Ghost networks remain a major problem in health care – and behavioral health is no exception.
Broadly, ghost networks create barriers for individuals trying to access behavioral health services through their payer. There are many reasons why ghost networks exist, with payers and providers both contributing to the problem.
In this BHB+ TALKS conversation, senior editor Laura Lovett sits down with Sandeep Acharya, CEO of Octave, to delve into the pressing issue of ghost networks in behavioral health. This discussion highlights the challenges patients face in accessing in-network behavioral health care, the reasons behind these ghost networks and the regulatory landscape aimed at addressing them.
The conversation also touches on potential solutions, including the role of digital providers, the need for improved payer-provider communication and more.
The full transcript from BHB’s conversation with Acharya is below for BHB+ members.
Lovett: Hello. Welcome, everyone, to our very first BHB+ TALKS conversation. I’m Laura Lovett. I’m the senior editor over at Behavioral Health Business. Today, I’m so lucky to be joined by Octave CEO Sandeep Acharya to talk about ghost networks in behavioral health and some other topics in behavioral health as well. TALKS is one of our many perks that we’re going to be offering to BHB+ members.
As part of our new exclusive membership program, you’ll have access to TALKS, more deep-dive reporting, exclusive networking opportunities, event discounts and more. We’re launching BHB+ with the support of our exclusive sponsor, The Braff Group, a go-to source for us when it comes to all things behavioral health M&A. Quick shout out to The Braff Group before we get started.
Sandeep, can you kick us off? Can you tell us a little bit about Octave?
Acharya: Yes. Octave is a mental health practice I founded in 2018. Our goal from the inception is to figure out a way to improve the quality of behavioral health care offered in network because we think that’s the opportunity for patients to be able to access care … . Our mission is accessibility through insurance, effectiveness through providing better care and sustainability for providers. We’ve operated as a practice. We operate in 17 states with about 1500 therapists. We’ve begun recently offering technology services to help payers offer value-based care programs to their own [members] across their patient population.
As a quick note, in addition to accepting insurance, we’ve worked with large national payers since inception on programs where we are accountable for the quality of care being provided. We are on the hook for outcomes and actually getting patients better. Not just being accessible, a large part of our strategy. Beginning to work on technology with payers has really exposed me to some of the challenges that we’re going to talk about today, ghost networks. It’s been interesting having spent some time recently with the regulators to understand how a problem can look so different depending on where you sit. I look forward to sharing more.
Lovett: Great. I also wanted the audience to know they can submit questions. We want to keep this conversational and really open. I know ghost networks are a really hot topic now. There’s a lot of questions coming in about that, so please feel free to ask. All right. Let’s get the conversation started about ghost networks. What are they, first of all? What’s the scope of the issue?
Acharya: Yes. Ghost networks are this notion that insurance directories and insurance companies often come and they advertise how large their provider networks are. There’s a third factor that insurance executives talk about when they go and they talk to employers about their networks – hundreds of thousands of providers in the network. When a patient goes to access care, they’re not finding the care. They’re either finding a couple of things.
One, providers that they contact don’t accept the insurance. Two, they have inaccurate contact information. What’s really astonishing is how prevalent this is. The Senate finance committee did a study and found that in behavioral health, this issue is particularly a problem. About 80% of the providers they contacted were not reachable or accessible or accepted the insurance. It’s a lot larger than it is in other verticals. Although it does affect other specialties.
Our team actually recently did a study and found that more people use their insurance directories to find care than you’d think. About 37% of patients that we interviewed looked for care in their insurance directories. That was even more than folks who said they’d Google it; only 31% use Google search. That’s the issue. Happy to talk about why, but first wanted to make sure I gave the answer to that question. It’s really about finding accurate providers that take your insurance.
Lovett: Yes, I’d love to dive into some of those points – the why this is happening. Are there certain populations that are really disproportionately impacted by this?
Acharya: Yes. First of all, in terms of why it happens, I think a couple of things. One, in behavioral health, it’s a lot more challenging to be able to provide information on which providers accept your insurance, and that’s for a couple of reasons. First of all, the average doctor sees 1,500 to 2,000 patients. That’s a lot of patients a single doctor sees. The average behavioral health provider may only see 30 to 50 patients, and within any individual insurance company, it might be 10.
The second is, those providers on average can get paid more by taking cash-pay patients. That’s why over half of therapists nationwide are not accepting insurance in some shape or fashion. Actually, those that do often quit. We did a study of claims analysis looking at an all-payer claims data set this year and found that over 20% of providers who were billing the insurance system in 2022 weren’t billing it in 2023, which means that they were quitting the system, and that’s been a persistent trend. The thing is, when they do that, they don’t necessarily notify the insurance company.
There’s no reason to. There’s not necessarily a carrot or a stick to be able to keep those records up to date. Providers who by and large are working in solo practice don’t have much incentive to. What happens is often insurance companies don’t have that information. When they do have that information, often they can be slow due to some legacy systems to actually update that information. Those factors make it very, very hard to keep an accurate record. The fragmentation, the superior compensation, the ghosting that occurs when providers do quit, and then the inaccuracy and the payment systems in the payer systems to be able to keep those records up to date.
In terms of populations that are disproportionately affected, having looked at this data, we’ve seen that it’s worse amongst government payers Medicare and Medicaid that tend to have smaller networks. There are some commercial plans that have a little bit more of a challenge here. Not going to single you out, you know, who they are. We’ve done some research and found that some plans just have more problems with this than others.
Lovett: Yes, that’s really interesting. I think we’ve seen a lot, particularly the Medicaid space. What things can payers do to actually help combat these ghost networks? Are there actually any incentives out there for payers to do this? What can motivate payers to be working towards this?
Acharya: Yes. In terms of things that payers do, what I’ve seen folks announce is payers have contracted with large provider groups and behavioral health, and then announced their sites and provided direct linkings to their sites to go and provide some navigation to patients. They also invest in some technology to uplevel their directories. Payers have done that on occasion to be able to do that work. Those are a couple of the examples in terms of what they’re doing to maintain accuracy and to try to combat the gross networks.
There’s also companies that offer services to maintain the accuracy of your networks using AI and others or startups that specialize in that. I think the most interesting, though, is really helping patients navigate to care. Because ultimately, it doesn’t matter if you have 100 providers that are accurate if you can’t find two that are really appropriate for you. I think that extra step would be better for payers to take.
Your point, I don’t think there’s necessarily a strong incentive for payers to create accurate directories. First off, the way insurance accounting works, they can pass off things as medical costs to the underlying patient, but administrative services come out of a very, very tight margin. They can only keep 15 cents on the dollar for administrative services and it often costs multiple millions of dollars to be able to upgrade this capability set.
By and large, they push that off. I will mention in one conversation with an insurance executive, in a moment of candor, this person said to me, “Believe it or not, we kind of benefit from this insurance being confusing because people will not pursue their benefit and, therefore, we save money.” Unfortunately, that view does exist, not amongst all payers, but I’ve heard it from time to time, where people benefit from people giving up and not seeking care from ghost networks.
Lovett: Interesting as behavioral health costs in a lot of health insurance plans have really risen in the last few years. Wonder if it was readily accessible, what that would actually do to margins.
Acharya: Yes, I think that’s part of the reason that you see a reticence to make it easier to find behavioral health care.
Lovett: Super interesting. I also want to let the audience know if you do have a question, submit it in the Q&A button and we’ll be going from there. I see some hands raised. I just want to make sure we’re putting it in the Q&A.
Let’s talk a little bit about regulation. Another big part of this – we talked about incentives, maybe some disincentives. Recently, there was a study published in BMC Health Services Research that found, despite regulations like the No Surprise Act, ghost networks have not been substantially impacted. Where are we at with regulation around ghost networks, and what can be done? What would be your future hope for regulation around these ghost networks?
Acharya: Yes, that’s a great question. It’s one that I’ve been spending a lot of time, particularly this year, working on because there actually is a lot of activity focused on this, both broadly and in behavioral health. There’s two federal laws that are being debated and discussed, and one at the state level in California. California always goes the farthest. We’ll talk about the federal ones first. The Requiring Enhanced & Accurate Lists of Health Providers Act, also known as the REAL Health Providers Act, introduced by Sen. Thom Tillis and a couple of Democratic senators, is one that basically looks at all Medicare Advantage enrollees.
The idea is that if you as an insurance company have someone listed on your directory, and then someone goes and gets care on that directory, and then finds out later that what would’ve cost $10 if they were in network instead cost $1,000, the insurance company is liable for that difference. They have to honor the cost share. Now, where that is, it’s got bipartisan support, and I was speaking with some of the policy folks behind it. They feel confident that they believe that they can get it passed this year.
Although I think we’ve got something going on with the federal election right now that’s keeping folks a little busy, but I think the handicap is that they believe in a December lame-duck session, there might be an opportunity for that to be passed, but we’ll see. It’s early.
There’s another even more progressive one that’s focused on behavioral health – the Behavioral Health Network and Directory Improvement Act. That was introduced by Sen. Ron Wyden and Sen. Tina Smith in March. That speaks specifically to behavioral health that provides for civil penalties for not maintaining accurate directories.
That one is early. That was put out for some discussion. It’s not exactly clear that there’s bipartisan support. The earliest that that would get passed is later this year. I think actually the most interesting, potentially aggressive bill on ghost networks has been introduced in California, AB 236. There is a bill that is requiring accuracy in directories, and it’s actually quite specific and it’s quite consequential. The bill is actually in the middle of assembly right now, has good momentum, and may pass in this session.
If passed, it would require insurance directories to be 60% accurate by July of 2025 or 95% accurate in the next three years, or face penalties that would amount to up to $5 an employee. That’s multiple millions of dollars for not complying. Now, when you read the bill, you understand the underlying rationale for it, but in behavioral health, it’s particularly a problem because there are requirements that get passed down to providers where providers have to respond within 5 or 10 days to requests from insurance companies.
While it’s well-meaning, I think the challenge is that it will end up potentially increasing the administrative costs for providers to operate and take insurance. If you choose to stop accepting insurance, you could actually face penalties from insurance companies, and insurance companies will figure out a way to pass that burden down. I worry that it might have a counterproductive impact of actually reducing insurance participation because providers say, “Hey, look, I don’t want to have to do more compliance or paperwork to be able to maintain accuracy.”
Because there’s not only timing requirements, but there’s also some availability expectation that providers maintain that might be relevant for a primary care provider that sees 2,000 patients, but for a behavioral health provider, that might be totally full one week and totally empty the next week could actually be really challenging. We are engaging at the federal and state level to try to create a different passageway for ghost networks. We think that 100% accuracy is a valiant goal, but what we’d rather see is the payers invest in an ecosystem to be able to direct folks to providers who have availability.
In other words, instead of trying to make sure all 100 providers are accurately listed, what we think is appropriate is to make sure that the payers do more work to ensure that providers with availability are shown on the payer website. We’ve actually spoken about Safe Harbor provisions that we like to see at the federal level that would allow payer plans to comply by offering a scheduling system. We’d like to see something similar at the state level, although we have not started conversations yet with those state legislators behind the bill.
Lovett: I know you come from a digital provider. Do you see digital having a role to play in curbing these networks?
Acharya: Yes. I’d say first of all, more virtual care availability creates less friction. If you’re looking for providers in Modesto, it sure helps to be able to access providers in San Diego. A lot of the providers in the digital health realm improve access simply by being available everywhere. I think the other thing is a lot of digital health providers are obtaining a scale that makes it easier. If you’re a solo or sub-five clinician practice, it’s hard to be able to maintain accuracy when someone leaves and be aware of all the compliance requirements. Larger providers are better able to keep their credentialing rosters up to date, so that’s better eco infrastructure. Honestly, a lot of this is just brass tacks. It’s simple, it’s about being able to have good communication – back and forth between the insurance companies and providers, which often fails, unfortunately.
Lovett: How can we get some more of that back and forth between providers and insurance companies? Where are we really breaking down there, and how can those conversations just be bolstered?
Acharya: I think for one thing, I would like to see payers invest in shared infrastructure. Why do payers have separate credentialing systems when effectively we’re going through the same process for all different payers? I know that there’s a startup that’s been working on trying to get the payers to convene around a shared credentialing application. I think that would be incredibly helpful as an example. As I mentioned, I think shared availability systems would be very, very helpful if payers did a little bit more work to make sure that there were shared availability systems that providers could use. I think it could go a long way towards solving this problem.
Let’s just be really clear. In a perfect world, you can log into your insurance system. You have a simple questionnaire that you’re asked and you answer to gather your preferences, and then you get matched with an experienced provider who actually accepts your insurance. We think that the work required for payers to do that is actually not that much from the perspective of cost. Actually, we’ve seen an example, Cigna as an insurance company this year launched a bookable directory system, and kudos to them for investing in doing that. For the first time ever, people can log onto the Cigna portal and actually book with a provider, which is powerful.
Lovett: Really interesting. We do have a question from the audience. They’re asking, what about providers’ responsibility to share information with a plan, especially in behavioral health? There are more small and solo practices. This can be a challenge for many practices. Also, are plans looking for more hybrid providers and feel they have enough virtual providers?
Acharya: Absolutely. First off, and this question also points out that CAQH is a standard application process that forms the basis, but yet there’s still some custom work that every payer needs. It’s slightly different for every payer. I think there’s opportunities to add more standardization. I think, to answer your question, plans are absolutely looking for more hybrid providers. We hear overwhelmingly that in the pandemic, there’s been a huge move towards virtual providers.
Providers went from not wanting to be virtual to many of them being more comfortable virtual. Many of them have given up their in-person spaces, but plans are overwhelmingly looking for folks who can see folks in person. We’re hearing that, that there’s been an overreaction in some ways of our providers to only offer care virtually. Then, 100% agree with the point here, which is that for the most part, regulators that are looking at ghost networks are saying we’re only going to penalize the plans but the reality is the enforcement mechanisms that go to the plans actually end up trickling down to providers.
This can be a challenge for small practices. Providers will have increased responsibility if these laws get passed. We can only watch them and try to make sure that they’re appropriate. For instance, I don’t think it’s appropriate for providers to have to provide within five days of view of whether they’re closed or open because that’s incredibly difficult for a practice. I do think it’s fair for a provider to let a plan know when they fully opted out of accepting that insurance within a certain period of time. There’s some nuance. The devil’s unfortunately in the details, and it varies by the legislation required.
Lovett: I’ve been hearing a lot about interstate compacts. I know the Biden administration did something on interstate social worker compacts recently. Could this help combat ghost networks at all? Is that connected at all or am I just picking this out of air?
Acharya: Yes, it’s a great question. I think what it will absolutely do is improve access. Having providers being able to see folks in multiple states increases the number of providers that can see you. It doesn’t necessarily improve accuracy because, unfortunately, the problem is still that providers quit, that there’s no notification. I think it’s actually more important to tackle access than accuracy. If there’s one thing you take away from this and if you’re a regulator or you’re a plan, I think it’s more important to tackle access than accuracy. A lot of the legislation is focused on accuracy when I think access is more important.
Lovett: That’s really interesting. I wanted to switch gears a little bit and pick your brain about value-based care. I know Octave has ventured into the space a little bit. What value-based care contracts have you worked with? What are you looking at right now?
Acharya: I would say more than ventured – our first insurance contract was a contract that put us at risk for outcomes. We had that contract about five years ago, and we’re still all in on the idea that providers should be accountable for providing higher quality care. We’re continuing to look for ways to innovate on that.
We think that sharing more information and using measurement-based care ultimately allows payers to value quality. Providers are often reluctant to use measures. I think the argument I’ve heard is that they can be reductive. A client isn’t necessarily all better because their depression symptoms have been reduced. It is true that without any measure, there’s no way to differentiate between what good care and bad care is.
We’ve been able to demonstrate that 75% of our patients are experiencing remission or have been improving in severity levels on depression and anxiety. Being able to share that with payers, to demonstrate that there’s return on the money that they’re spending, is really critical to get folks to spend more. You are right. You mentioned recently, Laura, that payers are seeing huge increases in their behavioral health spend already. The question they’re asking themselves are, what is the return I’m getting from that?
The more that we are equipped to be able to speak to the outcomes, I think the better we’re able to justify an increase in spend on behavioral health, which is inevitable. We should be spending more on behavioral health. In terms of the kinds of contracts we’ve worked with, by and large, we don’t go under capitated contracts or episode-related contracts. These are relatively simple contracts that provide bonuses or shared savings for upside opportunities and some downside risk if for achieving specific goals around outcomes.
Lovett: We have another question: One issue is behavioral health improvements often impact that physical health side. How do you address those outcomes? Do you get any credit for those outcomes? What happens with that?
Acharya: Yes. The only tricky thing about demonstrating physical cost of care based outcomes is scale and size of population. We have one payer with whom we work with to measure total cost of care. After two, three years working with that payer, we’ve realized that we still are not big enough as a company to be able to look at populations. What I would say is we absolutely do account for that by including a provision on measuring total cost of care. That’s a long-term goal as we attain scale as a mental health organization and as the system attains scale.
The other couple pieces are, for us as a company, we are focusing more on both integrating with primary care doctors. That’s about a quarter of our inbound volume of patients. Also, increasingly focusing on specific conditions where we think we can have impact. There are about four or five chronic diseases where we’ve seen a huge differential for patients who have comorbid BH versus who don’t have comorbid BH, and just to call those out – diabetes, CKD, CHF, COPD, chronic pain.
I would also add, although it’s not a chronic disease, cancer is another one that we hear from the plans as well are really critical to ensure good care. It’s those things. It’s trying to make sure that you have an eye towards those, working with payers where possible to try to develop measurement, and then it’s trying to focus more on integrating with the medical system.
Lovett: Yes. I also want to go back, and we often hear a lot that behavioral health isn’t ready for value-based care. Obviously, you’ve been doing it for quite a while. What are some of those main barriers that folks are seeing and how have you actually overcome those?
Acharya: First of all, I don’t think that the behavioral health value-based care should look like value-based care in accountable care organizations. We shouldn’t be putting providers under capitation. I don’t think we’re ready for that. What we’ve done is we’ve kept it simple, as I’ve mentioned, fee-for-service arrangements that provide enhancement for providing good quality care. In terms of barriers, there are a couple of things.
One, a lot of the standard technology infrastructure that behavioral health providers use to chart doesn’t even have measures integrated. There’s a number of EMRs that don’t have measurement-based care integrated. Just simply having the tools is a portion of it. Then two, there’s a chicken-egg question where as providers begin adopting tools, plans, then having programs in place that reward folks has been the issue. Now what we’re seeing is that more and more folks are using measurement-based care.
The next step is really convincing the plans to see the value from using measurement-based care. From my perspective, the couple of things that we argue to the plans are, one, when you use measurement-based care, it’s more likely the client stays in care. Clients are triaged to a better provider that’s a better match for them, and they stay in care to be able to demonstrate effectiveness.
Two, in the long run, we think it can actually be used to be able to demonstrate when a client doesn’t need to stay in care – when a client is better able to self manage. I think the chief barrier from my perspective beyond tools is just getting the payers to really value and pay a premium for value-based care. Some plans increasingly have started to, and we think there needs to be more over time.
Lovett: Yes. We have a question about, how do you do value-based reimbursement in a fee-for-service reimbursement payer environment? Do you get higher reimbursement CPT codes for an outcome of all contracts up-front? How does it look? What are the main differences there?
Acharya: Yes. That is what I answered. We do it through fee-for-service, and we get an enhanced rate based on outcomes. That’s how we do it.
Lovett: Great. I know we have a lot of questions here, so trying to get to everybody. There’s one, how do you see non-traditional providers like coaches or AI apps and Wobot increasing access overall?
Acharya: That’s a great question. Well, first of all, I would say I’m an early believer in coaching. In 2019, we actually had a coaching program and we were one of the first to get it reimbursed through an insurance contract. I think there was a huge role because we found that there were very, very good outcomes from the use of coaching appropriately when supervised. I think the two or three things that led to us discontinuing it or pausing it was, one, we kept getting denied because there wasn’t a reimbursement pathway.
Two, when clients have a choice between a coach for the same copay as a therapist, they will choose a therapist because they believe the higher licensure correlates with better service. That’s not necessarily true, but in some instances, a coach may actually provide a better approach. I absolutely think that there is an opportunity for us to increase via coaching through just really focusing on providing better reimbursement pathways.
In terms of AI apps like Wobot, I think that there’s definitely impact. There was a study that just got announced that showed an AI therapist getting similar results to humans. There has also been studies in the past that even just say plain old digital interventions getting similar results to humans. I think the challenge is just risk tolerance.
At what point does the artificial intelligence overcome the risk of potentially messing up that conversation in terms of where AI can have impact? Behavioral health wouldn’t be the first specialty I’d look at for using artificial intelligence for direct clinical care provision, but I absolutely think that we just have to keep monitoring and ensuring that we keep track of how the relative performance looks like.
What I’d say is this, I think adjunctively, the use of AI tools under supervision of providers can be powerful. Again, just trying to make sure we’re minimizing the risk of negative occurrence. I think it’s part of the reason I want to make sure that we don’t completely remove humans from the process.
Lovett: Yes, that makes a lot of sense. It’s really interesting. We do have one other question: Why are some providers so resistant to measurement-based care, value-based care? I guess it’s asking who’s the barrier here – providers, payers?
Acharya: I would say I’m spoiled. In order to join Octave, you have to be comfortable with measurement-based care, and providers we’ve seen adopted willingly when provided an ecosystem. I don’t want to blame the providers. What I will say, the one tricky thing about measurement-based care is this notion that you’re being individually evaluated based on an outcome and any provider who’s seen a patient knows that any specific patient encounter is very complex.
A patient may look like they’re better by a measurement-based care tool, but still need a lot of help or they may not have achieved remission, but made a lot of progress. I would say I’m sympathetic to the hesitation that providers feel, but in my opinion, it’s actually more about tooling and incentives. If we had appropriate tooling and incentives across the ecosystem, I think we would see very, very high adoption of measurement-based care.
Lovett: Now, I wanted to wrap up by focusing on Octave. What’s on tap for the rest of the year from your team?
Acharya: I think we continue to focus on expanding. We just launched in 10 more new states. We’re expanding in those states. We are continuing to focus on ways to use the measurement-based care data set that we’ve accumulated over the years to better help us identify when high-quality care is being present and better QA, the quality of work that’s being done.
Increasingly, we are really, really focused on improving the experience and quality of matching and the process of identifying a high-quality provider. All this claims data has helped us identify who the highest quality providers are in the country, and we’d like to create an opportunity to celebrate those providers.
Look for some more from us to both improve the quality of the matching that we’ve got at Octave where we’re rolling out some things that’ll kind of help better understand the providers before you actually work with them. Then we want to do more to publicly celebrate high-quality providers because we think that the backbone of the insurance system today is high-quality providers and they often go without being recognized for their work, so we’d like to create opportunities for that.
Lovett: Great. Then looking ahead at 2025, what are some of the long-term goals that you’re looking at for the company?
Acharya: I’d say they’re always pointed towards access, sustainability, and effectiveness. From our perspective, we’d like to be in all 50 states. We are currently in 17 states. We want to continue to raise the bar on what quality looks like. For care, we’d like to see our outcomes continue to improve from where they are today. I would just say, we’re continuing to evolve on how we’re doing things, but by and large, our focus is just going to remain with high-quality care at scale across the country with as many payers as possible – so to improve access and effectiveness.
Companies featured in this article:
