top of page

Telehealth has never yet shaken up provider market share. But could it? Two reasons to think yes

With stakes as high as ‘maybe healthcare won’t always be so local’, we should all pay close attention to even the softest of signals that a game-changing telehealth disintermediation may be in progress.

Most people who work in healthcare know that while telehealth has long held the theoretical power to disrupt market share and utilization patterns—well, it hasn’t done it yet.

Here’s a recap of all the opportunities that telehealth has been given, without big disruption coming to pass:

  • Pre-pandemic: First-mover providers who invested in telehealth capabilities did not win significant market share from their competitors. Nor did telehealth reduce avoidable utilization or replace visits. Direct-to-consumer (D2C, as it’s known) telehealth was a tiny blip on the massive healthcare service radar screen.

  • During the pandemic: Telehealth usage went through the roof and D2C investment surged and, extraordinarily, neither of these things disrupted market share. Patients used telehealth to interact with the providers they already had. D2C telehealth remained small.

  • Immediately post-pandemic: The whole industry waited with bated breath to see if disruption would occur. But again, it did not, despite the new telehealth precedent, and the zillion telehealth-related D2C healthcare companies working for a share of the market. Data from 2021 and 2022 showed that, with the exception of virtual behavioral health, traditional, local providers still owned the vast majority of post-pandemic telehealth visits. D2C telehealth remained a quirky, tiny corner of the market, with usage concentrated in behavioral health, urgent-care-type services outside of PCP office hours, or online prescribing for things people felt uncomfortable talking to their PCP about (1).

What’s different now? Reasons to look twice

And yet, when I look at the market, I think the situation is not stable; and that non-traditional, non-local entities might be about to shake up both utilization and share.

Two key reasons:

  1. Virtual-first primary care and virtual-first health plans are packed with disruptive potential—and poised to grow.

  2. The data supporting the whole ‘providers still own the lion’s share of visits’ assertion feels like thin ice. As an industry, we have plenty of data about the share of visits that are telehealth; but it doesn’t reliably show what delivery model those visits belong to, nor what kinds of entities are operating those models. That means if there were disintermediation by non-traditional, non-local providers going on, it might go unmeasured for some time.

To avoid being surprised by this hypothetical, yet massively consequential, change in the healthcare market, let’s unpack this idea a bit more—and then you can tell me what you think.

‘Look twice’ reason #1: VHCs and VHPs are packed with disruptive potential

To have this conversation, we need a couple key vocabulary terms.

  • Virtual-first primary care (VHC*): A primary care model in which patients are encouraged by the delivery model to consult the provider team virtually before showing up at a brick-and-mortar facility. The practice itself may not even have a brick-and-mortar presence. There is also virtual-forward primary care, in which the virtual layer is a little more optional, and there is definitely a traditional, local office component. *If you’re wondering why the abbreviation for virtual-first primary care is VHC and not VFPC, we don’t know why. What the ‘C’ in VHC stands for…Clinic? Care? Sure, let’s go with that. If you want to get really cute, you can also call virtual-first care ‘V1C’.

  • Virtual-first health plans (VHP): Insurance products that use plan design, incentives (such as $0 copay virtual visits), and communication to encourage enrollees to use virtual care before showing up for in-person care at a physical location. Basically, it’s the insurance design complement to the VHC delivery model.

VHP and VHC don’t have to go together; a VHC could be a covered option in a traditional health insurance plan. But one can imagine that VHP and VHC impact would be greater together (more on impact in a moment). And, if VHPs are taken up, that would certainly accelerate growth in use of VHCs.

Both VHP and VHC aim to solve the same constellation of healthcare market problems:

  • Reducing payer cost: A virtual-first (or forward) delivery model should be lower-cost to operate. It should also filter out avoidable low-value utilization through better upfront triage and coordination: preventing avoidable ED and inpatient admissions and strategically directing patients to lower-cost, lower-priced sites of care. These savings should translate into lower total spending, premiums, and out-of-pocket costs.

  • Improving patient access and convenience: Patients receive 24/7 access to advice and guidance. In theory, this would especially benefit patients in remote, provider-shortage areas (though in reality, this has not been the case because of rural bandwidth problems; telehealth usage to date is concentrated in more metropolitan areas).

Guessing at the disruption potential of VHC/VHP

The VHC/VHP model is new and therefore lacks great data on its power to disrupt healthcare service utilization and market share. But there are soft indicators suggesting the impact could be significant.

For the Medicare population, VHCs/VHPs could significantly reduce avoidable utilization. As one anecdotal piece of the puzzle, UPMC’s virtual-first clinic reports that it reduced ED use by 50% compared to in-person PCP visits (2). To be fair, UPMC’s clinic is all-payer; but I am personally assuming that most of that shift happened among the Medicare population, because there almost certainly isn’t that amount of preventable ED utilization to wring out of a commercial population.

For the commercial/employer-sponsored population, VHCs/VHPs could accelerate site-of-care shifts and drive downstream share to lower-priced providers. In an employer population, less of the spend-reduction opportunity resides in avoiding low-value utilization (with the exception of unneeded surgeries; more on this in a moment). Instead, the goal is more about price arbitrage. To save employers on total spend, the VHC would have to shift care into lower-cost settings, direct patients to lower-negotiated-rate acute care providers, and/or encourage patients to utilize out-of-market centers of excellence for scheduled surgeries (via episodic bundles, which may cost less for the episode, but also is where the filtering out of avoidable surgeries comes in). All these shifts would impact downstream providers—positively, if they are the lower-priced destination players, and negatively, if they are the local high-priced choice.

In terms of total spending, the VHC/VHP combo is forecasted to achieve major reductions for employers. Caveat: most of this forecasting is VHP-cheerleading, not based on actual data. But if the estimates are even directionally true, the impact of these programs would be big. For example, Trustmark Health Benefits (which has a VHP and a partnership with Teladoc’s VHC) forecasts a 10-30% reduction in total employer healthcare spend (3) – which, again, it would have to achieve by shifting care to lower-cost settings and tilting the acute care, specialty playing field to lower-priced providers in any given market.

And while there is no impact track record—only forecasts—it stands to reason that any VHC that has the power to reduce avoidable ED use by 50% should have the power to make those other, price-affecting, downstream market share-shifting changes too.

The critical ‘who’ of VHCs/VHPs: If insurer-aligned, greater downstream disruption would follow

The degree and kind of downstream service disruption that would be caused by a VHC/VHP would depend greatly on who is operating it. Basically, the more aligned the VHP/VHC is with an insurer, the more likely it is to actively shift care to lower-cost sites and redirect acute share to lower-priced providers.

Insurers are currently positioned to dominate the VHP/VHC landscape. Yes, providers are developing, or even operating, their own VHCs; a 2022 survey found that 70% of health systems had identified virtual-first primary care as priority (4). And provider systems that have insurance arms are also offering their own VHPs, which presumably integrate with their own system-operated VHCs (5).

But it seems fair to assume that insurer-offered VHPs/VHCs will be the majority of what employers and Medicare Advantage consumers are going to see. In theory, a national insurer could offer a VHP that is designed to cover enrollee use of any VHC, including those offered by traditional, local providers. But in reality, if a VHP is offered by a national insurer, it will likely be integrated with an insurer-designated, non-local, non-traditional provider VHC that is built into the plan. A major part of the fuel for insurers standing up and pushing VHPs is the perceived opportunity to marry plan design to the telehealth subsidiaries within insurers’ ‘diversified services’ arms, which are an increasingly important source of overall profitability for these types of organizations (6).

Bottom line: In a high-VHP/VHC adoption scenario, a chunk of the provider market share game would shift to the insurers’ home turf—and that would significantly alter follow-on utilization and market share.

VHPs uptake outlook: Poised to grow?

All major insurers are now offering VHPs. Those insurers, together with telehealth companies and benefits consultants, are currently pushing VHPs hard (7).

Are employers buying? Some information is available but it’s incomplete (see next section on data blind spots).. What we know about uptake in the commercial world is this: most large employers already cover some form of telehealth and a substantial minority are specifically covering (but not forcing) some kind of VHC. Uptake of VHPs is still small overall (estimated at 7%), but it's projected by some to double by 2024 (keeping in mind this rapid growth would be on a small base).

How big could VHP uptake get?

I could see VHPs growing significantly, as long as the VHPs/VHCs are operated well enough not to cause a 90s-HMO-type enrollee backlash about choice. Employers say they want a healthcare plan that reduces both total spend and employee spend without raising deductibles. In theory, the VHP does all these things (again remembering that the most bullish cost-savings projections are coming from companies that are trying to play in the VHP market).

According to survey data, “nearly half of large employers say they need to better understand this new model before they could consider it […] and 39% are not interested or don’t believe it would be a good fit for their organizations.” (8) Employers are conservative buyers and the model is very new. If VHPs reduce spending and please consumers, it’s hard to see why those waiting for track-record data—and even a significant portion of those holdouts who today say it’s not for them—wouldn’t be won over down the road.

On the consumer side, prospective enrollees are being offered the chance to pay less for the VHP premium: low or no copay/deductibles for visits and generic medications, and a range of other perks. For example, in the case of United/Optum’s VHP NavigateNow!, the plan comes with a one-year membership to the Peloton app (9). Survey data suggests that a substantial number of consumers would be attracted to a virtual-first plan, especially if it reduced their out-of-pocket costs. Again, as long as the VHPs (and the associated VHCs) avoid alienating enrollees with heavy-handed gatekeeping, it seems hard to see the argument against.

Reason to look twice #2: Step out onto the thin ice of today’s data on telehealth disruption

Now let’s explore the idea that if a telehealth-enabled disintermediation did happen, we wouldn’t know about it for a while—because the industry lacks good, current data on this topic.

Missing data on the ‘what’ of virtual visits in general

Relative to its strategic importance, we currently have very little ability to spot uptake of VHC amid the sea of data on telehealth usage. For example, we have data on percentage of all visits that take place virtually vs. in person, which can be cut by specialty. That tells us that 8% of ‘general medicine’ visits were virtual in 2021 (10). But not only is that data not especially fresh; it also says nothing about how much of that general medicine volume was VHC vs. regular primary care (or some other kind of general medicine).

Meanwhile, the most-cited data source on ‘who’–the one used to show that traditional, local providers, not disruptors, still dominate telehealth market share–is RAND’s survey that asks respondents how often telehealth visits happened with “your own doctor.” Results show that the share of visits described as with ‘own physician’ are mostly in the high 90% range (depending on type of visit) (11). The problem with this construction is that it assumes the patient’s ‘own doctor’ is the one they also see in person. In a virtual-first environment, the patient’s ‘own doctor’ could very well start to be one who is not traditional, or local.

To get at what share of visits are happening with traditional, local providers vs non-traditional, non-local providers, we are going to need more studies like this one (12) that ask whether recent telehealth visits were had with a provider that the patient also sometimes sees in person. Until and unless we have lots of that kind of data, telehealth market intelligence could start to seriously undercount the market share that may be migrating over to virtual-first providers, no matter who those providers are. Which brings us to the next point...

Missing data on the ‘who’ of VHCs/VHPs

The VHC/VHP uptake data that we currently have is silent on the key distinction between insurer-aligned vs traditional, local provider-aligned VHPs. This lack of visibility is a huge blind spot for figuring out how much disintermediation is under way.

Faulty assumption: The 25% ‘ceiling’ on the share of visits that could reasonably convert to telehealth

This point is a little different in kind, but conventional wisdom tends to lean on a data point that says that across specialties (with various highs and lows among them), about 25% of all visits could be virtual.

In general, traditional providers have found this type of ceiling statistic reassuring. For one thing, everyone knows the heftiest portion of that 25% is behavioral health, and few providers are looking to compete on behavioral health due to its unfavorable economics. And while behavioral health certainly should be well integrated with primary care—which everyone knows is the competitive battleground for downstream market share—to date, that integration hasn’t happened. Instead, behavioral health is most often functionally adjacent to primary care in the delivery system. So if that part of the clinical service portfolio tipped to telehealth, and even if it tipped to non-traditional, non-local providers, many traditional providers have decided they can live with it.

But I think this ceiling is weak, because of the elastic nature of health care services. If VHC/VHP really took off, it would likely add a new initial layer of services onto today’s continuum of care. If that happened, we would grow the pie, ending up with, say, 35% total telehealth share of visits—because 25% of what we had before went virtual, plus, we added another 10% at the front of the legacy pattern.

In sum: Keep a sharp eye out for change

It seems unwise to take providers’ ownership of telehealth market share for granted, especially when it comes to virtual primary care. The industry needs to start putting more feelers out for data about VHC/VHP impact, who the dominant VHC/VHP providers are, and VHC/VHP uptake. And if the answer that starts to reveal itself is, “insurer-aligned VHCs/VHPs are growing, saving money, and pleasing employers and consumers”—then we should all expect some major downstream traditional, local provider market share effects.


Union is a researcher-founded insight company. We understand the history, incentives, and forces that drive the healthcare industry. We also have a passion for learning and teaching others about how the whole thing works. Want to chat? Reach out at


(1) In 2019, urinary tract infection (53%), erectile dysfunction (21%), and contraception (13%) together comprised 87% of D2C visits; compare to 2.3% of conventional PCP visits. Mehrotra, A. et al. “Receipt of Out-of-State Telemedicine Visits Among Medicare Beneficiaries During the COVID-19 Pandemic,” JAMA Health Forum, September 16, 2022. Available at:

(2) Siwicki, B. “How UPMC's virtual-first healthcare model works,” Healthcare IT News, February 28, 2023. Available at:

(3) AHA. “3 Takeaways from the Rise of Virtual-First Health Plans,” Available at:

(4) Landi, H. “HIMSS 2022: Health systems see virtual care companies as biggest competitive threat: report.” Fierce Healthcare, March 16, 2022. Available at: contains link to Refers to Chartis survey data; public version of source report shows that health systems are working on virtual first clinics (and has interesting data about health systems perceiving Amwell and Teladoc as competitors) but does not share specific data about implementation stage re: virtual-first clinics.

(5) For example: BusinesWire. “Priority Health Lowering Costs for Virtual-First Health Plans in 2022,” Devember 13, 2021. Available at:

(6) Frank, R. and Milhaupt, C. “Medicare Advantage spending, medical loss ratios, and related businesses: An initial investigation,”Brookings, March 24, 2023. Available at: Hat tip to this earlier coverage of the Brookings research: Minemeyer, P. “Brookings: A look at profitability in Medicare Advantage,” July 13, 2022. Available at:

(7) Fouhy, R. and Umlad, B. “Virtual-First Health Plans, Explained” Mercer Health News. June 9, 2022. Available at:; PWC. “The Evolving Model of Virtual-First Plans”. Available at:; AHA. “3 Takeaways from the Rise of Virtual-First Health Plans,” Available at: “Trustmark believes that depending on the plan design, virtual-first programs can drive savings of between 10% and 30% of an employer’s overall health plan spend”

(8) Fouhy and Umlaut (see above)

(10) Bestsennyy, O. et al. “Telehealth: A quarter-trillion-dollar post-COVID-19 reality?” McKinsey, July 9, 2021. Available at:

(11) Fischer, S. et al. “Patients Log On to See Their Own Doctors During the Pandemic,” January 7, 2021.

(12) Mehrotra, A. et al. “Receipt of Out-of-State Telemedicine Visits Among Medicare Beneficiaries During the COVID-19 Pandemic,” JAMA Health Forum, September 16, 2022. Available at:

See also:

Methodology for D2C market estimates:

  • Medicare: According to Mehrota et al (citation above), 5% of Medicare patient telehealth visits, Jan-June of 2021, were out of state. Of that 5%, 63% were with a provider the patient had seen in-person in the last two years. Thus Medicare D2C estimate of roughly 2% total.

  • Commercial: According to Jain et al (citation above), 87% of D2C visits are for things that represent only 2.3% of conventional PCP visits


Join our mailing list to see future posts

Thanks for submitting!

bottom of page