ASAM continued stay documentation: writing six-dimension updates that survive concurrent review

At each concurrent review, a US payer reads a substance use disorder (SUD) program’s continued-stay note and decides whether the patient continues at the current level of care, steps down, or steps up. Medicaid agencies, managed care plans, and commercial payers that apply the ASAM Criteria make that call against ASAM’s transition and continued service criteria. ASAM continued stay documentation has to give them what that decision needs: what changed since the last review, what risk remains in each of the six dimensions, and why the next level down is not yet safe. Dimensions 1 through 5 inform the level-of-care recommendation, and Dimension 6 captures barriers, preferences, and the patient’s own view of the next level.^[1]

A denied day, a peer-to-peer call that exposes a thin note, or a missed review window raises the question of whether the weakness sits in the notes, in the process around them, or in the system they are written in. One pattern to test against your own denial log is an admission that was justified and initial days that were authorized, followed by generic progress notes that cannot show why the next level down is still unsafe. The cost is administrative denials, forced peer-to-peer reviews, and uncompensated days in residential, partial hospitalization (PHP), and intensive outpatient (IOP) settings.

The sections below are written for clinical directors, utilization review (UR) specialists, and primary counselors, who each own part of the note and part of the process. They cover how concurrent review works, what each dimension update needs to show, and how to audit the workflow behind it. For level definitions and the Fourth Edition changes, see blueBriX’s ASAM criteria levels of care guide.

What does the reviewer decide, and what can they see?

A central question in review is whether the note shows why the patient cannot yet safely move to a lower level. A note that records activity (attended groups, participated, tolerated the program) without recording risk gives the reviewer little to approve. The standard behind that question is the least intensive level of care that is still safe and effective. CMS does not require states to use ASAM, but states with Medicaid section 1115 SUD demonstrations must have providers assess treatment needs with a multidimensional tool, and states most often meet that requirement by using the ASAM Criteria. The Fourth Edition also labels Level 2.5 high-intensity outpatient, so this guide says PHP where a program or payer still uses that label.^[2]

A reviewer can weigh only what is submitted or said on the call, so the note is the primary evidence.

Which ASAM edition’s dimension language belongs in the note?

The Fourth Edition reorders the dimensions, and which edition a reviewer applies depends on the payer and the state. Table 1 shows the crosswalk.

Table 1. Third Edition to Fourth Edition dimension crosswalk. Source: CMS webinar slides.

Read across the rows, Dimensions 1 through 3 keep their core subject areas under updated names, and Dimension 1 now explicitly includes addiction medications. Readiness to change no longer stands alone and is considered within each dimension. The Third Edition’s Dimensions 5 and 6 became Dimensions 4 and 5, and the new Dimension 6 covers barriers to care, patient preferences, and the need for motivational enhancement. The crosswalk is a reordering, so a Third Edition note cannot simply be renumbered.

Which edition applies moves on each payer’s own schedule. Optum Behavioral Health began implementing the Fourth Edition for adult SUD treatment in November 2023, unless state, federal, or contractual requirements say otherwise. Its changes take effect for a facility only after a fully executed updated contract, and until then it directs facilities to follow Third Edition guidelines.3 ^[3] The working rule is to write dimension headings in the edition each payer’s contract applies, and to ask each payer which edition that is before the first review. A note labeled to the wrong edition forces the reviewer to translate it. Triggers also differ by plan type. Ambetter from Sunflower Health Plan, a Kansas marketplace plan, told providers in May 2026 that Fourth Edition medical necessity review would apply to adult SUD admissions on and after June 21, 2026, with child and adolescent requests staying on the Third Edition. The same notice lists, among the Fourth Edition changes, that the Dimensional Admission Criteria may recommend a recovery residence in addition to an outpatient level of care.4 ^[4] A trigger tied to a signed contract, an admission date, or an age group means one facility can be reviewed under two editions at the same time.

Where do you find the ASAM level of care review cadence for each payer?

Cadence is set by payer, plan, contract, state Medicaid program, and level of care. The authorization letter, the payer’s provider manual, and the outcome of each approval are where it is written down, and public state documents show how differently it can be set. Montana’s 2025 Medicaid manual for ASAM Level 3.1 allows prior authorization for up to 90 days and includes a continued stay review requirement of up to 30 days.^[5] Nevada Medicaid’s guide for residential SUD treatment in an institution for mental disease, updated September 30, 2026, requires continued service requests to reach the state by the last authorized date and recommends sending them 5 to 15 days earlier.6 ^[6] Neither figure carries over to another payer. For how reauthorization clocks differ by level of care, see blueBriX’s prior authorization software guide for behavioral health programs.

Commercial and Medicare Advantage schedules sit in each plan’s contract and provider manual, so ask each payer for its schedule in writing instead of assuming a Medicaid figure applies. Because the cadence lives in documents rather than in the chart, record the same fields each time an approval arrives: the authorization dates and last covered day, the request window, the required packet contents, the ASAM edition and form version, the reviewer or portal path, and who files the request and who covers when that person is out.

The review date is only as reliable as whoever tracks it, and the reassessment, plan update, and packet all have to land before it.

What do the Fourth Edition’s risk ratings and reassessment schedule change?

The Fourth Edition ties its decision rules to a risk rating assigned to each subdimension. ASAM’s FAQ describes each rating as the minimum level of care at which the patient should be treated to address the needs in that subdimension, with the initial recommendation based on Dimensions 1 through 5 and Dimension 6 handled afterward through shared decision-making with the patient. ASAM describes these ratings in the context of admission and the initial recommendation. Reviewers may weigh one unresolved risk, such as an unsafe recovery environment, the same way at continued stay while the other dimensions are stable, so confirm how each payer’s review tool applies the ratings and check with your clinical reviewer. The same FAQ says the standards for each level discuss how often the treatment plan is reassessed, that Levels 2.1, 2.5, 3.1, and 3.5 recommend monthly formal reassessments, that treatment planning should be continuous and updated when new information changes a patient’s needs, and that assessment frequency is also subject to state regulations. It adds that ASAM is not a regulatory body, so states and payers decide whether and when to implement the standards.7 ^[7] For the note, that means naming the subdimension and the specific risk instead of stopping at the dimension. The text of the Fourth Edition’s continued service and transition criteria sits in ASAM’s licensed textbook and is not reproduced here, so pair this guide with that text and with your payer’s own review tool.

Does ASAM publish a continued service request form?

Yes. ASAM publishes a Continued Service Request Form for the Fourth Edition, alongside an admission service request form. Both are organized by dimension and subdimension and capture risk ratings, dimensional drivers, and level-of-care adjustments, and the continued service form also captures summaries of progress and transition plan information. ^[8] A payer may accept the ASAM form, require its own, or ask for both, so confirm that with each payer before the first review. Even when a payer uses its own form, the ASAM form shows what a complete update contains: a rating and a driver for each subdimension, in place of a narrative that restates admission.

How do you show change since the last review?

A note that repeats the admission picture cannot show progress or its absence, so a defensible update is built around the review interval: what changed, what did not change and why that matters, and what risk remains and why this level is still the right one. Colorado’s Medicaid utilization management training shows the Third Edition framework that reviewers in states and plans not yet on the Fourth Edition still apply.^[9] It states that ASAM does not support fixed treatment lengths such as a 28-day rehab, and that length of stay follows assessed clinical need. It lists three Third Edition conditions for keeping a patient at the present level:

• The patient is making progress but has not reached the treatment plan goals.
• The patient is not yet making progress but has the capacity to resolve the problems and is actively working on the goals.
• New problems have been identified that are appropriately treated at the present level.

The training also lists progress statuses a reviewer can map a note to for IOP and residential continued stay: satisfactory progress toward goals, some progress with a plan that needs modification, no progress with a plan that must be modified, and regression with a plan that must be modified. Each status calls for a different addition to the note: which goals remain open and what at this level is producing the progress, the specific plan change and why greater gains depend on it, the new interventions and evidence that the patient is working on the goals, and what worsened and whether the current level can still address it. Those additions are this guide’s guidance, not Colorado’s.

It also says assessments should be included in continued stay requests when available to document changes in condition, amended treatment plans should be included, and updated ASAM dimensions should go with any concurrent review request. It adds that treatment plans should be tied to the ASAM assessment, that a dimension with significant issues and no matching goal may signal the patient is not receiving appropriate care, and that a patient in an unsafe living situation should have both short-term and long-term goals for it in the plan (that dimension is Dimension 6 in the Third Edition and Dimension 5 in the Fourth).

How do you write the residual-risk statement?

This is often the pivotal sentence in the review. It names a specific, observable risk tied to a subdimension and to an event in the review interval, and it says what would need to be true to step down, which turns the note into a roadmap. Illustrative example, not drawn from a patient record. A weak version reads, “Needs continued structure.” A stronger version reads, “On day 17 he reported strong cravings after a call with a relative who uses (Dimension 4, likelihood of risky substance use), and he stayed in treatment after a safety-plan review with staff. At a lower level he would return to that household before housing is confirmed (Dimension 5, safety in the current environment), and he has not yet used his coping plan without staff present. He can step down to IOP when recovery housing is confirmed and two community uses of the coping plan are documented.”

How do you keep the note, plan, and dimensional picture consistent?

Goals should be measurable within the authorization window so each review can show movement, and the plan should be updated at each continued-stay review. The level of care named in the note should match the level authorized, and any difference is worth resolving with the payer before the review rather than during it. Doing all of this takes interval information from several people, a current treatment plan, and a known review date, all in hand before the call.

How do you document the clinical picture as it is?

The note records what the clinician assessed, including improvement. When criteria for a lower level are met, the note says so and the transition plan follows. A note written backward from a desired approval misrepresents the chart. The level of care is set by the clinical assessment, and the documentation records that assessment.

What if the patient is getting worse?

A step-up note uses the same dimensional structure. The same Colorado training’s transfer and discharge criteria point to a more intensive level when problems have intensified or new problems can be treated effectively only at that level. State which dimension worsened, what changed in the interval, and what the current level cannot provide. Timing differs by payer, so confirm with each payer how an off-cycle request for a higher level of care is submitted.

Table 2 shows illustrative weak and strong entries for each dimension, written for this guide and not drawn from patient records. Payer request forms show what reviewers look for. Humana’s SUD checklist for its Oklahoma Medicaid plan asks that residential continued stay requests carry, updated since the last review date, updates to all ASAM dimensions and the anticipated discharge plan, the level of engagement, progress toward goals, and the specific objectives that need additional days. Its initial-request items (stage of change, relapse history, and living situation) match the Third Edition’s meanings for Dimensions 4 through 6, so confirm which edition the plan applies now, and add a line mapping the labels if a Fourth Edition note goes to that reviewer.^[10]

Dimension 1: intoxication, withdrawal, and addiction medications

Withdrawal and intoxication risk each need a status since the last review: resolved, persisted, or returned. Medication belongs in the same entry: started, adjusted, tolerated, adherence, and the plan for continuity at the next level of care. Any scale result, such as CIWA-Ar or COWS, needs the clinician’s interpretation beside it, because a bare score does not connect to a level-of-care decision. A stable Dimension 1 sets context for the other dimensions and does not decide a stay or a step-down by itself.

Dimension 2: biomedical conditions

Current conditions, monitoring needs, medication management, and outside appointments belong in this entry, along with a judgment on whether a lower level could safely manage them. “No medical concerns” leaves the reviewer with little status to weigh, so state the specific status and who coordinates it.

Dimension 3: psychiatric and cognitive conditions

Symptom trajectory, safety and risk status, and medication changes in the interval come first. Cognitive functioning follows, with its effect on following program structure and retaining skills, and the entry closes with whether lower-intensity supports would hold.

Dimension 4: substance use-related risks

Cravings, recent use or lapses, high-risk situations, and the patient’s response when one arose are the core of this entry, along with functional impairment tied to substance use. Because readiness to change is now considered within each dimension rather than rated separately, record engagement as behavior (attendance, participation, skills used) in place of the label “motivated.”

Dimension 5: recovery environment interactions

Housing stability, supports, safety, and the people and settings the patient returns to belong here. Pair each barrier with the step being taken to resolve it, since a barrier that never changes reads as an open-ended stay. A high-risk environment can drive the level of care, and because the Fourth Edition allows a recovery residence alongside an outpatient level, as noted earlier, the note should say which of the two the risk calls for and why. A pending housing placement may not justify a continued stay on its own with some payers, so pair it with the clinical risk, such as the Dimension 4 craving episode in the example, and document the steps being taken to resolve it.

Dimension 6: person-centered considerations

The patient’s own words about the next level, preferences, and barriers to care (including social determinants of health) go here, together with the shared decision on how to get there. That keeps the plan for the next level realistic as well as clinically correct.

Table 2. Weak and strong entries by dimension (illustrative).

Six entries should read as one dated, interval-focused narrative that ends in a level-of-care statement, a residual-risk statement, and step-down criteria. Illustrative composite, review 3, day 21: D1, withdrawal resolved by day 5 with no symptoms since; buprenorphine-naloxone stable at the current dose, taken as prescribed, with an outpatient prescriber confirmed. D2, hypertension controlled on one medication; primary care visit booked before transfer. D3, anxiety improved since the last review, no safety concerns, sleep still disrupted. D4, one craving episode on day 17 after a family call; used the coping plan with staff present, not yet alone. D5, would return to a household where a relative uses; recovery housing application submitted day 14, decision pending. D6, prefers IOP and has agreed that recovery housing is a condition for stepping down. Level of care: continue at the current level. Residual risk: the coping plan has not been used outside the unit and housing is unconfirmed. Step-down criteria: recovery housing confirmed and two community uses of the coping plan documented.

Doing this at every review across a caseload is where it strains: each dimension draws on a different clinician’s observations, time pressure favors copying prior text forward, and the assembly has to happen on a schedule that does not depend on any one person’s memory.

Public policies show how literal the rules can be. Delaware’s continued stay policy for state-contracted SUD programs serving uninsured and underinsured adults treats a request received seven or more days after the last covered day as late and not accepted, with no exceptions, and it sets review at least every 30 days based on level of care, progress, and the documentation provided. The same policy lists what the packet must contain: documentation that treatment cannot be rendered at a lower level of care, treatment goals, progress and group notes since the last review, a relapse prevention plan, toxicology results, medication administration record changes, discharge planning, collateral contacts, and CIWA-Ar or COWS scores when available.^[11] Nevada Medicaid shows the edition version of the same literalism. After revising its SUD prior authorization request form to align with the Fourth Edition, it told providers to use the new form beginning August 1, 2026. Requests on the old form are pended back for resubmission through August 30, 2026 and denied from September 1, 2026.12 ^[12]

Table 3 maps documentation gaps to published payer requirements, and marks where an entry is blueBriX practice judgment.

Table 3. Documentation gaps, basis, and fixes.

How do you write the improving patient and the planned step-down?

Progress is the clinical goal, and gains reported without the supports producing them can read as readiness to leave. Document gains together with what they depend on at the current level. When criteria are met, the note carries the transition plan, aftercare, and medication continuity at the next level, coordinated with that level’s authorization so no gap opens between levels.

How do you prepare for a peer-to-peer review?

• Build a one-page dimensional summary from the chart.
• Use dated events instead of general impressions.
• Name the specific lower level under consideration and why it is not yet safe.
• Send the clinician with direct knowledge of the patient to take the call, and document the outcome afterward.

Several of these triggers come from gaps in process, such as a missed window, a copied note, or a level mismatch, which is why the next section audits the workflow before the notes.

Who documents what?

Four roles each own a piece of the update, and a gap in any one of them reaches the reviewer as a thin note, a late packet, or a mismatched level. The primary counselor owns interval observations and engagement recorded as behavior, and when that piece is missing the reviewer sees activity without risk or change. The prescriber or medical staff own medication and medical status across Dimensions 1 through 3, so medication changes and scale interpretation that never reach the packet leave those dimensions blank. The UR specialist owns packet assembly and payer translation, which is where late packets and the wrong form or edition come from. The clinical director owns note quality and the review of patterns, and without that review, copied text and level mismatches go unnoticed until a denial.

What does a vendor-neutral self-audit look like?

Sort recent denials, downgrades, and peer-to-peer calls by root cause: documentation gap, missed window, level mismatch, or edition mismatch. A dominant cause points to a specific fix, and a scattered pattern points to a systemic one. For each case, ask who knew the review date, where the treatment plan was, and how long the packet took to assemble.

How do teams run continued-stay documentation today?

There are three broad ways to run this today. Generic templates in a general EHR are familiar and low-friction, but they often lack a native recurring review cycle, so dates and dimensional context may live elsewhere. Spreadsheets and a UR calendar outside the chart are fast to set up and flexible, but they can be a single point of failure that depends on whoever last updated them. Workflow embedded in a behavioral health EHR keeps review dates, the plan, and authorization status with the clinical record, at the cost of a system change and the configuration work that comes with it.

What should you bring to leadership?

Three items frame the internal conversation: the root-cause pattern, the days at risk by level of care, and the review dates that depend on one person’s memory.

What documentation structure should the EHR support?

• Treatment plans tied to the program and level of care the patient is enrolled in.
• Goals and interventions that reflect the dimensional needs driving the authorization.
• Assessments and scores tracked over time instead of scattered across notes.

How should review dates and authorization status appear in the clinical workflow?

• Authorization status and expiration visible to the clinician, not only in the billing module.
• Configurable rules with active and passive alert options, so critical prompts interrupt and routine ones do not cause alert fatigue.
• Alerts for approaching review dates and documentation gaps.
• Level-of-care changes carried through to the authorization requirement.

How should AI-assisted drafting be governed?

Medical necessity is a clinician’s judgment. AI-assisted drafting and summarization belong in the workflow only when a clinician reviews and finalizes the output before anything reaches a payer. AMA’s account of its 2025 statement to the Senate HELP Committee says physicians must remain at the forefront of AI decision-making and be actively engaged in reviewing and validating AI outputs, and it quotes the statement as saying physicians’ expertise “cannot be substituted.”.^[13] ASAM sets its own limit: it states that inputting the ASAM Criteria and other ASAM intellectual property into AI systems is strictly prohibited, and its versioning tracker shows that language added across its Fourth Edition assessment guide, service request forms, and treatment planning template (the assessment guide update is dated December 3, 2025),^[14] so drafting tools should work from the chart and the program’s own templates, and clinicians should not paste licensed ASAM text or forms into them. For more on AI drafting in behavioral health EHRs, see blueBriX’s AI clinical documentation guide.

How should you test it in a live demo?

Ask any vendor to show, without slides, one patient moving from a continued-stay update to a review-date alert to a step-down, and ask what the system did with the authorization when the level of care changed. Also ask which ASAM edition the six-dimension assessment follows, and whether Third Edition and Fourth Edition labels can run side by side by payer.

Counselors, UR specialists, and clinical directors each need a different part of the continued-stay workflow, and blueBriX’s residential EHR for SUD programs is designed to keep those pieces in one record. Each capability below links to the page that documents it, and how it is configured depends on your programs and payers.

How does blueBriX support the primary counselor?

Treatment plans can be configured to link to the program and subprogram the patient is enrolled in, so goals and interventions can reflect the dimensional needs driving the authorization, as described in blueBriX’s ASAM criteria levels of care guide. The SUD program page also lists CIWA-Ar and COWS withdrawal scales. Reminders for upcoming review dates and documentation gaps can be configured to surface within the care team’s existing workflow.

How does blueBriX support the utilization review specialist?

Authorization and concurrent review tracking, with payer-specific deadlines, lapse alerts, and review-cycle management by program and payer, can be configured during implementation alongside the treatment plan, as listed on the SUD program page. A reassessment can be configured to prompt the transition workflow and flag authorization documentation for staff to update. Expiration alerts are designed to surface in the dashboards staff already use (see blueBriX’s prior authorization software guide). For psychiatric residential programs serving youth, see blueBriX’s guide to what causes PRTF prior authorization delays.