Utilization-Review Denials Are a Documentation Problem

When a payer denies a day of treatment, it feels like a judgment on the care. It usually is not. The reviewer never saw the care. They saw the record, and the record did not make the case.

The reviewer reads, they do not observe

Utilization review happens twice: at admission, and again in concurrent reviews during the stay. Both are decided entirely on documentation. The clinical work in the room can be excellent, the counselor attentive, the medication management sound, and the day still gets denied because the note never established medical necessity in the terms the payer evaluates. The record is the argument. If the argument is not on the page, it does not exist for the reviewer sitting on the other end of the fax machine or portal.

This is not a hypothetical compliance concern. Federal parity requirements mean behavioral health benefits are not supposed to face more restrictive review than comparable medical or surgical care, and Medicaid guidance specifically requires managed-care organizations to disclose their medical-necessity criteria and explain the reason for a denial when asked (Medicaid parity guidance). A 2025 federal parity-enforcement report found the opposite still happening in practice: more frequent utilization review, earlier or repeated reviews, and shorter authorization periods for mental health inpatient care than for comparable medical stays (parity report to Congress). Operators do not control that imbalance. What they control is whether their own documentation gives the reviewer nothing to push back on.

The four ways the argument fails

Most denials trace to a handful of recurring documentation gaps, and they repeat across payers and levels of care:

• Medical necessity not stated in the payer's language. A diagnosis alone is not the standard. The record has to connect diagnosis, functional impairment, risk, and treatment response into a case for the specific level of care requested, on the specific dates requested. CMS policy for Medicare Advantage plans requires coverage criteria to draw on current evidence-based treatment guidelines when no national coverage determination applies, which means a bare diagnosis code without functional detail rarely clears the bar (CMS policy fact sheet).
• A level of care the notes do not support. If the chart never explains why outpatient is insufficient and why inpatient is not overkill, the reviewer has no basis to approve the requested tier.
• An authorization that lapsed because the concurrent review was late, not because the patient stopped needing care.
• A continued stay that documents activity but not necessity. Notes that say the patient "continues to benefit" or "remains engaged" describe attendance, not medical necessity. They do not answer what changed since the last authorization or what remains unresolved.

Build necessity in, do not bolt it on

The durable fix is to make medical necessity a byproduct of ordinary documentation rather than a scramble at review time. That means every element the payer will ask for is captured as part of routine clinical work, not reconstructed under deadline pressure:

• Diagnosis, symptom severity, and objective findings, current and specific.
• Functional impairment: effects on safety, daily living, work or school, relationships, housing, and treatment participation.
• Risk and protective factors: suicidality, withdrawal, overdose potential, psychosis, and the current mitigation plan.
• Why this level of care, now: why a lower level is insufficient and a higher level is unnecessary.
• Treatment intensity: frequency, modality, staffing, and the specific interventions actually delivered.
• Measurable progress against baseline, using the same validated measures at consistent intervals so a reviewer can see change rather than a repeated score.
• A continued-stay rationale that names what has changed and what criteria remain unmet.
• A discharge and transition plan with dates, follow-up appointments, and contingencies.

Assessments and reviews should speak in the dimensions the payer actually uses, most commonly the ASAM criteria for substance use levels of care. Objectives on the treatment plan need to be measurable, and progress notes need to document movement against them, not just presence in the program. When the record already answers the reviewer's questions before they are asked, the review becomes a formality instead of a fight.

Same-day notes matter more than most centers treat them. Retrospective reconstruction, written days later to satisfy a review deadline, is far more vulnerable to contradictions, missing dates, and language that sounds like advocacy rather than clinical observation.

Never lose a day to a calendar

Some denials are not clinical at all. An authorization that expires before the concurrent review is submitted is pure, avoidable loss, and it happens constantly during census surges or staff turnover. Beginning January 1, 2026, CMS's interoperability and prior-authorization rule requires impacted payers to issue standard decisions within 7 calendar days and expedited decisions within 72 hours (CMS interoperability rule, cited in GAO report). Medicare Advantage plans generally work on 14-day standard and 72-hour expedited timelines today (GAO-25-107342). Those clocks are knowable in advance. A center that tracks admission time, request time, submission time, payer receipt, and decision deadline in one place can escalate before a deadline passes instead of appealing after it has.

Approval rates on their own can mask the risk. Washington's 2025 prior-authorization report recorded 16,506 behavioral health service requests with a 97.7% approval rate (Washington 2025 report), and GAO found prior-authorization approval rates for behavioral health services ranging from 83% for outpatient substance use services up to 94% for inpatient psychiatric hospitalization (GAO-25-107342). High aggregate approval rates still leave a meaningful tail of denials, and when those denials hit high-cost, authorization-dependent, multi-week stays, the revenue impact is disproportionate to the percentage.

Staffing makes this harder, not easier to ignore

Workforce scarcity compounds all of this. HRSA reported that as of December 2025, roughly 137 million people, 40% of the U.S. population, lived in a federally designated mental health professional shortage area (HRSA workforce brief). Thin staffing means the person who usually catches a missing discharge barrier or an unclear frequency note is out sick, or has left. A documentation process that depends on one experienced utilization reviewer will fail exactly when census is highest and turnover is worst. The design goal is a workflow that produces a defensible record regardless of who is on shift that week.

Where to put ownership

Treat the record as an operational product with owners, deadlines, and a review clock, not a clinical afterthought that gets patched before an appeal. Track denial rates, overturn rates, days from service to submission, and authorization gaps by payer and by clinician. The dates that matter should be visible and should prompt someone before they pass, not after. A system that surfaces an expiring authorization turns a denial into a task instead of a loss.

For the denial patterns and how to prevent them, see reducing utilization-review denials.