Nothing had prepared Jeffrey Harris for this. Then again, what would have prepared anyone for the most widespread pandemic since the Spanish flu afflicted a third of the world’s population a century ago?
Like so many others last winter, Harris watched anxiously as the novel coronavirus discovered in China hitchhiked its way across the globe. At the beginning of March—before the spring spike that led to lockdowns of American cities—he was still seeing patients at a community health center in Los Angeles while on sabbatical from MIT. Although Harris has been an economics professor since 1977, he’s also been a practicing physician for nearly 50 years and has worked exclusively with underserved populations at various community clinics abroad and in the US since 2005. For people who rely solely on Medicare or Medicaid, or don’t even have health insurance, these are the only places where they can afford to see a doctor.
Yet Harris knew it was just a matter of time before covid-19 engulfed Los Angeles County and made it impractical to see every patient in person. As the weeks went on, cases of coronavirus piled up, and it became crucial to prevent people from gathering in crowded waiting areas and exam rooms. That’s when he and his fellow providers turned to their phones.
“We’ve had regular phone calls with patients—sometimes just by voice, sometimes by video—since the start of the epidemic. Before that, it really wasn’t something we did on a regular basis,” Harris says. “I’ve been practicing medicine my whole life, but this is definitely new for me.”
By the summer, with Los Angeles County a major center of covid-19 in California, Harris sometimes spent whole shifts conversing with patients remotely—usually 20 to 24 of them. Before the pandemic, he’d typically top out at around 18 patients a day, seeing all of them on site.
The experience sold him on telemedicine: “It really has shown us other ways that we can deliver medical care,” he says. And many of his fellow physicians clearly agree. Last January, Medicare patients had fewer than 3,000 virtual primary-care visits a week, according to the Centers for Medicare and Medicaid Services; by April, virtual visits had spiked to more than 1.7 million a week. Forrester Research estimated that in 2020 alone, one in 10 general visits for routine and chronic care—more than 260 million appointments—were conducted virtually in the US. They calculated that nearly another 30 million telemedicine visits were conducted for patients with covid-19.
“Telehealth has certainly been a big part of the response to the pandemic, and it’s been highly successful,” says Ford Professor of Economics Jonathan Gruber ’87, who specializes in health-care economics. Writing in Newsweek in April, he argued that telemedicine, not the emergency room, should be the front line of pandemic care, since it could serve people who might be infected while protecting nurses and doctors from exposure.
“We have seen an incredible expansion in virtual care across the industry as a result of the pandemic,” says Forrester analyst Arielle Trzcinski. “It is something we expect to become a mainstay of health care going forward.”
On campus, covid-19 also spurred the rapid adoption of telemedicine at MIT Medical. Before the clinic shut down on March 16for all but essential visits, its telehealth offerings “were largely nonexistent,” says Brian Schuetz, MIT Medical’s executive director.
“The launch of telehealth services was very much on our to-do list, but with a 12- to 18-month development and rollout timeline,” hesays. “The urgent need created by the pandemic required us to put the tools in the hands of our clinicians immediately.”
The clinic outfitted every provider on staff with the necessary hardware andbegan offering telehealth services on March 29. Staff and patients took swift advantage: there were 1,138 telemedicine visits in April, 1,564 in May, and almost 2,000 in June. (Since MIT Medical ramped up in-person services again in July, it has continued to offer telemedicine, racking up an average of 2,175 visits a month through November.)
Telemedicine’s sudden popularity is no surprise to professors and researchers at MIT who have studied its utility for years. But conventional wisdom has long dictated that to really benefit from a doctor visit, the patient has to be in the room.
“Patients thought it wasn’t effective, and providers thought they couldn’t get paid,” says Gruber. Now, he adds, they see that neither of those things has to be true.
Still, the post-pandemic future of virtual care is uncertain. American health care, a labyrinth of private, state, and federal payers with different reimbursement policies, makes covering telehealth visits a headache. Medical licensing laws determined by each state often mean doctors can’t conduct virtual visits across state lines. The basic technology to administer virtual care is available, but updating and sharing patient information among providers remains a problem. Not to mention that not all patients will be equally adept at navigating the various apps and video tools required—assuming they even have smartphones or computers in the first place.
Overcoming all these barriers requires going beyond the quick measures taken in response to the pandemic, MIT researchers and professors say. What’s really needed? A set of long-term policies that make telehealth a cornerstone of American health care.
Upending medical dogma
Clinicians and public health officials in the US were quick to call for expanded telehealth options as a way to stop covid-19 transmission: Nancy Messonnier, director of the National Center for Immunization and Respiratory Diseases at the Centers for Disease Control and Prevention, urged hospitals to expand these services at the end of February. Of particular concern was the health of older Americans, since even when little else was known about the virus, it was clear that it was more deadly for people over 60.
When the Trump administration declared covid-19 a public health emergency, on March 13, a host of telemedicine-related provisions for Medicare beneficiaries went into effect. Gone were restrictions that previously made it difficult to conduct a virtual visit with a doctor. Pre-pandemic, Medicare could pay for telehealth only if a patient lived in a designated rural area and went to a medical facility to consult with doctors based elsewhere. But now Medicare would reimburse clinicians for telemedicine services that patients accessed at any health-care facility in the US, or even by dialing in from home. Audio-only calls, typically excluded from telehealth coverage, were also included under the national emergency provisions—a recognition that not everyone has access to video technology.
Expanding telehealth opportunities struck White House coronavirus advisor Anthony Fauci, director of the National Institute of Allergy and Infectious Diseases, as particularly good policy.
“I believe telemedicine is a very important component,” he said at a congressional hearing in June, in response to a question about whether virtual visits could help protect vulnerable people. “As we look forward in the future, I think you’re going to see a lot more of that.”
To Amar Gupta, SM ’80, the new federal measures, while welcome, were steps that should have been taken before—and Fauci’s position was one he had adopted a while ago. A researcher at MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL), Gupta is a longtime advocate for telemedicine. He coauthored a seminal paper challenging the constitutionality of laws hindering the practice of telemedicine across state borders in the US, and he designed a popular MIT course, Telemedicine and Telehealth for Enhancing Global Health, which he taught from 2016 to 2018. His published research includes examples showing that telemedicine can reduce health-care costs with no negative effect on outcomes. One study coauthored by Gupta and published last July in the Journal of Urology shows that travel and wait times accounted for 98.4% of the total time pediatric patients spent visiting urologists for postoperative care. Shifting to virtual care meant that children missed fewer school days and parents saved money. (The opportunity cost of missing work to take a child for an in-person visit was found to be $23.75 per minute of face time with a physician; with virtual visits, the cost fell to $1.14 per minute.)
Gupta’s ideas about telemedicine often fell on deaf ears. “Doctors were insistent that the consultation needed to happen face to face,” he says. “They used to mock me when I said it can be done through telemedicine.”
Yet real-world experience taught him otherwise. During a three-year stint early last decade at Pace University in New York City, Gupta was instrumental in introducing telemedicine concepts in New York and played a pivotal role in establishing the Telehealth Intervention Program for Seniors (TIPS) in Westchester County. The program uses trained personnel who visit patients in senior centers and apartment buildings to gather vital signs such as oxygen-saturation levels, pulse, and blood pressure, which are later reviewed by a nurse. If the readings point to trouble like abnormally high blood pressure, the nurse alerts the senior’s primary-care physician, who can then set up a virtual visit to determine the next steps. In the first few months after this health maintenance program was rolled out, it helped reduce the number of ambulance calls in one part of the county by 75%, Gupta says. As the program continued, hospital visits for TIPS Medicare patients fell by 50% and their short-term readmissions dropped by 76%.
Virtual care can also support doctors’ own well-being. In a multiyear telemedicine project, Gupta and his coauthors interviewed teams of several doctors and nurses from Emory University who had been transferred to Australia for weeks or months at a time so that patients in Atlanta could receive care at night from well-rested clinicians. (Gupta and colleagues had first proposed such a strategy in 2010.) The team in Australia worked a 12-hour daylight shift and then handed things off to clinicians in Atlanta during the daytime in the US. The study focused on how the schedule affected the clinicians in Australia, and the data showed that “the doctors and nurses were far happier, far more relaxed,” says Gupta. Arguably, the clinicians in Australia were also in a better position to provide good care during Atlanta’s night shift. Doctors in the US might work up to 40 hours with no sleep, as Gupta points out, and are more prone to mistakes when they have to grind away in such long shifts.
Since the pandemic upended the medical dogma about telehealth, it has quickly become the preference for many providers. At the end of April, the US recorded 36,400 confirmed covid-19 cases in a single day—nowhere near December’s 200,000-plus daily case counts, but a record single-day high at the time. That same month, the health information company IQVIA conducted a survey of about 300 doctors—primary-care physicians and specialists alike—on their telemedicine usage.
During one week of the widespread US lockdowns last spring, more than half of their interactions with patients were via telemedicine, up from 9% before the pandemic. And the doctors surveyed said they expected to continue using telehealth for more than 20% of their patient interactions when the pandemic ends.
The trend may be key to keeping small clinics afloat, says Mei Wa Kwong, executive director of the Center for Connected Health Policy, a nonprofit that has been working toward integrating telehealth into America’s health-care system for over a decade.
“When the pandemic hit and people were afraid to go to the clinic, they had to pivot toward telehealth, and that actually saved a lot of them,” she says. “Telehealth was a lifeline to them in being able to keep their doors open.”
Going forward, “I think there is no doubt we’ll have a lot more telehealth than we had before the pandemic,” says Gruber. “The question is sort of how much and to what extent does it continue to replace in-person health care?”
Even among those who agree that telehealth is sticking around, crucial questions remain about the best ways to implement the technology and deal with the laws currently standing in its way.
Retsef Levi, a professor at the MIT Sloan School of Management, is familiar with some of these barriers firsthand: like Harris, he is a physician in a community health center. Following the emergency declaration in March, he and fellow Sloan professor Simon Johnson, PhD ’89, convened the Covid-19 Policy Alliance, a team of experts in medicine, logistics, and computing. Very quickly, the team released two ambitious policy papers outlining ways to deploy telemedicine at the state and federal levels—including the creation of a national covid-19 center that would make telehealth available to everyone in the country.
The proposals weren’t enacted in any systemic way, but Levi says they underscore an ongoing worry that the typical regulations around reimbursement and cross-state practicing impede the expansion of telehealth. What happens once the public health emergency ends, and various laws that were relaxed presumably revert to their original form?
“We need to ask ourselves what kind of system we want to have. What is the right way to deliver health care and manage health?” Levi says. “Put it this way: I don’t think that, with the pandemic, we started to deliver anything that we couldn’t deliver before through telehealth. It just forced us to do the right thing.”
Ushering in a new wave of telemedicine requires meeting some key technological challenges. In the US, one is interoperability, which “continues to be a total disaster,” Gupta says. It’s a major question of how doctors update and share information, not only between different hospitals but also among doctors working at the same hospital. Three specific issues need to be addressed: how to securely transport patient data; how to format that data; and how to ensure that different types of health-care providers—doctors, nurses, pharmacists, lab technicians—know what that data is saying.
Right now there are no standards for cataloguing something as simple as a heartbeat recorded via a digital stethoscope. The stethoscope might track it by the half-minute, but the provider might want that information by the minute. Unless the provider knows the correct unit, the situation is akin to an American telling a French person it’s a chilly 40 degrees outside.
The digital gadgetry that’s used to securely share patient data also needs to be improved. While working in Australia, the Emory doctors and nurses used the university system to transmit sensitive information, but they had to deal with long processing delays. Gupta says providers have to resist using unencrypted devices—like their own cell phones—in the interest of speed.
Even once these problems are solved, advocates for telehealth say, a more fundamental challenge is to change the thinking about what this technology can offer. “There is a tendency, as there is with every technology, to think about telehealth as just being a better way of doing something that we did in the past,” says Micky Tripathi, PhD ’00, an expert in health IT who headed the nonprofit Massachusetts eHealth Collaborative and now is an executive at the health-care data company Arcadia.
Four ways telehealth can help fight covid
The Covid-19 Policy Alliance, led by Sloan professors Vivek Farias, Simon Johnson, PhD ’89, Kate Kellogg, and Retsef Levi, identified four ways telemedicine could help effectively allocate scarce health-care resources to address and contain the pandemic:
1) Facilitate safe mass testing.
Phone or video interviews can be used to determine who needs a test and schedule appointments in a way that balances the load across testing sites.
2) Prevent hospitals from becoming infected.
3) Treat mild cases.
4) Replace routine office visits.
See high-risk patients (like those in nursing homes) remotely to prevent infections and overwhelming ICU capacity.
Telemedicine, Tripathi says, can be much more than just a way to consult a doctor by phone or video. Think of the fingertip pulse oximeters many Americans have turned to during the covid-19 pandemic to keep an eye on their own oxygen levels. Now imagine digitally connected devices like electronic stethoscopes and telemetry-capable electrocardiographs that could transmit data on a patient’s heartbeat, respiration, and blood-oxygen levels. Such devices could be distributed to people who need them, with data collected via smartphone app. Or nurses could make house calls with portable ultrasound machines and other medical equipment to collect patients’ information from the comfort of their homes.
“[Telemedicine] allows us to do so many other things,” Tripathi says. “It won’t just be a replacement for an on-site visit.”
But while regularly collecting data at home could improve care, the insurance system doesn’t always cover medical gear for patients. Hypertension, for instance, can be monitored remotely, but blood-pressure cuffs aren’t covered. “In our health-care system, everything hinges on who’s going to pay for it and what’s covered by insurance, and that represents a significant barrier,” says Harris, who became a professor emeritus in July and continues to practice in Los Angeles.
Now that the pandemic has prompted doctors to rethink telemedicine, a concerted effort to integrate these services into the US health-care system seems more likely.
For Harris, telemedicine hasn’t been just a way to keep his clinic operational during the pandemic: it’s offered his patients benefits they never had before. Now that they can talk to doctors from home, getting the care they need doesn’t mean negotiating scheduling conflicts or taking a day off work. And older patients can get help from family members when it comes to reading their prescription bottles or describing their symptoms.
Perhaps the biggest change Harris notices is how much more relaxed and comfortable his patients are. That leads to better communication, better conversations, and better information about how to look after their own health.
“Telemedicine has been an enormous event,” he says. “I hope it is here to stay.”
This new data poisoning tool lets artists fight back against generative AI
The tool, called Nightshade, messes up training data in ways that could cause serious damage to image-generating AI models.
Rogue superintelligence and merging with machines: Inside the mind of OpenAI’s chief scientist
An exclusive conversation with Ilya Sutskever on his fears for the future of AI and why they’ve made him change the focus of his life’s work.
The Biggest Questions: What is death?
New neuroscience is challenging our understanding of the dying process—bringing opportunities for the living.
Data analytics reveal real business value
Sophisticated analytics tools mine insights from data, optimizing operational processes across the enterprise.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.