Overview
Late last week, we spoke to a behavioral health executive (an LCSW) that works for a private, multi-state behavioral health medical group. Like our initial Field Check - Behavioral & Mental Health I Is Teladoc Immune to HC Labor Market Trends? CLICK HERE for the note, our contact noted very strong demand for mental health services, and a very tight hiring environment. However, the feedback on Teladoc's (TDOC) BetterHelp (and thus MyStrength) was markedly more critical, as well as for Talkspace (TALK). Based on these comments and others, we see risks with the BetterHelp model that could be exacerbated by the tight labor market for therapists. We think BetterHelp is at risk for service disruption, lower margins, and likely higher churn among clinicians and members. Even if they have a data advantage, what happens next is a big question. We also continue to work on Lifestance (LFST), which sits on our Long bench and has felt the labor impact two quarters in a row (management is hiring feverishly to offset it), and we continue to look at inpatient behavioral providers like UHS and ACHC (Short Bench). As always, if you have questions or feedback: .
Highlights
- Patient demand is high, driving high demand and wages for licensed therapists
- Labor shortages are likely to persist as education and training lead times can be long
- BetterHelp therapist hourly pay rates vary with volume and look unprofitable assuming industry customer acquisition costs
- BetterHelp may be more interested in the collected data to develop additional tools
Call Notes
Background: Our contact is an LCSW at a privately-held behavioral health medical group and has 5+ years of experience in revenue cycle, collaborative care models, and shift work (hospital setting) - all in/around behavioral and mental health; presently oversees a couple of states for the mental health group and is responsible for staffing, provider relationships/partnerships, and more.
What can you tell us about collaborative care and the relationship between primary care providers (PCPs) and behavioral? We hear a lot about "whole-person" care and companies like One Medical are adding behavioral/mental health coverage...
- Collaborative care is an interesting model - everything is executed and billed under the PCP, who assumes responsibility for the encounter(s). It's not billed to the behavioral side, so you don't have to go through credentialing for therapists (some people get stuck with that).
- Basically, you expect that the PCP is doing the right thing - it's intended that the provider is considered collaborating but is not executing interventions or the services rendered.
- You don't have to worry about denials, but there are therapy limits and caps (it's a capitated month - the codes cut off at two hours per month).
- Please note, we found this quick read informative: Why you need to integrate behavioral health into your primary care practice (Physicians Practice) - it includes the relevant codes.
- Not all Medicaid plans have adopted this - they have in ~20 states.
For the doctor, how lucrative is adding mental health to a primary care practice?
- There are different approaches and rules/regulations around fee splitting (Stark and Anti-Kickback), but the PCP bills everything under their ID, and the claim cleanly returns back, you invoice for the differential, etc. Some players will front the salary for the care manager/therapist because you can get a larger margin of the claim, but a PCP isn't going to get rich off of this (it's maybe an extra $10k per year; however, it ties them to the patient better... the PCPs are more aware of their patients' mental health issues).
- Based on the codes, the amount for the two hours differs. It depends on the region of the country (state or county). Some states are much higher/lower than others (e.g., what's $200-300 in CT might be $180 in AZ, and then every county in CA has a different fee schedule).
In terms of behavioral health coaches... can you make one from scratch? How much training is there or how hard are they to find? Can they be anywhere? It seems like the number of people willing to work at the prevailing wage is shrinking...
- First, we don't use coaches or BA-level, non-licensed – so all of the people are Master’s prepared or more qualified. It depends on the state. Some require full licensure (e.g., an associate or LMSW vs. LCSW). Most states do require a Master’s degree – sit for the exam, 2-3 years of supervised fellowship to practice on your own. It does vary state-to-state or health plan, from fully licensed w/ thousands of hours, to just have the degree and pass the test. AZ and CA, for example, you get the license when you're done.
- If you're thinking of hiring and scaling, you want the independent license - you can't open a practice and charge $200/hr. if you don't have the license. Many of these professionals are well-trained and can be fantastic, and the labor force is full of people looking to build their experience along the pathway.
- What kind of professional does BetterHelp employ? They have to do only independently licensed - so, everyone is of the "expensive" variety.
Do you know how the comp works with BetterHelp vs. the cost/what people pay for sessions? It looks to us like they charge $80 or the $320/mo.?
- Yes. They offer clinicians flexibility, a set schedule, no billing or revenue cycle worries, cash pay patients, no audits, etc. However, everything is on the clinician's license. Said differently, the clinician is responsible for the risks.
- My understanding is that the model scales based on the number of patients - for example, if you work 10 hours per week, it's $35/hr. Then something like $45/hour if you're at 15-20 it's $45/hr., 30-35 it's $60/hr., and 40+ it’s $70/hr. - there's an incentive to take as many patients as is possible.
If you're a patient, do you know if you see the same person?
You should, but the algo and way patients are introduced to the platform doesn't guarantee the same person. Also, if a therapist changes his/her schedule, the likelihood of things lining up is slimmer.
- We think, based on the app data, that the average treatment duration is about 5 weeks... does that make sense? That's probably about right. Traditionally, only 60% of patients make it to a first visit w/ a specialist, and then only 40% make it to a second. That said, self-referred or cash pay, up to 70% will stay for 5-6 months, and very few people make it beyond 4 visits. So, that's probably about par for the market.
Who can work where and can a coach substitute for an LCSW?
- Clinical social workers and PsyDs work with Medicare, other counselors may not take Medicare or see patients at a Medicare-certified facility (no hospice, for example). From there, all LCSWs and PsyDs can work in more places. LCSW is a big bin that everyone is looking at. There are more social work programs than professional counseling programs. The master's level social work program has outpaced any other master’s level degree over the past decade.
- The supply varies widely by state - I think in MA there's 1 clinician for every 148 people, but in Alabama, there's 1 for every 1,000.
What can you tell us about the demand side of the equation?
- Take an average PCP panel of ~1,500 patients plus or minus and you'll see 10% - 15% of people experiencing at least moderate depression - on average.
- The demand is huge.
Do you think BetterHelp is making money?
- Yes, but it could run out. We call it "fast food therapy." People drive by and get a coping skill. This hits a wall at some point for almost everyone (patients and clinicians).
- The thing that’s most concerning/disturbing is they make it clear in the terms and conditions that anything said and executed belongs to BetterHelp (for data). I wonder about provider/patient confidentiality. There's a lot of texting going on - it started as a texting platform - and everything belongs to Teladoc. I and others think they are attempting to build an AI therapist. They are taking all the data points pushed to the platform - a month of therapy data times however many patients - and that's a lot of conversations/messages. They probably can build an AI therapist.
- But you’d have to have an authentic discussion, right? Isn't human interaction necessary? The last thing I thought would be automated is/was mental health; however, post-pandemic, people seem to be willing to give in. I never would have thought people would text a therapist as they do.
When we try to think through the margin on that $320, it doesn't seem like there's much left, especially at $70/hr. How can they scale it and how's churn on the provider side?
- I think they use that scale to try and get a therapist to think about their schedule, but it's challenging. The industry is predominantly female and most, unfortunately, don't have great business skills and are willing to let BetterHelp cover the EMR, CAC, etc.
- My opinion, it's not a great provider experience and we see turnover there after about 3 months - so, there's patient churn, and providers churn because it's "quick emotional burnout" (never get to see people get better, they just dump their worst emotional experiences on you and literally disappear).
- I think there's a substantial exit interview - but I'm not sure if there's a notice to patients when a clinician leaves.
- There's also risk - I think clinicians are more aware of the risks and don't want to lose a state license (can't get that back).
- Clinicians have the bargaining power - I know I wouldn't sign anything I didn't like (NDAs, non-competes, anything that suggested I might have to pay back a signing bonus or limits my ability to take on other work). It's the largest "silent strike" in history.
- The labor pool is only so big - you probably have a year or two at most before churning through enough therapists, and then the data mining is becoming more public. Many therapists don't want to contribute to that.
- Do you think they’ll get there…? The AI therapist? Yes, I do. BetterHelp tries to sell it as it’s video and/or phone, so you don’t have to do the texting thing, but still, what's said and done are data points that they own (I don't think recorded, but they must get data out of the sessions).
OK, and your thoughts on customer acquisition costs for BetterHelp and margins?
- Agree that you're potentially losing money at the higher pay rates per hour. That's why I think the true business isn’t actually teletherapy. It's data collection. They want the millions of data points to leverage or sell at a scale never seen before - or to build the AI doc/therapist.
- Think about the Travis Scott concert and Teladoc/BetterHelp offering a free month. They want that for a month – imagine all the data points to sell to whoever after. Data points on what a licensed professional would do after such a tragedy.
- Note: the press/social media coverage is very interesting: Inside Travis Scott’s Strange Post-Astroworld Partnership With BetterHelp, the ‘Tinder of Psychotherapy’(Daily Beast).
Assuming digital/virtual providers are delivering good, quality therapy, what's the impact on the pipeline for inpatient? Do you have a view there?
- Part of the value prop of collaborative care is it's high-touch and can keep people out of the ER. Look at the IMPACT study. People hit the emergency department less, blood pressure improved, etc. It's all about whole-person care.
- Also, this is another reason you're seeing social workers hired by health plans or value-based companies. Health plans want the social worker to do that risk adjustment for them. It helps.
Can you share your opinion of other providers? We've been looking at Lifestance, hearing a lot about Lyra, there's Talkspace, etc.? Who are the top providers and who should go away?
- Just my opinion, but if Talkspace and BetterHelp went away, I think everyone would be better off.
- Tia is really interesting – women-focused, PCP, OB/GYN, etc. All female staff, very women's health focused. People are willing to pay more for that experience, so what they charge for therapy is more comparable to the traditional model.
- There is so much competition and variation - Cerebral, Alma, AbleTo, Sondermind. I like things that improve access/steering, but some are just scraping 15% off the claims. Those can last for quite some time because therapists will work for them and save the overhead.
- BetterHelp with a fee of $80/visit = below market comp for therapists. Therapists get $80/hour (Medicaid) and for Medicare you get paid like $100-110. For commercial it can vary from $90-$150.
How do you escalate something or deal with a high-acuity issue if you're at a BetterHelp? Do they have the resources?
- No. it’s a big concern. If a patient tells me he/she is suicidal, how does the company support me, the therapist? It's my license on the line. It seems like a dangerous model in that light. You're an independently licensed clinician taking on a suicidal person - I have to resolve that, and it's a scary thing. Some peers have run into it, but the really scary part is that you may have to report it to the police and the person entered fake information on the portal. What do you do then?
- What's the incidence of higher acuity? Normally 10-15% of patients. On average, about 12% of patients are suicidal, for example.
Where is demand right now?
- So, there are some local differences - e.g., in some areas you need more PCPs, in others, you need more PsyDs to meet demand. In general, in an area where Medicaid will support an inpatient stay (up to 15 days in AZ, for example), you have this social determinants issue (the homeless know they can find housing in Inpatient Psych and will come back every month). Also, parity isn't set up correctly - the end-to-end risk adjustment is tricky. Hospitals can twist what does exist.
- That said, no matter how imperfect, early mental health care moves the needle. Many of these apps make care more accessible and raise awareness. Plus, many of them are affordable. There's more online/virtual treatment of mild-to-moderate, which helps and leaves severe to the psych hospitals. So, I'm fine with BetterHelp spending millions on advertising - it helps us, and the more people that get treatment when in a mild-to-moderate state, the better.
Please reach out to with feedback or inquiries.
Thomas Tobin
Managing Director
Twitter
LinkedIn
