Virtual Primary Care—Is Its Expansion Due to COVID-19 All Upside? | Infectious Diseases | JAMA Health Forum | JAMA Network
[Skip to Content]
Curated health policy research and original commentary from across the JAMA Network
[Skip to Content Landing]

Virtual Primary Care—Is Its Expansion Due to COVID-19 All Upside?

  • 1NYU School of Medicine, New York, New York
  • 2Department of Population Health, NYU School of Medicine, New York, New York
  • 3Geriatric Cardiology Program, NYU School of Medicine, New York, New York
  • 4Center for Healthcare Innovation and Delivery Science, NYU Langone Health, New York, New York
  • 5Division of Healthcare Delivery Science, Department of Population Health, NYU School of Medicine, New York, New York

Virtual primary care has long been promoted as the future of health care in the US.1 Proposed benefits include improved access for rural populations, more convenient care, and reduced overhead expenses. Despite this promise, progress until recently had been slow, in part because of limited reimbursement, restrictions on where telehealth could be delivered, interstate care prohibitions, discomfort with technology, privacy concerns, and limited public enthusiasm.

The US response to coronavirus disease 2019 (COVID-19) has swept many of these barriers away.2 Medicare and private insurers now cover virtual visits, even without video. Patient costs have plummeted. Aetna, for instance, has eliminated co-payment for telemedicine visits. Platforms that are noncompliant with the Health Insurance Portability and Accountability Act, such as Skype and Zoom, can now be used. Medicare has relaxed its rules about interstate care provision and service location. Unease over seeing a clinician virtually has vanished—superseded by fears of contracting COVID-19 in a waiting room. In many ways, these changes represent a major step forward, and one that we hope will persist beyond the pandemic. But, with progress also comes a set of questions for policy makers and researchers. How well can virtual care replace physical care? What are the risks?

Lessons may be learned from the prior experience of the UK National Health Service (NHS) with Babylon Health. Babylon, a health care start-up, launched its service in fall 2017. Using a small primary care practice in West London as a point of entry, Babylon flooded Facebook, bus stops, and the subway with advertisements promising free, 24/7, virtual-first primary care and rapidly expanded to serve nearly 60 000 patients across London and Birmingham. A high-profile, technology start-up entered a market where there was no advertising (and no “poaching” of patients from other practices) and a long history of clunky and failing health information technology. Unsurprisingly, it generated an enormous amount of controversy.3 By offering video and telephone appointments, it also disrupted the traditional geographic link between primary care practices and patients. Thanks to an independent evaluation, the NHS experience with Babylon Health also generated evidence on who used it and for what.4 Much of this learning will be relevant for the US as we try to understand how virtual care is used during the pandemic and as we look to its continued use as the pandemic wanes.

Who used Babylon Health? Compared with the average NHS practice, patients were younger (94% were younger than 45 years), relatively healthy (only 10% had 1 of 5 major long-term conditions, which was less than the age-adjusted London average), wealthier, and more educated (more than 80% were educated to at least college level). This pattern was arguably by design; the NHS restricted patients with chronic medical conditions and those who were pregnant from signing up, citing safety concerns about these groups using a model based primarily on virtual consultations.4 But it also largely reflects what we already know about digital health care—that there are disparities in engagement driven by ethnicity, age, and socioeconomic status and that these disparities can lead to widening health inequalities.5,6

Disparity in digital access is a particular risk in the context of COVID-19, which is already disproportionately affecting low-income and minority groups both directly through health effects and indirectly through economic hardship and loss of health insurance. Further research is needed, but it seems likely that the virtual care rollout has been more robust at large academic medical centers, which typically serve primarily privately insured patients, than at public hospitals, which by design serve patients who are insured by Medicaid or uninsured.7,8 Virtual care also does not consistently address individuals’ own barriers to accessing care—for example, lack of broadband access and interpreter services.

As well as thinking about who virtual care works for, we need to consider what conditions it works for. The evaluation of Babylon Health showed that patient experience was largely very positive, even for those with long-term conditions (though less so for those with particularly complex needs), and patients felt that Babylon Health offered higher-quality care than their previous practice.4 However, the rate of turnover was high (around a quarter of patients deregistered), with some suggestion this was driven by changing health needs and a desire to be able to directly book a face-to-face appointment. It may be that virtual-first care was inappropriate or inconvenient for certain conditions.

Evidence from other countries, notably Sweden,9 on what patients consulted for may shed light on this issue (Babylon did not share data on this, citing commercial confidentiality). This showed that the most common condition among women was urinary tract infection, and among men, it was upper respiratory infection. In the US, prior to the COVID-19 pandemic, telehealth was primarily used for psychiatry and urgent care. Now, with almost all care moving online, the number of conditions treated with telehealth has widened.10 For some, it will work well; for some, it will be simply adequate but worth the trade-off with the risks of in-person visits; for others, it will be ineffective or possibly dangerous. Discerning which conditions fall into which category will be essential to building an effective, safe, post–COVID-19 telehealth world. To accomplish this goal, health systems will need to ensure that they are collecting the data on who is using their telehealth services, for what reasons, and—to the extent possible—what their quality of care and outcomes are. When contracting with virtual care providers, insurers need to ensure that they, unlike the NHS in Babylon Health’s case, can access the data collected. Analysis of these data can wait, but data collection cannot.

The COVID-19 pandemic has rapidly accelerated virtual primary care in the US health care system. Almost overnight, many regulatory and attitudinal barriers have been overcome, and the US now stands ready to benefit from these opportunities. However, if done poorly, there is a considerable risk that those who are most in need of virtual consultations will not be able to access them. Without collection of rich data now, the learning and experience of clinicians and patients on what works and what does not work in telehealth will be lost. Following the unprecedented expansion of telehealth, it is imperative that this expansion be evidence based and fair.

Article Information

Corresponding Author: Holly Krelle, MPhil, NYU School of Medicine, 550 1st Ave, New York, NY 10016 (

Conflict of Interest Disclosures: None reported.

Duffy  S, Lee  TH.  In-person health care as option B.   N Engl J Med. 2018;378(2):104-106. doi:10.1056/NEJMp1710735PubMedGoogle ScholarCrossref
Webster  P.  Virtual health care in the era of COVID-19.   Lancet. 2020;395(10231):1180-1181. doi:10.1016/S0140-6736(20)30818-7PubMedGoogle ScholarCrossref
Oliver  D.  David Oliver: lessons from the Babylon Health saga.   BMJ. 2019;365:l2387. doi:10.1136/bmj.l2387PubMedGoogle ScholarCrossref
Salisbury  C; Ipsos MORI and York Health Economics Consortium. Evaluation of Babylon GP at Hand: final evaluation report. Accessed June 29, 2020.
Weiss  D, Rydland  HT, Øversveen  E, Jensen  MR, Solhaug  S, Krokstad  S.  Innovative technologies and social inequalities in health: a scoping review of the literature.   PLoS One. 2018;13(4):e0195447. doi:10.1371/journal.pone.0195447PubMedGoogle Scholar
Gray  DM  II, Joseph  JJ, Olayiwola  JN.  Strategies for digital care of vulnerable patients in a COVID-19 world—keeping in touch.   JAMA Health Forum. 2020;1(6):e200734. doi:10.1001/jamahealthforum.2020.0734Google Scholar
Hollander  JE, Carr  BG.  Virtually perfect? telemedicine for COVID-19.   N Engl J Med. 2020;382(18):1679-1681. doi:10.1056/NEJMp2003539PubMedGoogle ScholarCrossref
Hong  Y-R, Lawrence  J, Williams  D  Jr, Mainous  A  III.  Population-level interest and telehealth capacity of US hospitals in response to COVID-19: cross-sectional analysis of Google Search and National Hospital Survey data.   JMIR Public Health Surveill. 2020;6(2):e18961. doi:10.2196/18961PubMedGoogle Scholar
Ekman  B, Thulesius  H, Wilkens  J, Lindgren  A, Cronberg  O, Arvidsson  E.  Utilization of digital primary care in Sweden: descriptive analysis of claims data on demographics, socioeconomics, and diagnoses.   Int J Med Inform. 2019;127:134-140. doi:10.1016/j.ijmedinf.2019.04.016PubMedGoogle ScholarCrossref
Poppas  A, Rumsfeld  JS, Wessler  JD.  Telehealth is having a moment: will it last?   J Am Coll Cardiol. 2020;75(23):2989-2991. doi:10.1016/j.jacc.2020.05.002PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words