GUBMENTPlain talk · policy frontier
Filings / Mental health / Sources / §5 Findings: Medicaid rates, take-up,
GBMT-11 · Research record · No. 11

§5 Findings: Medicaid rates, take-up, and the safety-net market

mental-health/research/ws05-medicaid-rates.md
This is a working research document from the mental health filing, published as written — including the parts later corrected. It is the underlying record for Whitepaper No. 11, not a summary of it.

Date: 2026-08-11. Adjudicates H5; verifies protocol anchor 9 (CCBHC access). Cross-cites §2 on Medicaid acceptance (H2) — rates/paperwork are candidate mechanisms for that wedge, not a re-litigation of acceptance levels.

PRISMA-lite search log

Field Logged
Strings Medicaid psychiatrist fee schedule Medicare index Health Affairs; Medicaid mental health reimbursement elasticity; CCBHC demonstration Mathematica evaluation wait time PPS; FQHC behavioral health same-day billing NACHC; Allegheny Medicaid rate setting mental health supply; Decker Medicaid fee bump participation; Cunningham Medicaid administrative burden
Date range Prefer 2015–2025 fee schedules / natural experiments; CCBHC DY1–DY4 evaluations
Inclusion State fee-schedule comparisons; well-identified or quasi-experimental rate→supply papers; official ASPE/Mathematica CCBHC reports; FQHC integration financing barriers with primary sources. Exclude stigma-only surveys as sole participation evidence.
Searched 2026-08-11

1. Fee schedules — levels and dispersion

Zhu, Breslau & McConnell (or Zhu et al.), Health Affairs 2022 — Medicaid FFS schedules for a common psychiatrist MH service basket (2022):

Implication: low Medicaid fees are real and highly state-dispersed, but level comparisons alone do not identify participation. That pushes H5 toward changes / payment-design experiments, not static maps.

Allegheny County analytics paper (2025; MCO rate-change design over many codes): supply elasticity ≈ 0.16 (95% CI ~0.11–0.20) — a 20% rate increase → ~3.2% more services, mostly to existing patients rather than new panel openings. Responses strongest when post-change rates approach 90–100%+ of Medicare.

2. Natural experiments and paperwork (attitude-only fails)

Evidence Frame Result for H5
Allegheny MCO rate changes Within-market staggered code-level increases Positive but modest quantity response — rates bind, weakly
ACA primary-care fee bump literature (Alexander & Schnell 2019 vs Decker 2018) National temporary Medicare-parity fees for primary care Mixed: some access gains; Decker finds no association with physician-reported Medicaid participation — rates without admin fix often fail
Cunningham / CMS MMRR PCP interviews Why physicians limit Medicaid Rates + delayed payment + billing requirements co-cited (majority cite multiple reasons) — stigma/attitude is not the measured object
NACHC BHI / CHC behavioral fact sheets (2025–26) FQHC integration financing Same-day billing restrictions, inadequate team-based BH reimbursement, documentation burden named as binding barriers alongside workforce

Attitude-only path: provider surveys that stop at “stigma” or willingness language without holding payment/admin constant do not explain the §2 acceptance wedge. When rates move and paperwork stays hard (Decker-style nulls; FQHC same-day billing), participation/access response is muted — consistent with H5’s “rates and paperwork” claim, not rates alone.

3. Anchor 9 — CCBHC evaluations (verified)

Official ASPE/Mathematica (+ RAND) national evaluation of the §223 CCBHC demonstration (PPS payment + required service scope including crisis):

Metric Finding
Clients served (original states, DY1→DY2) Aggregate ~9% increase (304,988 → 332,135 across seven reporting states; PA up to 23%)
Time to initial evaluation (adults) Mean 9.0 days (DY1) → 5.4 days (DY2); share evaluated within 10 business days improved in most states
DY1→DY4 access (RTC 2023) Adults within-10-days stable ~69–73%; mean days 9.1 → 8.4; child/adolescent mean days 9.9 → 7.2
Same-day / open access 94% of CCBHCs reported open-access or same-day scheduling (2023 RTC survey window)
Claims DID (MO / OK / PA) Heterogeneous: MO +5.7% BH ambulatory visits vs comparison; PA −13% BH-related ED visit count (no change in any-ED probability); hospitalizations/ED often null

Verified for anchor table: CCBHC demonstration states show improved clinic-reported access metrics (wait to evaluation; client volume; open-access prevalence) under cost-based PPS. Downstream ED/hospital impacts are state-heterogeneous, not a uniform national win.

GAO-21-104466: early demonstration states reported mixed spending effects; CMS guidance gaps on aligning PPS rates with costs / avoiding duplication — payment design is load-bearing and perishable.

4. FQHC behavioral expansion

NACHC: health centers are a major BH delivery surface; >60 dual CHC–CCBHC certifications reported; telehealth and partnerships with CMHCs/CCBHCs are the expansion modes. Binding friction called out in NACHC BHI work: Medicaid same-day billing bans in some states force clinics to choose which encounter to bill or split clinically integrated visits — paperwork/payment rules, not stigma.

Treat FQHC as supporting terrain for architecture #8 (primary-care BH integration) and #2 (CCBHC), not as a separate national acceptance series.

5. H5 adjudication

H5 — Rates and paperwork bind before “stigma.” Medicaid fee levels and administrative burden predict participation better than provider-attitude surveys.

Criterion Result
Participation elasticities or natural experiments on rate changes show material supply response in ≥2 studies Met with magnitude caveat. (1) Allegheny elasticity 0.16; (2) CCBHC PPS + certification package → shorter waits / more clients / open-access (Mathematica). “Material” = detectable supply/access response, not large enough alone to close the §2 acceptance gap.
and attitude-only explanations fail when rates are held constant Met. Cross-state fee levels ≠ participation (Zhu); fee bumps without admin relief often null on reported participation (Decker); surveys co-cite billing delays/prior auth; FQHC same-day billing binds at constant clinical intent.
Refute path (rate increases produce no participation response in well-identified settings) Does not fire — Allegheny and CCBHC show response; refute would require clean nulls in both rate and PPS designs
Indeterminate if fee schedules cannot be linked to participation Schedules are linkable (Zhu; Allegheny); cross-section link is weak, change designs are informative

Adjudication: Supported — rates and paperwork bind; elasticities are modest; stigma-only stories fail the hold-rates-constant test. Do not promise that a fee bump alone opens panels at H2 scale.

Implications for scorecard

Confidence: Moderate–strong

Strong on fee dispersion and CCBHC access metrics (primary federal evaluation). Moderate on Allegheny (one MCO market). Weak that any single elasticity travels nationally — disclose local design.

← All Mental health research documents Sources digest Read the whitepaper