The Centers for Disease Control and Prevention was long revered for its methodical and meticulous scientific approach. Agencies in other nations modeled themselves after the world’s most highly regarded public health authority, even adopting the name.

At the outset of the pandemic, the C.D.C. moved at its accustomed pace. But this time, with a novel virus moving so quickly, the country paid a price: Testing and surveillance lagged as the agency tried to implement dated approaches with creaky infrastructure. Officials were late to recommend masking, in part because federal scientists took too long to recognize that the virus was airborne.

Now the contagious Omicron variant is pushing the C.D.C. into uncharted territory. Because decisions must be made at a breakneck pace, the agency has issued recommendations based on what once would have been considered insufficient evidence, amid growing public concern about how these guidelines affect the economy and education.

The agency’s director, Dr. Rochelle P. Walensky, has sometimes skipped much of the traditional scientific review process, most recently in shortening the isolation period for infected Americans.

After the Trump administration’s pattern of interference, President Biden came to office promising to restore the C.D.C.’s reputation for independence and rigorous science. The challenge now for Dr. Walensky is figuring out how to convey this message to the public: The science is incomplete, and this is our best advice for now.

For a bureaucracy staffed primarily by medical professionals, the change has not been easy.

In recent interviews, some officials at the C.D.C. privately described the decisions as demoralizing, and worried about Dr. Walensky’s increasing reliance on a small group of advisers and what they saw as the White House’s heavy political influence on her actions.

Yet others outside the agency commended Dr. Walensky for short-circuiting a laborious process and taking a pragmatic approach to managing a national emergency, saying she was right to move ahead even when the data was unclear and agency researchers remained unsure.

There are policy considerations in a pandemic that are “not the sole purview of C.D.C.,” said Dr. Richard E. Besser, who served as interim chief of the agency during the H1N1 influenza virus outbreak of 2009. But, he added, “I think we need some more clarity” when policy and economics drive agency recommendations.

As of Sunday, more than 800,000 Americans on average are infected daily, according to data gathered by The New York Times. Many schools and businesses are struggling to remain open; hospitals in nearly two dozen states are nearing capacity.

At the end of December, Dr. Walensky announced that infected Americans would need to isolate for only five days, not 10, if they were no longer experiencing symptoms, and that a negative test result would not be required to end the isolation period.

Critics complained that the virus might spread as contagious people were allowed to return to offices and schools. Many pointed out that the research supporting a shortened isolation period for Omicron infections was scant.

But the recommendation had an important advantage: It could help keep hospitals, businesses and schools afloat through the worst of the Omicron surge.

The recommendations for isolation are “basically correct,” said Dr. Thomas R. Frieden, who led the agency under President Barack Obama. “The problem is, they were not explained.”

Dr. Walensky and the C.D.C. declined requests for comment on new tensions in the agency’s decision-making. But the director has frequently cited rapidly evolving science as justification for recommendations that proved to be confusing or unpopular.

Testifying before the Senate on Tuesday, Dr. Walensky said that the agency’s new recommendations for shortened isolation periods represent “swift science-based action to address the very real possibility of staffing shortages.”

It has been something of a mantra for the director.

Last March, the C.D.C. said that schoolchildren could safely sit three feet apart in classrooms, instead of six feet, although there was virtually no research to back up the recommendation. But the move did make it easier for administrators to consider opening schools.

In May, Dr. Walensky cited scientific data when she told vaccinated people that they could take off their masks and mingle freely, much to the consternation of experts who said that the move ignored the possibility of breakthrough infections. (Those arrived with the Delta variant.)

In August, Dr. Walensky joined President Biden in supporting booster shots for all Americans, well before scientists at the Food and Drug Administration or at her own agency had a chance to review the data on whether they were needed.

The most recent example, the isolation advice, left turmoil within the agency over the way in which it was established and announced.

On the Sunday night after Christmas, Dr. Walensky called an emergency meeting of the agency’s Covid response leaders. She told them the agency would shrink the recommended isolation period and would drop a negative test result as a requirement for leaving isolation, according to an official familiar with the video call who spoke on condition of anonymity because the individual was not authorized to speak on the matter.

The new guidance would be made public the next day, Dr. Walensky said, and officials were not to discuss it until then.

Stunned, the scientists scrambled to gather the limited data to support the recommendations and to rewrite the hundreds of pages on the agency’s website that touch on quarantine and isolation.

Before publishing a new recommendation, federal researchers normally pore over data, write a draft and fine-tune it based on comments from others. There was so little evidence for shortened isolation — and even that was based mostly on the Delta variant — that the “science brief” that typically accompanies guidance was downgraded to a “rationale” document.

Some researchers bristled at being left out of the decision-making process and were enraged by the agency’s public statement the next day that the change was “motivated by science.”

While some believed the new five-day cutoff was arbitrary, they also knew of data suggesting that rapid tests might miss some Omicron infections, and so mostly agreed with Dr. Walensky’s decision not to require a negative test result before ending isolation.

But when Dr. Walensky informed staff of the new recommendations in the emergency meeting on Dec. 26, they were far from ready. Over the next week, C.D.C. scientists struggled to adjust hundreds of guidance documents on the agency’s website.

About 2,000 health officials, public health lab directors and epidemiologists at the state and city levels join a weekly call with C.D.C. officials.

On the call on Monday, Dec. 27, just hours before the C.D.C. released its statement, state and local officials peppered agency scientists with questions about the plans for isolation guidance for the general public.

Under strict orders to not talk about the new recommendations, C.D.C. staff members were silent.

“We would have appreciated more opportunity for input and heads up,” said Scott Becker, chief executive of the Association of Public Health Laboratories.

Dr. Walensky’s supporters said the pivot by the C.D.C. was inevitable and that she had made the right calls. The agency is a behemoth, filled with researchers accustomed to taking their time, and the pandemic needed more urgent solutions.

“There are people at C.D.C. who really don’t get it,” Dr. Frieden said.

During his tenure, he said, he was frequently confronted with “in some ways charming, but in some ways problematic, cluelessness on the part of C.D.C. staff that their recommendations, their guidance, their statements could have big implications.”

Several outside experts said Dr. Walensky had become a scapegoat for people who were weary and frustrated by a virus that seemed repeatedly to have retreated only to return in a horrific new form in short order.

Leading the C.D.C. is challenging even at the best of times, they said. But Dr. Walensky took the reins in the middle of a pandemic, in a politically charged climate and at a low point in the agency’s credibility and staff morale.

And agency researchers are still working remotely — “almost an unthinkable hurdle to overcome,” Dr. Besser said.

“I am concerned about C.D.C. I am concerned about the nation’s trust in public health,” Dr. Besser said. “But I think it’s really unfair to put that on the shoulders of Dr. Walensky.”

Dr. Walensky has explained the rationale for her decisions at news briefings held by the White House. But last week, responding to wide criticism about muddled messaging, she and other agency scientists held a briefing of their own, answering questions from reporters about the isolation guidance, the rising rate of hospitalizations among young children and the agency’s plans for a fourth shot of the coronavirus vaccine.

The briefing was a welcome step toward rebuilding trust in the C.D.C. and clarifying its decisions, some experts said.

“Separating out public health considerations from political considerations is very important,” Dr. Besser said. “And by doing briefings from C.D.C., she’ll be able to lift up C.D.C. scientists and experts.”

Some of the current conflict at the C.D.C. predates the pandemic and Dr. Walensky’s leadership. Tension between the agency and the National Institutes of Health, represented by Dr. Anthony S. Fauci, festered even during previous public health crises, some health officials noted.

In the most recent instance, Dr. Fauci and Surgeon General Dr. Vivek Murthy gave assurances on television that the C.D.C. would revisit its recommendations for isolation — when the agency had no plans to do so — and irritated senior C.D.C. scientists.

Ideally, the secretary of health and human services, Xavier Becerra, should smooth things over, said Dr. Peter Hotez, dean of the National School of Tropical Medicine at Baylor College of Medicine.

In a rare appearance, Mr. Becerra last week defended Dr. Walensky in a CNN interview, saying she had “a medical license and a degree in public health. She doesn’t have a degree in marketing.”