With the rapid expansion of telemedicine in response to the COVID-19 pandemic, researchers have begun to explore how a new emphasis on virtual care might exacerbate persistent racial and ethnic disparities in U.S. health care — and, perhaps, help repair them.
Recent research and anecdotal evidence point to significantly lower levels of telemedicine usage among Black and Hispanic patients, particularly those over age 65, as compared with white patients, since the pandemic’s onset. Ellerie Weber, an assistant professor in population health science and policy at Mount Sinai’s Icahn Medical School, saw it firsthand during the initial COVID-19 surge in New York City.
“There were these huge lines for the ER, and for testing,” says Weber, lead author of the study “Characteristics of telehealth users in NYC for COVID-related care during the coronavirus pandemic.”
Many of those waiting in line were Black and Hispanic, she says, “and people who were higher [socioeconomic status], younger, more white, more Asian, you just heard, anecdotally, that they were calling their doctor or getting a telehealth appointment to at least say, ‘Hey, do I have COVID? Do I need a test?’”
Telemedicine is the delivery of clinical services using telecommunications technology, including video conferencing and telephone calls. (Telehealth is a broader term that generally encompasses clinical services as well as nonclinical services such as provider training.)
Despite evidence of lower rates of telemedicine use among racial and ethnic minorities, other research suggests that seeing patients in a standardized way on a screen might reduce providers’ biases and the resulting disparities in health care. Take, for example, the results of a recent study in which researchers analyzed electronic health records of patients in New York City at the pandemic’s initial peak, titled “Telemedicine and health care disparities: a cohort study in a large health care system in New York City during COVID-19.”
The results suggest that suspected COVID-19 diagnoses among Black patients, who on average were sicker when they sought care, were more likely to happen in telemedicine visits than in-person.
“It seems like, on the telemedicine system, it’s maybe a more equal playing ground for the Black patients and white patients to get treated the same,” says the study’s lead author, Rumi Chunara, an assistant professor at New York University’s Tandon Engineering School, whose research aims to inform the development of telemedicine tools and platforms.
Historically, stark differences in the way people of color access, experience, and receive health care, as compared with their white counterparts, have been well-documented. And while progress has been made in increasing insurance coverage and improving overall population health in recent years, the data show that some gaps in access to and quality of care have widened, according to the 2018 National Healthcare Quality and Disparities (NHQD) Report.
Released by the Agency for Healthcare Research and Quality, part of the US Department of Health and Human Services, the NHQD Report outlines persistent disparities for Black patients, as compared with whites, including the number of new HIV cases per 100,000 people over age 13 and hospital admissions for asthma per 100,000 population, among children age 2-17.
Likewise, past research on the so-called digital divide has shown that much of the world is still cut off from the internet. People who lack access to broadband internet – due to cost, availability or the technological literacy to leverage online tools and information — are at a disadvantage in terms of health care and other aspects of life.
With the explosion in telemedicine in recent months, due to the convergence of social distancing guidelines and emergency government policies aimed at facilitating virtual care, researchers are just beginning to learn what happens when telemedicine is introduced at scale.
“There hasn’t been much research into how the digital divide influenced telemedicine use in particular,” explains Weber, “because prior to COVID, telemedicine use was actually not that widely taken up. People were trying for a long time; people kept saying we were on the verge of a telehealth revolution, but it wasn’t really used much for a variety of reasons.”
Health care digital divide
In San Francisco earlier this year, a team of clinicians and researchers affiliated with the University of California San Francisco, led by Dr. Sarah Nouri, noticed that after their primary care practices quickly shifted almost entirely to remote patient visits because of the pandemic, Black and Hispanic patients used telemedicine at significantly lower rates than white patients. They attributed the differences to disparities in access to technology and digital literacy.
“This has alarming implications for inadequate chronic disease management that may result in increased disparities in clinical outcomes as some patients disproportionately lose frontline access to primary care,” Nouri and her colleagues write in a May commentary published in the journal Innovations in Care Delivery.
To address access concerns, some insurers, including Medicare, have started covering audio-only visits conducted via telephone at the same rates as in-person and video visits. However, Weber’s research suggests that even when taking these telephone visits into account, Blacks and Hispanics, particularly in older age brackets, are still less likely to use telemedicine.
Moreover, audio-only visits might sometimes be inadequate, even if they are reimbursed at the same rates. During the pandemic’s initial surge, as emergency federal policies facilitated increased reimbursement for telemedicine services, the combined use of video and audio-only telehealth among Medicare beneficiaries skyrocketed nearly 13,000% over pre-pandemic levels, according to the Centers for Medicare & Medicaid Services (CMS).
But a study by Eric Roberts, an assistant professor of health policy at the University of Pittsburgh, and Dr. Ateev Mehrotra of Harvard Medical School, finds that more than 26% of Medicare enrollees living in the community (as opposed to residential facilities such as nursing homes) have neither a smartphone with a data plan nor an internet-connected personal computer. That figure is more than 37% among Black patients on Medicare, according to their research, published in August in JAMA Internal Medicine.
In addition to expanding broadband access through policies at the federal, state, and local level, other efforts have been shown to help narrow the digital divide, which impedes telemedicine access among lower-income and older adults, as well as minority groups. These include increasing digital literacy and subsidies for internet access to low-income consumers, and providing low-cost digital devices to qualifying patients.
In terms of understanding how health outcomes might vary across different population groups as a result of the quick pivot to telemedicine, it could take “several years” before academics have access to the health insurance claims data that would enable such research, says Roberts, who studies issues surrounding care access and quality among marginalized patients enrolled in public health care programs.
“It’s not like somebody who has a video visit or who’s only able to have a voice-only visit today is going to have a hospitalization for exacerbation of their diabetes tomorrow,” Roberts says. “It’s going to take a while for those disruptions in care to have measurable health effects that one could detect in claims.”
Beyond documented factors such as access to technology and variations in health and digital literacy, some researchers suggest there may be other community-specific social factors, including mistrust of the health care system and skepticism about health technology, driving what early research suggests is inequitable access to telemedicine-delivered care.
The role of policymakers and health care providers
Increasingly, hospitals themselves are creating programs and centers devoted to rooting out the causes of health disparities and finding ways to deliver equitable care in their own communities. Based on recent events in the United States and ongoing demands for racial justice, these centers are becoming even more proactive, says Dr. Gezzer Ortega, the lead faculty for research and innovation for equitable surgical care within the Center for Surgery and Public Health at Brigham and Women’s Hospital in Boston.
Ortega and a group of colleagues recently analyzed the emergency federal policies put in place to facilitate the explosion in telemedicine use since the pandemic began. The result was “Telemedicine, COVID-19, and disparities: Policy implications,” a paper outlining a series of policy prescriptions and practice recommendations, published in September in the journal Health Policy and Technology.
Most critical among the authors’ recommendations is the need to create equitable and overarching federal policies surrounding insurance coverage and financial reimbursement for telemedicine services, says Ortega.
“Early on [in the pandemic], states adopted various policies and health insurance companies had different policies, and that variation made it a little bit challenging depending on what health insurance you had to access some of these services,” he says. “I think providing access via health insurance and reimbursement is critical at the federal level.”
Moving forward, the development of policies that standardize personal and data privacy will need to strike a careful balance between being user-friendly and secure, while keeping equitable access (by making platforms multilingual, for example) at the forefront, Ortega adds. “I think transparency is key,” he says. “Patients need to be aware of where their information is being stored and to know that this is an encrypted means of communication, and other technology needs to be adapted to that.”
Addressing mistrust
Finding ways to address deeply rooted mistrust of the health care system will require more time and effort, says Ortega. Based on past traumas, such as the exploitation of Henrietta Lacks and the decades-long Tuskegee Syphilis Study, and more recently with allegations of unnecessary medical procedures occurring at US border with Mexico, communities of color have historically distrusted the health care system.
“I think, in order to regain that trust, one of the things we need to do is one diversify the workforce,” says Ortega. “People are more willing to trust if there’s some level of concordance… I also think that reaching out to these communities that the health care centers serve, making them part of the health care system, making them part of the billboards, part of the population that they’re working to attract to these centers, is important. And more community engagement, partnering with trusted community leaders in these communities to provide health care services is important.”
Collecting data on modes of health care used and outcomes for various patient groups is another important function for health care organizations in addressing health disparities, Ortega says, particularly as telemedicine is expected to retain its more prominent role in the system post-pandemic, at least to some degree. Armed with the numbers, health care professionals can begin to craft outreach and education programs in concert with community partners.
“It’s complex,” Ortega says. “I think it’s going to take different approaches at multiple levels in order for us to address the inequities, but it’s important work that needs to be done so that the health of our society will improve.”
Significant promise
Despite the risk that telemedicine might be out of reach for some disadvantaged groups and potentially exacerbate persistent racial and ethnic health disparities, the use of telemedicine also holds significant promise, including bringing higher quality care to people who live far away from top health centers.
Additionally, telemedicine provides opportunities for real-time professional consults by distant specialists and health education for large audiences who can participate in webinars and workshops remotely. Remote patient monitoring, wearable devices, and the Internet of Things offer new ways providers can interact with patients, with the potential to reduce costs and improve outcomes. In a way, technology can also help transport diversity and inclusion.
“I was actually on a call yesterday with someone from central Michigan, and they had never seen a Latino physician in their lives, because they don’t have any where they work,” Ortega says. “And I was just talking to this person – we were talking about some non-health-related activities — but it presented them with the idea that, ‘Oh, wow. I’m actually having a conversation with someone that looks like me, who’s a health care provider, whereas I live in a community where that does not exist.’
“I think that this is an example of where technology can help mitigate these disparities,” he adds. “There is a ton of potential there with technology and what we can do to improve a lot of the disparities and equity.”
What follows is a curated list of recent peer-reviewed research studies on racial and ethnic disparities related to the use of telemedicine, along with summaries of their main findings. We also offer five tips for journalists reporting on inequities in telemedicine-delivered care.
————
Rumi Chunara, et al., Journal of the American Medical Informatics Association, August 2020
This study analyzed health records for 140,184 patients in New York City who sought care either in person or via telemedicine between March 19 and April 30, 2020, at one of New York University’s Langone Health system’s four hospitals and more than 500 ambulatory locations.
Black patients were about half as likely as white patients to seek care for possible COVID-19 symptoms via telemedicine within the NYULH system, where the majority of patients have insurance coverage and access to broadband internet.
When Black patients did use telemedicine, they were more likely to receive a suspected diagnosis of COVID-19 and subsequently test positive for COVID-19, as compared with white patients, suggesting that Black patients “may be sicker when seeking care through telemedicine compared to white patients.”
The researchers also note, however, that these results could be due to an increased likelihood of getting tested among white patients.
In contrast, when both telemedicine and in-person visits were taken into account, Black patients were less likely to receive a suspected COVID-19 diagnosis than white patients, suggesting that more equitable care occurs over telemedicine. This could be because “the online platform normalizes interactions between the clinician and patient through the virtual medium and decreases healthcare diagnosis biases, offering a possible mitigating role for telemedicine with respect to healthcare disparities.”
With the goal of informing the design of telemedicine tools and platforms, the researchers suggest including diverse perspectives in a “participatory design” process and incorporating “cultural and linguistic adaptations,” including building multilingual platforms. Ensuring representativeness among providers offering care via telemedicine, “removing as much individual discretion of providers as possible,” and targeting outreach to communities identified as experiencing barriers to care are also encouraged.
Characteristics of Telehealth Users in NYC for COVID-related Care during the Coronavirus Pandemic
Ellerie Weber, Sarah J. Miller, Varuna Astha, Teresa Janevic, Emma Benn, Journal of the American Medical Informatics Association, November 2020.
This study used a Mount Sinai COVID-19 database that was quickly made available to researchers, without patients’ names, to analyze the demographic characteristics of patients who sought care via telemedicine as an entry point to a single health care system during the pandemic’s initial peak in New York City.
Analyzing 39,229 initial contacts with the health care system, the researchers found that 38.3 percent were via emergency room and 38.5 percent were via telehealth. The rest occurred at an outpatient office. Patients older than 65 had the lowest likelihood of telemedicine use, including audio-only telemedicine visits.
Black patients had 4.3 times higher odds of seeking care at the ER than white patients, and Black and Hispanic patients over the age of 65 were most likely to visit the ER, even when controlling for other health conditions.
The researchers hypothesize that “patients without a source of usual care (e.g., a primary or specialty care physician) would be more likely to go to the ER, and less likely to seek telehealth treatment through previously established care relationships.”
Severity of COVID symptoms, which the data did not indicate, might also have contributed to the findings. The researchers also note that: “We include race as a predictor in our model because documenting racial/ethnic differences for public awareness is a necessary first step in reducing disparities, however we view race as a social construct.”
The Digital Divide in Health-Related Technology Use: The Significance of Race/Ethnicity
Uchechi A. Mitchell, Perla G. Chebli, Laurie Ruggiero, Naoko Muramatsu, Genterologist, November 2018
This study analyzes data from 1,336 white, Black, and Hispanic adults age 54 and older who completed the 2014 technology module of the University of Michigan Institution for Social Research Health and Retirement Study to identify differences in technology usage among different population groups. The research did not look at telemedicine specifically, due to its relatively low use prior to the pandemic. Compared with whites, older Black and Hispanic respondents were “less likely to use technology for health-related purposes after accounting for demographic characteristics, education, and health conditions.”
The disparities among whites and other racial and ethnic groups in the use of health management systems, social media use and searching online for health information became evident around age 62, which corresponds with people born prior to 1952. “Although the disparity widened with age for Blacks, it remained relatively similar at all ages for Hispanics,” the researchers wrote.
However, the study found a lag in usage rates such that Hispanics at age 62 have levels of technology use similar to whites who are 20 years older.
“Differences were not apparent for the use of text messaging, email, or health-related mobile applications,” the study authors add.
The effects of telemedicine on racial and ethnic disparities in access to acute stroke care
Michael J. Lyerly, Tzu-Ching Wu, et al., Journal of Telemedicine and Telecare, June 2015
This study sought to uncover whether previously reported racial and ethnic disparities in access to emergency or urgent stroke care were present in the use of telemedicine among racial and ethnic minorities in Texas. They found that the availability of telemedicine service gave Black patients similar access to acute stroke services, compared with white patients.
The researchers interviewed hospital personnel to determine the availability of acute care and emergency telemedicine services for stroke patients and “summing the population that could reach a primary stroke center or (PSC) or telemedicine spoke within specified time intervals using validated models.”
They found that 1.5 million more residents were able to access emergency or urgent-care stroke expertise through telemedicine, with the odds of Black patients being able to access stroke care within 60 minutes similar to that of white patients, even controlling for urbanization.
Rapid implementation of outpatient teleneurology in rural Appalachia: barriers and disparities
Roy E. Strowd, et al., Neurology Clinical Practice, July 2020
This study analyzed 1,011 telehealth neurology patient visits among patients in Appalachia, 44% of which were video and 56% of which were phone-only. Patients who chose to complete a video visit were younger, more likely to be female, white, and not have Medicare or Medicaid insurance. The most common barrier to scheduling video visits was technology limitations (46%). While patients from rural and urban communities were equally likely to be scheduled for video visits, patients from rural communities were more likely to consider future telehealth visits.
This paper was covered succinctly in NeurologyToday.
Assessment of Disparities in Digital Access Among Medicare Beneficiaries and Implications for Telemedicine
Eric Roberts and Ateev Mehrotra, Health Care Policy and Law, August 2020
For this research letter, Roberts and Mehrotra analyzed 638,830 responses to the 2018 American Community Survey by community-dwelling Medicare beneficiaries to determine the proportion who had access to a smartphone with a data plan or a desktop computer with high-speed internet access.
They examined those who had access to neither method of internet connection on a number of demographic and social dimensions. They found that individuals who reported having less than a high school education were the least connected, with 44.8% reporting no personal internet access. Among those respondents who were widowed, 40.6% reportedly did not have access to a connected smartphone or personal computer.
Among Black and Hispanic respondents, the proportions that did not have internet access were 37.3% and 31.6%, respectively. Among the total number of respondents, the proportion without any form of internet access was 26.3%.
For these patients, audio-only visits may be the only telehealth option, and although very little research has been done on the efficacy of such visits, “audio-only visits might be inadequate in some situations, such as when visual monitoring or diagnosis is important for care,” the researchers write.
“Although Medicare’s payment for audio-only visits at the same rate as video and in-person visits may be associated with improved access to care for those without digital access,” they add, “the inability to have a video visit may be associated with increased disparities in access to care.”
Telemedicine, COVID-19, and disparities: Policy implications
Gezzer Ortega, et al., Health Policy and Technology, August 2020
In this paper, a team of physicians reviews recent federal policy changes put in place to facilitate the rapid growth of telemedicine as a way to promote social distancing following the onset of the COVID-19 pandemic. They outline recommendations for government officials and health care centers designed to reduce the risk telemedicine exacerbating persistent racial and ethnic health care disparities.
They contend that access to telemedicine is driven by three main barriers: access to broadband internet; financial barriers to reimbursement; and lack of institutional commitment to equity in telemedicine.
The authors urge a national unified effort at the federal level to expand broadband internet coverage, as well as overarching policies to standardize reimbursement rates for telemedicine services, to the extent possible.
Federal policies that protect data security for patients are also critical, especially in the context of community of color, which have a history of mistrust in the health care system, the authors write. They also advocate for hospitals and other health care organizations taking a lead role in lobbying for necessary policies, promoting digital access and literacy, and developing “versatile, user-friendly, and culturally tailored tools” for patients.
“If equity is made an integral part of telemedicine now, it will help ensure that all can benefit from these innovative technologies moving forward.”
The Association Between Physician Race/Ethnicity and Patient Satisfaction: an Exploration in Direct-to-Consumer Telemedicine
Kathryn A. Martinez, et al., Journal of General Internal Medicine, July 2020
Bias against non-white physicians, as measured by patient satisfaction, was the subject of this study, which analyzed data on patients seeking care on a large, nationwide direct-to-consumer (DTC) telemedicine platform between July 2016 and July 2018 and their physicians.
Patients ranked their physicians on a scale of 1 to 5 stars, with 5 being the highest, and the rankings were made public for other patients to use in deciding on a telehealth provider. Increasingly, these types of patient experience measures are being used to determine physician compensation, the researchers explain, and lower ratings have been found to negatively affect physician job satisfaction.
“Physician bias against non-white patients has been widely documented,” the researchers write, “while patient bias against non-white physicians has been reported, but not well-studied.”
The sample included 119,016 encounters with 390 physicians, 233 of whom were classified as white American. The researchers controlled for patient age, gender and geographic region, based on those defined by the US Census Bureau. They also controlled for physician gender, specialty and geographic region.
Encounters with South Asian or East Asian physicians were less likely than those with white American physicians to result in 5-star ratings for patient satisfaction. Compared with encounters with white American physicians, those with Black American physicians, South Asian physicians, and East Asian physicians were more likely to result in reports of patient dissatisfaction (2 stars or fewer). The odds of a patient reporting dissatisfaction were the same with Middle Eastern and Hispanic physicians as they were with white physicians.
The effects were persistent even when adjusting for whether the patient received a prescription, “which has been shown to be the strongest predictor of satisfaction with telemedicine care,” the researchers write.
“Our findings indicate that in DTC telemedicine, patients express somewhat less top-box [5 star] satisfaction but considerably more dissatisfaction with some groups of non-white American physicians,” the study authors conclude. “This may have implications for these physicians’ compensation, reputation, and professional well-being.”
Five tips for journalists reporting on racial and ethnic disparities in telemedicine
1. Seek out experts focused on equitable health care and patient representatives. Hospitals and other health care organizations, particularly those associated with academic institutions, are increasingly establishing centers specifically designed to ensure equitable health care within their communities. Experts in health care disparities and the barriers to access for specific populations groups can be useful sources when reporting on research findings or other stories related to health care and communities of color.
Dr. Gezzer Ortega is the lead faculty for research and innovation for equitable surgical care within the Center for Surgery and Public Health at Brigham and Women’s Hospital in Boston. He says centers such as the one where he works are growing in number and becoming more proactive in their efforts to address inequities in health care.
Additionally, many hospitals have set up patient advisory councils that are designed to be representative of the patient population. Members of these councils can be good sources for stories. Journalists can also consider asking for comments from professional associations that represent physicians and other health care professionals from specific communities, such as the National Medical Association, which is the largest and oldest organization of Black physicians in the United States.
2. Ask researchers about the choices they made when conducting a study. To increase sensitivity in reporting on issues of racial and ethnic disparity in health care or along other dimensions, it’s important for journalists to ask and report on the reasons why researchers and others break out specific patient groups to highlight inequity.
“When someone does a disparities-related study or puts out a disparity statistic,” Ortega says, “I think it’s important to understand and to put upfront why they’re looking at that. So, when you look at a specific underserved population and you’re examining their outcomes, why did you look at that population? What was the rationale for categorizing a certain group in a certain way?”
3. When reporting on telemedicine tools or technology systems, ask whether the potential for bias was considered. Software developers in the past have admitted to creating widely used platforms without ever having considered the possibility that the design of an interface could inadvertently facilitate user bias. Putting users’ profile pictures and names front and center has been shown to increase bias.
Journalists can serve readers by finding out whether specific telemedicine tools or systems are multilingual or adaptable to multiple styles of interaction by users, says Rumi Chunara, an engineering professor at NYU, whose research focuses on finding ways to use data to improve public health. It can also be useful to identify whether participatory design was employed and whether user testing involved a diverse group of testers that were representative of the patient population.
4. Explore how community outreach is conducted. How do health care institutions and medical or mental health practices inform prospective users of telemedicine tools and systems about their availability and how they work? It’s important to understand whether outreach is occurring in partnership with community leaders and organizations that marginalized groups can trust and where they feel comfortable.
5. Ask about reimbursement. Policies regulating the availability and reimbursement of telemedicine services vary by state. Medicare is governed by federal policy, but states oversee Medicaid guidelines and decisions about Medicaid expansion, and commercial insurers’ policies can vary substantially.
For further reference, the Center for Connected Health Policy recently issued an updated State Telehealth Laws and Reimbursement Policies Report.
For more on covering the topic of telemedicine, see our roundup of research on telemedicine for substance use disorder treatment.
Expert Commentary