Ishani Ganguli, MD, MPH

Cascades of care are common and can lead to significant harms for patients, clinicians, and the health care system at large. In this commentary, we argue that there are 2 ways to reduce cascades: decrease the use of unnecessary services that often initiate cascades (ie, close the floodgates) and mitigate cascades once they begin (ie, slow the flow through the floodgates). So far, most efforts to address cascades have focused on identifying, measuring, and educating clinicians on low-value services, with only modest success. We explore potential solutions for both closing the floodgates and slowing a cascade once the floodgates have been opened, including information to assist patients and clinicians in making better decisions, relationships that enable shared decision-making, and policy changes. Ultimately, reducing cascades while maintaining our commitment to high-quality care requires equipping patients and clinicians with the information, tools, and support needed to embrace uncertainty.

Reducing cascades while maintaining our commitment to high-quality care requires equipping patients and clinicians with the information, tools, and support to embrace uncertainty.

Addressing harmful cascades of care requires strategies to decrease the use of unnecessary, low-value services that often initiate cascades and also stop cascades once they begin.

Decreasing the use of low-value care is challenging, but there is an opportunity to implement policies that incentivize patients and clinicians to put recommendations around avoiding low-value services into practice.

To mitigate cascades once they begin, we need information to assist patients and clinicians in making better decisions, relationships that enable shared decision-making, and accompanying policy changes.

 refers to a succession of medical services that are often fueled by the desire to avoid even the smallest risk of a serious condition.

 They can follow from incidental or marginal findings on screening and diagnostic tests, some of which are themselves of low value. Regardless of their origins, they can be difficult to stop once the floodgates are opened, often leading to harms for patients, clinicians, and the health care system at large.

In certain cases, further evaluation of unexpected findings can serve a useful purpose, such as uncovering a dangerous yet treatable cancer.

 More often, these pursuits are ultimately futile and costly. For example, consider an unwarranted preoperative electrocardiogram prior to a low-risk procedure like cataract surgery, where an inconclusive result may prompt additional referrals, diagnostic testing, and catheterization that rarely confirms significant cardiovascular disease.

 The process can be emotionally and physically exhausting for patients without offering them any real benefits.

Cascades of care are extremely common. In a recent survey of US internists, more than 99% of respondents reported that they had experienced cascades from incidental findings. One-third experienced cascades with no clinically important or intervenable outcome on at least a monthly basis. Notably, more than half of respondents said cascades led to physical, psychological, and financial harm to patients and wasted time, frustration, and anxiety for physicians.

There are 2 ways to reduce cascades: (1) decrease the use of unnecessary services that often initiate cascades (ie, close the floodgates) and (2) mitigate cascades once they begin (ie, slow the flow through the floodgates). Both are critical to preventing these floods from causing catastrophic damage.

So far, most efforts to address low-value care, including the Choosing Wisely campaign and the US Preventive Services Task Force (USPSTF) rating system, have focused on identifying, measuring, and educating patients and clinicians on low-value services.

 However, these initiatives have had only modest success in reducing the use of low-value care.

 One reason for this might be the lack of policies that incentivize clinicians or patients to put recommendations into practice. Of note, Section 4105 of the Patient Protection and Affordable Care Act authorizes Medicare plans to remove coverage for preventive care services that are deemed harmful or unnecessary by the USPSTF (ie, D rating).

 By pushing implementation of this provision, the new administration has an opportunity to meaningfully address the prevalence of low-value care.

Closing the floodgates makes sense when it is possible to identify a specific service as low value. However, this can be difficult in many cases, given the poor consensus around how to incorporate clinical nuance, patient preferences, and cost-benefit trade-offs into this designation.

 Moreover, cascades are often triggered by initial tests that are clinically appropriate.

 Given these factors, it is also important to mitigate cascades once they begin.

Slowing the flow may be more challenging; it is easier to avoid discovering an “incidentaloma” than to unsee it. Clinicians order that second or third test because they fear missing something important, fear being sued, feel an obligation to follow practice and community norms, or believe they are responding to patient preferences.

 To slow a cascade once the floodgates have been opened, we need (1) information to assist patients and clinicians in making better decisions, (2) relationships that enable shared decision-making, and (3) policy changes.

Information to help patients and clinicians make better decisions. 

Physicians often feel frustrated but compelled to pursue additional testing for incidental findings when results are unfamiliar and are not accompanied by information to contextualize their clinical importance.

 Yet, cascades may occur even when such information is provided. For example, a multicenter study showed that including data on the prevalence of age- and sex-specific abnormalities in healthy adults on spine imaging reports (thereby “normalizing” these abnormalities) did not decrease subsequent spine-related utilization.

This speaks to the importance of providing the right information at the right time. In the spine imaging study, the information included in reports was based on population-level estimates, making it challenging to generalize to individual patients, and did not include recommendations for follow-up. For information to be useful to clinicians, it should be clear, specific, and provided at the point of care. The information should also be guided by high-quality evidence rather than just expert opinion, which tends to be more conservative and promote cascades. To this end, there is a need for more research on when a laboratory “abnormality” does or does not require further testing and on best practices for evaluation of various incidentalomas.

In the wake of the 21st Century Cures Act, patients increasingly have access to their test results, and any deviation from “normal” can provoke anxiety. Ideally, test results would be presented alongside interpretative comments in language that is easily understood. Another solution might be allowing patients to review imaging results with radiologists directly. In general, the public should be better sensitized to the idea that medical tests and procedures carry harms and that attempts to “prevent” devastating but rare outcomes can have risks of their own.

Trusting patient-clinician relationships. 

Using reliable, point-of-care information, we can leverage trusting patient-clinician relationships to engage patients in shared decision-making to minimize the harms of cascades. When faced with a dilemma (eg, after an incidental finding is found, should testing be pursued to rule out a rare but life-threatening diagnosis?), clinicians need to understand and factor in what is important to patients. In an open conversation, clinicians can explain the clinical implications of findings and potential harms and benefits of further testing; similarly, patients can share their concerns and reasons for and against further testing. These conversations should start early—when deciding to order a test in the first place—and may reassure clinicians who pursue cascades due to fears of litigation.

Clinicians need training and resources to navigate these conversations. Although most clinicians report having conversations about cascades with patients, they are often limited to specific use cases (eg, pulmonary nodules incidentally found on chest radiography).

 Education on overuse should be baked into the medical curriculum such that clinicians feel comfortable talking to patients about uncertainty. Decision aids and scripts can be integrated into the electronic health record to help guide these conversations.

It is important to consider the legal and policy context in which these conversations are happening. For instance, the shift to value-based payment may incentivize clinicians to reduce unnecessary medical tests and treatments, although evidence remains mixed.

 Alongside efforts to reduce the number of low-value tests ordered, future models should invest in point-of-care solutions to stem low-value cascades once they are triggered.

Medical malpractice reform is another solution that is appealing in theory, although reforms to date have had limited impact on low-value care utilization. A fear of malpractice lawsuits can propel cascades

—in one study, the possibility of a lawsuit increased the intensity of care that patients received, although this extra testing and treatment did not improve outcomes.

 One promising approach is the creation of “safe harbor” laws that protect clinicians from liability if they adhere to clinical practice guidelines based on rigorous evidence.

 There is also evidence that specific communication behaviors, such as educating patients on the purpose of tests and procedures, soliciting patients’ opinions, and checking understanding, are associated with fewer malpractice claims.

 Thus, safe harbor laws that encourage clinicians to follow guidelines and explain the context and implications of incidental findings may mitigate cascades motivated by legal concerns.

Ultimately, reducing cascades while maintaining our commitment to high-quality care requires becoming more comfortable with uncertainty.

 Although the increasing attention to low-value care is an important step forward, to really stop the flood, we need strategies that provide patients and clinicians with the information, tools, and support to embrace the uncertainty inherent in medicine.


 Harvard Medical School (PC, IG), Boston, MA; Department of Internal Medicine and Center for Value-Based Insurance Design (AMF), Ann Arbor, MI; Division of General Internal Medicine and Primary Care, Department of Medicine, Brigham and Women’s Hospital (IG), Boston, MA.

 Dr Fendrick has been a consultant for AbbVie, Amgen, Centivo, Community Oncology Alliance, Covered California, EmblemHealth, Exact Sciences, Freedman Health, GRAIL, Harvard University, Health & Wellness Innovations, Health at Scale Technologies, MedZed, Merck, Montana Health Cooperative, Penguin Pay, Risalto, Sempre Health, State of Minnesota, US Department of Defense, Virginia Center for Health Innovation, Wellth, Yale–New Haven Health System, and Zansors; has performed research for the Agency for Healthcare Research and Quality, Arnold Ventures, Boehringer Ingelheim, Gary and Mary West Health Policy Center, National Pharmaceutical Council, Patient-Centered Outcomes Research Institute, PhRMA, Robert Wood Johnson Foundation, and State of Michigan/CMS; and holds outside positions as co-editor-in-chief of 

®, member of the Medicare Evidence Development & Coverage Advisory Committee, and partner in V-BID Health, LLC. Dr Ganguli has received a grant from the Agency for Healthcare Research and Quality on cascades of care and has presented on cascades at conferences. Ms Chandrashekar reports no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

 Concept and design (PC, AMF, IG); drafting of the manuscript (PC, AMF, IG); critical revision of the manuscript for important intellectual content (PC, AMF, IG); and supervision (AMF, IG).

 Ishani Ganguli, MD, MPH, Brigham and Women’s Hospital, 1620 Tremont St, 3rd Floor, Boston, MA 02120. Email: iganguli@bwh.harvard.edu.

1. Mold JW, Stein HF. The cascade effect in the clinical care of patients. 

. 1986;314(8):512-514. doi:10.1056/NEJM198602203140809

2. Ganguli I, Simpkin AL, Lupo C, et al. Cascades of care after incidental findings in a US national survey of physicians. 

. 2019;2(10):e1913325. doi:10.1001/jamanetworkopen.2019.13325

3. Ganguli I, Lupo C, Mainor AJ, et al. Prevalence and cost of care cascades after low-value preoperative electrocardiogram for cataract surgery in fee-for-service Medicare beneficiaries. 

. 2019;179(9):1211-1219. doi:10.1001/jamainternmed.2019.1739

4. Korenstein D, Chimonas S, Barrow B, Keyhani S, Troy A, Lipitz-Snyderman A. Development of a conceptual map of negative consequences for patients of overuse of medical tests and treatments. 

. 2018;178(10):1401-1407. doi:10.1001/jamainternmed.2018.3573

5. Choosing Wisely. Accessed December 22, 2020. https://www.choosingwisely.org

6. U.S. Preventive Services Task Force Ratings. U.S. Preventive Services Task Force. Accessed April 1, 2021. https://www.uspreventiveservicestaskforce.org/uspstf/us-preventive-services-task-force-ratings

7. Ganguli I, Lupo C, Mainor AJ, et al. Assessment of prevalence and cost of care cascades after routine testing during the Medicare annual wellness visit. 

. 2020;3(12):e2029891. doi:10.1001/jamanetworkopen.2020.29891

8. Mafi JN, Reid RO, Baseman LH, et al. Trends in low-value health service use and spending in the US Medicare fee-for-service program, 2014-2018. 

. 2021;4(2):e2037328. doi:10.1001/jamanetworkopen.2020.37328

9. Compilation of Patient Protection and Affordable Care Act. HHS. 2010. Accessed April 1, 2021. https://www.hhs.gov/sites/default/files/ppacacon.pdf

10. Beaudin-Seiler B, Ciarametaro M, Dubois RW, Lee J, Fendrick AM. Reducing low-value care. 

. September 20, 2016. Accessed April 1, 2021. https://www.healthaffairs.org/do/10.1377/hblog20160920.056666/full/

11. Zafar HM, Bugos EK, Langlotz CP, Frasso R. “Chasing a ghost”: factors that influence primary care physicians to follow up on incidental imaging findings. 

. 2016;281(2):567-573. doi:10.1148/radiol.2016152188

12. Jarvik JG, Meier EN, James KT, et al. The effect of including benchmark prevalence data of common imaging findings in spine image reports on health care utilization among adults undergoing spine imaging: a stepped-wedge randomized clinical trial. 

. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713

13. Schwartz AL, Chernew ME, Landon BE, McWilliams JM. Changes in low-value services in year 1 of the Medicare Pioneer Accountable Care Organization Program. 

. 2015;175(11):1815-1825. doi: 10.1001/jamainternmed.2015.4525

14. Ganguli I, Simpkin AL, Colla CH, et al. Why do physicians pursue cascades of care after incidental findings? a national survey. 

. 2020;35(4):1352-1354. doi:10.1007/s11606-019-05213-1

15. Frakes MD, Gruber J. Defensive medicine: evidence from military immunity. National Bureau of Economic Research working paper No. 24846. July 2018. Accessed December 21, 2020. https://www.nber.org/system/files/working_papers/w24846/w24846.pdf

16. Mello MM, Kachalia A, Studdert DM. Medical liability — prospects for federal reform. 

. 2017;376(19):1806-1808. doi:10.1056/NEJMp1701174

17. Levinson W, Roter DL, Mullooly JP, Dull VT, Frankel RM. Physician-patient communication. the relationship with malpractice claims among primary care physicians and surgeons. 

. 1997;277(7):553-559. doi:10.1001/jama.277.7.553

18. Colla CH, Kinsella EA, Morden NE, Meyers DJ, Rosenthal MB, Sequist TD. Physician perceptions of Choosing Wisely and drivers of overuse. 

Articles

Stopping the Flood: Reducing Harmful Cascades of Care

The coronavirus disease 2019 (COVID-19) pandemic forced health systems to offer video and telephone visits as in-person visit alternatives. Although video visits offer some benefits compared with telephone visits, they require complex setup, which may disadvantage some patients due to the “digital divide.” Our objective was to determine patient and neighborhood characteristics associated with visit modality.

 This was a cross-sectional study across 1652 primary care and specialty care practices of adult patients at an integrated health system from April 23 to June 1, 2020.

 We used electronic health record and administrative data. Our primary outcome was visit modality (in-person, video, or telephone), which was captured using billing codes. We assessed predictors of using video vs telephone using multivariable logistic regression. We used hierarchical logistic regression to determine the contribution of patient-, physician-, and practice-level components of variance in the choice of video or telephone visits.

 We analyzed 231,596 visits by 162,102 patients. Sixty-five percent of the visits were virtual (31.7% telephone, 33.5% video). Patients who were older than 65 years (adjusted odds ratio [AOR], 0.41; 95% CI, 0.40-0.43), Black (AOR, 0.60; 95% CI, 0.57-0.63), Hispanic (AOR, 0.76; 95% CI, 0.73-0.80), Spanish-speaking (AOR, 0.57; 95% CI, 0.52-0.61), and from areas with low broadband access (AOR, 0.93; 95% CI, 0.88-0.98) were less likely to use video visits. Practices (38%) and clinicians (26%) drove more of the variation in video visit use than patients (9%).

 Telemedicine access differences may compound disparities in chronic disease and COVID-19 outcomes. Institutions should monitor video visit use across demographics and equip patients, clinicians, and practices to promote telemedicine equity.

The coronavirus disease 2019 pandemic prompted a transition to telemedicine, including video and telephone visits. Video visits offer some benefits, but they may disadvantage some patients due to the “digital divide.” Using data from a large, integrated health system, we determined patient, clinician, clinic, and neighborhood characteristics associated with visit modality.

Patients who were older than 65 years, Black, Hispanic, Spanish-speaking, nonportal users, or from areas with low broadband access were less likely to use video visits.

Practice (38%) and clinician (26%) factors largely drove the variation in video visit use.

Institutions should promote equitable access to all telemedicine modalities.

During the coronavirus disease 2019 (COVID-19) pandemic, social distancing requirements compelled health systems, practices, clinicians, and patients to use video and telephone visits as alternatives to in-person visits.

 This shift was further supported by policy changes—for example, CMS offered temporary payment parity across in-person, telephone, and video visits.

 There are concerns that the shift to telemedicine may exacerbate disparities in health care access. Early evidence on telemedicine use overall has demonstrated less use among female and non–English-speaking patients compared with other groups, but no differences among racial/ethnic groups.

Given key trade-offs between telemedicine modalities, there may be additional, potentially concerning differences in access to video and telephone visits. Compared with telephone visits, video visits allow some aspects of a physical examination and a more personal connection between clinicians and patients. However, video requires more complex setup and broadband internet access, which may present barriers for older adults, racial/ethnic minorities, and those with limited English proficiency (LEP) who are on the wrong side of the “digital divide.” The digital divide refers to the gap between those who have access to and can meaningfully engage with technology and those who cannot. For example, at least 21 million people lack broadband internet access in the United States, many from underserved communities that also face worse health outcomes.

 Further, a recent study suggests that 13 million older adults are not ready for video visits due to difficulties using technology.

 Video visits may also present technical and logistical challenges for some practices and clinicians to set up and incorporate into workflows. However, we know little about differences in and drivers of the use of video vs telephone visits. Therefore, we analyzed detailed electronic health record (EHR) data from a large, integrated health system to fill these gaps.

Understanding drivers of differences in use of telephone and video visits would help to inform rapidly changing telemedicine payment policies and practices. In this study, we examined trends in use of outpatient care, including telephone and video visits across 1652 outpatient primary care and specialty practices in a large, integrated health system.

We used data from Mass General Brigham (MGB) in Boston, Massachusetts, a large, integrated health system with 16 member organizations across New England. We extracted EHR data from the enterprise data warehouse, which includes patient demographic information collected during registration and detailed clinical encounter and billing data.

We identified a cohort of adult patients 18 years or older who were attributed to an MGB primary care physician (PCP) (defined by the EHR “PCP” field) as of March 1, 2020. We assessed all outpatient encounters occurring between March 1 and June 1, 2020. We excluded encounters that required in-person care (eg, dialysis, procedures, chemotherapy or radiation administration).

Our primary outcome was visit modality (in-person, video, or telephone visit), captured using billing codes. Telemedicine visits were identified by code GT for video visits and GPH for telephone visit. For visits with billing codes that did not clearly indicate modality, we included the visits in total visit counts but not in analyses about types of modalities.

We chose patient characteristics that have been previously demonstrated to influence telemedicine use: age (18-44, 45-64, ≥ 65 years), gender (male, female), race/ethnicity (White, Black, Hispanic, Asian, other), preferred language (English, Spanish, other), insurance type (commercial, Medicaid, Medicare, veteran/military, missing/no insurance), and patient portal access (activated, not activated).

 We merged patient zip codes with the American Community Survey data to capture zip code–level broadband internet access (quartiles based on percentage of the population with broadband internet access), educational attainment (quartiles based on percentage of the population with a high school diploma), and income (quartiles based on median household income).

We performed descriptive statistics to show trends in use of telephone, video, and in-person visits between March 1 and June 1. We then focused on the period April 23 to June 1 (the study period), which began 1 month after the shift to telemedicine (Massachusetts issued a stay-at-home advisory on March 23). We chose this period to bypass transition challenges, including rapid workflow and technical requirements, that could affect initial uptake of video visits. During this period, we examined trends in the percentage of all telemedicine visits done via video by patient sociodemographic categories. We assessed predictors of video vs telephone use with bivariate analyses using χ

 tests as appropriate for the previously described patient- and zip code–level variables. We then built a multivariable logistic regression model in which the outcome was video or telephone visit. Finally, we built a hierarchical logistic regression model to determine the relative contribution of patient-, physician-, and practice-level components to variance in the choice of video or telephone visits. Hierarchical models account for the fact that patients are nested within clinicians and clinicians are nested within clinics. In a sensitivity analysis, we repeated our main analyses among primary care practices alone. The MGB Institutional Review Board approved this study and waived the use of informed consent. Analyses were performed using R statistical software (version 3.5.3).

We examined 746,356 patients with a PCP at MGB. From March 1 to June 1, telephone visits were more common in the early weeks of telemedicine expansion and were then overtaken by video visits (

). During the study period, we found lower video visit use among Black and Hispanic patients (compared with White and Asian patients), among Spanish-speaking patients (compared with English-speaking patients), and among patients 65 years and older (compared with those aged 18-64 years) (

). These trends developed early and grew steadily over time.

During the study period, 162,102 patients (21.7% of all attributed patients) attended 231,596 visits to primary care and specialty practices. Of all visits during this period, 65.2% were telemedicine visits (73,353 [31.7%] telephone and 77,530 [33.5%] video) (

). Among patients with a telemedicine visit, 53,999 (33.3%) patients had at least 1 video visit, 50,576 (31.2%) patients had at least 1 telephone visit, and 8187 (5.0%) patients had at least 1 of each.

In bivariate analyses, patients using video visits were more likely to be White, portal users, enrolled in commercial insurance, and living in areas with higher income and broadband access. In adjusted analyses, patients who were older than 65 years (adjusted odds ratio [AOR], 0.41; 95% CI, 0.40-0.43), Black (AOR, 0.60; 95% CI, 0.57-0.63), Hispanic (AOR, 0.76; 95% CI, 0.73-0.80), Spanish-speaking (AOR, 0.57; 95% CI, 0.52-0.61), and not active patient portal users (AOR, 0.39; 95% CI, 0.38-0.40) were less likely to use video visits (

). Additionally, patients living in areas with the lowest broadband internet access (AOR, 0.93; 95% CI, 0.88-0.98), lowest median income (AOR, 0.49; 95% CI, 0.46-0.52), and lowest educational attainment (AOR, 0.84; 95% CI, 0.80-0.88) were less likely to use video visits. The hierarchical model revealed that practice (38%) and clinician (26%) factors accounted for more of the variance in video visit use than patient-level factors (9%), whereas 27% of the variance was unexplained. In our sensitivity analysis examining primary care visits, we found similar results (

At a large, integrated health system scaling telemedicine visits in response to the COVID-19 pandemic, we found lower use of video vs telephone visits among older, Black, Hispanic, and Spanish-speaking patients that developed almost immediately and persisted. We also found that clinicians and practices largely drove this variation in the use of video vs telephone visits, suggesting an important target for intervention.

Various factors likely influence the choice between a telephone and video visit, including patient access to technology, an organization’s telemedicine infrastructure, and patient and clinician preferences. To the extent that video is better than telephone, it is important to expand access to video visits. At the same time, given that telephone visits may be more accessible for some and are often reasonable substitutes, it will be important to offer patients flexibility and ensure payment parity for both visit types.

We found that Spanish-speaking patients were less likely to use video visits. This extends prior findings that showed decreased telemedicine use among patients with LEP.

 For example, Eberly et al found that non–English-speaking patients were 50% less likely to use telemedicine, although they did not find differences in video visit use.

 Given these findings, practices might develop video visit workflows that integrate interpreter services and assign dedicated language-concordant guides to train patients in the use of video visits.

We found that patients in areas with lower levels of broadband internet access were less likely to engage in video visits, corroborating prior work showing that internet access predicts telemedicine use and patient portal adoption.

 Importantly, internet access is necessary yet not sufficient for use of video visits. Patients must also have a video-enabled device, as well as digital literacy, or the ability to navigate telehealth platforms, which underserved and older patients often lack.

Clinician- and practice-level factors contributed to the gaps in video visit use. Clinicians and practices that serve primarily underserved patients may be less equipped to provide video visits given additional implementation requirements. Thus, a focus on supporting video visit infrastructure and training clinicians on using video visits may help address these challenges. Due in part to implicit biases, clinicians may also offer telephone over video visits for some patients based on assumptions about their ability or desire to engage in video visits.

 A structured approach to screening patients for video visit readiness can help check these assumptions. Further, practices should develop efficient workflows to make it easy for both patients and clinicians to use video visits.

Addressing differences in use the of visit modalities will require a multilevel approach. To ensure that greater telephone visit use among underserved populations does not contribute to decreased access to medical care, payment parity for telephone and video visits is critical. In March 2020, CMS extended parity for both telephone and video relative to in-person visits; these changes should persist beyond the pandemic, and commercial payers should follow suit. In addition, policy initiatives to narrow digital divides, such as investments in telemedicine and broadband infrastructure, are essential. Policy makers should establish universal broadband laws and funding to ensure that patients have essential digital access and to mitigate pressures faced by health systems to connect their patients. The recent COVID-19 Telehealth Program, which provided $200 million for health care organizations to expand their telemedicine infrastructure, is a start.

 At the organizational level, effective workflows are important to ensure that patients and providers feel satisfied with care in the digital setting. Additionally, collaboration with community organizations already focusing on digital literacy will address digital divides in local communities. Finally, EHR and telemedicine vendors should develop inclusive platforms that account for the varied literacy, numeracy, and usability needs of underserved and older populations.

We found differences in telemedicine access that may compound existing racial, ethnic, and language-based disparities in chronic disease outcomes and COVID-19 case rates and mortality. With the pandemic still ongoing, institutions should carefully monitor video visit use across patient demographics and equip patients, clinicians, and practices to promote equitable access to all telemedicine modalities.

The authors would like to thank Harrison T. Hubbell, BS, of Mass General Brigham, for his help with data extraction.

 Harvard Medical School (JAR, JRB, TDS, IG), Boston, MA; Division of General Internal Medicine and Primary Care, Brigham and Women’s Hospital (JAR, TDS, IG), Boston, MA; Division of General Internal Medicine, Massachusetts General Hospital (JRB), Boston, MA; Mass General Brigham (TDS), Boston, MA.

 The authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

 Concept and design (JAR, JRB, TDS, IG); acquisition of data (JAR, JRB, TDS, IG); analysis and interpretation of data (JAR, IG); drafting of the manuscript (JAR, IG); critical revision of the manuscript for important intellectual content (JAR, JRB, TDS, IG); statistical analysis (JAR, IG); and supervision (IG).

 Jorge A. Rodriguez, MD, Division of General Internal Medicine and Primary Care, Brigham and Women’s Hospital, 1620 Tremont St, Boston, MA 02120. Email: jarodriguez1@partners.org.

1. Mehrotra A, Ray K, Brockmeyer DM, Barnett ML, Bender JA. Rapidly converting to “virtual practices”: outpatient care in the era of Covid-19. 

NEJM Catalyst: Innovations in Care Delivery

. Published online April 1, 2020. doi:10.1056/CAT.20.0091

2. Trump administration makes sweeping regulatory changes to help U.S. healthcare system address COVID-19 patient surge. News release. CMS; March 30, 2020. Accessed March 30, 2020. https://www.cms.gov/newsroom/press-releases/trump-administration-makes-sweeping-regulatory-changes-help-us-healthcare-system-address-covid-19

3. Eberly LA, Khatana SAM, Nathan AS, et al. Telemedicine outpatient cardiovascular care during the COVID-19 pandemic: bridging or opening the digital divide? 

. 2020;142(5):510-512. doi:10.1161/CIRCULATIONAHA.120.048185

4. Verma S. Early impact of CMS expansion of Medicare telehealth during COVID-19. 

. July 15, 2020. Accessed August 17, 2020. https://www.healthaffairs.org/do/10.1377/hblog20200715.454789/full/

5. 2019 broadband deployment report. Federal Communications Commission. May 29, 2019. Accessed August 17, 2020. https://www.fcc.gov/reports-research/reports/broadband-progress-reports/2019-broadband-deployment-report

6. Lam K, Lu AD, Shi Y, Covinsky KE. Assessing telemedicine unreadiness among older adults in the United States during the COVID-19 pandemic. 

. 2020;180(10):1389-1391. doi:10.1001/jamainternmed.2020.2671

7. Park J, Erikson C, Han X, Iyer P. Are state telehealth policies associated with the use of telehealth services among underserved populations? 

. 2018;37(12):2060-2068. doi:10.1377/hlthaff.2018.05101

8. American Community Survey (ACS). United States Census Bureau. Accessed August 17, 2020. https://www.census.gov/programs-surveys/acs

9. Reed ME, Huang J, Graetz I, et al. Patient characteristics associated with choosing a telemedicine visit vs office visit with the same primary care clinicians. 

. 2020;3(6):e205873. doi:10.1001/jamanetworkopen.2020.5873

10. Rodriguez JA, Lipsitz SR, Lyles CR, Samal L. Association between patient portal use and broadband access: a national evaluation. 

. Published online January 10, 2020. doi:10.1007/s11606-020-05633-4

11. Wilcock AD, Rose S, Busch AB, et al. Association between broadband internet availability and telemedicine use. 

. 2019;179(11):1580-1582. doi:10.1001/jamainternmed.2019.2234

12. Anderson M. Mobile technology and home broadband 2019. Pew Research Center. June 13, 2019. Accessed March 15, 2020. https://www.pewresearch.org/internet/2019/06/13/mobile-technology-and-home-broadband-2019/

13. Nouri S, Khoong EC, Lyles CR, Karliner L. Addressing equity in telemedicine for chronic disease management during the Covid-19 pandemic. 

. Published online May 4, 2020. doi:10.1056/CAT.20.0123

14. Anthony DL, Campos-Castillo C, Lim PS. Who isn’t using patient portals and why? evidence and implications from a national sample of US adults. 

. 2018;37(12):1948-1954. doi:10.1377/hlthaff.2018.05117

15. Paradies Y, Truong M, Priest N. A systematic review of the extent and measurement of healthcare provider racism. 

. 2014;29(2):364-387. doi:10.1007/s11606-013-2583-1

16. Chapman EN, Kaatz A, Carnes M. Physicians and implicit bias: how doctors may unwittingly perpetuate health care disparities. 

. 2013;28(11):1504-1510. doi:10.1007/s11606-013-2441-1

17. van Ryn M, Saha S. Exploring unconscious bias in disparities research and medical education. 

. 2011;306(9):995-996. doi:10.1001/jama.2011.1275

18. COVID-19 Telehealth Program. Federal Communications Commission. Accessed August 17, 2020. https://www.fcc.gov/covid-19-telehealth-program

During the coronavirus disease 2019 (COVID-19) pandemic, we found lower use of video vs telephone visits among older, Black, Hispanic, and Spanish-speaking patients, driven largely by clinician and practice factors.

Ishani Ganguli, MD, MPH

Articles

Stopping the Flood: Reducing Harmful Cascades of Care

Differences in the Use of Telephone and Video Telemedicine Visits During the COVID-19 Pandemic