At a certain point, it was no longer a matter of if the United States would reach the gruesome milestone of 1 in 500 people dying of covid-19, but a matter of when. A year? Maybe 15 months? The answer: 19 months.
Given the mortality rate from covid and our nation’s population size, “we’re kind of where we predicted we would be with completely uncontrolled spread of infection,” said Jeffrey D. Klausner, clinical professor of medicine, population and public health sciences at the University of Southern California’s Keck School of Medicine. “Remember at the very beginning, which we don’t hear about anymore, it was all about flatten the curve.”
The idea, he said, was to prevent “the humanitarian disaster” that occurred in New York City, where ambulance sirens were a constant as hospitals were overwhelmed and mortuaries needed mobile units to handle the additional dead.
The goal of testing, mask-wearing, keeping six feet apart and limiting gatherings was to slow the spread of the highly infectious virus until a vaccine could stamp it out. The vaccines came but not enough people have been immunized, and the triumph of science waned as mass death and disease remain. The result: As the nation’s covid death toll exceeded 663,000 this week, it meant roughly 1 in every 500 Americans had succumbed to the disease caused by the coronavirus.
While covid’s death toll overwhelms the imagination, even more stunning is the deadly efficiency with which it has targeted Black, Latino, and American Indian and Alaska Native people in their 30s, 40s and 50s.
Death at a younger age represents more lost years of life. Lost potential. Lost scholarship. Lost mentorship. Lost earnings. Lost love.
Neighborhoods decimated. Families destroyed.
“So often when we think about the majority of the country who have lost people to covid-19, we think about the elders that have been lost, not necessarily younger people,” said Abigail Echo-Hawk, executive vice president at the Seattle Indian Health Board and director of the Urban Indian Health Institute. “Unfortunately, this is not my reality nor that of the Native community. I lost cousins and fathers and tribal leaders. People that were so integral to building up our community, which has already been struggling for centuries against all these things that created the perfect environment for covid-19 to kill us.”
Six of Echo-Hawk’s friends and relatives — all under 55 — have died of covid.
“This is trauma. This is generational impact that we must have an intentional focus on. The scars are there,” said Marcella Nunez-Smith, chair of President Biden’s COVID-19 Health Equity Task Force and associate dean for health equity research at Yale University. “We can’t think that we’re going to test and vaccinate our way out of this deep pain and hurt.”
The pandemic has brought into stark relief centuries of entwining social, environmental, economic and political factors that erode the health and shorten the lives of people of color, putting them at higher risk of the chronic conditions that leave immune systems vulnerable to the coronavirus. Many of those same factors fuel the misinformation, mistrust and fear that leave too many unprotected.
Take the suggestion that people talk to their doctor about which symptoms warrant testing or a trip to the hospital as well as the safety of vaccines. Seems simple. It’s not.
Many people don’t have a physician they see regularly due in part to significant provider shortages in communities of color. If they do have a doctor, it can cost too much money for a visit even if insured. There are language barriers for those who don’t speak English fluently and fear of deportation among undocumented immigrants.
“Some of the issues at hand are structural issues, things that are built into the fabric of society,” said Enrique W. Neblett Jr., a University of Michigan professor who studies racism and health.
Essential workers who cannot avoid the virus in their jobs because they do not have the luxury of working from home. People living in multigenerational homes with several adult wage-earners, sharing housing because their pay is so low. Even the fight to be counted among the covid casualties — some states and hospitals, Echo-Hawk said, don’t have “even a box to check to say you are American Indian or Alaskan Native.”
It can be difficult to tackle the structural issues influencing the unequal burden of the pandemic while dealing with the day-to-day stress and worry it ignites, which, Neblett said, is why attention must focus on both long-term solutions and “what do we do now? It’s not just that simple as, ‘Oh, you just put on your mask, and we’ll all be good.’ It’s more complicated than that.”
The exacting toll of the last year and a half — covid’s stranglehold on communities of color and George Floyd’s murder — forced the country to interrogate the genealogy of American racism and its effect on health and well-being.
“This is an instance where we finally named it and talked about structural racism as a contributing factor in ways that we haven’t with other health disorders,” Neblett said.
But the nation’s attention span can be short. Polls show there was a sharp rise in concern about discrimination against Black Americans by police following Floyd’s murder, including among White Americans. That concern has eroded some since 2020, though it does remain higher than years past.
“This mistaken understanding that people have, almost this sort of impatience like, ‘Oh, we see racism. Let’s just fix that,’ that’s the thing that gives me hives,” Nunez-Smith said. “This is about generational investments and fundamental changes in ways of being. We didn’t get here overnight.”
Exactly 300 years ago, in 1721, Benjamin Franklin and his fellow American colonists faced a deadly smallpox outbreak. Their varying responses constitute an eerily prescient object lesson for today’s world, similarly devastated by a virus and divided over vaccination three centuries later.
As a microbiologist and a Franklin scholar, we see some parallels between then and now that could help governments, journalists and the rest of us cope with the coronavirus pandemic and future threats.
What was new, at least to Boston, was a simple procedure that could protect people from the disease. It was known as “variolation” or “inoculation,” and involved deliberately exposing someone to the smallpox “matter” from a victim’s scabs or pus, injecting the material into the skin using a needle. This approach typically caused a mild disease and induced a state of “immunity” against smallpox.
Even today, the exact mechanism is poorly understood and not muchresearch on variolation has been done. Inoculation through the skin seems to activate an immune response that leads to milder symptoms and less transmission, possibly because of the route of infection and the lower dose. Since it relies on activating the immune response with live smallpox variola virus, inoculation is different from the modern vaccination that eradicated smallpox using the much less harmful but related vaccinia virus.
Known primarily as a Congregational minister, Mather was also a scientist with a special interest in biology. He paid attention when Onesimus told him “he had undergone an operation, which had given him something of the smallpox and would forever preserve him from it; adding that it was often used” in West Africa, where he was from.
Inspired by this information from Onesimus, Mather teamed up with a Boston physician, Zabdiel Boylston, to conduct a scientific study of inoculation’s effectiveness worthy of 21st-century praise. They found that of the approximately 300 people Boylston had inoculated, 2% had died, compared with almost 15% of those who contracted smallpox from nature.
The findings seemed clear: Inoculation could help in the fight against smallpox. Science won out in this clergyman’s mind. But others were not convinced.
Stirring up controversy
A local newspaper editor named James Franklin had his own affliction – namely an insatiable hunger for controversy. Franklin, who was no fan of Mather, set about attacking inoculation in his newspaper, The New-England Courant.
One article from August 1721 tried to guilt readers into resisting inoculation. If someone gets inoculated and then spreads the disease to someone else, who in turn dies of it, the article asked, “at whose hands shall their Blood be required?” The same article went on to say that “Epidemeal Distempers” such as smallpox come “as Judgments from an angry and displeased God.”
In contrast to Mather and Boylston’s research, the Courant’s articles were designed not to discover, but to sow doubt and distrust. The argument that inoculation might help to spread the disease posits something that was theoretically possible – at least if simple precautions were not taken – but it seems beside the point. If inoculation worked, wouldn’t it be worth this small risk, especially since widespread inoculations would dramatically decrease the likelihood that one person would infect another?
Franklin, the Courant’s editor, had a kid brother apprenticed to him at the time – a teenager by the name of Benjamin.
Historians don’t know which side the younger Franklin took in 1721 – or whether he took a side at all – but his subsequent approach to inoculation years later has lessons for the world’s current encounter with a deadly virus and a divided response to a vaccine.
That he was capable of overcoming this inclination shows Benjamin Franklin’s capacity for independent thought, an asset that would serve him well throughout his life as a writer, scientist and statesman. While sticking with social expectations confers certain advantages in certain settings, being able to shake off these norms when they are dangerous is also valuable. We believe the most successful people are the ones who, like Franklin, have the intellectual flexibility to choose between adherence and independence.
Perhaps the inoculation controversy of 1721 had helped him to understand an unfortunate phenomenon that continues to plague the U.S. in 2021: When people take sides, progress suffers. Tribes, whether long-standing or newly formed around an issue, can devote their energies to demonizing the other side and rallying their own. Instead of attacking the problem, they attack each other.
Franklin, in fact, became convinced that inoculation was a sound approach to preventing smallpox. Years later he intended to have his son Francis inoculated after recovering from a case of diarrhea. But before inoculation took place, the 4-year-old boy contracted smallpox and died in 1736. Citing a rumor that Francis had died because of inoculation and noting that such a rumor might deter parents from exposing their children to this procedure, Franklin made a point of setting the record straight, explaining that the child had “receiv’d the Distemper in the common Way of Infection.”
Writing his autobiography in 1771, Franklin reflected on the tragedy and used it to advocate for inoculation. He explained that he “regretted bitterly and still regret” not inoculating the boy, adding, “This I mention for the sake of parents who omit that operation, on the supposition that they should never forgive themselves if a child died under it; my example showing that the regret may be the same either way, and that, therefore, the safer should be chosen.”
A scientific perspective
A final lesson from 1721 has to do with the importance of a truly scientific perspective, one that embraces science, facts and objectivity.
Inoculation was a relatively new procedure for Bostonians in 1721, and this lifesaving method was not without deadly risks. To address this paradox, several physicians meticulously collected data and compared the number of those who died because of natural smallpox with deaths after smallpox inoculation. Boylston essentially carried out what today’s researchers would call a clinical study on the efficacy of inoculation. Knowing he needed to demonstrate the usefulness of inoculation in a diverse population, he reported in a short book how he inoculated nearly 300 individuals and carefully noted their symptoms and conditions over days and weeks.
The recent emergency-use authorization of mRNA-based and viral-vector vaccines for COVID-19 has produced a vast array of hoaxes, false claims and conspiracy theories, especially in various social media. Like 18th-century inoculations, these vaccines represent new scientific approaches to vaccination, but ones that are based on decades of scientific research and clinical studies.
We suspect that if he were alive today, Benjamin Franklin would want his example to guide modern scientists, politicians, journalists and everyone else making personal health decisions.Like Mather and Boylston, Franklin was a scientist with a respect for evidence and ultimately for truth.
When it comes to a deadly virus and a divided response to a preventive treatment, Franklin was clear what he would do. It doesn’t take a visionary like Franklin to accept the evidence of medical science today.
As the delta variant of the coronavirus spreads, especially among the unvaccinated, the Biden administration is gearing up for a new push to vaccinate the so-called “movable middle”—and some public health experts say FDA could advance that goal by fully approving Covid-19 vaccines.
Analysis reveals toll of US Covid-19 deaths among unvaccinated patients
According to an analysis by the Associated Press, nearly all recent Covid-19 deaths have occurred in unvaccinated individuals.
The AP analysis is based on data from CDC, although CDC has not itself released estimates of the share of Covid-19 deaths among unvaccinated patients.
According to the AP analysis, just 0.8% of Covid-19 deaths in May were among the fully vaccinated. Meanwhile, the share of hospitalized patients who were fully vaccinated was just 0.1% in May, with fewer than 1,200 fully vaccinated people hospitalized out of more than 853,000 hospitalizations.
Meanwhile, according to CDC, 54% of the U.S. population, including 66% of American adults, have received at least one dose of a Covid-19 vaccine, while 46.1% of the total population and 56.8% of American adults have received all required doses.
One hesitates to elevate obviously bad arguments, even to point out how bad they are. This is a conundrum that comes up a lot these days, as members of the media measure the utility of reporting on bad faith, disingenuous or simply bizarre claims.
If someone were to insist, for example, that they were not going to get the coronavirus vaccine solely to spite the political left, should that claim be elevated? Can we simply point out how deranged it is to refuse a vaccine that will almost certainly end an international pandemic simply because people with whom you disagree think that maybe this is a good route to end that pandemic? If someone were to write such a thing at some attention-thirsty website, we certainly wouldn’t want to link to it, leaving our own readers having to figure out where it might be found should they choose to do so.
In this case, it’s worth elevating this argument (which, to be clear, is actually floating out there) to point out one of the myriad ways in which the effort to vaccinate as many adults as possible has become interlaced with partisan politics. As the weeks pass and demand for the vaccine has tapered off, the gap between Democratic and Republican interest in being vaccinated seems to be widening — meaning that the end to the pandemic is likely to move that much further into the future.
Consider, for example, the rate of completed vaccinations by county, according to data compiled by CovidActNow. You can see a slight correlation between how a county voted in 2020 — the horizontal axis — and the density of completed vaccinations, shown on the vertical. There’s a greater density of completed vaccinations on the left side of the graph than on the right.
If we shift to the percentage of the population that’s received even one dose of the vaccine, the effect is much more obvious.
This is a relatively recent development. At the beginning of the month, the density of the population that had received only one dose resulted in a graph that looked much like the current density of completed doses.
If we animate those two graphs, the effect is obvious. In the past few weeks, the density of first doses has increased much faster in more-Democratic counties.
If we group the results of the 2020 presidential contest into 20-point buckets, the pattern is again obvious.
It’s not a new observation that Republicans are less willing to get the vaccine; we’ve reported on it repeatedly. What’s relatively new is how that hesitance is showing up in the actual vaccination data.
A Post-ABC News poll released on Monday showed that this response to the vaccine holds even when considering age groups. We’ve known for a while that older Americans, who are more at risk from the virus, have been more likely to seek the vaccine. But even among seniors, Republicans are significantly more hesitant to receive the vaccine than are Democrats.
This is a particularly dangerous example of partisanship. People 65 or older have made up 14 percent of coronavirus infections, according to federal data, but 81 percent of deaths. That’s among those for whom ages are known, a subset (though a large majority) of overall cases. While about 1.8 percent of that overall group has died, the figure for those aged 65 and over is above 10 percent.
As vaccines have been rolled out across the country, you can see how more-heavily-blue counties have a higher density of vaccinations in many states.
This is not a universal truth, of course. Some heavily Republican counties have above-average vaccination rates. (About 40 percent of counties that preferred former president Donald Trump last year are above the average in the CovidActNow data. The rate among Democratic counties is closer to 80 percent.) But it is the case that there is a correlation between how a county voted and how many of its residents have been vaccinated. It is also the case that the gap between red and blue counties is widening.
Given all of that, it probably makes sense to point out that an argument against vaccines based on nothing more than “lol libs will hate this” is an embarrassing argument to make.
Of 26 health systems surveyed by MedCity News, nearly half used automated tools to respond to the Covid-19 pandemic, but none of them were regulated. Even as some hospitals continued using these algorithms, experts cautioned against their use in high-stakes decisions.
A year ago, Michigan Medicine faced a dire situation. In March of 2020, the health system predicted it would have three times as many patients as its 1,000-bed capacity — and that was the best-case scenario. Hospital leadership prepared for this grim prediction by opening a field hospital in a nearby indoor track facility, where patients could go if they were stable, but still needed hospital care. But they faced another predicament: How would they decide who to send there?
Two weeks before the field hospital was set to open, Michigan Medicine decided to use a risk model developed by Epic Systems to flag patients at risk of deterioration. Patients were given a score of 0 to 100, intended to help care teams determine if they might need an ICU bed in the near future. Although the model wasn’t developed specifically for Covid-19 patients, it was the best option available at the time, said Dr. Karandeep Singh, an assistant professor of learning health sciences at the University of Michigan and chair of Michigan Medicine’s clinical intelligence committee. But there was no peer-reviewed research to show how well it actually worked.
Researchers tested it on over 300 Covid-19 patients between March and May. They were looking for scores that would indicate when patients would need to go to the ICU, and if there was a point where patients almost certainly wouldn’t need intensive care.
“We did find a threshold where if you remained below that threshold, 90% of patients wouldn’t need to go to the ICU,” Singh said. “Is that enough to make a decision on? We didn’t think so.”
But if the number of patients were to far exceed the health system’s capacity, it would be helpful to have some way to assist with those decisions.
“It was something that we definitely thought about implementing if that day were to come,” he said in a February interview.
Thankfully, that day never came.
The survey Michigan Medicine is one of 80 hospitals contacted by MedCity News between January and April in a survey of decision-support systems implemented during the pandemic. Of the 26 respondents, 12 used machine learning tools or automated decision systems as part of their pandemic response. Larger hospitals and academic medical centers used them more frequently.
Faced with scarcities in testing, masks, hospital beds and vaccines, several of the hospitals turned to models as they prepared for difficult decisions. The deterioration index created by Epic was one of the most widely implemented — more than 100 hospitals are currently using it — but in many cases, hospitals also formulated their own algorithms.
They built models to predict which patients were most likely to test positive when shortages of swabs and reagents backlogged tests early in the pandemic. Others developed risk-scoring tools to help determine who should be contacted first for monoclonal antibody treatment, or which Covid patients should be enrolled in at-home monitoring programs.
MedCity News also interviewed hospitals on their processes for evaluating software tools to ensure they are accurate and unbiased. Currently, the FDA does not require some clinical decision-support systems to be cleared as medical devices, leaving the developers of these tools and the hospitals that implement them responsible for vetting them.
Among the hospitals that published efficacy data, some of the models were only evaluated through retrospective studies. This can pose a challenge in figuring out how clinicians actually use them in practice, and how well they work in real time. And while some of the hospitals tested whether the models were accurate across different groups of patients — such as people of a certain race, gender or location — this practice wasn’t universal.
As more companies spin up these models, researchers cautioned that they need to be designed and implemented carefully, to ensure they don’t yield biased results.
An ongoing review of more than 200 Covid-19 risk-prediction models found that the majority had a high risk of bias, meaning the data they were trained on might not represent the real world.
“It’s that very careful and non-trivial process of defining exactly what we want the algorithm to be doing,” said Ziad Obermeyer, an associate professor of health policy and management at UC Berkeley who studies machine learning in healthcare. “I think an optimistic view is that the pandemic functions as a wakeup call for us to be a lot more careful in all of the ways we’ve talked about with how we build algorithms, how we evaluate them, and what we want them to do.”
Algorithms can’t be a proxy for tough decisions Concerns about bias are not new to healthcare. In a paper published two years ago, Obermeyer found a tool used by several hospitals to prioritize high-risk patients for additional care resources was biased against Black patients. By equating patients’ health needs with the cost of care, the developers built an algorithm that yielded discriminatory results.
More recently, a rule-based system developed by Stanford Medicine to determine who would get the Covid-19 vaccine first ended up prioritizing administrators and doctors who were seeing patients remotely, leaving out most of its 1,300 residents who had been working on the front lines. After an uproar, the university attributed the errors to a “complex algorithm,” though there was no machine learning involved.
Both examples highlight the importance of thinking through what exactly a model is designed to do — and not using them as a proxy to avoid the hard questions.
“The Stanford thing was another example of, we wanted the algorithm to do A, but we told it to do B. I think many health systems are doing something similar,” Obermeyer said. “You want to give the vaccine first to people who need it the most — how do we measure that?”
The urgency that the pandemic created was a complicating factor. With little information and few proven systems to work with in the beginning, health systems began throwing ideas at the wall to see what works. One expert questioned whether people might be abdicating some responsibility to these tools.
“Hard decisions are being made at hospitals all the time, especially in this space, but I’m worried about algorithms being the idea of where the responsibility gets shifted,” said Varoon Mathur, a technology fellow at NYU’s AI Now Institute, in a Zoom interview. “Tough decisions are going to be made, I don’t think there are any doubts about that. But what are those tough decisions? We don’t actually name what constraints we’re hitting up against.”
The wild, wild west There currently is no gold standard for how hospitals should implement machine learning tools, and little regulatory oversight for models designed to support physicians’ decisions, resulting in an environment that Mathur described as the “wild, wild west.”
How these systems were used varied significantly from hospital to hospital.
Early in the pandemic, Cleveland Clinic used a model to predict which patients were most likely to test positive for the virus as tests were limited. Researchers developed it using health record data from more than 11,000 patients in Ohio and Florida, including 818 who tested positive for Covid-19. Later, they created a similar risk calculator to determine which patients were most likely to be hospitalized for Covid-19, which was used to prioritize which patients would be contacted daily as part of an at-home monitoring program.
Initially, anyone who tested positive for Covid-19 could enroll in this program, but as cases began to tick up, “you could see how quickly the nurses and care managers who were running this program were overwhelmed,” said Dr. Lara Jehi, Chief Research Information Officer at Cleveland Clinic. “When you had thousands of patients who tested positive, how could you contact all of them?”
While the tool included dozens of factors, such as a patient’s age, sex, BMI, zip code, and whether they smoked or got their flu shot, it’s also worth noting that demographic information significantly changed the results. For example, a patient’s race “far outweighs” any medical comorbidity when used by the tool to estimate hospitalization risk, according to a paper published in Plos One. Cleveland Clinic recently made the model available to other health systems.
Others, like Stanford Health Care and 731-bed Santa Clara County Medical Center, started using Epic’s clinical deterioration index before developing their own Covid-specific risk models. At one point, Stanford developed its own risk-scoring tool, which was built using past data from other patients who had similar respiratory diseases, such as the flu, pneumonia, or acute respiratory distress syndrome. It was designed to predict which patients would need ventilation within two days, and someone’s risk of dying from the disease at the time of admission.
Stanford tested the model to see how it worked on retrospective data from 159 patients that were hospitalized with Covid-19, and cross-validated it with Salt Lake City-based Intermountain Healthcare, a process that took several months. Although this gave some additional assurance — Salt Lake City and Palo Alto have very different populations, smoking rates and demographics — it still wasn’t representative of some patient groups across the U.S.
“Ideally, what we would want to do is run the model specifically on different populations, like on African Americans or Hispanics and see how it performs to ensure it’s performing the same for different groups,” Tina Hernandez-Boussard, an associate professor of medicine, biomedical data science and surgery at Stanford, said in a February interview. “That’s something we’re actively seeking. Our numbers are still a little low to do that right now.”
Stanford planned to implement the model earlier this year, but ultimately tabled it as Covid-19 cases fell.
‘The target is moving so rapidly’ Although large medical centers were more likely to have implemented automated systems, there were a few notable holdouts. For example, UC San Francisco Health, Duke Health and Dignity Health all said they opted not to use risk-prediction models or other machine learning tools in their pandemic responses.
“It’s pretty wild out there and I’ll be honest with you — the dynamics are changing so rapidly,” said Dr. Erich Huang, chief officer for data quality at Duke Health and director of Duke Forge. “You might have a model that makes sense for the conditions of last month but do they make sense for the conditions of next month?”
That’s especially true as new variants spread across the U.S., and more adults are vaccinated, changing the nature and pace of the disease. But other, less obvious factors might also affect the data. For instance, Huang pointed to big differences in social mobility across the state of North Carolina, and whether people complied with local restrictions. Differing social and demographic factors across communities, such as where people work and whether they have health insurance, can also affect how a model performs.
“There are so many different axes of variability, I’d feel hard pressed to be comfortable using machine learning or AI at this point in time,” he said. “We need to be careful and understand the stakes of what we’re doing, especially in healthcare.”
Leadership at one of the largest public hospitals in the U.S., 600-bed LAC+USC Medical Center in Los Angeles, also steered away from using predictive models, even as it faced an alarming surge in cases over the winter months.
At most, the hospital used alerts to remind physicians to wear protective equipment when a patient has tested positive for Covid-19.
“My impression is that the industry is not anywhere near ready to deploy fully automated stuff just because of the risks involved,” said Dr. Phillip Gruber, LAC+USC’s chief medical information officer. “Our institution and a lot of institutions in our region are still focused on core competencies. We have to be good stewards of taxpayer dollars.”
When the data itself is biased Developers have to contend with the fact that any model developed in healthcare will be biased, because the data itself is biased; how people access and interact with health systems in the U.S. is fundamentally unequal.
How that information is recorded in electronic health record systems (EHR) can also be a source of bias, NYU’s Mathur said. People don’t always self-report their race or ethnicity in a way that fits neatly within the parameters of an EHR. Not everyone trusts health systems, and many people struggle to even access care in the first place.
“Demographic variables are not going to be sharply nuanced. Even if they are… in my opinion, they’re not clean enough or good enough to be nuanced into a model,” Mathur said.
The information hospitals have had to work with during the pandemic is particularly messy. Differences in testing access and missing demographic data also affect how resources are distributed and other responses to the pandemic.
“It’s very striking because everything we know about the pandemic is viewed through the lens of number of cases or number of deaths,” UC Berkeley’s Obermeyer said. “But all of that depends on access to testing.”
At the hospital level, internal data wouldn’t be enough to truly follow whether an algorithm to predict adverse events from Covid-19 was actually working. Developers would have to look at social security data on mortality, or whether the patient went to another hospital, to track down what happened.
“What about the people a physician sends home — if they die and don’t come back?” he said.
Researchers at Mount Sinai Health System tested a machine learning tool to predict critical events in Covid-19 patients — such as dialysis, intubation or ICU admission — to ensure it worked across different patient demographics. But they still ran into their own limitations, even though the New York-based hospital system serves a diverse group of patients.
They tested how the model performed across Mount Sinai’s different hospitals. In some cases, when the model wasn’t very robust, it yielded different results, said Benjamin Glicksberg, an assistant professor of genetics and genomic sciences at Mount Sinai and a member of its Hasso Plattner Institute for Digital Health.
They also tested how it worked in different subgroups of patients to ensure it didn’t perform disproportionately better for patients from one demographic.
“If there’s a bias in the data going in, there’s almost certainly going to be a bias in the data coming out of it,” he said in a Zoom interview. “Unfortunately, I think it’s going to be a matter of having more information that can approximate these external factors that may drive these discrepancies. A lot of that is social determinants of health, which are not captured well in the EHR. That’s going to be critical for how we assess model fairness.”
Even after checking for whether a model yields fair and accurate results, the work isn’t done yet. Hospitals must continue to validate continuously to ensure they’re still working as intended — especially in a situation as fast-moving as a pandemic.
A bigger role for regulators All of this is stirring up a broader discussion about how much of a role regulators should have in how decision-support systems are implemented.
Of the hospitals surveyed by MedCity News, none of the models they developed had been cleared by the FDA, and most of the external tools they implemented also hadn’t gone through any regulatory review.
“My experience suggests that most models are put into practice with very little evidence of their effects on outcomes because they are presumed to work, or at least to be more efficient than other decision-making processes,” Kellie Owens, a researcher for Data & Society, a nonprofit that studies the social implications of technology, wrote in an email. “I think we still need to develop better ways to conduct algorithmic risk assessments in medicine. I’d like to see the FDA take a much larger role in regulating AI and machine learning models before their implementation.”
Developers should also ask themselves if the communities they’re serving have a say in how the system is built, or whether it is needed in the first place. The majority of hospitals surveyed did not share with patients if a model was used in their care or involve patients in the development process.
In some cases, the best option might be the simplest one: don’t build.
In the meantime, hospitals are left to sift through existing published data, preprints and vendor promises to decide on the best option. To date, Michigan Medicine’s paper is still the only one that has been published on Epic’s Deterioration Index.
Care teams there used Epic’s score as a support tool for its rapid response teams to check in on patients. But the health system was also looking at other options.
“The short game was that we had to go with the score we had,” Singh said. “The longer game was, Epic’s deterioration index is proprietary. That raises questions about what is in it.”
NEW DELHI — More than a year after the pandemic began, infections worldwide have surpassed their previous peak. The average number of coronavirus cases reported each day is now higher than it has ever been.
“Cases and deaths are continuing to increase at worrying rates,” said World Health Organization chief Tedros Adhanom Ghebreyesus on Friday.
A major reason for the increase: the ferocity of India’s second wave. The country accounts for about one in three of all new cases.
It wasn’t supposed to happen like this. Earlier this year, India appeared to be weathering the pandemic. The number of daily cases dropped below 10,000 and the government launched a vaccination drive powered by locally made vaccines.
But experts say that changes in behavior and the influence of new variants have combined to produce a tidal wave of new cases.
India is adding more than 250,000 new infections a day — and if current trends continue, that figure could soar to 500,000 within a month, said Bhramar Mukherjee, a biostatistician at the University of Michigan.
While infections are rising around the country, some places are bearing the brunt of the surge. Six states and Delhi, the nation’s capital, account for about two-thirds of new daily cases. Maharashtra, home to India’s financial hub, Mumbai, represents about a quarter of the nation’s total.
Mohammad Shahzad, a 40-year-old accountant, was one of many desperately seeking care. He developed a fever and grew breathless on the afternoon of April 15. His wife, Shazia, rushed him to the nearest hospital. It was full, but staffers checked his oxygen level: 62, dangerously low.
For three hours, they went from hospital to hospital trying to get him admitted, with no luck. She took him home. At 3:30 a.m., with Shahzad struggling to breathe, she called an ambulance. When the driver arrived, he asked if Shahzad truly needed oxygen — otherwise he would save it for the most serious patients.
The scene at the hospital was “harrowing,” said Shazia: a line of ambulances, people crying and pleading, a man barely breathing. Shahzad finally found a bed. Now Shazia and her two children, 8 and 6, have also developed covid-19 symptoms.
From early morning until late at night, Prafulla Gudadhe’s phone does not stop ringing. Each call is from a constituent and each call is the same: Can he help to arrange a hospital bed for a loved one?
Gudadhe is a municipal official in Nagpur, a city in the interior of Maharashtra. “We tell them we will try, but there are no beds,” he said. About 10 people in his ward have died at home in recent days after they couldn’t get admitted to hospitals, Gudadhe said, his voice weary. “I am helpless.”
Kamlesh Sailor knows how bad it is. Worse than the previous wave of the pandemic, like nothing he’s ever seen.
Sailor is the president of a crematorium trust in the city of Surat. Last week, the steel pipes in two of the facility’s six chimneys melted from constant use. Where the facility used to receive about 20 bodies a day, he said, now it is receiving 100.
“We try to control our emotions,” he said. “But it is unbearable.”
We’re a year into the coronavirus pandemic, so the math that undergirds its risks should by now be familiar. We all should know, for example, that the ability of the virus to spread depends on it being able to find a host, someone who is not protected against infection. If you have a group of 10 people, one of whom is infected and nine of whom are immune to the virus, it’s not going to be able to spread anywhere.
That calculus is well known, but there is still some uncertainty at play. To achieve herd immunity — the state where the population of immune people is dense enough to stamp out new infections — how many people need to be protected against the virus? And how good is natural immunity, resistance to infection built through exposure to the virus and contracting covid-19, the disease it causes?
The safe way to increase the number of immune people, thereby probably protecting everyone by limiting the ability of the virus to spread, is through vaccination. More vaccinated people means fewer new infections and fewer infections needed to get close to herd immunity. The closer we get to herd immunity, the safer people are who can’t get vaccinated, such as young children (at least for now).
The challenge the world faces is that the rollout of vaccines has been slow, relatively speaking. The coronavirus vaccines were developed at a lightning pace, but many parts of the world are still waiting for supplies sufficient to broadly immunize their populations. In the United States, the challenge is different: About a quarter of adult Americans say they aren’t planning on getting vaccinated against the virus, according to Economist-YouGov polling released last week.
That’s problematic in part because it means we’re less likely to get to herd immunity without millions more Americans becoming infected. Again, it’s not clear how effective natural immunity will be over the long term as new variants of the virus emerge. So we might continue to see tens of thousands of new infections each day, keeping the population at risk broadly by delaying herd immunity and continuing to add to the pandemic’s death toll in this country.
But we also see from the Economist-YouGov poll the same thing we saw in Gallup polling earlier this month: The people who are least interested in being vaccinated are also the people who are least likely to be concerned about the virus and to take other steps aimed at preventing it from spreading.
In the Economist-YouGov poll, nearly three-quarters of those who say they don’t plan on being vaccinated when they’re eligible also say they’re not too or not at all worried about the virus.
That makes some perverse sense: If you don’t see the virus as a risk, you won’t see the need to get vaccinated. Unfortunately, it also means you’re going to be less likely to do things like wear a mask in public.
Or you might be more likely to view as unnecessary precautions such as avoiding close-quarter contact with friends and family or traveling out of state.
About a quarter of adults hold the view that they won’t be vaccinated when eligible. That’s equivalent to about 64 million Americans.
Who are they? As prior polls have shown, they’re disproportionately political conservatives. At the outset of the pandemic, there was concern that vaccine skepticism would heavily be centered in non-White populations. At the moment, though, the rate of skepticism among those who say they voted for Donald Trump in 2020 and among Republicans is substantially higher than skepticism overall.
That shows up in another way in the Economist poll. Respondents were asked whose medical advice they trusted. Among those who say they don’t plan to get the vaccine, half say they trust Trump’s advice a lot or somewhat — far more than the advice of the Centers for Disease Control and Prevention or the country’s top infectious-disease expert Anthony S. Fauci.
If we look only at Republican skeptics, the difference is much larger: Half of Republican skeptics say they have a lot of trust in Trump’s medical advice.
The irony, of course, is that Trump sees the vaccine as his positive legacy on the pandemic. He’s eager to seize credit for vaccine development and has — sporadically — advocated for Americans to get the vaccine. (He got it himself while still president, without advertising that fact.) It’s his supporters, though, who are most hostile to the idea.
Trump bears most of the responsibility for that, too. Over the course of 2020, worried about reelection, he undercut containment efforts and downplayed the danger of the virus. He undermined experts such as Fauci largely out of concern that continuing to limit economic activity would erode his main argument for his reelection. Over and over, he insisted that the virus was going away without the vaccine, that it was not terribly dangerous and that America should just go about its business as usual — and his supporters heard that message.
They’re still listening to it, as the Economist poll shows. One result may be that the United States doesn’t reach herd immunity through vaccinations and, instead, some large chunk of those tens of millions of skeptics end up being exposed to the virus. Some of them will die. Some may risk repeat infections from new variants against which a vaccine offers better protection. Some of those unable to get vaccinated may also become sick from the virus because we haven’t achieved herd immunity, suffering long-term complications from covid-19.
Trump wants his legacy to be the rollout of the vaccine. His legacy will also probably include fostering skepticism about the vaccine that limits its utility in containing the pandemic.
About 1 in 10 nursing homes in California and nationwide are owned by private equity (PE) investors, and new research suggests that death rates for residents of those facilities are substantially higher than at institutions with different forms of ownership.
Researchers from New York University, the University of Chicago, and the University of Pennsylvania found that the combination of subsidies from Medicare and Medicaid alongside incentives for PE owners to increase the value of their investments “could lead high-powered for-profitincentives to be misaligned with the social goal of affordable, quality care [PDF].” The researchers — Atul Gupta, Constantine Yannelis, Sabrina Howell, and Abhinav Gupta — reported that nursing homes owned by private equity entities were associated with a 10% increase in the short-term death rate of Medicare patients over a 12-year period. That means more than 20,000 people likely died prematurely in homes run by PE companies, according to their study, which was published in February by the National Bureau of Economic Research (NBER).
In addition to the higher short-term death rates, these homes were found to have sharper declines in measures of patient well-being, including lower mobility, increased pain intensity, and increased likelihood of taking antipsychotic medications, which the study said are discouraged in the elderly because the drugs increase mortality in this age group. Meanwhile, the study found that taxpayer spending per patient episode was 11% higher in PE-owned nursing homes.
There’s nothing new about for-profit nursing homes, but private equity firms are a unique subset that in recent years has made significant investments in the industry, Dylan Scott reported in Vox. PE firms typically buy companies in pursuit of higher profits for shareholders than could be obtained by investing in the shares of publicly traded stocks. They then sell their investments at a profit, often within seven years of purchase. They often take on debt to buy a company and then put that debt on the newly acquired company’s balance sheet.
They also have purchased a mix of large chains and independent facilities — “making it easier to isolate the specific effect of private equity acquisitions, rather than just a profit motive, on patient welfare.” About 11% of for-profit nursing homes are owned by PE, according to David Grabowski, professor of health care policy at Harvard Medical School. The NBER study covered 1,674 nursing homes acquired in 128 unique transactions.
While the owners of many nursing homes may not be planning to sell them, they also have strong incentives to keep costs low, which may not be good for patients. A study funded by CHCF, for instance, found that “early in the pandemic, for-profit nursing homes had COVID-19 case rates five to six times higher than those of nonprofit and government-run nursing homes. This was true of both independent nursing homes and those that are part of a corporate chain.”
Given the dramatic increase in PE ownership of nursing facilities coming out of the COVID-19 pandemic, the higher death rates are troubling. The year-over-year growth between 2019 and 2020 is especially striking. Before the pandemic, 2019 saw 33 private equity acquisitions of nursing homes valued at just over $483 million.In 2020, there were 43 deals valued at more than $1.5 billion, according to Bloomberg Law reporter Tony Pugh.
And PE interest in health care is not restricted to nursing homes, explained Gretchen Morgenson and Emmanuelle Saliba at NBC News. “Private equity’s purchases have included rural hospitals, physicians’ practices, nursing homes and hospice centers, air ambulance companies and health care billing management and debt collection systems.” Overall, PE investments in health care have increased more than 1,900% over the past two decades. In 2000, PE invested less than $5 billion. By 2017, investment had jumped to $100 billion.
Industry advocates argue that the investments are in nursing homes that would fail without an influx of PE capital. The American Investment Council said private equity firms invest in “nursing homes to help rescue, build, or grow businesses, often providing much-needed capital to strengthen struggling companies and employ Americans,” according to Bloomberg Law.
The Debate Over Staffing
A bare-bones nursing staff is implicated in poorer quality at PE-owned nursing homes, both before and during the COVID-19 pandemic. Staff is generally the greatest expense in nursing homes and a key place to save money. “Labor is the main cost of any health care facility — accounting for nearly half of its operating costs — so cutting it to a minimum is the fastest profit-making measure owners can take, along with paying lower salaries,” journalist Annalisa Merelli explained in Quartz.
Staffing shrinks by 1.4% after a PE purchase, the NBER study found.
The federal government does not set specific patient-to-nurse ratios. California and other states have set minimum standards, but they are generally “well below the levels recommended by researchers and experts to consistently meet the needs of each resident,” according to the journal Policy, Politics, & Nursing Practice.
According to nursing assistant Adelina Ramos, “understaffing was so significant [during the pandemic] that she and her colleagues . . . often had to choose which dying or severely ill patient to attend first, leaving the others alone.”
Ramos worked at the for-profit Genesis Healthcare, the nation’s largest chain of nursing homes, which accepted $180 million in state and federal funds during the COVID-19 crisis but remained severely understaffed. She testified before the US Senate Finance Committee in March as a part of a week long look into how the pandemic affected nursing homes. “Before the pandemic, we had this problem,” she said of staffing shortages. “And with the pandemic, it made things worse.”
$12.46 an Hour
In addition, low pay at nursing homes compounds staffing shortages by leading to extremely high rates of turnover. Ramos and her colleagues were paid as little as $12.46 an hour.
Loss of front-line staff leads to reductions in therapies for healthier patients, which leads to higher death rates, according to the NBER study. The effect of these cuts is that front-line nurses spend fewer hours per day providing basic services to patients. “Those services, such as bed turning or infection prevention, aren’t medically intensive, but they can be critical to health outcomes,” wrote Scott at Vox.
Healthier patients tend to suffer the most from this lack of basic nursing. “Sicker patients have more regimented treatment that will be adhered to no matter who owns the facility,” the researchers said, “whereas healthier people may be more susceptible to the changes made under private equity ownership.”
Growing Interest on Capitol Hill
In addition to the Senate Finance Committee hearings, the House Ways and Means Committee held a hearing at the end of last month about the excess deaths in nursing homes owned by PE. “Private equity’s business model involves buying companies, saddling them with mountains of debt, and then squeezing them like oranges for every dollar,” said Representative Bill Pascrell (D-New Jersey), who chairs the House Ways and Means Committee’s oversight subcommittee.
The office of Senator Elizabeth Warren (D-Massachusetts) will investigate the effects of nursing-home ownership on residents, she announced on March 17.
The hope is that the pandemic’s effect on older people will bring more attention to the issues that lead to substandard nursing home care. “Much more is needed to protect nursing home residents,” Denise Bottcher, the state director of AARP’s Louisiana office, told the Senate panel. “The consequence of not acting is that someone’s mother or father dies.”
LIMA, Peru — The doctor watched the patients stream into his intensive care unit with a sense of dread.
For weeks, César Salomé, a physician in Lima’s Hospital Mongrut, had followed the chilling reports. A new coronavirus variant, spawned in the Amazon rainforest, had stormed Brazil and driven its health system to the brink of collapse.Now his patients, too, were arriving far sicker, their lungs saturated with disease, and dying within days. Even the young and healthy didn’t appear protected.
The new variant, he realized, was here.
“We used to have more time,” Salomé said. “Now, we have patients who come in and in a few days they’ve lost the use of their lungs.”
The P.1 variant, which packs a suite of mutations that makes it more transmissible and potentially more dangerous, is no longer just Brazil’s problem. It’s South America’s problem — and the world’s.
In recent weeks, it has been carried across rivers and over borders, evading restrictive measures meant to curb its advance to help fuel a coronavirus surge across the continent. There is mounting anxiety in parts of South America that P.1 could quickly become the dominant variant, transporting Brazil’s humanitarian disaster — patients languishing without care, a skyrocketing death toll — into their countries.
“It’s spreading,” said Julio Castro, a Venezuelan infectious-disease expert. “It’s impossible to stop.”
In Lima, scientists have detected the variant in 40 percent of coronavirus cases. In Uruguay, it’s been found in 30 percent. In Paraguay, officials say half of cases at the border with Brazil are P.1. Other South American countries — Colombia, Argentina, Venezuela, Chile — have discovered it in their territories. Limitations in genomic sequencing have made it difficult to know the variant’s true breadth, but it has been identified in more than two dozen countries, from Japan to the United States.
Hospital systems across South America are being pushed to their limits. Uruguay, one of South America’s wealthiest nations and a success story early in the pandemic, is barreling toward a medical system failure. Health officials say Peru is on the precipice, with only 84 intensive care beds left at the end of March. The intensive care system in Paraguay, roiled by protests last month over medical shortcomings, has run out of hospital beds.
“Paraguay has little chance of stopping the spread of the P.1 variant,” said Elena Candia Florentín, president of the Paraguayan Society of Infectious Diseases.
“With the medical system collapsed, medications and supplies chronically depleted, early detection deficient, contact tracing nonexistent, waiting patients begging for treatment on social media, insufficient vaccinations for health workers, and uncertainty over when general and vulnerable populations will be vaccinated, the outlook in Paraguay is dark,” she said.
How P.1 spread across the region is a distinctly South American story. Nearly every country on the continent shares a land border with Brazil. People converge on border towns, where crossing into another country can be as simple as crossing the street. Limited surveillance and border security have made the region a paradise for smugglers. But they have also made it nearly impossible to control the variant’s spread.
“We share 1,000 kilometers of dry border with Brazil, the biggest factory of variants in the world and the epicenter of the crisis,” said Gonzalo Moratorio, a Uruguayan molecular virologist tracking the variant’s growth. “And now it’s not just one country.”
The Brazilian city of Tabatinga, deep in the Amazon rainforest, where officials suspect the virus crossed into Colombia and Peru, is emblematic of the struggle to contain the variant. The city of 70,000 was swept by P.1 earlier this year. Many in the area have family ties in several countries and are accustomed to crossing borders with ease — canoeing across the Amazon River to Peru or walking into Colombia.
“People ended up bringing the virus from one side to the other,” said Sinesio Tikuna Trovão, an Indigenous leader. “The crossing was free, with both sides living right on top of one another.”
Now that the variant has infiltrated numerous countries, stopping its spread will be difficult. Most South American countries, with the exception of Brazil, adopted stringent containment measures last year. But they have been undone by poverty, apathy, distrust and exhaustion. With national economies battered and poverty rising sharply, public health experts fear more restrictions will be difficult to maintain. In Brazil, despite record death numbers, many states are lifting restrictions.(SOUND FAMILIAR)
That has left inoculation as the only way out. But coronavirus vaccines are South America’s white whale: often discussed, but rarely seen. The continent hasn’t distributed its own vaccine or negotiated a regional agreement with pharmaceutical companies. It’s one of the world’s hardest-hit regions but has administered only 6 percent of the world’s vaccine doses, according to the site Our World in Data. (The outlier is Chile, which is vaccinating residents more quickly than anywhere in the Americas — but still suffering a surge in cases.)
“We should not only blame the policy response,” said Luis Felipe López-Calva, the United Nations Development Program’s regional director for Latin America and the Caribbean. “We have to understand the vaccine market.”
“And there is a failure in the market,” he said.
The vaccine has become so scarce, López-Calva said, that officials are imposing restrictions on information. It’s nearly impossible to know how much governments are paying for doses. Some regional blocs, such as the African Union and the European Union, have negotiated joint contracts. But in South America, it has been every country for itself — diminishing the bargaining power for each one.
“This has been harmful for these countries, and for the whole world to stop the virus,” López-Calva said. “Because it’s never been more clear that no one is protected until everyone is protected.”
Paulo Buss, a prominent Brazilian scientist, said it didn’t have to be like this. He was Brazil’s health representative to the Union of South American Nations, which negotiated several regional deals with pharmaceutical companies before the coronavirus pandemic. But that union came apart amid political differences just before the arrival of the virus.
“It was the worst possible moment,” Buss said. “We’ve lost capacity and our negotiation attempts have been fragmented. Multi-lateralism was weakened.”
Vaccine scarcity has led to line-jumping scandals all over South America, but particularly in Peru. Hundreds of politically connected people, including cabinet ministers and former president Martín Vizcarra, snagged vaccine doses early. Now people are calling for criminal charges.
As officials bicker and the vaccination campaign is delayed, the variant continues to spread. P.1 accounts for 70 percent of cases in some parts of the Lima region, according to officials. Last week, the country logged the highest daily case count since August — more than 11,000. On Saturday, the country recorded 294 deaths, the most in a day since the start of the pandemic.
Peruvians have been stunned by how quickly the surge overwhelmed the health-care system. Public health analysts and government officials had believed Peru was prepared for a second wave. But it wasn’t ready for the variant.
“We did not expect such a strong second wave,” said Percy Mayta-Tristan, director of research at the Scientific University of the South in Lima. “The first wave was so extensive. The presence of the Brazilian variant helps explain why.”