfield’s main protector and an avid political chess player, was running out of moves.

The appointment of McGowan as CDC chief of staff had been a norm-busting move: The 34-year-old was the first political appointee in memory to hold the influential post. He told senior scientists, “I know you think I’m a spy, but I’m really not.”

McGowan had managed campaigns for Georgia Congressman Tom Price, who’d received a 100% rating from the American Conservative Union. When Trump appointed Price as HHS Secretary, McGowan followed him. Six months after Price resigned, McGowan was named to the CDC post. He soon won the trust of CDC career staff. “There was a sense that he’d gone native,” a senior scientist said.

Before getting on the phone with his fellow political appointees in Washington, he’d call CDC scientists. “What can you live with?” he’d ask, according to people familiar with these conversations.

But McGowan and the CDC were often on the losing side. One of their prime tormentors was Michael Caputo, a political fixer handpicked by Trump himself to oversee communications at HHS. A proud protégé of convicted dirty trickster Roger Stone, Caputo had served as an adviser for Russian politicians, worked for Trump’s campaign and promoted conspiracy theories. Soon after arriving at HHS in April, Caputo began riding herd over CDC communications seen as conflicting with Trump’s political message. He made it clear that anyone who dared talk to a journalist without approval could be fired.

McGowan warned his CDC colleagues to be careful what they put in writing. “They can read your email,” he told them.

McGowan became increasingly protective of the CDC’s senior scientists, particularly Schuchat, whose office was adjacent to his. She was viewed as the defender of the agency’s principles, the one immortalized as a disease-hunter on screen. With a close colleague McGowan shared worries that she had become a target of the administration’s wrath, a symbol of the “deep state” bureaucrats the Trump die-hards believed were bent on destroying the president. She attracted the administration’s ire with her blunt assessments in media interviews.

During a June 29 interview with the editor of the medical journal JAMA, Schuchat said that what used to keep her up at night was a fear of an influenza pandemic like the one that struck the U.S. in 1918.

The current pandemic, she said, is similar to “that 1918 transformational experience.” And when asked about the rising case numbers in the United States, she said, “I think there was a lot of wishful thinking around the country, that, ‘Hey, summer, everything’s gonna be fine. We’re over this,’ And we are not even beginning to be over this.”

Schuchat had contradicted Trump’s message that life was returning to normal. McGowan told a colleague that he was hearing rumbles that Caputo and others were trying to fire Schuchat. It had come to this: A world famous scientist was in jeopardy for telling the truth.

“Should I be worried, Kyle?” she asked McGowan, according to a person familiar with the conversation, who said McGowan replied: “Not yet.”

McGowan reached his breaking point when Redfield asked him to stop the deportation of a dog, according to people who worked closely with him.

In late June, a Peace Corps volunteer evacuated from West Africa was told that the rabies vaccine of her dog, a terrier mix named Socrates, was not valid. Rabies vaccines are marked with pink dye, and a photo of Socrates’ vaccination showed a clear liquid, a CDC email said. Border authorities said Socrates had to be sent back to Africa, revaccinated and quarantined there for 28 days before returning. The Peace Corps volunteer sparked a #SaveSocrates outcry on social media.

CDC experts told McGowan that the last foreign dog with rabies that slipped through had cost more than $500,000 in public health charges, including shots for 44 people who had been near the animal, an email shows. Making an exception threatened to render the policy unenforceable for the 500 animals that are deported every year.

At a time when the pandemic had killed nearly 130,000 Americans, McGowan spent an hour and a half on the phone with the HHS general counsel and other senior officials to figure out how to make an exception for a dog. All the while, he told colleagues, his mind kept returning to the fact that the same administration was using the CDC’s quarantine power to deport thousands of children at the border with Mexico.

Later that day, Brian Harrison, the HHS chief of staff and a former labradoodle breeder, announced the liberation of Socrates. Secretary Azar tweeted out the news with the hashtag #SaveSocrates.

Privately, McGowan fumed.

“He was sad, downtrodden and defeated,” a colleague said. “This was really the final straw for him: How we are going to let dogs in, but basically we’re going to require children to be carted off and out of the country? And all in the name of public health.”

McGowan resigned in August.

The following month, Caputo took a medical leave after he hosted a live video on his personal Facebook in which he accused “deep state scientists” of “sedition” and warned his followers to stock up on ammunition in anticipation of political upheaval. In that rant, which was reported by The New York Times, Caputo said CDC scientists had only changed out of their sweatpants to meet at coffee shops and plot “how they’re going to attack Donald Trump next.”

In Atlanta, lawn signs popped up: “I SUPPORT Sweatpants, Coffee Shops and the CDC.”

Longtime CDC employees confess that they have lost trust in what their own agency tells the public.

In August, the CDC stunned infectious disease doctors everywhere when it recommended that people who had close contact with a COVID patient didn’t necessarily need testing if they didn’t have symptoms. Even Butler, one of the highest ranking scientists at the agency, began signing his emails to state and local health departments, “Keep testing, Jay.”

Another dismayed veteran who works with local health officials did something he had never done before. He told them to ignore his own agency’s guidance. The agency reversed the much-criticized recommendation about testing a month later, but the damage was done. After more than a decade at the CDC, the veteran decided to quit.

“It’s just a disappointment,” he said. “People’s reaction now at other agencies, at state and local public health agencies, when the CDC comes out with a recommendation, they are going to ask: ‘Is that the truth? Or is that what you were told to say?’”

Some longtime senior scientists at the CDC are grappling with whether they are too tainted to lead the rebuilding of trust.

“Many of us who might be viewed as complicit need to decide whether we need to leave,” one of them said, “Or can we be part of the ‘never again’ so that the agency never gets this kind of political interference again?”

ProPublica is a Pulitzer Prize-winning investigative newsroom. Sign up for The Big Story newsletter to receive stories like this one in your inbox.