At Serif Health, we believe that price transparency is only useful if it provides not just data - but answers. That is why we have spent 3+ years building the gold standard for price transparency; data that is cleaned, validated, enhanced, and understandable. And it is hard to find a clinical area where economic answers are more important than in behavioral health. 

In the backdrop of an ongoing mental health crisis in the US, the behavioral health space is undergoing a massive transformation: teletherapy is now mainstream, therapists are becoming more accessible through Management Services Organizations (MSOs), and insurance companies are more generously covering therapy services. However, reimbursement rates continue to vary widely across payers, regions, and providers. Understanding behavioral health data within price transparency requires domain expertise, pattern recognition, and ultimately reading between the lines. That is where Serif Health comes in.

In this post, we explore and analyze behavioral health reimbursement rates using 2025 machine-readable files, with examples from real payer-MSO relationships.

Context: What are behavioral health MSOs?

Behavioral health MSOs function as intermediaries between clinicians and payers. While they don’t deliver care directly, they handle credentialing, billing, collecting, and contract negotiations with payers.

Here’s a simplified view of how billing typically works:

1. A clinician sees a patient through the MSO platform (e.g., a standard 60-minute psychotherapy session, CPT^TM code 90837)
2. The MSO submits the claim to the patient’s insurance
3. Insurance reimburses the claim based on the clinician’s licensure and the MSO’s negotiated rate
4. The MSO and clinician each receive a share of the reimbursement

This model allows clinicians to see more patients without extra administrative burden; providing patients easier access to therapists when they need it. It’s a win-win for everyone.

Are behavioral health MSO provider networks growing?

There’s been massive growth over the last year in the behavioral health MSO space - and you can see these trends clearly using price transparency data. Payers are required to post the NPIs of individual clinicians associated with each rate in their dataset, and at Serif Health we track  these provider networks over time. We analyzed the growth of 8 of the top behavioral health MSOs (in terms of credentialed NPIs) and found the following:

• The count of distinct NPIs across MSOs has increased over 57% in the past 12 months alone. Year-to-date growth is over 22%.

• On average, each of the MSOs recruited 36% of their roster in the past year.

• Over 32% of these clinicians are credentialed with more than 1 MSO at once.

These numbers tell a clear story: behavioral health MSOs are rapidly scaling their networks (and increasingly competing for the same clinicians) at a pace that’s reshaping the landscape of mental health care delivery.

So what do reimbursement rates look like?

As mentioned, reimbursement rates and methodologies can vary widely across payers, providers, and regions - and the data is nuanced - so let’s dive in! 

We focus in on CPT^TM code 90837 (60-minute psychotherapy session), the most commonly used CPT^TM code for behavioral health visits. An important consideration when looking at reimbursement rates is a clinician’s licensure: 90837 reimbursement rate for a PhD-level clinician is typically different from that of a Master’s-level clinician. This “rate tiering” is crucial to understanding and analyzing reimbursement trends and comparisons. 

Example 1 - Aetna

A simple 90837 search in Signal for Aetna’s rate in NY negotiated with the MSO Alma produces 750+ rows of data:

How should we interpret 750 rows of data and find the right rate based on licensure? It’s not immediately clear. But if you have been following Serif Health for a while, you have probably read our FAQ on how to pick between multiple rates, which notes that the rates with the most affiliated NPIs are often the critical records to help filter through the noise. 

Each of the rates seen below has the highest NPI list length (4,781 clinicians): 

This takes us from 750+ rows down to 4 rows. But the NPI list is exactly the same across these rates (e.g., NPI 1164761623 shows up in all 4 rows of data). So how do we know which is the right rate? Well, at Serif Health, we know from years of working with Aetna’s data that Aetna will often paste in the same NPI list for multiple rate tiers. This seemingly looks like they have multiple rates for the same NPIs, but we know from experience that this is Aetna’s rate ladder.

Aetna’s policy states that their NP-level rate is 85% of the PhD-level rate, and their Master’s-level rate is 75% of the PhD-level rate. Quick math tells you that:

• 85% of $202.31 = $171.96
• 75% of $202.31 = $151.74

So with that knowledge in mind - we know that Aetna’s rates with Alma are as follows:

• $202.31 - PhD / MD rate
• $171.96 - NP rate
• $151.74 - Master’s rate

With a little expertise and some quick tricks, you can filter out the noise of 750+ rates to get to the answers.

Example 2 - CareFirst

Let’s dive into another example that is a bit trickier. When searching 90837 rates for CareFirst BlueCross Blue Shield with the MSO Sondermind, we see the following negotiated rates in Signal:

At first it may seem like Sondermind has the same rates across all licensures and provider taxonomies. Unlike Aetna, the affiliated NPI lists are different for each row of data, so it seems logical to conclude that there is no rate tiering.

But here at Serif - we like to dig a little bit deeper…

It didn’t sit right with us that CareFirst would reimburse the same rate regardless of licensure. So we next checked our historical database to review previous postings. Here’s what CareFirst posted back in February 2024:

You can see the same $137.01 - but with clear rate tiering! 

Guess what 85% of $137.01 is? $116.46 (NP rate)

75% of $137.01? $102.76 (Master’s rate)

Serif’s historical tracking helped surface rate ladders not visible in a single file. While it is not totally clear why the rate tiering was removed in CareFirst’s more recent postings, it takes a robust historical data library, knowledge of specific payer policies and patterns, and attention to detail to truly cut through the noise and find the right rates.

Conclusion

Accurate benchmarking of behavioral health reimbursement by healthcare organizations is essential for negotiating fair contracts, setting benefit design, and supporting provider recruitment and care delivery. Price transparency data gives a new window into these dynamics.

Behavioral health price transparency data is full of signals, but it takes domain knowledge to accurately decode them. At Serif Health, we don’t just collect the data. We validate it, interpret it, and transform it into insights that support smarter decisions for our partners.

If you’re a forward-thinking health plan, provider, MSO, or employer trying to understand behavioral health reimbursement in your market, now is the time to dig in. Reach out to us at hello@serifhealth.com to explore benchmarking opportunities tailored to your populations.