Sorry! I left out a word by mistake: most US workers do not have a realistic choice between a higher-paying no-insurance job and a lower-paying insured job.
With that corrected, I’m still getting stuck on the “addressable market” and using BetterHelp’s US market penetration to draw inferences about how big BetterHelp (or a comparable service) might fare in other countries. It seems to me that BetterHelp’s market penetration is highly dependent on it being a cash-pay service in a country with an arguably unusual mix of competing insurance-subsidized products. Unfortunately, I think that limits the generalizability of your findings.
To take an extreme example, Medicaid is almost 100% subsidized for the consumer, and by definition Medicaid consumers are poor (otherwise they would not qualify). Because of these limitations, I’d model the universe of Medicaid consumers with mental illness as outside of BetterHelp’s addressable market altogether under its current model, just like you’ve excluded non-Internet households. Then you have most other patient populations for which alternatives to BetterHelp are heavily subsidized by insurance, but not as that great an extent. It’s hard to assess how many of those consumers are even plausible recruits for the current version of BetterHelp.
US consumers might have a number of reasons for using BetterHelp rather than an insurance-compatible option, such as:
They don’t have insurance.
Their insurance is pretty bad—such as requiring a $5,000/yearly deductible before it pays anything—or at least bad enough that BetterHelp is somewhat cost-competitive (e.g., a high per-session copay).
They have concerns about their insurer gaining access to their mental-health information.
Their insurer will not cover teletherapy.
The choice of telehealth-based therapists within their insurance network is sharply limited, or are not accepting new patients in a reasonable time (the latter of which I believe was a major issue during the pandemic period).
There are specific elements of the BetterHelp platform that they value beyond individual sessions (and not just teletherapy, since lots of insurance-covered therapists offer that nowadays).
I don’t think BetterHelp’s current US market penetration tells us much without considering how it stacks up against other options in the other country. For example, in a country like the UK, issues like lack of insurance, huge insurance deductibles/copays, highly restrictive provider networks, etc. shouldn’t exist on the NHS. In other countries, subsidized options may not be available to most consumers or might not be decent substitutes for BetterHelp.
BetterHelp could start joining insurance networks in the US and thus start competing more effectively with insurance-subsidized traditional providers. However, its market penetration would become a much less reliable estimate of the potential size of the overall US teletherapy market than it is of the cash-pay market at present. I think it’s likely that most insured patients would stick with solo / small-group therapists rather than migrate to a big platform.[1]
And I’m not convinced that joining insurance networks would be a good play for BetterHelp anyway. The insurance companies want discounts off your cash rate to list you in their network, and there’s a significant amount of overhead involved in US insurance for the provider. So insurance-using patients are likely to be much less profitable. While you could think of these costs as a form of marketing spend, there’s a deeper problem lurking here.
I suspect many current (or potential future) cash-pay patients have insurance but are choosing not to go through it for one of the reasons I listed above. For most of those use cases, a consumer would prefer BetterHelp + insurance over BetterHelp + cash-pay. For example, even if you have a $5K deductible and the insurance isn’t going to actually pay for your therapy, you could still take advantage of the insurer’s negotiated discount for BetterHelp by going through your insurance. So BetterHelp would have to figure out a way not to accidentally convert its cash-pay patients into much less profitable insurance-using patients.
I agree with most of this tbh, I am probably stretching with the generalisability of the findings. What I was roughly trying to get at was:
The U.S. is a uniquely tractable market for out-of-pocket teletherapy due to the reasons you listed above
BetterHelp are really, really struggling to acquire new customers in spite of this
On balance, we can surmise that they’ve probably hit some kind of inherent ceiling in the acceptability of teletherapy rather than an issue with access
So we can extrapolate this ceiling to other populations where we might be interested in delivering a more cost-competitive or free teletherapy intervention
(Even without that specific extrapolation it was useful for me to just see what [mental health prevalence] × [internet access] looked like globally)
But I don’t think I did a great job of communicating that as the through-line.
Sorry! I left out a word by mistake: most US workers do not have a realistic choice between a higher-paying no-insurance job and a lower-paying insured job.
With that corrected, I’m still getting stuck on the “addressable market” and using BetterHelp’s US market penetration to draw inferences about how big BetterHelp (or a comparable service) might fare in other countries. It seems to me that BetterHelp’s market penetration is highly dependent on it being a cash-pay service in a country with an arguably unusual mix of competing insurance-subsidized products. Unfortunately, I think that limits the generalizability of your findings.
To take an extreme example, Medicaid is almost 100% subsidized for the consumer, and by definition Medicaid consumers are poor (otherwise they would not qualify). Because of these limitations, I’d model the universe of Medicaid consumers with mental illness as outside of BetterHelp’s addressable market altogether under its current model, just like you’ve excluded non-Internet households. Then you have most other patient populations for which alternatives to BetterHelp are heavily subsidized by insurance, but not as that great an extent. It’s hard to assess how many of those consumers are even plausible recruits for the current version of BetterHelp.
US consumers might have a number of reasons for using BetterHelp rather than an insurance-compatible option, such as:
They don’t have insurance.
Their insurance is pretty bad—such as requiring a $5,000/yearly deductible before it pays anything—or at least bad enough that BetterHelp is somewhat cost-competitive (e.g., a high per-session copay).
They have concerns about their insurer gaining access to their mental-health information.
Their insurer will not cover teletherapy.
The choice of telehealth-based therapists within their insurance network is sharply limited, or are not accepting new patients in a reasonable time (the latter of which I believe was a major issue during the pandemic period).
There are specific elements of the BetterHelp platform that they value beyond individual sessions (and not just teletherapy, since lots of insurance-covered therapists offer that nowadays).
I don’t think BetterHelp’s current US market penetration tells us much without considering how it stacks up against other options in the other country. For example, in a country like the UK, issues like lack of insurance, huge insurance deductibles/copays, highly restrictive provider networks, etc. shouldn’t exist on the NHS. In other countries, subsidized options may not be available to most consumers or might not be decent substitutes for BetterHelp.
BetterHelp could start joining insurance networks in the US and thus start competing more effectively with insurance-subsidized traditional providers. However, its market penetration would become a much less reliable estimate of the potential size of the overall US teletherapy market than it is of the cash-pay market at present. I think it’s likely that most insured patients would stick with solo / small-group therapists rather than migrate to a big platform.[1]
And I’m not convinced that joining insurance networks would be a good play for BetterHelp anyway. The insurance companies want discounts off your cash rate to list you in their network, and there’s a significant amount of overhead involved in US insurance for the provider. So insurance-using patients are likely to be much less profitable. While you could think of these costs as a form of marketing spend, there’s a deeper problem lurking here.
I suspect many current (or potential future) cash-pay patients have insurance but are choosing not to go through it for one of the reasons I listed above. For most of those use cases, a consumer would prefer BetterHelp + insurance over BetterHelp + cash-pay. For example, even if you have a $5K deductible and the insurance isn’t going to actually pay for your therapy, you could still take advantage of the insurer’s negotiated discount for BetterHelp by going through your insurance. So BetterHelp would have to figure out a way not to accidentally convert its cash-pay patients into much less profitable insurance-using patients.
I agree with most of this tbh, I am probably stretching with the generalisability of the findings. What I was roughly trying to get at was:
The U.S. is a uniquely tractable market for out-of-pocket teletherapy due to the reasons you listed above
BetterHelp are really, really struggling to acquire new customers in spite of this
On balance, we can surmise that they’ve probably hit some kind of inherent ceiling in the acceptability of teletherapy rather than an issue with access
So we can extrapolate this ceiling to other populations where we might be interested in delivering a more cost-competitive or free teletherapy intervention
(Even without that specific extrapolation it was useful for me to just see what [mental health prevalence] × [internet access] looked like globally)
But I don’t think I did a great job of communicating that as the through-line.