Negotiate behavioral health payer contracts with outcomes data

Behavioral health organizations in the US negotiate payer contracts with outcomes data by converting clinical results into the measures payers already score, then presenting them by payer, against a benchmark, and across more than one reporting cycle. The measures include Follow-Up After Hospitalization for Mental Illness (FUH), Initiation and Engagement of Substance Use Disorder Treatment (IET), and PHQ-9 remission. Evidence in that form can support requests for higher rates, performance riders, lighter prior authorization requirements, and value-based bonuses.

Most contract conversations begin with an imbalance. The provider arrives with clinical conviction and a mission statement. The payer arrives with actuarial models, HEDIS results, and a scorecard that already ranks its network. Plans are working those scorecards actively: in NCQA’s 2026 Health Plan Ratings, released in September, behavioral health was among the strongest-performing domains, and FUH was among the most improved measures, rising 8.9 percentage points in commercial plans, 9.0 in Medicaid, and 11.9 in Medicare Advantage.^[1]

Three arrangements now reward providers who can document results: performance riders layered onto fee-for-service rates, quality bonus payments such as those in the Certified Community Behavioral Health Clinic (CCBHC) Medicaid demonstration^[2], and value-based models such as the CMS Innovation in Behavioral Health Model^[3]. This guide is written for the Contracting Director, CFO, Clinical Director, or Quality Manager preparing for a renewal. It covers what payers score, what your data must be able to show, how results become contract terms, and a six-test checklist to run before you sit down.

Payers score providers on structured, reportable measures. A provider that arrives with narrative claims gives the contract nothing that separates it from other in-network providers.

The documentation gap payers can see

A statement such as “we help people” gives a payer’s quality team nothing to score. Network reviews and quality bonus calculations often run on structured, reportable measures, and state Medicaid agencies have been required to report the behavioral health measures on the CMS Adult Core Set since 2024.^[4] The plans and agencies across the table from you are measured on follow-up, treatment engagement, and depression screening whether or not your clinicians think those measures capture their work. A provider that cannot produce the same measures, under the same definitions, looks to the payer like any other in-network provider.

Fee-for-service as a ceiling

Under a flat fee-for-service contract with no quality attachment, the contract has no mechanism for telling a high-performing provider from a low-performing one. Both negotiate from the same baseline, and the rate conversation can default to what the payer sees in its own file: cost trends. A request for more money without a data mechanism gives the payer no reason to move, because the contract has nothing for an increase to attach to.

What leverage means in a payer negotiation

Real leverage sounds like this: “we perform above your network average on the metrics you already use to score us.” In our framework, every version of that sentence falls into one of three categories, and each calls for a different kind of evidence.

Network adequacy leverage rests on the payer’s need for enough in-network capacity to meet access standards. The evidence that answers it is capacity, wait-time, and treatment-initiation data, such as time to first appointment and IET initiation.

Quality bonus pool leverage rests on the payer’s own HEDIS, Star, or state quality bonus scores. The evidence is provider-level results on the same measures, above the payer’s network rate, such as FUH, FUM, FUA, and CDF.

Member retention and complaint reduction leverage rests on member continuity, satisfaction, and lower grievance volume. The evidence is engagement and continuity data by payer, such as completion, engagement, and no-show rates.

The scorecard also reaches you through different channels depending on who pays: state-reported Core Set measures for Medicaid managed care, and HEDIS results and NCQA Health Plan Ratings for commercial plans. Ask each payer which scorecard its contract uses, then choose the leverage category your data can support and build the rest of the conversation around it.

The measures covered below span follow-up after hospitalization or an ED visit (FUH, FUM, FUA), SUD treatment initiation and engagement (IET), depression screening and remission, antipsychotic adherence, readmissions, and access to a first appointment.

The table maps the measures that come up most often in behavioral health rate and quality conversations to their status and to the leverage argument where each tends to carry the most weight. The leverage mapping is our editorial judgment, not a classification published by any payer or agency, so confirm which measures your own contracts name.

Symptom and clinical outcome measures

PHQ-9 and GAD-7 trend data carry the most weight when the payer has agreed in advance how improvement is defined. Because no single definition applies across payers, set a cohort rule before collecting data (for example, the share of patients with an elevated baseline score whose score falls by a stated amount by a stated visit) and ask whether the payer’s bonus tiers already use one. The CMS159 specification cites a five-point PHQ-9 drop as the minimal clinically significant difference, which gives a defensible starting point for that rule. A repeatable trend across a defined cohort is the form of evidence a payer analyst can audit.

Remission at 12 months works differently. In CMS159, the index event is the first PHQ-9 above 9 paired with a depression or dysthymia diagnosis, and a follow-up score counts only if it falls between 10 and 14 months after that date. Patients with bipolar disorder, schizophrenia or psychotic disorders, and certain other diagnoses are excluded, and the specification asks for the payer to be recorded for every patient. That last detail matters for negotiation, because the measure itself expects payer-level segmentation. The timing matters as well: a cohort indexed this month cannot produce a measure-grade 12-month remission rate until the follow-up window opens, between roughly August and December 2027.

NCQA’s HEDIS Depression Remission or Response measure (DRR-E) offers a shorter runway. It counts members 12 and older with a diagnosis of depression and an elevated PHQ-9 (above 9) who show remission (a PHQ-9 below 5) or response (a reduction of at least 50 percent) on a follow-up PHQ-9 documented 4 to 8 months later, so an October 2026 index produces a result between roughly February and June 2027.^[12] Whether a payer’s contract names DRR-E, CMS159, or neither is a contract question, so ask which depression outcome measure its quality program uses.

CDF tests whether a positive screen is followed by a documented plan, as its name states: screening for depression and follow-up plan. PDS remains provisional: CMS kept it voluntary for 2027 reporting after states raised concerns about clinical data access and code use, and states have not yet reported it. Treat it as a differentiator for providers serving perinatal populations, not as a current requirement.

Continuity-of-care and follow-up measures

FUH is tracked at 7-day and 30-day windows after discharge. FUM and FUA apply the same logic to emergency department visits, and all three have been mandatory for state reporting on the Adult Core Set since 2024. These measures test whether patients reach outpatient care after a crisis, before the next one. FUI extends the same follow-up logic to patients leaving high-intensity SUD care.

IET is the SUD access measure, and for a provider making a network adequacy argument for SUD services it is the closest match. The two windows start at different points. Initiation must occur within 14 days of the episode date, and engagement must occur within 34 days after the initiation event, so one episode can take up to 48 days to resolve. Capture both the episode date and the initiation date.

Medication and medical management measures

SAA measures whether adults with schizophrenia stay on their antipsychotic medication, and SSD measures diabetes screening for people with schizophrenia or bipolar disorder who take antipsychotics. Together they give payers a view into how well providers manage the medication and metabolic side of serious mental illness care. SSD also sits on the mandatory 2026 Adult Core Set, alongside a related glycemic status measure for people with serious mental illness (HPCMI-AD). These measures matter most to programs serving serious mental illness populations, where a rate conversation turns on whether the program is managing the physical health risks that accompany antipsychotic treatment.

Utilization, cost, and access measures

The 30-day psychiatric readmission rate is a central cost metric in inpatient psychiatric contracts. CMS calculates a claims-based, risk-standardized version for inpatient psychiatric facilities from Medicare fee-for-service claims. A payer can calculate its own version from its own claims, so any readmission rate you present should reconcile with the payer’s definition, and residential programs should agree with the payer on a comparable readmission definition in writing.

ED use for behavioral health conditions among attributed members can be read as an inverse quality signal: high use may suggest outpatient care is not managing crisis risk. No single definition governs it, so agree on the numerator, denominator, and attribution method before presenting a result. It fits shared-savings and member retention arguments.

Time to first appointment after referral is the network adequacy measure for access, and it is where access evidence meets regulation. Under CMS’s 2024 Medicaid managed care rule, states must establish and enforce routine outpatient mental health and SUD appointment wait times, within state-established timeframes of no more than 10 business days where those services are covered in the plan’s contract, and a plan is deemed compliant when secret shopper results show appointments available at least 90 percent of the time. The wait-time standard applies from the first rating period for contracts beginning on or after July 9, 2027, and the independent annual secret shopper surveys that test it apply one year later.^[13] Telehealth appointments count toward compliance only if the surveyed provider also offers in-person appointments to the plan’s enrollees, and survey results report the two separately. Your own request-to-appointment data is what lets a plan show it meets these expectations, which gives it weight in a network adequacy conversation.

Federal parity law adds a related obligation: limits on mental health and SUD benefits may be no more restrictive than those applied to comparable medical and surgical benefits. The parity statute is unchanged, but the rules around it are in flux. In May 2025 the Departments of Labor, Health and Human Services, and the Treasury announced that they will not enforce the new provisions of the 2024 parity final rule while related litigation proceeds and for 18 months afterward, and that they are reconsidering the rule.^[14] On September 8, 2026, the Department of Labor’s Field Assistance Bulletin 2026-03 restated that nonenforcement position and named three enforcement priorities for non-quantitative treatment limitations: separate treatment limitations and exclusions; medical necessity standards and review processes, including prior authorization; and network adequacy standards, including network admission standards and provider reimbursement methodologies.^[15] The bulletin is internal Labor Department enforcement guidance for group health plans and issuers, so it bears on commercial contracts and does not set Medicaid managed care requirements. Check the current status before citing parity in a negotiation.

Total cost of care for attributed behavioral health members (ED, inpatient, and pharmacy combined) is the denominator behind every shared-savings conversation. A provider usually cannot calculate it without payer claims data, so the contract should state the attribution method and a data-sharing schedule.

Engagement and completion measures

Treatment completion, engagement, no-show, and cancellation rates have no standardized HEDIS definition, which makes them easier to produce and easier to dispute. Define each rate in writing (what counts as completion, which cancellations are excluded, how the denominator is set), propose the definitions to the payer, and apply them at every site. These rates fit member retention arguments, and it is worth checking whether any of your value-based contracts use them as process metrics.

The metric mismatch problem

Internal dashboards typically track clinician productivity and operational throughput, while payer scorecards track quality and outcomes. The negotiation works when internal data is translated into the payer’s measurement language instead of being presented in the provider’s native format. The table shows the translation step for five common cases.

Outcome data needs the same validated tool at set intervals, a payer and contract tag on every record, a named benchmark, and patient-level records a payer analyst can trace.

Standardized capture at the point of care

Outcome data has to come out of the clinical workflow because it is difficult to reconstruct credibly afterward. The same validated tool must be administered at consistent intervals (intake and defined checkpoints) by every provider at every site. The Joint Commission’s standard CTS.03.01.09 for accredited behavioral health organizations already expects this: a standardized instrument administered at multiple intervals during care, with the data aggregated for performance improvement. Retrospective chart pulls at renewal time can produce incomplete and inconsistent data, and a payer’s analysts can expose the gaps by checking your figures against their own claims.

Where discharge and ED dates come from

FUH, FUM, and FUA are built on hospital discharges and emergency department visits, and an outpatient program often does not see the other facility’s encounter. Three sources can supply the dates. First, hospitals and psychiatric hospitals whose record systems conform to CMS’s content standard must send electronic admission, discharge, and transfer notifications and make a reasonable effort to reach applicable post-acute providers, the patient’s established primary care practitioner, and any other practitioner or practice the patient identifies as primarily responsible for their care, subject to patient privacy preferences and applicable law.^[16] Ask your hospital partners to route notifications to your organization, and record your organization as the responsible practice at intake. Second, ask the payer for claims-based discharge and ED files for your attributed members and agree on a refresh schedule. Third, check whether your state health information exchange offers encounter notifications. Record the source of each qualifying event on the patient record so a payer analyst can see how it was found. Patient-level lists and discharge or ED feeds that involve substance use treatment can fall under 42 CFR Part 2, HIPAA, and state confidentiality laws. CFR Part 2 now permits a single patient consent to cover future treatment, payment, and health care operations disclosures, and compliance with the rule was required by February 16, 2026.^[17] Have your compliance team confirm what you may share with a payer, in what form, and under which consent or agreement. blueBriX’s integration page lists connections to HIEs, labs, and payers through FHIR, HL7, and an open API, and whether a particular notification or payer claims feed can be set up depends on configuration and on the payer’s data-sharing terms.

Benchmarking against population and industry norms

A raw internal average carries little weight without a comparison point. A claim such as “above your network average on FUH” is a statement the payer can check against its own results. Four sources can supply the comparison: the payer’s own HEDIS results (ask for them during the negotiation); NCQA’s free national averages for individual HEDIS measures, which require contact details on a measure page and are published for internal use only, so confirm NCQA’s terms before showing them to a payer or use licensed Quality Compass data instead;^[18] the state-level Core Set results CMS publishes for measures reported by at least 25 states that meet data-quality standards;^[19] and your own prior-period cohorts. The free NCQA report currently carries measurement year 2024 data, with 2025 data due in February 2027. Name the source and year of every benchmark where you use it.

Segmenting outcomes by payer and contract

A blended all-payer average is unlikely to work as negotiating evidence. Each payer evaluates performance against its own covered population under its own contract terms, so the number to bring is the one calculated for that population. The measure specifications point the same way: CMS159 asks for the payer to be recorded for every patient, and states are not required to include dual eligible beneficiaries or members with other primary coverage in mandatory Core Set reporting, so a Medicaid figure that includes them will not match the state’s. Tag every patient record with payer and contract at intake and keep the tag through every report.

The credibility problem

The source of the data matters as much as the number. Expect a payer’s analytics team to test three things: whether the numerator and denominator rules can be reproduced, whether each reported result traces to a patient-level record with a timestamp, and whether the data is consistent across sites. Spreadsheet-assembled figures can struggle with all three. A structured system of record with audit logs makes it easier to answer each test without a manual rebuild, and it gives the payer’s analysts a record they can trace.

Outcomes become contract terms through pay-for-performance riders, shared savings, or bundled and case rates, each needing defined measures, attribution, and a data source written into the agreement.

Pay-for-performance: tying bonus payments to measurable thresholds

Pay-for-performance usually enters as a rider on top of a standard fee-for-service base rate, with bonus payments tied to pre-agreed thresholds. Two federal programs show the pattern. In the CCBHC Medicaid demonstration, a state that makes quality bonus payments pays a clinic only after it attains or exceeds a state-defined threshold for the measure, and the PPS-2 and PPS-4 payment structures require those bonuses. CMS’s Innovation in Behavioral Health Model pairs per-person-per-month payments with additional performance-based payments during its 2028 to 2032 implementation period, and it directs investment toward interoperability and EHR tools meant to improve quality reporting. As of October 2026, the IBH Model is state-based: the participating states are Michigan, New York, and South Carolina, so it applies only to eligible practices in those states.

These programs also show a limit. Where a state sets the base rate, as it does for many Medicaid fee-for-service payments, outcomes data works through the design of quality bonuses and measure thresholds more than through the rate table itself, and no amount of evidence obliges a payer to change a rate.

The metrics chosen for a rider have to be ones the provider can already produce reliably. A rider tied to an aspirational measure puts bonus dollars at risk for reasons unrelated to care quality. Before signing, define each of the ten terms below in writing.

Outcomes data can also count against you. A payer that receives provider-level results can use them to judge your network position, and a rider tied to a measure you cannot yet produce reliably puts bonus dollars at risk. Share only measures you have validated, agree in writing how the payer may use the data outside the contract, and have counsel review the data-sharing and confidentiality clauses before signing.

Shared savings and bundled or case rate models

Shared savings rewards a provider for holding spending below a benchmark while meeting quality performance standards, the structure CMS uses in the Medicare Shared Savings Program.^[20] A bundled payment covers the combined cost of a defined episode of care in a single payment.^[21] Outcomes data supplies the evidence for the savings side of either arrangement: avoided readmissions and avoided escalation to higher levels of care. Case rate and bundled models also require per-episode outcome tracking. Without it, a provider takes on downside financial risk with no way to prove it earned the upside.

The savings argument reduces to a formula a payer’s finance team will recognize: avoided cost equals the drop in readmission rate, multiplied by the number of discharges, multiplied by the payer’s average cost of a readmission. The first two inputs come from your data and the third should come from the payer’s claims, so agree on the cost figure before presenting a result. Any dollar figure shown without that agreement is an estimate and should be labeled as one.

Illustrative example (placeholder figures, not benchmarks). A provider discharges 500 members a year for one payer. Its 30-day readmission rate falls from 18 percent to 15 percent, which is 15 fewer readmissions (500 x 0.03). If the payer agrees that an average readmission costs $10,000, the avoided cost is $150,000 (15 x $10,000). The ask is then a stated share of that figure, with the share, the attribution method, and the payment date written into the contract. Before presenting anything, replace each input with your own discharge count, rate change, and payer-agreed cost per event, and confirm the change holds on the same defined measure and cohort.

Building the rate increase narrative for a contracting director

A Contracting Director or CFO on the payer side needs the argument in financial terms even when the underlying evidence is clinical. Translate each clinical result into the payer-side consequence and the request it supports:

• Higher 7-day follow-up after discharge means a better plan-level FUH score and may reduce avoidable readmissions, which supports a performance rider or bonus tier tied to FUH.
• A shorter request-to-appointment time helps the plan show compliance with access standards, which supports preferred network status or a rate adjustment.
• Faster SUD initiation and engagement supports the plan’s IET performance, and a SUD rate review or case rate.
• A lower readmission rate can lower total cost of care for attributed members once attribution and claims data confirm it, which supports shared savings or a higher per-diem or case rate.
• A lower no-show rate and higher engagement may indicate better continuity, which supports a care coordination payment tied to engagement.

Benchmarks move, and the narrative has to account for that. Plan-level FUH rates rose across all three lines of business in NCQA’s 2026 Health Plan Ratings (see the figures at the start of this guide), so a provider rate that beat an older benchmark may only match a current one. Present the comparison against the most recent benchmark available, and show the direction of your own trend as well as its level.

Using outcomes data to ask for lighter prior authorization

Prior authorization is a separate request from a rate, and it calls for different evidence. The data that matters is your authorization history: approval rate by service, denial reasons, appeal outcomes, and days from request to decision. Under CMS-0057-F, Medicare Advantage organizations, Medicaid and CHIP fee-for-service programs, Medicaid and CHIP managed care plans, and Qualified Health Plan issuers on the federally facilitated exchanges were required to post aggregated prior authorization metrics on their websites by March 31, 2026.^[22] Those figures are reported at the contract, plan, state, or issuer level, so use them as context for your own numbers and not as a like-for-like comparison. Then propose one specific change that your data supports, such as a longer authorization window, or no concurrent review for a level of care where your approvals are consistently high. Other commercial plans, such as employer-sponsored plans, fall outside this federal reporting requirement, so check state requirements and ask those payers for the same data directly. A payer is not obliged to agree, and a request tied to your own data is easier for the payer to assess. CMS has proposed further changes to this reporting in CMS-0062-P, and the proposal is not final.

Timing the ask: renewal cycles and the data runway

Payers set renewal and renegotiation windows on their own calendars, so the first step is to read the notice and reopener clauses in each contract. A strong position needs a trend across more than one reporting cycle, not a single period assembled the month before the call. Measure definitions also set the earliest date at which a result can exist, as the table shows for events starting in October 2026.

Derived from the timing rules in the cited measure specifications. A rate stable enough to negotiate on also needs enough volume per payer, which depends on your panel.

The practical consequence is a sequencing decision. An organization with a renewal in mid-2027 can lead with continuity and access measures, add a depression response measure such as DRR-E if the payer uses one, and treat 12-month remission as a commitment for the next cycle.

Run these six tests before the payer’s renewal window opens. Each one has a pass condition, the evidence to bring, and a next move if the test fails.

Passing five or six tests supports opening with a rate and rider request. Passing three or four supports a narrower request built around the measures where you pass. Passing two or fewer points to using this cycle to build the data record and negotiating for reporting-based terms. These bands are our editorial guidance, not a standard published by any payer.

How to use the checklist in a payer meeting or contract renewal

Sort what you found into two lists before the meeting: evidence that is proven and current-cycle, and commitments you will demonstrate next cycle. A payer’s analysts will check the first list against claims and will judge the second on whether it has a plan and a date.

Bring as proven, current-cycle evidence the FUH, FUM, and FUA rates for the last four quarters by payer, request-to-first-appointment data by program, and the assessment completion rate by site. Commit to demonstrating next cycle the twelve-month depression remission rate for a cohort indexed this fall, postpartum depression screening and follow-up for perinatal patients, and total cost of care once the payer shares attribution and claims data.

The same six tests can structure an RFI response, with one section for data infrastructure, one for clinical outcomes, and one for financial and contracting evidence.

What will the payer say, and how do you answer?

Expect five objections. The answers below are editorial guidance built from the measures and contract terms above, not a payer-published standard.

If the payer says your sample is too small, show the denominator for every rate, combine reporting periods, and lead with the measures where your volume supports a stable result. Offer reporting-only terms for the rest.

If the payer says your patients are not comparable to its network, stratify by diagnosis, age, and level of care, define the comparison cohort in writing, and offer to use the payer’s risk-adjustment logic where one exists.

If the payer says its claims do not match your numbers, reconcile to claims before the meeting, bring a patient-level list under the contract’s data-use terms, and agree which source controls in a conflict (rider term 5).

If the payer says the state sets the rate, move to quality bonus design, measure thresholds, access standards, and prior authorization terms, where outcomes data has more purchase.

If the payer says to revisit next cycle, ask for a reporting-based first year with named measures and a dated reopener written into the contract.

blueBriX’s product pages list assessment, payer-level reporting, and benchmarking capabilities that map to these requirements. Which outcome measures can be reported by payer depends on configuration.

Each capability below is linked to the page that describes it.

For clinical workflow and compliance owners

The blueBriX assessment library includes 200+ pre-built assessments, among them the PHQ-9, GAD-7, and C-SSRS, linked to treatment plans with trend tracking and early-intervention alerts. That supports the consistency tests in the data foundation section, provided every site administers the tools on the same schedule. For CCBHCs, the CCBHC and CMHC page lists quality measure dashboards for FUH, FUA, FUM, IET, OUD pharmacotherapy, and depression and experience tracking, with quality bonus payment thresholds, and its FAQ says measure logic is calculated from the clinical and claims data already in the system. Which measures can be calculated in a given deployment depends on the clinical and claims data feeds connected, including hospital and ED encounter data from outside your organization. For organizations with several locations, blueBriX advanced analytics consolidates data from multiple sites into one dashboard and lists outcome benchmarking against industry standards. That is one comparison point, so confirm each benchmark’s source and year and keep the payer’s own network rate in view.

For economic and operations owners

The care coordination software page lists contract-specific KPI dashboards that show performance measures by payer, real-time HEDIS and Star Ratings dashboards, and support for RAF, Stars, HEDIS, or contract-specific metrics. It also states that care programs, cohorts, and workflows can be configured by funder, contract, or payer without coding, and that blueBriX tracks outcomes, engagement, and savings by cohort or care program. Savings tracking of that kind supports the avoided-cost worksheet described earlier, although the cost per event still has to come from the payer’s agreement. Which outcome measures can be reported by payer in a given deployment (PHQ-9 remission, for example) depends on how programs, payers, and data sources are configured, so confirm each measure you plan to bring to a negotiation during implementation scoping.

Two blueBriX case studies describe adjacent work in production: value-based care administration and billing accuracy. Neither describes an outcomes-based rate negotiation. The Arkos Health case study describes Arkos Health, a value-based care organization, using Arkos 360 with blueBriX across 750+ healthcare providers and a patient population above 350,000 across five or more states. The Laurel Life case study reports 20% fewer claim rejections, 40% faster payment cycles, and a 20% increase in revenue after claim scrubbing, automated eligibility verification, and payer contract management that submits claims in line with negotiated contract terms. These are company-reported billing results.

One data layer for both sides of the negotiation

On its care coordination software page, blueBriX describes a tool it calls blue-Intelligence, which aggregates data from various sources into a data lake so that insights by funder, population, site, or care model can be generated in any BI tool; the page also lists audit-ready activity logs. blueBriX advanced analytics lists connections to Power BI and Tableau, using blueBriX’s data lake or the customer’s own, along with role-based access controls. Activity logs address traceability, which is one of the credibility tests described earlier. Keeping assessment data and payer-level reporting on one platform is intended to reduce the reconciliation between a clinical record and a contract report, although payer claims data and payer-defined measure logic still have to be supplied or agreed with each payer.

No custom build required to start

The assessment library is pre-built, and the CCBHC page lists no-code form and report builders that run without IT, so an organization can begin standardized capture without a custom development project. Implementation still involves onboarding, integration, and workflow design, which blueBriX describes as part of its dedicated onboarding and data migration support. Timing matters here: a negotiating position depends on trend length, and each month without standardized capture shortens the trend available at renewal. For related reading, see blueBriX’s pieces on behavioral health clinic metrics to track in 2026, value-based care readiness for behavioral health EHRs, and documentation for value-based care.