A study published in Health Affairs and conducted by the RAND Corporation made a big splash this week with a bold claim: That telemedicine doesn’t actually reduce healthcare costs because the increased convenience leads to increased utilization, which ultimately costs more than in-person care would have.

The study looked at claims data from a cohort of 300,000 employees with access to Teladoc through their employer. Researchers compared a cohort of telemedicine users to a cohort of non-telemedicine users and found that in the telemedicine users, visits to primary care doctors barely decreased, meaning that the Teladoc visits were mostly additive (visits that otherwise would not have occurred), rather than substitutive (visits that otherwise would have occurred in person). They found that 88 percent of visits were additive, and only 12 percent replaced in-person visits. The result: telemedicine cost the payer $45 per patient more than a plan without telemedicine would have.

Health Affairs and RAND are some of the most well-respected names in health policy research, and rightly so. The methodology of the study was strong, the sample size was large, and the conclusion bears careful consideration, and follow-up research. But there's also a lot to be said in response to the study, both by those who think the data is misleading and by those who think there's a bigger picture here than the one the study paints.

First of all, as might be expected, telemedicine vendors have some quibbles with the study.

Teladoc CEO Jason Gorevic says that Teladoc surveys every patient at the moment they seek care and asks them what they would have done if telemedicine wasn’t available. Their data -- even for that particular cohort -- is almost exactly the reverse of the Health Affairs study: 87 percent said they would have gone somewhere else to get care, while only 13 percent was additional utilization. The RAND study dismisses that survey data as subject to a number of psychological biases, but Teladoc's approach of trusting the patients themselves over statistical modeling certainly has an appeal. And for what it's worth, Gorevic added, the payer that was the subject of the study has not only continued to use Teladoc, but has added more members to the platform in the intervening years.

Roy Schoenberg, the CEO of Teladoc competitor American Well, pointed out that the data is at least four years old at this point, and changes in the technology and level of consumer trust in the technology since then might make the data invalid (not to mention that the study is of phone visits and a lot of telemedicine is offered by video). While in 2013 people were using telemedicine tentatively and still visiting their primary care provider when they were really concerned, now that the technology is more established and more often offered by their provider, consumers might be more likely to use telemedicine as a replacement. 

Both Schoenberg and Gorevic pointed to other research studies and published case studies that have shown the opposite of the Health Affairs study, or at least shown that overall telemedicine saves money. Research and consulting firm Advisory Board has some other thoughts about possible shortcomings.

I also spoke with health economist and blogger Jane Sarasohn-Kahn, of Health Populi. The biggest omission, to her mind, is that the study doesn’t consider the longterm impact of increased utilization of preventative care. If those extra visits keep patients out of the hospital because they catch a respiratory infection early and it prevents a more serious condition like pneumonia, those savings could make up for the cost of extra visits.

“The big question that goes unanswered in this study is what costs were averted because people used cheaper telehealth,” Sarasohn-Kahn said. “So all they’re looking at is the numerator, not the denominator. But on the other end, we don’t know what people saved by taking advantage of this quote-unquote ‘extra session’. We have to look at the whole equation, and this study only looks at one side of the equation.”

This becomes especially apparent, she added, when we consider populations for whom the alternative to telemedicine is not a primary care doctor (the cohort in the Health Affairs study was fairly affluent), but an emergency room or urgent care center, which represent a much steeper jump in price. In those cases, the one telemedicine visit that catches a big problem down the road makes up for a lot of visits that don’t.

Sarasohn-Kahn's plea to look at the bigger picture raises some even larger ideological considerations.

The whole idea of “overutilization” includes some implicit assumptions: One, that there is some appropriate amount of healthcare a person should seek, and, two, that that amount is whatever people were seeking under the status quo of in-person visits. Is it paternalistic to suppose patients can’t be trusted to know when they should and shouldn’t see a doctor? Rather than concluding that patients are overusing telemedicine because it’s convenient, could we just as easily conclude from the same data that up until now patients have been underusing their doctor because seeing them was too cumbersome?

“Talking to a doctor is not a recreational activity,” is how Teladoc’s Gorevic put it. “People talk to a doctor when they’re sick, and when they need care, and when they need help.”

More than that, evaluating telemedicine in a vacuum is just as problematic as comparing it only to the status quo. The healthcare system of the future will very likely be a tiered one, where telemedicine is only one of a series of escalating steps. Patients might start with an automated app or online symptom-checker, move to a chat or messaging with a nurse or care manager, see a doctor by telemedicine if that isn’t enough, and come in in person only as a last resort. If every step of that process is weighted equally as “a doctor visit”, certainly technology will appear to increase the number of doctor visits. But in reality, it will decrease the strain on the system where it matters most.

“One of the biggest mistakes of the way people think of telehealth is they think of it as just another way for people to get to the same kind of healthcare,” Schoenberg told me. “If you think of it that way, overutilization is going to be an immediate byproduct and it’s going to be a terrible thing. But the perspective around telehealth of most of the payers and most of the brokers and most of the employers out there is that it changes the way patients utilize healthcare and they become more discerning, almost intuitively, about what can be handled by basic forms of triage before they have to go to a telehealth visit.”

Payers and providers who are using or considering telemedicine shouldn’t ignore the Health Affairs study, or any other research that calls the perceived benefits of new technology into question. But they should recognize that utilization is a broad, complicated issue with many facets. And that when it comes to innovation, investigators always run the risk of judging next year’s technology by last year’s metrics.