As Ebola Spread in Dallas, New York Honed Protocol

This time was different.

When Craig Spencer, a young doctor just back from treating patients with Ebola in Guinea, fell ill with the virus in New York on Thursday, the paramedics who went to get him were dressed in protective suits. He entered Bellevue Hospital through a rear door, far from the busy emergency room, and was taken to a state-of-the-art isolation ward that was locked and guarded.

The carefully planned response was a world apart from the scene that unfolded in a Dallas hospital last month when a Liberian man, Thomas Eric Duncan, became the first person to test positive for Ebola in the United States.

It was Ebola's first surprise encounter with a modern medical system, and it was, by all accounts, a bumpy one. Duncan lay in an emergency department room at Texas Health Presbyterian Hospital for hours, with a fever and frequent diarrhea, while the same doctor and nurses treated patients in nearby rooms, medical workers said. A lab technician had to comb through his inbox to find an email from the government on procedures for Ebola blood samples, only to find that he had handled them improperly, the workers said.

Rules for protective gear seemed to change with the weather, and some of Duncan's nurses, unsure of what to wear, resorted to looking for answers on the website of the Centers for Disease Control and Prevention, according to a health worker involved in the response.

The often rudderless response lasted two weeks, and in the end, two nurses, Nina Pham and Amber Joy Vinson, fell ill with the virus. Both have recovered, but the searing experience stunned experts, and shook Americans' confidence in their health care system.

"I'm just gobsmacked," said David P. Fidler, a law professor at Indiana University and a fellow at Chatham House Centre on Global Health Security. "Ebola was supposed to be something we could handle easily."

The response also prompted a national reckoning, both by the CDC, the federal agency that contributed to the wobbly response in Dallas, and among hospitals nationwide that are now scrambling to prepare, having learned from mistakes in Texas that many say could have happened to any of them.

"There had to be a first hospital, and unfortunately for Texas Presbyterian, it was them," said Dr. Sean P. Elliott, medical director of infection prevention at the University of Arizona Health Network. His Tucson hospital, Elliott said, is "burning through our supplies of protective gear and putting in tons of people hours" to prepare.

"Dallas has touched a very sensitive nerve," he said.

Dallas also shattered the fundamental assumption among most American health officials that almost any large U.S. hospital could safely treat Ebola. Many of the procedures that have been put in place in New York and elsewhere since Dallas reflect that change in thinking - for example, the designation of specific hospitals, like Bellevue or Elliott's, as point places for Ebola treatment.

If there was ever a moment for redemption, Spencer's case was it. Even before the result of his Ebola test, a CDC team was on its way to New York from the agency's base in Atlanta. It was whisked there on a Defense Department plane, which turned right around with a sample of Spencer's blood that it delivered to a CDC lab around 4 a.m. More CDC disease trackers followed on Friday morning for a total of seven people.

The initial team that deployed to Dallas last month was bigger - 10 people - but was hampered by inconsistent protocols for gear at the hospital and a lack of leadership on the ground, for which the team itself was partly responsible. And far from being zipped to a lab on a military flight, the blood sample for Duncan did not arrive at a state laboratory in Austin until almost 48 hours after he got to the emergency room.

What is more, the CDC arrived in Dallas only after Duncan's test came back positive, on Sept. 30, two days after the hospital first called the agency.

It is too soon to tell whether the response to Spencer's infection will continue as smoothly as it began. His condition worsened on Saturday, though he remained awake and communicative, health officials said.

Spencer was "entering the next and more serious phase of his illness, as anticipated with the appearance of gastrointestinal symptoms," Ana Marengo, a spokeswoman for the city's public hospital system, said in a statement. In a brief telephone interview from his hospital room, Spencer spoke in a neutral tone. "I'm still undergoing treatment," he said.

The case of Spencer, a health worker who was diligently monitoring his own temperature, was always going to be different from that of Duncan, a welder from Monrovia visiting the United States, who arrived unannounced and told health workers he had not been exposed.

Even so, Spencer's case, as it unfolded across New York, with disease detectives chasing his trail from a Brooklyn bowling alley to his Harlem apartment, seemed to ring in a new era.

"We are all much smarter now," said Dr. William Schaffner, an infectious-disease specialist at Vanderbilt University Medical Center in Nashville, Tennessee. "We have all learned from the Dallas episode."

Shifting Protocols

The first time Duncan went to the emergency room at Texas Health Presbyterian feeling ill, shortly after 10:30 p.m. on Sept. 25, a nurse noted in his record: "Patient recently came from Africa." But that did not raise any flags.

A series of tests was ordered, and at 3:18 a.m., Duncan was released.

But shortly after 10 a.m. on Sept. 28, he was back, this time in an ambulance. One of Duncan's relatives said she had warned the ambulance crew that he might have an infectious disease.

"Nurses were saying: 'No. Could it be Ebola? No,'" said one hospital nurse, who spoke on the condition of anonymity because of concerns about job repercussions.

The next 72 hours would be crucial. According to the CDC it was during this time that Pham and Vinson most likely became infected.

The hospital had been assuring the public that it was ready. Dr. Edward Goodman, an epidemiologist at the hospital, said on Sept. 30: "We have had a plan in place for some time now."

But a review of hospital records and interviews with hospital workers, federal officials and relatives of Duncan evoke a picture of worry and uncertainty in which health workers, in the absence of any training and facing rules that were constantly changing, had to improvise while racing to try to save Duncan's life. "We'd see pictures of Dr. Frieden in Africa, and we were like, 'If this is Ebola, we shouldn't be wearing gowns, right?'" one nurse said, referring to Dr. Tom Frieden, director of the CDC, who traveled to West Africa in August. "We were looking at these pictures and seeing what other people were wearing and saying, 'We're not wearing anything like that,'" said the nurse,

who asked not to be identified because the hospital told employees not to speak with reporters.

"There was no training, no explanation, nobody around to help them," said Dr. Pierre Rollin, an Ebola expert at the CDC who arrived on Oct. 12 to help train the nurses after Duncan's death. "They had the impression that they were running after the ideal personal protective equipment, and they didn't have it. It was like a race."

Duncan's first days in the hospital's intensive care unit, where he was moved at 4:40 p.m. Sept. 29, more than 30 hours into his hospital odyssey, highlighted the problem. That first night in the ICU, nurses wore two gowns to cover front and back, shoe covers, a surgical mask, two layers of gloves and a face shield that went to the chin, Wendell Watson, a hospital spokesman, said. By the next night, after Duncan tested positive for Ebola, they added Tyvek suits with hoods and respirators. Then came plastic aprons and triple layers of bootees and gloves. The CDC later said too much gear may have increased risk.

The hospital says it was consulting with Emory University Hospital in Atlanta, which had treated Ebola patients, as well as county health officials and the CDC on protective gear.

Watson said in an email that health workers at the Dallas hospital immediately isolated Duncan from other patients when he arrived in the emergency room, and that it emptied an entire 24-bed unit to dedicate to Duncan's care once he was admitted, even erecting temporary walls around his hallway for better protection. Ebola training had recently started when the virus struck, Watson said.

Pham, who spent two hours watching over Duncan on his first night in the ICU, also cared for him in the days that followed.

By Oct. 1, the patient felt better, so much that he asked to watch an action movie. But after midnight Saturday, his organs failing, Duncan was intubated and put on a respirator.

On Oct. 6, Duncan's eighth day at the hospital, his mother, sister and nephew arrived in Dallas, having driven through the night from their North Carolina home. They hurried to the hospital, which arranged for them to see and talk to Duncan, at that point in a medically induced coma, via Skype.

From a different floor of the hospital, the relatives recalled last week, they glimpsed Duncan in his bed, his head elevated on a pillow. His mother, Nowai Gartay, told him to be strong as she looked at his face for the last time.

"I'm ready to hug you right now, my baby," she told him. "Can you hear me?"

Less than two days later, on Oct. 8, Duncan was dead.

Late on the night of Oct. 11, Pham tested positive for Ebola, sending shock waves through the hospital, and jolting the government into action.

Frieden immediately sent 16 more experts, including Rollin. Frieden reflected publicly a few days later that he wished he had sent a bigger team sooner.

"That was for me the moment at which something that we had hoped would not happen, had happened," he said in an interview. "The protocols that we had recommended and that the hospital had tried to implement, had not worked."

One problem was that its protective gear protocols had been developed for use in field hospitals in Africa, where patients are given only basic care, like intravenous fluids, not for use in intensive care units of hospitals in modern medical systems, where large teams perform high-risk procedures that can involve lots of bodily fluids. The agency has substantially rewritten those protocols.

"This is an entirely different setting," said Dr. Anthony S. Fauci, the director of the National Institute of Allergy and Infectious Diseases, which treated Pham and on Friday released her. "Nobody realized that when the patient was admitted."

Another crucial question - who was in charge - seemed to have no answer, with several layers of government and the hospital all playing roles.

Even with "the best-laid plans," Dr. David L. Lakey, commissioner of the Texas Department of State Health Services, said in an interview, "you see how they are being implemented on the ground during the events and you have to make hard, real-time decisions on gaps that you see." One thing he learned, he added, was the importance of "having someone in charge."

The hospital has started a new system in its emergency department to immediately move high-risk patients to an isolation unit. And the CDC, with other federal agencies, has been in talks with designers of the moon suits, which were not originally intended for medical care, to make them safer to take off, for example by placing the zipper across the shoulder with a large ring that makes it easy to tug in gloves.

"You can go back and, with 20-20 hindsight, say the CDC's recommendations were wrong, and perhaps they were," Frieden said. "But you also have to say that if you are using the equipment you are familiar with - face shields, gowns, gloves - and you are doing it meticulously, then your risk of infection is minimized."

Testing New Skills

Bellevue, New York City's flagship public hospital, had started preparing for the possibility of Ebola patients in July, after an alert from the CDC. But after Duncan tested positive in Dallas, the city intensified its preparations, ordering 911 emergency service workers to screen callers for travel to West Africa.

On Oct. 16, Gov. Andrew Cuomo announced that eight hospitals statewide, including Bellevue, had volunteered to be Ebola treatment centers. Ebola drills would also be conducted in subways and on other mass transit, he said, as well as college campuses. Testers posing as Ebola patients were sent into emergency rooms to determine whether staff members would respond appropriately. Hospital employees practiced putting on and taking off protective gear, and officials reached out to West African communities like "Little Liberia," on Staten Island, trying to build relationships.

At Bellevue, officials set up four single-bed isolation rooms to receive high-probability or confirmed Ebola cases. Those rooms were carved out of a tuberculosis ward, which is equipped with decontamination rooms and cameras that can monitor patients remotely, to avoid unnecessary trips into the rooms.

"Unfortunately, we have an advantage from watching what happened in Dallas," Cuomo told thousands of health care workers Tuesday at an Ebola training session at the Jacob K. Javits Convention Center. "Because Dallas didn't have a chance to prepare like we have a chance to prepare."

Two days later, his boast was put to the test, as Spencer, 33, reported a fever of 100.3.

A day later, Spencer was in stable condition and in good enough spirits to talk to his family and friends by phone.

Still, it was far from clear that he would get better, and the possibility that experimental treatment could be needed was being discussed. Through it all, officials have been working hard to tamp down fear. City workers handed out informational leaflets on his block and near subway stations in his neighborhood.

"After the Texas incident, they reviewed every procedure from top to bottom," said Mark Levine, the city councilman for the neighborhood. "New York City could not have been more prepared for an incident like this."