January 10, 2019 — For many Americans, a visit with their doctor no longer requires an actual trip to the doctor’s office. More physicians are offering so-called “telemedicine” services, where they provide care to patients via smart phones, tablets, and computers. And while telemedicine is being more used more frequently across the U.S., it’s still relatively uncommon, according to a new study led by Michael Barnett, assistant professor of health policy and management at Harvard Chan School. In this week’s episode we speak to Barnett about the rise of telemedicine, ways to further increase its use, and how this technology could improve health care for patients in the years to come.
MICHAEL BARNETT: It actually can represent multiple different types of care delivery. I think the umbrella for telemedicine is really any medical care that’s delivered remotely. And so I think the classic form of telemedicine that we most often hear about and that policymakers think about is actually a live video conference. So it’s basically exactly the same as a doctor’s visit except you’re talking with a doctor on a video screen instead of in-person, and of course the doctor can’t examine you. So that’s kind of the most, quote unquote, “classic” form of telemedicine. But there are many other kinds of technologies that could be considered telemedicine.
The one that’s most relevant for the paper we’re talking about is called direct to consumer telemedicine. And this is basically a live video conference or an interactive chat you can have the doctor through an app on your phone or on the computer, you know, through a website. And that’s basically where the patient themselves seek out a doctor and say, you know, I think I have a sinus infection. I’m going to log onto this website, find a doctor, and tell them about it so I can figure out what I need to do.
Other forms of telemedicine that we didn’t necessarily capture in our paper but also are under the umbrella include what some people call e-visits, which is basically a visit where you just submit a bunch of information maybe through a patient portal by email to your physician. And then they give you advice on what to do. And then there’s other kinds of telemedicine that can be more on the provider end, like PCPs to specialists. And that’s something I also study, something called e-consults, where basically the specialist is doing almost a telemedicine visit with a specialist to figure out what to do for their patients.
NOAH LEAVITT: Some of the background for this paper is that many states have passed these parity laws which mandate that insurers cover these different types of telemedicine visits. So can you explain more about what these laws do and I guess what has driven this legislation? I think it’s like 30 something states.
MICHAEL BARNETT: The telemedicine industry and a lot of rural medicine advocates have wanted to create a more fertile environment for telemedicine to grow in states. And one of the problems is before these laws were passed, insurers and Medicare wanted to basically pay less for telemedicine services because they don’t necessarily involve quite the same level of intensity they would potentially argue. The point of these laws was to say if you’re delivering an equivalent service, whether it’s remotely through telemedicine or in-person, they should be reimbursed the same amount. Not every state has a parity law that is that strict some of the states have laws which actually just mandate that telemedicine is covered by insurance so that if you receive a telemedicine service, they’re not mandating what price it should be. But they’re saying at least that should be covered. And the goal with these laws is really to help nudge providers into actually providing services to make it more financially attractive.
NOAH LEAVITT: And so the thinking was that if you’re a doctor and you’re not being reimbursed the same for a telemedicine visit, you’d basically have less incentive to offer that to your patients.
MICHAEL BARNETT: Yeah, I mean, you know, whenever a physician needs to decide how to pick up a new– make a new investment in their practice, they’re going to look at, what can I make if I actually buy all the equipment necessary and get all the necessary licensure and certifications to offer telemedicine services? And if it doesn’t reimburse that well, then you know, they’re probably not going to offer it unless there’s some other very compelling kind of mission-driven reason why they might set up a system. But that’s not really a recipe for telemedicine to scale nationally.
NOAH LEAVITT: And so I know in this study you looked at this large kind of insurance database to basically kind of assess what does the adoption of telemedicine look like across the US. And so what did you find there?
MICHAEL BARNETT: So some of the key findings– one was that telemedicine has been growing very steadily over the past 15 years or so with a particularly steep rise in 2016 that I’ll talk about. Even though it’s been growing quite steadily– in fact, it’s sort of a 50% annual compound growth– you know, telemedicine was really quite rare more than 10 years ago. And so we’re only at a use rate of around 7 people per 1,000 across the country in this particular insurance database. The other patterns that we saw are actually starting in 2016, there’s this huge spike in growth. And it was almost all driven by the use of primary care telemedicine. And it coincided with a few major insurers starting coverage for this direct to consumer telemedicine I talked about. So some of the big companies, some of their names are Doctor On Demand or American Well are a couple of large providers. And there are many others. A few other interesting trends that we saw– we had just talked about parity laws. And in fact, there was stronger growth in telemedicine for mental health in states that passed parity laws than those that didn’t. And actually one, of the more interesting findings was a very simple one, which is that when you looked at the scope of telemedicine that was being offered in this insurer across the whole country, the vast majority of it was actually either for primary care or for mental health. And only a small sliver, like around 8% or so, was actually for telemedicine for every other specialty out there. And the reason why that’s surprising is because telemedicine is often touted and thought of as a way to get very specialized care out to rural settings or to people who can’t otherwise access it. And so to see basically non-mental health specialty care be such a small slice was actually quite surprising to us. In terms of mental health telemedicine, though, it hasn’t grown quite as quickly in the past two years as primary care telemedicine, but it’s grown pretty steadily. And its adoption has very different patterns than primary care telemedicine. So mental health telemedicine is much more common in regions without psychiatrists and where patients are more likely to– in regions where the average patient has lower education. It’s a poor area, more rural, the kind of scenario that a lot of people think of as the test case for telemedicine. On the other hand, for primary care telemedicine, a lot of which is this direct to consumer kind of app-based telemedicine, its growth was basically explosive everywhere, and it grew kind of irrespective of how rural your region was or how wealthy the region was or how many PCPs were in your area. It just seemed to grow really quickly everywhere.
NOAH LEAVITT: That’s interesting because, I mean, I guess if you live in a rural area where there likely are in a lot of mental health care providers, you would assume that in those same areas, you wouldn’t have as many primary care physicians as well. Or is that generally not the case where you tend to see just more primary care physicians? I mean, do you have a sense of why you kind of saw those differences between mental health and primary care?
MICHAEL BARNETT: You know, I think physician supply is– I think, you know, the number of specialists in any given area is probably– they’re all pretty correlated with one another because there are places that physicians want to live. And there are places where physicians are less likely to want to live there unless they’re from there or they have some really compelling reason to go there. In terms of the difference between mental health and primary care, I think for mental health care, it seems to be more driven by the need to just see any doctor, right? So if you have no mental health provider in your area and there’s a group of patients who just really desperately need to have an interaction with a mental health provider, telemedicine is filling a key gap there. On the other hand, for primary care telemedicine, what we’ve learned from some research over the past few years from my collaborator Ateev Mehrotra, who’s at Harvard Medical School, is that it seems like a lot of direct to consumer telemedicine probably just adds more utilization, more care, on top of what people are already getting. And so it just makes it very easy to check in with a doctor about a cold or sinusitis or you’re feeling nauseous. And so I think people aren’t using it to replace a primary care physician but as a more convenient option when they might have otherwise just kind of toughed it out or thought about not going to the office.
NOAH LEAVITT: Yeah, it’s interesting because I think of it as more of a convenience thing. Like, it’s like an Uber Eats or just food delivery. It’s just another kind of convenience based app versus more than mental health kind of being more of– like, like filling that critical gap of care.
MICHAEL BARNETT: Exactly, exactly.
NOAH LEAVITT: And so I know we’ve talked in the past about, for example, the use of telemedicine to address kind of issues such as substance use disorders. So how– I guess how is telemedicine currently being used in that regard? And I guess how could it be used more effectively to address opiate addiction, things like that?
MICHAEL BARNETT: Yes, so telemedicine for substance use disorder is a big topic. It also featured pretty prominently in the Support Act, which was recently signed into law, which includes a wide range of policies to try to promote access to treatment for substance use disorder. So it seems like at least in this population that substance use treatment via telemedicine is still quite uncommon. So it’s actually several fold less common than even mental health telemedicine. Again, my colleague Ateev Mehrotra had just published a paper looking at substance use disorder through telemedicine specifically in this population. It’s provided by the people you think would provide it. So it’s very heavily weighted towards social workers and psychiatrists but also family practitioners who we’re guessing are likely buprenorphine providers, buprenorphine being one of the most common treatments for opioid use disorder. And you know, right now, I think it’s still very much in the nascent stage. It’s something I think is going to grow very quickly. I think now that Medicare is going to pay for it and more payers are trying to create more options for substance use disorder treatment, I think we’re going to see more activity in the commercial space, more start-ups, insurers trying to get into it to be able to capture this kind of key treatment gap.
NOAH LEAVITT: And from the patient perspective, is it a case too where patients maybe don’t know it’s an option in some cases, and that’s why they haven’t been taking advantage of it?
MICHAEL BARNETT: I’m sure that– I mean, I think it’s both providers and patients. I think a lot of providers don’t realize they could actually provide the service if they needed to. And also I think for provider, you know, how do you attract patients, right? So if I’m in– let’s just take an example. Let’s say I’m in Alaska, and I’m in Anchorage. And I want to offer telemedicine. How do I communicate that I can do telemedicine in some city that’s a six hour drive away? You know, I think that’s probably, you know, a bit of a barrier for providers. And then for patients, how do they learn about providers? It’s kind of a classic marketing problem, right? Like, how do you actually get everyone to realize that this service is available and actually they can create something that makes everybody happier by doing it?
NOAH LEAVITT: When it comes to kind of increasing adoption of telemedicine, what are some of the technical barriers? I mean, whether it’s internet access, whether it’s in a low income area, people might not have a smartphone to download the app. So what are some of those kind of technical barriers, and are there ways to overcome any of those?
MICHAEL BARNETT: So for direct to consumer telemedicine, I actually don’t think smartphone availability is really much of a gap anymore because smartphones are– you know, smartphone penetration is huge in almost every market. And you know, even in poor areas, there are very cheap smartphones that can really run almost any app. So actually, I don’t– so that’s an area– I think that’s part of the reason why we see adoption increasing so much because consumers basically can just decide for themselves. And so many people have smartphones, and it’s really quite simple. On the provider side, there are a lot more technical challenges there. So a lot of states have fairly elaborate licensing and certification guidelines for being able to set yourself up as a telemedicine provider. The clinic that you are interfacing with needs to have broadband access that can support video if the definition of telemedicine you’re using is kind of live video.
And so there’s just the technology, but actually, we think it’s probably the licensing and the regulatory aspects that may be a bigger barrier because anybody can kind of log online and buy the technology and have some come and install it. But I think actually getting through the bureaucracy and regulations can be a very protracted process and takes up quite a bit of energy. And I think a lot of clinicians do not relish the idea of going through that.
NOAH LEAVITT: Is the licencing and regulation– is that kind of more like a privacy data concern because just anytime it’s a new service type of care, it’s just complicated?
MICHAEL BARNETT: Yeah, exactly. It’s just– you know, departments of public health and health and human services, whenever a clinic offers some kind of new service you, need to be certified that your staff can actually do it, that you know how to manage problems with it. I mean, you know, we could argue whether or not these regulatory requirements are really necessary for safety. But I think the idea is more safety and competence so that a provider is not offering a service for which they are not able to safely operate it. So for example, in my clinic, we have a microscope, and we do– sometimes, we do– we look at slides to diagnose certain conditions. And we have to take like a competency test every 6 to 12 months to certify that we can actually interpret the images in the slides. And that’s necessary for regulations to certify that the physicians in the office actually know what they’re doing. Otherwise, we might not be able to use a microscope.
NOAH LEAVITT: And you spoke in the beginning that these parity laws seem like a step in the right direction in terms of encouraging more providers to offer telemedicine because of the changes in reimbursement. But what else can be done, either if it’s marketing or from a policy perspective, to increase adoption of telemedicine, both providers offering it and then patients being able to take advantage of it and use it?
MICHAEL BARNETT: Yes, so I think what we learned in our paper was that you have to think about adoption of different kinds of telemedicine differently. So we don’t need to do anything for primary care, direct to consumer. I mean, it’s exploding on its own. The market seems to really be ready for that convenience care. And whether or not that is a good or a bad thing for our health system is kind of a conversation for another time and needs some other research. But I don’t think we need to really work on that too much. For the mental health side, you know, parity laws are good. But I think if we think it’s an important policy goal to expand telemedicine we probably have to, you know, inject more resources into actually making it happen. So for instance, a lot of places in the country that would benefit from telemedicine may not have broadband access. They’re rural enough that they actually really don’t have a good broadband infrastructure yet.
And a lot of these places may just not have a provider who knows about it or is willing to invest the amount of time and energy to make it happen. And an analogy I think of is back when Obama first became president and we were in the midst of a recession, the Obama administration was basically looking for, quote unquote, “shovel ready projects” to inject resources into the economy to just kind of try to get things moving along. And from that came something called the HITECH Act, which is basically an enormous incentive program for hospitals to adopt electronic health records. And that really changed the game in terms of adoption of electronic health records. And now we’re at a point where almost every hospital in the country has an electronic health record. And it was really the injection of resources from the HITECH Act that seemed to have really catalyzed that.
NOAH LEAVITT: Is there any concern on the patient end about am I still getting the same quality of care when I have a telemedicine visit, or do patients generally seem kind of open to this and kind of willing to try it? Is there any hesitation on the patient end, I guess?
MICHAEL BARNETT: I’m not an expert on the patient experience for telemedicine, so I can’t speak directly to that from sort of research that I’ve done. I think what I can say, really, from my perspective as a primary care physician is I think patients really do value seeing a doctor in person. I think that’s one of the reasons why telemedicine is still being used pretty uncommonly. You know, it’s still less than a percent of people every year in our dataset. And I think in America, we do have a lot of doctors, and people are used to seeing them in person. And studies on the quality of care of telemedicine have been actually pretty positive. I think there is obviously a lot of value to having that human connection and being able to get to know your doctor face to face. But actually, the physical exam, actually laying hands on a patient, can be very helpful in many clinical circumstances. But most of the time, you don’t really get that much from it. You don’t have to examine a patient to come up with a perfectly adequate plan of care.
And so I don’t think that quality is the main issue. I think it’s really comfort level with having medical care without that personal connection, with a person who you feel like you can trust. It’s a very personal thing, receiving medical care. You’re really opening up to someone who is basically a stranger.
NOAH LEAVITT: So why don’t you just finish up by kind of going back saying you talked a bit earlier where, I mean, the telemedicine that’s currently being provided is overwhelmingly mental health and primary care and not kind of these other specialties. You know, like with substance use disorders, are there other kind of opportunities for telemedicine in those other specialties that you would see if there’s greater adoption of telemedicine? Like, what are the potential future benefits of this wider adoption?
MICHAEL BARNETT: So I do think there’s a lot of potential. I think the conditions that telemedicine could be best for are, I think, conditions where more frequent contact with clinicians is probably better for your health outcomes. So there are a lot of conditions– I think there’s a tendency for us to think seeing the doctor more is better for your health or something like that, right? That it’s better to see your doctor more often if you need to.
But sometimes, a lot of people see the doctor too often, or they’re not really getting much from seeing the doctor. There are certain conditions where it’s probably pretty good to check in with a clinician of some kind pretty regularly because things can get out of whack without– you know, before the patient realizes it. So examples of conditions like that– one could be congestive heart failure. It’s a common one. We need to actually check people’s weight and follow their medication adherence. There have been a lot of different telemonitoring studies for heart failure that have had mixed results, but I that remains a pretty good one. Another one is chronic pain. So people who are chronic pain sufferers, you know, their medication needs kind of fluctuate. And as things happen in their lives and their health goes up and down, they need a lot of help with managing their pain medication. I think that could be another big opportunity. The other thing about chronic pain is actually, there’s very good randomized control trial data that chronic pain can be managed just as effectively remotely as in person. Another example could be, for instance, in neurology for a follow-up for survivors of stroke or people with epilepsy to be able to titrate seizure medications. Another example could be, say, in pulmonology, patients with severe emphysema. So these are all conditions basically where things can kind of get bad enough quickly enough that people need to go to the hospital multiple times a year. But we think we can potentially prevent some of those exacerbations if people check in with a doctor on a more regular basis.
Use of telehealth rising—slowly (Harvard Chan School news)