Behavioral Health Integration Billing Went From Zero to Millions Overnight
Medicaid spent just $1.4 million on Behavioral Health Integration codes in 2018. By 2024, that number was $14.2 million. The entire growth curve is striking, but one year stands apart: 2023, when spending more than doubled in twelve months.
That kind of acceleration doesn't happen by accident. It reflects a reimbursement pathway that, after years of slow uptake, finally reached a critical mass of providers willing to bill for integrating mental health care into primary care visits. Whether patients are getting meaningfully better care as a result is a separate question. The billing data alone tells a story worth examining.
A Single Year Rewrote the Trajectory
From 2018 through 2022, BHI spending grew steadily but modestly, adding between $472 million and $1.1 billion annually in percentage terms that ranged from 14% to 59%. Then 2023 happened. Spending jumped from $4,373,301 to $10,149,919, a $5,776,618 increase representing 132.1% growth in a single year. 2024 added another 39.7%, bringing the total to $14,178,939.
That 2023 inflection point coincides with a provider base that had been quietly expanding for years. The number of organizations billing at least one BHI code grew from 118 in 2019 to 246 in 2022 to 332 by 2024. Individual providers went from 25 to 77 to 118 over the same period. More providers billing more codes to more patients is the mechanical explanation for the spending curve. The policy question is whether that expansion is distributed evenly, or concentrated in a handful of states.
Arizona Spends 27 Times More Per Enrollee Than Colorado
It's not evenly distributed. Not even close.
| State | 2023 BHI Spending | Medicaid Enrollees | Per-Enrollee Spend |
|---|---|---|---|
| Arizona | $1,265,245 | 2,297,737 | $0.55 |
| Utah | $215,450 | 461,831 | $0.47 |
| California | $5,495,837 | 14,652,304 | $0.38 |
| Tennessee | $474,426 | 1,760,802 | $0.27 |
| Kentucky | $314,671 | 1,491,534 | $0.21 |
| New York | $779,478 | 7,487,923 | $0.10 |
| Colorado | $26,294 | 1,510,729 | $0.02 |
Arizona's per-enrollee BHI spending of $0.55 is more than 27 times Colorado's $0.02. Both are large states with substantial Medicaid populations. California dominates in absolute dollars at $5,495,837, but its per-enrollee rate of $0.38 ranks it third, behind Arizona and Utah.
For Medicaid enrollees in low-utilization states, this gap is concrete: the infrastructure for integrated behavioral health care in primary care settings either doesn't exist, isn't being reimbursed, or isn't being billed. A patient in Colorado is statistically almost invisible in this dataset compared to a patient in Arizona.
The Billing Base Is Still Thin
Even at peak 2024 levels, only 332 organizations and 118 individual providers billed any BHI code across the entire Medicaid program. That's a remarkably small provider base for a federal reimbursement pathway designed to bring mental health care into every primary care visit.
The concentration matters. California's $5.5 million in 2023 BHI spending came from a state with over 14.6 million Medicaid enrollees. New York, with 7.5 million enrollees, generated just $779,478. Texas, with 5.7 million enrollees, billed only $144,804, a per-enrollee rate of $0.03. These are not states without primary care infrastructure. They are states where BHI billing has not taken hold at scale.
The total program grew from $1.4 million to $14.2 million over six years. That sounds like momentum. But spread across tens of millions of Medicaid enrollees nationally, it remains a fraction of what full adoption would produce. Arizona's $0.55 per enrollee, if applied to California's 14.6 million enrollees, would imply over $8 million in spending from California alone. Instead, California generated $5.5 million. The gap between what's theoretically possible and what's actually being billed is still enormous, and it's widest in the states with the most patients to reach.
Explore the data yourself
Run your own queries against 240M+ rows of federal health data using natural language — powered by AI.
Start analyzing