An outbreak would demand peak performance from America's medical professionals — especially in hospitals. But many of the facilities that may be on the front lines have well-documented histories of failing to prevent the spread of infectious diseases.

This article was first published on Tuesday, March 3, 2020 in ProPublica.

By Marshall Allen, Caroline Chen, J. David McSwane and Lexi Churchill

In early February, Royal Caribbean's Anthem of the Seas docked in Bayonne, New Jersey, in need of a hospital. The cruise ship was carrying patients who had traveled from China, where an outbreak of COVID-19 had taken root. Four passengers needed to go somewhere for further medical observation.

The obvious next step was University Hospital in Newark, a major academic medical center equipped with isolation rooms. "The hospital is following proper infection control protocols while evaluating these individuals," Gov. Phil Murphy said in a statement. The patients tested negative, but the governor was clear. The state's first coronavirus cases would go to University.

That's a hospital that has struggled in recent years with a critical skill essential to battling COVID-19: controlling the spread of infection.

Less than two years ago, a deadly bacteria made its way through the facility. Three babies in the neonatal intensive care unit got infected and died. Government inspectors cited the hospital for being short of staff; failing to maintain a sanitary environment, including improper hand hygiene and sterilization; and inadequately isolating patients with respiratory conditions. They determined the hospital had put patients in "immediate jeopardy."

Today, the state's former health commissioner, Dr. Shereef Elnahal, is in charge of the hospital. He told ProPublica its infection control problems are a thing of the past. The violations cited in the report were corrected with increased screening of patients, improved handwashing and equipment and focused leadership, Elnahal said. "I'm proud of our progress since then," he said. Murphy's office directed inquiries to the state's Department of Health, where a spokeswoman said there is "complete confidence" in the hospital's ability to manage the coronavirus response.

But infection control has been a recurring problem at some of the very hospitals that would likely be called upon to treat COVID-19 patients, a ProPublica review of hundreds of hospital inspection reports found. This raises concerns that they could become hotbeds for disease, putting patients at risk and rendering infected workers unable to care for others.

"Health care workers are my top worry," said Dr. Peter Hotez, dean of the National School of Tropical Medicine at Baylor College of Medicine in Houston. He noted that in China, so far, about 15% of infected hospital workers have become severely ill. "If this takes place in the U.S., and we see those numbers of workers sent home or in the ICU, being taken care of by their colleagues, things will start to unravel. This is the soft underbelly of our preparedness system right now."

At least two health care workers in Northern California had preliminary positive tests for COVID-19 at NorthBay VacaValley Hospital, pending CDC confirmation. The hospital had treated a patient who later tested positive for the disease. Other health care workers who came into contact with the patient are also in quarantine.

There is no list of designated centers to handle the most critical COVID-19 patients, experts said. But the Centers for Disease Control and Prevention, during the 2014-16 Ebola outbreak, named 55 hospitals it considered to be in the first tier of treatment centers to handle that kind of crisis — mostly large, urban teaching hospitals capable of complex care like blood transfusions and ventilation.

ProPublica analyzed five years of federal hospital inspection reports for these facilities and found violations for infection control failures or other factors that could hamper the response to an outbreak at more than half of them. About 1 in 5 of the facilities had four or more violations; the analysis found more than a hundred overall. It's not clear by looking at the reports how many of the violations led to patient infections. Problems that get cited on the inspection reports are required to be corrected as part of the regulation process.

But it's also true that inspections only flag a small number of the actual problems in hospitals. American hospitals, overall, are so bad at preventing infections that hospital-acquired infections are considered a leading cause of death in the United States. The hope would be that the sites designated as specialized infection-control centers would do better.

MedStar Washington Hospital Center in Washington, D.C., says it's ready to screen coronavirus patients. Inspectors have cited the facility more than a dozen times since 2017, including for infection control failures. Among the violations: Staff did not wear and dispose of masks according to federal guidelines. Short staffing caused scores of patients to go without respiratory treatments. There were sewage leaks in operating rooms.

In an email, a spokeswoman said the hospital has addressed the failures: "We maintain a constant state of readiness for treating complex illnesses, including the coronavirus."

Montefiore Medical Center in the Bronx, New York, says on its website COVID-19 patients will be immediately isolated. But it got written up back-to-back, in 2016 and 2017, for violating infection control protocol. Among the shortcomings: "Chronic overcrowding" in its emergency room, not isolating a Hepatitis B-positive patient and contaminated supplies. Infection control breaches put patients and staff at risk, one inspection report said. Hospital officials did not respond to requests for comment.

Medical experts say they wonder: if hospitals can't control the spread of pathogens under normal conditions, what happens if they face a rush of patients with a disease as contagious and serious as the one caused by COVID-19?

During the SARS outbreak in the greater Toronto area, 44% of the total cases were among health care workers. A retrospective study, published in the journal Emerging Infectious Diseases in 2004, hypothesized that "lapses in infection control measures may be responsible," noting that caregivers were particularly at risk during procedures like intubation.

Though COVID-19, with its estimated 2% fatality rate, is far milder than SARS, which killed about 10%, it is thought to have a similar method of transmission and will require similar methods of protection to prevent the disease from spreading throughout hospitals. Without a proven treatment or vaccine, infected patients would need to be handled with the utmost caution. They would be isolated, and caregivers would don protective gear, including gloves, goggles, gowns and masks.

Medical providers across the country told ProPublica that they're worried about their safety and their hospitals' lack of preparation. They spoke on the condition of anonymity because they were not authorized to speak on behalf of their hospitals.

The coronavirus arrived in Washington state "like a slap in the face," a nurse in the Seattle area told ProPublica. Two weeks ago, her hospital was talking about the virus as something it was "watching, but with no big alarm." Now, the state has had the first deaths in the United States and 18 confirmed cases as of Monday. The hospital is "desperately trying to get more supplies," she said, particularly of masks and gowns. She fears that morale will drop. Already, she's heard staff grumbling that only certain units are being allocated higher-protection masks.

An acute-care nurse in Rockford, Illinois, said that just last week, a severely ill patient on her floor initially tested negative for the flu but after nearly a week retested positive. In that period, nurses were in and out of his room, and what's known as "droplet precautions" weren't always taken — for example, sometimes the patient didn't have a mask on, meaning staff members were exposed. "How easily this happened gives me serious concerns about the much more serious infection we face with COVID-19," she said.

Another nurse, at a high-level hospital in western New York that is likely to handle severely ill patients, said the only information he's received is via the hospital's internal newsletter. "Management has just said, 'We're monitoring the situation and we'll keep you updated.' It's ridiculous. They haven't verbalized a specific plan, and that increases the anxiety of a lot of the care providers." He said he and his co-workers are "resigned to the fact that we're all going to get the coronavirus."

The risk to hospital workers could have a dangerous cascading effect, said Dr. Lance Peterson, the recently retired director of clinical microbiology and infectious disease research at the NorthShore University Health System in Evanston, Illinois. He said that hospitals often keep staffing to a minimum, and that could become a problem if there's prolonged spread of the virus. "If hospital workers start getting sick," he said, "you don't want them to come to work."

Some hospitals are more prepared than others for a potential outbreak, said Dr. William Schaffner, professor of preventive medicine and infectious diseases at Vanderbilt University Medical Center. Some have spent recent weeks running drills. Officials at University Hospital in Newark and MedStar in Washington, D.C., count themselves among those.

In many ways, the United States is better prepared than many other countries for an outbreak, Peterson said. Individual patient rooms are common, which makes it easier to isolate infected patients. The SARS and Ebola outbreaks prompted many hospitals to stock up on supplies like gloves and gowns and masks, he said. And The Joint Commission, which accredits hospitals, has been monitoring their level of preparation.

But it is true there are also going to be infection control problems, he said. "Whenever you have humans in the system, there will be lapses."

Ariana Tobin and Maya Miller contributed reporting.

Help us report on coronavirus by filling out this form or contacting our reporting team at coronavirus@propublica.org.

The bipartisan proposal comes in response to a ProPublica story that showed how a personal trainer posed as a doctor to defraud prominent health insurers.

This article was first published on Friday, February 14, 2020 in ProPublica.

Four United States senators have introduced a bipartisan bill to close a little-known, but gaping, loophole that allows scammers to plunder commercial insurers by posing as licensed medical providers.

The Medical License Verification Act, introduced Thursday, comes in response to a July ProPublica story that showed how absurdly easy it is to commit health care fraud, which experts say could be costing Americans hundreds of billions of dollars a year. ProPublica told the story of David Williams, a Texas personal trainer and convicted felon, who represented himself as a licensed physician when he applied for National Provider Identifiers, the unique numbers required by the federal government to bill insurance plans.

The Centers for Medicare and Medicaid Services administers the NPI program. But ProPublica found that it did not verify whether applicants had valid medical licenses, which could be done easily by checking with state licensing boards. That missing step allowed Williams to collect at least 20 different NPI numbers and bill some of the nation's most prominent health insurers, Aetna, Cigna and UnitedHealthcare, for millions of dollars in bogus medical services.

The insurance companies auto-paid the claims without checking to see if Williams had a license. Over more than four years, Williams filed more than $25 million in false claims with the companies, reaping more than $4 million. Some insurers actually sent him letters noting he wasn't a doctor and was filing false claims, but they continued to pay new claims anyway. In 2018, a jury convicted him of health care fraud and he's now in federal prison.

At his trial, investigators for the insurers said that their companies assume people are being honest when they file claims. The insurers told ProPublica it wasn't easy to stop the fraud because Williams billed them using different organization names and tax identification numbers.

The proposed Medical License Verification Act would require Medicare to verify that an NPI applicant's license is valid and in good standing before approving an NPI number. Sen. Rick Scott, R-Fla., called it a "commonsense step" in a statement released by the legislators. Consumers pay for fraud losses, Scott said, so the proposal will "prevent bad actors from defrauding families."

Cracking down on unlicensed providers will prevent fraud and lower health care costs across the board, Sen. Chris Van Hollen, D-Md., said in the statement. The bill was also signed by Sen. John Cornyn, R-Texas, and Sen. Catherine Cortez Masto, D-Nev. The bill's next step is to be assigned to a committee, where it will be considered, according to Scott's spokeswoman.

 A Medicare official said the agency does not comment on proposed legislation. The trade group America's Health Insurance Plans, better known as AHIP, did not reply to a request for comment about the proposed legislation.

In the Williams case, the insurers and regulators were slow to stop the personal trainer's fraud even though his ex-wife and her father, Jim Pratte, had been reporting the criminal activity to them. Pratte said Thursday that he was pleased to hear about lawmakers' proposal to block scammers like Williams from easily getting NPI numbers. Politicians have talked for years about how to pay for the high cost of health care, Pratte said. "One major thing you can do without hurting anybody is cut out the fraud," he said.

The failure of insurers to take prompt action against Williams, and the ease of his fraud, led ProPublica to look at how aggressively such scams are pursued. Across the country, ProPublica found that insurance carriers not only made it easy for fraud to happen, they seldom pursued the fraudsters or referred them for prosecution.

ProPublica is a nonprofit newsroom that investigates abuses of power. Sign up to receive our biggest stories as soon as they're published.

Andy Jurtschenko told his children that he didn't want to be a burden on them. But after he suffered brain damage during a heart transplant at a New Jersey hospital, his medical team deflected their request for a DNR.

This article was first published on Tuesday, December 31, 2019 in MedPage Today.

By Caroline Chen

Three weeks after his heart transplant, Andrey Jurtschenko still had not woken up.

A towering figure at 6 feet, 3 inches, with salt-and-pepper hair and matching mustache, Jurtschenko — known to one and all as Andy — delighted friends and family with his seemingly endless supply of wisecracks and goofball humor. On April 5, 2018, he went into surgery at Newark Beth Israel Medical Center in Newark, New Jersey, for a new heart and what he hoped would be renewed energy. He dreamed of returning to his carpet business and to enjoying New York Mets games on the weekends after years of exhaustion and strain caused by congestive heart failure.

Typically, patients begin reviving within 24 hours after transplant surgery. Andy didn't. As the days passed, his children, Chris and Megan Jurtschenko, became increasingly concerned. On April 26, a neurologist called Chris and explained what an MRI the day before had shown: Andy's brain had likely been deprived of oxygen during the procedure. The doctor said he "would basically be in a vegetative state," Chris recalled in an interview. Chris asked to meet with the medical team the next day.

The devastated family took some comfort in knowing what Andy would have wanted. In several conversations before the surgery, he had made clear that "he did not want to be a burden on us, he did not want to live in an incapacitated form," Andy's older sister, Anna DeMarinis, said.

Now that their father could not speak for himself, Chris and Megan, as Andy's next of kin, had to be his voice. On April 27, they went to the hospital for the meeting. Still hoping Andy might recover, they did not seek to withdraw his feeding tube or medications. But they asked for a do not resuscitate order. If he were to stop breathing or have no pulse, a DNR order would direct doctors not to compress his chest, use a machine to force air into his lungs or give electric shocks to restart his heart.

If his heart stopped, "we weren't going to force him to stay," Megan said.

Dr. Margarita Camacho, the surgeon who had performed the transplant, deflected their request, the siblings said. She told them that it was too early for a DNR, and that they shouldn't give up hope because their father might recover, his children said. At Camacho's urging, Megan and Chris said, they let it go. No DNR order was signed that day. The family would continue to press the issue and finally secure a DNR more than a month later.

Megan and Chris Jurtschenko waived their privacy rights to allow the hospital to discuss their father's case with ProPublica. Asked directly about the meeting with the surgeon and why the family's wishes were not followed at the time, Linda Kamateh, a spokeswoman for Newark Beth Israel and Camacho said in an email: "Physicians are obliged to give their best medical advice based on a patient's medical condition. However, ultimately the decision to have a DNR resides with the patient. The hospital believes that it adhered to those principles in its discussions with the Jurtschenko family."

Except for "a very specific set of dire medical circumstances, in which a patient may require resuscitation," a DNR "does not otherwise affect ongoing care and treatment," Kamateh wrote in a separate email. "... These decisions are often revisited and reassessed within the course of treatment."

Andy's medical record doesn't mention the children's request for a DNR. "The family was able to express their concerns and decided to continue to see how PT [patient] progresses over the next few weeks," a social worker wrote.

Bearing out Camacho's prognosis, Andy would awaken and recover some cognitive ability — but only enough to attain the incapacitated state he had dreaded, not to become again the man that his children knew and loved. They remained adamant that, if his heart stopped, he would have preferred to die than to be resuscitated for such an existence.

The American Medical Association's code of medical ethics states, "The ethical obligation to respect patient autonomy and self-determination requires that the physician respect decisions to refuse care." Yet Newark Beth Israel's transplant team was often reluctant to sign DNR orders, according to four former employees and an audio recording of a staff meeting. While the team wouldn't outright refuse, especially when patients or their family members repeatedly asked, it often delayed or discouraged DNRs, especially before key dates tied to performance metrics such as the one-year survival rate, or the proportion of people undergoing transplants who are still alive a year after their operations, three of the ex-employees said.

The team also lacked a process for discussing beforehand whether patients would want CPR if their pulse or breathing stopped after their operations, the former employees said. Typically, the staff addressed the issue only if a patient's condition became critical and family members were insisting on a DNR.

Newark Beth Israel's DNR policies are consistent with best practices, Kamateh said. "These policies guide our clinical teams in support of the treatment decisions of our patients and their families, from the most routine procedures to the most complex and stressful situations," she said. "We strive to explain care options and deliver sound medical advice in ways that are timely and clear, yet also respectful and sensitive."

At least indirectly, the concern about DNRs may have stemmed from Newark Beth Israel's aggressive approach to transplants. Newark Beth Israel's heart transplant program is one of the top 20 in the U.S. by volume, having grown under Dr. Mark Zucker, its director for three decades, and Camacho, the main surgeon. As of November, the hospital had performed 1,096 heart transplants.

The program is known for taking on sicker patients who might be rejected at other programs. From 2014 through 2017, compared with its counterparts in New Jersey and nearby states, Newark Beth Israel's transplant team operated on a higher percentage of patients who were older, more overweight or obese, and who had been in an intensive care unit while awaiting transplant, according to the Scientific Registry of Transplant Recipients. (The registry is funded by the U.S. Department of Health and Human Services to track and analyze transplant outcomes.) In those years, Newark Beth Israel also had a higher percentage of patients who had been on a pump or some other support before transplant, which increases the difficulty of surgery. This stance filled an important gap in care and helped the program grow both in size and revenue; hospitals typically bill insurers about $1.4 million for a heart transplant.

"While the Advanced Heart Failure Treatment and Transplant Program at NBI does not seek out cases that are more complex than those handled by other prestigious transplant programs, patients from other programs have been referred to our care and been successfully transplanted," Kamateh said. "Our clinical decisions are driven by the best interest of our patients, including their personal preferences, not by statistical results."

Scores of grateful patients say they owe their lives to Zucker and Camacho.

"Dr. Zucker has saved my life again and again," said Mark Reagan, a retired AIG executive in Bluffton, South Carolina. Reagan received his heart transplant at Newark Beth Israel in March 2003 after suffering from congestive heart failure for eight years. Reagan said his arteries were initially too narrow for a transplant, but Zucker opened his arteries with an experimental treatment so he could get onto the waitlist. After his surgery, Reagan became part of the "Hearty Hearts" volunteers at the hospital who advocate for organ donations and lift the spirits of other transplant recipients. Through "Hearty Hearts," he said, he has met several transplant candidates who were turned away by other hospitals but "walked out of Newark Beth Israel with a new heart, because of Mark."

Accepting more difficult cases, though, can raise the risk of adverse outcomes. According to former employees and audio recordings of staff meetings, Newark Beth Israel's transplant team worried about its one-year survival rate, which would drop below the national average in 2019. That anxiety, the employees said, appeared to underlie the team's unwillingness to sign DNR orders, since resuscitation might be needed to keep a patient alive.

Besides Andy Jurtschenko's children, two former NBI employees, including one with firsthand knowledge, said that the transplant team initially balked at a DNR order for him. By ruling out extreme measures to revive him, a DNR could conceivably have hastened Andy's death and lowered the program's one-year survival rate. Whether or how much metrics influenced Camacho's rebuff of the DNR request is unclear. While DNR orders are documented in the medical record, unapproved requests — and the reasons behind those decisions — generally aren't.

There can be few greater points of contention between physicians and families, few so infused with emotion and anguish on both sides, than whether to resuscitate someone on the verge of death. Hospitals have been sued and nursing homes fined for resuscitating patients who had a DNR order on file. Or families may urge a medical team to initiate resuscitation that a physician believes is futile, or even torture, for a patient with a terminal diagnosis. The decision is inherently subjective, and ultimately, doctors are supposed to respect the wishes of patients — or, if they can't speak for themselves, their health care proxy.

A heart transplant itself is an act of resuscitation; there is a moment, after the old heart has been removed and the new organ not yet implanted, when the patient is only kept alive by a machine. If a surgery goes badly and a patient suffers serious complications, a request for a DNR can become a flashpoint, revealing the pressures that transplant teams are under to save patients, make sure that scarce donated organs don't go to waste and meet performance metrics.

More so than almost any other field of medicine, organ transplantation in the United States is tightly regulated. The U.S. Centers for Medicare and Medicaid Services can halt reimbursement if they find a program falling short of their standards. Transplant programs in the U.S. are constantly aware of the importance of keeping their survival rates at or above national medians, said a heart transplant surgeon at another hospital in the Northeast. "We are only judged on the numbers, on the results," said the surgeon, who requested anonymity to avoid jeopardizing future job opportunities. "My main job is to treat percentages, not patients."

In that context, medical staff can be reluctant to relinquish the option of resuscitation. "The 365-day problem is real, and if you gave truth serum to every transplant doctor in America, including me, and you asked if we didn't all keep an eye on it, and if we said we didn't, that would be a lie," said Dr. David Weill, a consultant to transplant programs and former director of the heart-lung transplant program at Stanford University Medical Center. As the one-year anniversary approaches, Weill added, he's seen doctors tell patients, "'It's too early [for a DNR], I've seen people recover, she has a strong will to live,' these kind of things."

Such attitudes extend beyond transplant wards to other types of cardiovascular surgery that judge success by survival metrics, though their key target is typically 30 days, not a year. An acute care nurse practitioner with more than two decades of experience in open-heart surgery care told ProPublica she's worked in several cardiac wards where families were discouraged from withdrawing life support or asking for a DNR before 30 days had elapsed.

"Families are lied to, they'll be told, 'It's too soon,' then after 30 days, doctors will say, 'Make them a DNR, go ahead,'" said the nurse, who requested anonymity because she feared retaliation from her employer. (She hasn't worked at Newark Beth Israel.) "Unless you have a medical directive and, if you're unable to speak for yourself, a relative with power of attorney, the hospital's driving the bus."

Metrics aren't the only reason that surgeons balk at DNR requests. They are trained to do everything they can to save lives. When a patient dies, while other doctors might blame a treatment protocol or the deceased's lifestyle, surgeons often feel personally responsible, according to Dr. Gretchen Schwarze, associate professor of vascular surgery at the University of Wisconsin School of Medicine and Public Health. In a 2010 study, Schwarze posed hypothetical scenarios to 10 surgeons, including a transplant specialist. One scenario described a patient who remained intubated and on a feeding tube a week after surgery. A family member then produces a previously undisclosed directive from the patient, asking not to be sustained by life support.

The surgeons "expressed significant emotional reaction," including "betrayal, unhappiness, disappointment," Schwarze reported. They felt that there was an implicit contract that patients entered into when signing up for surgery. One surgeon put it this way: "There is a commitment made by both the patient and the surgeon to get through the operation, as well as all of the post-operative issues that come up." While acknowledging that they were being "paternalistic" and contradicting the patient's directives, several of the surgeons in the study said they would refuse to withdraw life support in Schwarze's scenario.

One reason for conflicts over DNRs is that most patients aren't asked ahead of time about their desires for post-operative care. A 2010 study of admissions at two U.S. hospitals found that only one-third of seriously ill patients were asked what code status they wanted if they went into cardiac or respiratory arrest, and those conversations on average lasted one minute.

"We tend to wait too long to have a meaningful conversation. We wait until someone's been in the ICU for a while, until the point that the patient cannot participate in the conversation," said Luke Adams, a critical care nurse in Pennsylvania who founded a company called Advanced Care Solutions, which helps people write advance directives and navigate end-of-life decisions.

During a heart transplant operation, a patient cannot have a DNR order in place. The new heart may need to be shocked or stimulated with chest compressions to help it function properly. These actions would be considered "resuscitation."

After surgery, however, a patient should be able to ask for a DNR order at any time, and in the case of patients who can't speak for themselves, the request can come from a surrogate, usually the closest relative. A DNR order must be signed by a physician and put in the medical record. In New Jersey, DNR orders are typically entered as part of a form called a POLST, short for Practitioner Orders for Life-Sustaining Treatment, which also includes sections for other medical directives, including whether to use artificial nutrition and antibiotics.

Resuscitation is no assurance of either immediate or long-term improvement. A 2003 study of in-hospital resuscitation tracked 14,720 cardiac arrests and found that 44% of patients regained circulation, but only 17% recovered enough to be discharged from the hospital.

Darryl Young suffered brain damage during a heart transplant at Newark Beth Israel and never woke up. But, hardly consulting his family, doctors kept him alive for a year to avoid federal scrutiny.

Sometimes a DNR order is the obvious choice, said Dr. Perla Macip-Rodriguez, assistant professor of internal medicine at the Boston University School of Medicine, who specializes in palliative care and geriatrics. She gave the example of a patient with late-stage cancer that is no longer treatable. If the patient's heart stopped, CPR would not affect the terminal diagnosis. "It's just going to prolong their dying process."

Other cases are murkier. Andy Jurtschenko's brain was severely damaged. If his heart stopped and he was resuscitated, nobody could know what the outcome would be.

When the prognosis is uncertain, the medical team should focus on the patient's individual goals, said Dr. Jessica Zitter, an internist at Highland Hospital in Oakland, California, who practices both critical care and palliative care. "We should never, ever force treatment on a patient or their surrogate," she said. "There's no way you can ever insist on someone remaining on a machine or being resuscitated so long as the family has the patient's best interest in mind. That's called autonomy."

Megan Jurtschenko's fondest childhood memories involve visiting her father's carpet store, climbing on the stacks of samples and running her fingers over soft rugs while listening to him bantering and laughing with his clients.

The U.S.-born son of Eastern European immigrants, Andy Jurtschenko worked his way up through local furniture chains until he was able to open his own store, Route 46 Carpet, in West Paterson, New Jersey. His gift for patter made him a natural salesman. Divorced, he stayed close to his children. On his Facebook page, he described himself as "a caring American."

But in 2012, Andy began to feel tired and short of breath. Doctors diagnosed him with congestive heart failure. By 2014, as his heart became weaker, Andy had to stop working and go on disability. He was 57 years old. It frustrated him to rely on the government for financial support.

"All he ever wanted was to care for himself and his family. Losing that was not easy for him," Megan said.

Megan and Chris knew Andy would not want to be kept alive by extraordinary means. Just surviving wasn't his goal. His children wanted his caregivers to understand this as well.

On May 21, Andy was transferred from Newark Beth Israel to a rehab facility, JFK Johnson Rehabilitation Institute. Within a few days, a staff member of the facility called, asking for medical directives.

"They described him as very high risk for code, and they wanted to establish what we wanted," Megan recalled. If his heart were to stop, she instructed them, "they would not do any chest compressions, no electric shock. And if his lungs stopped functioning they would not put in a breathing tube. We wanted no artificial care."

Chris noted the difference between the rehab facility and the hospital. JFK "came to us and said we need to put this in place, where Newark kind of avoided it," he said.

Hackensack Meridian Health, a New Jersey network that includes JFK Johnson, "has comprehensive DNR policies that we consider an essential part of providing high-quality patient care," spokeswoman Mary Jo Layton said in an email. "Our teams work collaboratively with patients, their families and loved ones to ensure our patients' wishes are honored."

In September 2018, five months after Andy's surgery, another Newark Beth Israel patient, Darryl Young, suffered brain damage during a heart transplant. As ProPublica has reported, Zucker instructed his staff to keep Young alive and not to discuss palliative care options, such as hospice, with his family until the one-year anniversary of his surgery.

If Young were to die, the hospital's annual one-year survival rate, already at its lowest in a decade, would drop even further. In a previously unreported audio recording obtained by ProPublica, Zucker told his staff at a meeting in April 2019, "You can send him back to a rehab facility, but if you make him DNR at the rehab facility — they will make him DNR — it'll be a problem for us."

Zucker continued, "So, I don't really know what to do except to tell you that I recognize what I'm asking, I said myself I'm not sure that this is ethical, moral or right, but for the global good of the future transplant recipients, the others who come along, this program can't be put into an SIA."

SIA stands for systems improvement agreement, a process through which CMS can force a transplant program to get back into compliance. It typically costs a hospital at least $2 million. After the April meeting, Zucker's team never released Young to a rehab center, and he remains at the hospital. Young's sister, Andrea, did not request a DNR, and the team only discussed the option with her after ProPublica's article was published, she said.

The Centers for Medicare and Medicaid Services, the FBI, the New Jersey Department of Health and the State Board of Medical Examiners are investigating the hospital's treatment of Young. The hospital has placed Zucker on leave pending the results of its own review.

Newark Beth Israel and Zucker did not respond to questions about his comments regarding a DNR for Young.

On June 1, 2018, Megan Jurtschenko went to visit her father at the rehab facility, but his room was empty. After being transported to Newark Beth Israel the previous day for a checkup, he hadn't returned to JFK Johnson.

Panicked, Megan grabbed an Uber to the hospital. There, she raced to the fourth floor, sneakers pounding against the floor as she beelined for the cardiology unit. By that point, Andy had regained consciousness and could sometimes nod yes or no, but he easily became agitated, according to his children. Andy's medical record from that day described him as "non-verbal."

Megan went to her father's room first "to let him know that he wasn't alone" and then to the nurse's station, where two physicians met her. She said they told her that her father had a fever when he arrived for his checkup, so they kept him overnight and started a course of antibiotics. The hospital suspected pneumonia, according to his medical record.

Nobody at the hospital had called Megan, Chris or anyone else in the family to let them know that Andy was being admitted or treated, his children said. A few days later, a handwritten note in Andy's medical record said that the family was "upset" that they weren't informed.

Megan said she turned to a nurse to ask, "What's his code status?"

"Oh don't worry, he is at full code," the nurse responded, meaning that if Andy stopped breathing or his heart stopped, they would do everything possible to resuscitate him.

Her words riveted Megan to the floor.

"I said, 'Oh, no, that is the complete opposite of what we want.'"

Newark Beth Israel didn't respond to questions about this incident.

Andy needed to stay in the hospital until his infection cleared. That Friday, Megan refused to leave the hospital until she got a DNR order, so a cardiologist signed one. "Extensive discussion with daughter — Patient is DNR," a note in the medical record states. The family scheduled a formal meeting to discuss the order on the following Monday with a palliative care nurse practitioner and Zucker, the transplant program's director.

Chris was out of town, so Megan and her aunt, Anna DeMarinis, attended the meeting. Megan reminded Zucker that her father's goals for the transplant had been to resume his normal life, she said. She reiterated that the family wasn't asking to withdraw his feeding tube or medications. They just didn't want Andy to endure what they saw as extraordinary measures.

As Camacho had, Zucker urged caution. "He looked at me and said: 'You know, this is a very serious document. It's very extreme. Are you sure you want to do this?'" Megan recalled.

She wondered if Zucker was appealing to her emotions. "I got the impression he thought I was going to be the one who was going to say, 'No, I don't want to lose my Dad, let's keep trying!'"

Megan wasn't swayed. "I could see what was actually happening, which is — he was suffering and that's the only thing I saw." The palliative care nurse at the meeting revised and signed Andy's POLST form, indicating that Andy would continue full medical treatment unless he stopped breathing or had no pulse. In the CPR section, she checked the box for "Do not attempt resuscitation."

Even then, Zucker wouldn't let it go. After returning from his trip, Chris was sitting by his father's bedside when Zucker came by. Unprompted, Zucker brought up the DNR.

"He said, 'So, you're giving up on him.'" Chris recalled.

Chris didn't respond. "I was shocked, for a doctor to throw that in my face," he said.

Newark Beth Israel and Zucker did not respond to questions about the meeting with Megan or Zucker's conversation with Chris. "Mr. Jurtschenko's medical record confirms that his family's decision to implement a DNR in June rather than in April did not change his outcome," Kamateh said.

Andy Jurtschenko eventually returned to JFK Johnson. Over the next few months, during which he was moved to a different long-term care facility, Andy's condition improved modestly, until he was able to track people with his eyes and have short conversations.

But his gains soon plateaued. He was never able to stand, sit or eat on his own. He started hallucinating, seeing people who weren't there. Over time, his hands began to atrophy and curl in on themselves, despite attempts at physical therapy.

Andy's family members took turns visiting, making sure he had company nearly every day. Sitting by his side, they tried to engage him by flipping through photo albums, playing music from his favorite band, AC/DC, and watching sports on TV. They ached to see his misery. The spark in her younger brother's eyes was gone, DeMarinis said. Holding back tears, she said, "When he did become cogent, he just kept reiterating what we already knew — he was suffering. He was suffering terribly."

On the morning of Oct. 31, 2018, the rehab facility phoned Chris. Andy had died at sunrise. When his heart stopped, and his breathing faded, there was no attempt to resuscitate him.

Claire Perlman and Sophie Chou contributed research and analysis to this article.

Drug companies have paid doctors billions of dollars for consulting, promotional talks, meals and more. A new ProPublica analysis finds doctors who received payments linked to specific drugs prescribed more of those drugs.

This article first appeared on Friday, December 20, 2019 in ProPublica.

By Hannah Fresques

Doctors who receive money from drugmakers related to a specific drug prescribe that drug more heavily than doctors without such financial ties, a new ProPublica analysis found. The pattern is consistent for almost all of the most widely prescribed brand-name drugs in Medicare, including drugs that treat diabetes, asthma and more.

The financial interactions include payments for delivering promotional talks, consulting and receiving sponsored meals and travel.

The 50 drugs in our analysis include many popular and expensive ones. Thirty-eight of the drugs have yearly costs exceeding $1,000 per patient, and many topped the list that are most costly for the Medicare Part D drug program.

Take Linzess, a drug to treat irritable bowel syndrome and severe constipation. From 2014 to 2018, the drug's makers, Allergan and Ironwood, spent nearly $29 million on payments to doctors related to Linzess, mostly for meals and promotional speaking fees.

ProPublica's analysis found that doctors who received payments related to Linzess in 2016 wrote 45% more prescriptions for the drug, on average, than doctors who received no payments.

Those findings were repeated for drug after drug. In 2016, doctors who received payments related to Myrbetriq, which treats overactive bladder, wrote 64% more prescriptions for the drug than those who did not. For Restasis, used to treat chronic dry eye, doctors who received payments wrote 141% more prescriptions. The pattern holds true for 46 of the 50 drugs.

On average, across all drugs, providers who received payments specifically tied to a drug prescribed it 58% more than providers who did not receive payments.

Other research, including our own, has found a correlation between payments and overall prescribing. This new analysis expands upon past work by looking individually at a variety of popular drugs. "What clearly jumps out is how consistent the association is across drugs," said Aaron Mitchell, a medical oncologist at Memorial Sloan Kettering Cancer Center who has studied pharmaceutical payments for oncology drugs.

Our analysis looked at the relationship in two ways: whether those who received payments prescribed more of a drug, as well as whether those who prescribed a drug received higher payments than those who did not. We found that, on average, physicians who prescribed a drug received higher payments related to the drug that same year than those who didn't prescribe it. For Linzess, the value of payments was more than four times higher for providers who prescribed the drug than among those who did not. For Myrbetriq, it was three times higher, and for Restasis, it was twice as much. (Read our methodology for more about the analysis.)

Holly Campbell, a spokeswoman for the Pharmaceutical Research and Manufacturers of America, an industry trade group, said it stands to reason that doctors who have interactions with a company about a drug may prescribe more of it "because they have more information about the appropriate uses for the products."

Through spokespeople, Allergan (maker of Linzess as well as Alphagan P, Bystolic, Combigan, Lumigan, Namenda and Restasis), Janssen (maker of Invega, Invokana, Xarelto and Zytiga) and Novo Nordisk (maker of Levemir, Novolog and Victoza) described their interactions with physicians as important for sharing medical information.

Novo Nordisk added that prescribing data is not used to target physicians for speaking or other promotional interactions. Eli Lilly said in a statement that meals can take place in many contexts, including in doctors' offices, at speaker events and at conferences, but didn't answer other questions. GlaxoSmithKline, Ironwood, Astellas and Purdue declined to comment.

For some drugs that are household names, it was more common for prescribers to receive a payment than not to. More than half of doctors who prescribed Breo, an expensive asthma drug, to Medicare patients received payments involving the drug in 2016. This was also true for Invokana and Victoza, both of which are diabetes medications. For Linzess, nearly half of doctors who prescribed the drug had interactions with its maker.

More than one in five doctors who prescribed OxyContin under Medicare in 2016 had a promotional interaction with the drug's manufacturer, Purdue Pharma. The company did not respond to a request for comment.

"If there are physicians out there that deny that there is a relationship, they are starting to look more and more like climate deniers in the face of the growing evidence," said Aaron Kesselheim, a professor of medicine at Harvard Medical School and an expert in pharmaceutical costs and regulation. "The association is consistent across the different types of payments. It's also consistent across numerous drug specialties and drug types, across multiple different fields of medicine. And for small and large payments. It's a remarkably durable effect. No specialty is immune from this phenomenon."

Huey Nguyen, a gastroenterologist in southern Indiana, increased his prescribing of Linzess in recent years. From 2013 to 2015, Nguyen's Medicare patients had fewer than 60 claims per year for Linzess. In 2016 and 2017, that jumped to over 110 claims per year.

Over that time, Nguyen was a promotional speaker for Linzess. Allergan paid him $1,000 in 2013, over $4,000 in both 2016 and 2017, and $2,000 in 2018 to speak about the drug.

Though Linzess has been on the market since 2012, Ironwood and Allergan made a big push to promote the drug in 2016 and 2017. Spending on doctors reached $10 million in 2016 and nearly $8 million the following year, up from under $4 million in both 2014 and 2015.

In total, Nguyen has earned $25,000 from 2014 to 2018 related to six drugs from four pharmaceutical companies, excluding meals. In 2018, he was paid by two companies to promote competing drugs that treat irritable bowel syndrome.

ProPublica's analysis did not set out to examine, nor did it resolve, whether industry payments change doctors' behavior, or if patients receive inferior care from doctors who receive payments. Many factors can influence doctors' prescribing choices. Some patients, for instance, have conditions for which only brand-name treatments are available or for which other drugs have failed.

Nguyen said promotional speaking educates doctors about how a drug works, whether insurance covers it and when not to prescribe it.

"It's a way for the primary care physicians to have access to a gastroenterologist where they can ask one-on-one questions," Nguyen said. "I'm more educated towards the drug, because I have to be trained to speak on it, so I'm more comfortable prescribing it."

Experts are skeptical that interactions between companies and doctors benefit patients. "If there really were innovations and real benefits that were accruing to patients for a new treatment, it shouldn't take so much spending by the company to get the word out," said Stacie B. Dusetzina, associate professor of health policy at Vanderbilt University Medical Center, who advised ProPublica on the design of its analysis. "I wonder if promotion is really to try to push products that have a much less substantial benefit because they're not gaining the market share naturally."

Nguyen said he takes many things into account when prescribing a drug, including its approved uses, cost and side effects. "In my day-to-day practice, my patients still come first," Nguyen said. He said the speaking engagements do not influence his prescribing, "at least not consciously. Unconsciously, I don't know." He sees the public disclosure of industry payments to doctors as a way to help patients be active participants in their care.

Nguyen said he works with companies for the extra compensation but acknowledged that "it's perfectly reasonable for people to question my motives."

ProPublica's analysis matched doctors' prescribing in Medicare's prescription drug program to the industry payments doctors received. Drug and medical device companies are required to report these payments annually through the federal Open Payments program, and they are made public on a government website. More than 600,000 doctors receive payments annually. (Companies also report research payments and ownership interests, but these were excluded from our analysis.)

Some providers were paid thousands of dollars, often for promotional speaking. But the typical doctor took in much less. Most only received meals, typically worth less than $100 per year.

In 2016, ProPublica found a relationship between the total dollar value of a doctor's interactions with drug and device companies and the overall percentage of brand-name drugs he or she prescribed.

Other research has found correlations between industry interactions and prescribing for certain classes of drugs, including opioids, urology drugs, oncology treatments, inflammatory bowel disease treatments and heartburn medication. In one study, brand-name prescribing for certain classes of drugs was associated with receiving as little as a single pharmaceutical industry sponsored meal. A study of prostate cancer treatments did not find evidence of a connection.

Brand-name drugs are more expensive than generic options, both for patients and for Medicare. A recent report from the Department of Health and Human Services found that Medicare Part D and its beneficiaries could have saved almost $3 billion by switching from brand-name drugs to generics.

Linzess is an expensive drug, costing Medicare and patients an average of about $1,500 annually. A common alternative is the laxative Miralax, available over the counter as generic polyethylene glycol, which costs less than $200 annually if taken every day. Nguyen said he recommends Miralax to many patients, but that wouldn't show up in Medicare's data because Medicare doesn't cover over-the-counter drugs. He said he often prescribes Linzess to patients who have tried Miralax and not seen the symptom relief they hoped for.

For brand-name drugs that have good generic alternatives, "every time a doctor prescribes one of these brand-name medications, it's extra money transferred from the Medicare program to the manufacturer," said Michael Barnett, assistant professor of health policy at Harvard T. H. Chan School of Public Health. "Medicare spending is out of control. And drug costs are one of the major reasons."

Drug cost can have major consequences, not just for Medicare balance sheets but also for patients' well-being. "The newest, latest drug is often not any better than the old drugs" that treat the same condition, Mitchell said. "But the new drugs are always more expensive. That really hurts patients' pocketbooks. You've got physicians prescribing more expensive drugs and patients who aren't taking them as a result. A generic medicine that's cheaper that a patient does take is a whole lot more effective than an on-brand, expensive medicine that they don't take.