How Long Is Residential Mental Health Treatment?

There is no standard length for residential mental health treatment. No federal agency publishes a typical stay, no clinical body sets one, and the "30 to 90 days" on program websites describes how programs are built, not how long people actually stay. Four things decide a given stay: how the person is doing clinically, how many days the insurer will authorize, how the program is designed, and whether the next step is ready. For most people with insurance, the second one dominates. The insurer's clock runs in blocks of a few days, the program cannot bill without it, and few families can pay the difference.

Why authorization usually sets the pace

Clinical progress is what everyone says decides length of stay, and it should. But progress is judged by someone, and in an insured stay that someone is a utilization reviewer at the health plan, working from a written criteria set, deciding every few days whether the person still needs 24-hour care. The program's clinicians make the case; the plan decides whether to pay. An unauthorized night is a night the program either absorbs or bills to the family. That is why a discharge date so often moves the day after a review call.

Program design sets the outer shape. Some programs are 28- or 30-day tracks with a curriculum on that schedule; others use phases with no fixed calendar. Either way the model becomes the expectation before anyone has assessed the person against it. Discharge readiness is the factor that should decide, and the one families can actually watch. Most of the numbers you will find measure something else entirely.

The numbers that exist, and what each one measures

Hospital stays, not residential. The federal Healthcare Cost and Utilization Project reports that 2016 inpatient stays in community hospitals with a principal mental disorder diagnosis averaged 7.2 days: 6.1 for depressive disorders, 7.6 for bipolar disorders, 10.5 for schizophrenia and related disorders. Those are short-term general hospitals; freestanding psychiatric hospitals are excluded. They describe the crisis setting we compare in residential treatment vs. an inpatient psychiatric hospital, not residential care.

The closest data is about addiction treatment. SAMHSA's Treatment Episode Data Set counts discharges from substance use treatment reported by state agencies. For 2022, short-term residential, defined as 30 days or fewer, had a median stay of 21 days and a mean of 27, with the middle half of discharges between 8 and 28 days. Long-term residential, more than 30 days, had a median of 35 and a mean of 62, with the middle half between 12 and 85 days. That covers 135,099 short-term and 97,887 long-term discharges, measured from admission to last contact. Notice that "long-term" programs ended at a median of 35 days. Program labels and actual stays diverge.

Residential mental health: no clean figure. SAMHSA's facility census, the National Substance Use and Mental Health Services Survey, counts residential mental health facilities and their clients on a reference date; its 2024 report contains no length-of-stay figures. Peer-reviewed studies of adult residential mental health stays are single-program samples, often mixed with addiction treatment. One study of 1,400 adults at two private residential facilities treating co-occurring conditions found that older adults stayed nearly four days less than younger ones and that different factors predicted length of stay in each group. Useful for what drives a stay, not for how long one is.

So when a program quotes its average stay, ask what it counts: all admissions or only completers, and whether it includes people whose authorization ended early.

How authorization works day to day

Residential care needs prior authorization on nearly every commercial plan. The plan approves an initial block of days, then reviews the case at intervals. That is concurrent review, and our insurance coverage guide covers the vocabulary. What matters here is what the reviewer is asking. The question is never "is the person better?" It is "does the person still meet criteria for this level of care today?" Those are two different questions, and the second turns against a residential stay the moment the acute risk that justified admission has eased.

The criteria sets are public, at least in outline. For mental health conditions many plans and clinicians use LOCUS, the Level of Care Utilization System from the American Association for Community Psychiatry. It scores six dimensions: risk of harm; functional status; medical, addictive and psychiatric co-morbidity; stress and support in the recovery environment; treatment and recovery history; and engagement and recovery status. Higher scores generally point to more intensive care, and the AACP's own guide says a person's scores "typically change over time," which is exactly why a stay is re-reviewed. Where addiction is primary or co-occurring, The ASAM Criteria play the same role; ASAM describes them as standards for "placement, continued service, and transfer." Some insurers use proprietary guidelines instead.

That produces what families feel as step-down pressure. A person admitted with a high risk-of-harm score who is now sleeping, eating and attending groups scores lower on the dimension that carries the most weight, and the plan's position becomes that a partial hospitalization program can do the rest. Sometimes that is right. The program's job is to show, dimension by dimension, why it is not yet: a medication still being titrated, a home that has not changed, a history of collapsing after every previous discharge.

Federal rules put a floor under this. For employer plans, and for most individual and group plans that are not grandfathered, cutting off an ongoing course of treatment is an adverse benefit determination, and the plan must notify the person "sufficiently in advance" to allow an appeal before the benefit ends. A request to extend urgent care that reaches the plan at least 24 hours before the authorized period expires must be decided within 24 hours. You get at least 180 days to file an internal appeal, an urgent appeal must be decided within 72 hours, and you are entitled to the specific rule, guideline or criterion the plan relied on. Ask for it every time.

When the insurer says no

1. Peer-to-peer. The program's psychiatrist or clinical director talks directly to the plan's reviewing physician. A practice, not a legal right, but it resolves many denials and should happen the same day.
2. Internal appeal. A plan issued in New Jersey has two internal stages: review by a different clinician than the one who denied, then a panel that includes professionals trained in cases like this one. Ask for the expedited track while the person is still admitted.
3. External review. If the plan is fully insured and issued in New Jersey, or the person has NJ FamilyCare, the third stage is the Independent Health Care Appeals Program run by the state Department of Banking and Insurance. Since January 1, 2022, applications go to Maximus, the contractor that administers it. The request is normally due within four months of the final internal decision, the carrier pays for the review, and an independent review organization decides standard cases within 45 calendar days and expedited ones within 48 hours. "An admission, availability of care or continued stay" is a listed ground for expedited handling. The decision binds the carrier and the covered person alike. DOBI's program line is 1-888-393-1062.
4. Self-funded plans. If the employer pays claims itself, the plan is excluded from the state program, along with Medicare and policies issued in other states. Those plans use the federal external review process: four months to file, 45 days for a standard decision, 72 hours for an expedited one, also binding. Ask the plan which kind you have before a denial arrives.

Two more things bound how long a plan will pay. Federal parity law does not require a plan to cover residential care at all; our insurance page covers that. And for adults aged 21 to 64 on NJ FamilyCare, federal Medicaid rules generally bar payment for care in a mental health facility of more than 16 beds, with a managed care exception limited to 15 days in a month. For a Medicaid member in a larger facility, that 15-day figure can be the whole answer.

What "ready for discharge" means

Ready is not the same as calm. A person can be pleasant, attending every group, and three days into a new antidepressant with no idea yet whether it will work or what it will do to their sleep. Readiness is a judgment across the same dimensions the insurer scores: risk of harm low and stable for a defined stretch, not a single good day; the person able to manage a day's basic tasks; medications tolerated and the follow-up prescriber named; the home they are returning to changed, or a plan for the fact that it has not; and a reason to keep going that is their own.

Our guide to mental health discharge planning in New Jersey lists what the plan has to contain. The one point that belongs here is timing. The next level of care should be confirmed, with a start date, before the last authorized day, because the gap between residential care and a partial hospitalization program that starts "in a couple of weeks" is the gap where people fall.

Discharge readiness and the end of authorization are two different events. They often land on the same day. Ask directly: "Is the team recommending discharge, or has coverage ended?" If it is the second, the appeal steps above apply, and the program should be filing them with you, not after you leave.

Cut short, or stretched out

A stay can go wrong in either direction. Nothing below proves anything on its own; each item is a reason to ask a harder question.

A stay is probably being cut short if the explanation for the new discharge date is "insurance" rather than a clinical reason, if nothing has been scheduled yet at the next level of care, or if the person is partway through a medication change nobody has watched play out. The appeal steps above exist for exactly that.

The other failure is quieter, because no one is taking anything away. A stay may be running longer than the person needs if the recommended length was quoted on the admissions call, before any assessment, and happens to match the program's marketing; if the program cannot describe its phases or what moving between them requires; if discharge planning has not started by the second week; or if the extensions arrive mainly when the family is paying privately, each with a goal that was not in the original plan.

At each review, ask four things. Which criteria set was used, and which dimensions still support this level of care? How many more days were approved? What has to be true for the next review to go the same way? Has the step-down program accepted the person, and when does it start?

Long-term residential care and adolescents

Everything above describes a stay that ends when the acute need does. Long-term mental health residential facilities work on a different premise: months rather than weeks, treatment staged in phases, and a population whose short stays have repeatedly failed to hold. Commercial insurance still pays for those stays only in increments; open-ended supportive residences run on public funding.

For children and adolescents, the Medicaid rules for inpatient psychiatric care under age 21, which cover psychiatric residential treatment facilities, require a plan of care that includes post-discharge plans and must be "reviewed every 30 days by the team." A 2023 interview study of parents whose children entered psychiatric residential treatment named "inadequate lengths of stay" as one of three main hardships, shaped by whether the family had public or private coverage. Youth placements run through a separate system, and this article does not cover it.

What shortens a stay, and what lengthens it

None of this is medical or legal advice. Length of stay is decided case by case, by the treatment team and, for coverage, the plan itself.

Frequently Asked Questions

How long is residential mental health treatment on average?

No reliable national average exists. Federal data cover hospital psychiatric stays (7.2 days in 2016 community hospitals) and substance use residential programs (2022 medians of 21 days short-term and 35 days long-term), not residential mental health treatment. Ask each program for its own figures and what they include.

Is 30 days a real clinical standard?

No. It is a common program design and authorization pattern, not a finding about how long treatment takes. The criteria sets insurers use contain no fixed number of days.

Who decides when someone leaves residential treatment?

Three parties: the treatment team, which recommends; the insurer, which decides what it will pay for; and the person, who can leave a voluntary program at any time. Ask which of the three is driving a given date.

What if the insurer stops paying before the team thinks the person is ready?

The program should request a peer-to-peer review and file an expedited internal appeal while the person is still admitted. If that fails, a fully insured New Jersey plan or NJ FamilyCare member can go to the state's Independent Health Care Appeals Program, which decides expedited cases within 48 hours. Self-funded employer plans use the federal external review process.

Can the plan cut off coverage without warning?

Federal rules require notice of a reduction or termination of an ongoing course of treatment far enough in advance to allow an appeal before the benefit ends. If the notice arrives the day coverage stops, say so in the appeal.

Does Medicaid limit how long an adult can stay?

Often. For adults aged 21 to 64, federal Medicaid rules generally do not pay for care in a mental health facility of more than 16 beds, and the managed care exception covers no more than 15 days in a month. Ask the NJ FamilyCare plan which facilities it covers, and for how long, before admission.

Sources

• Owens PL, Fingar KR, McDermott KW, Muhuri PK, Heslin KC. Inpatient Stays Involving Mental and Substance Use Disorders, 2016. HCUP Statistical Brief #249. Agency for Healthcare Research and Quality; March 2019. Mean length of stay by principal diagnosis; community-hospital scope
• Substance Abuse and Mental Health Services Administration. Treatment Episode Data Set (TEDS) 2022: Admissions to and Discharges from Substance Use Treatment Services Reported by Single State Agencies. Table E-4, length of stay by type of service; discharge counts by service type
• SAMHSA. Treatment Episode Data Set Discharges (TEDS-D), 2022: Codebook. Definitions of short-term (30 days or fewer) and long-term (more than 30 days) residential; length-of-stay computation
• SAMHSA. National Substance Use and Mental Health Services Survey (N-SUMHSS): 2024 Data on Substance Use and Mental Health Treatment Facilities. Facility census; no length-of-stay data
• Morse SA, Watson C, MacMaster SA, Bride BE. “Differences between older and younger adults in residential treatment for co-occurring disorders.” J Dual Diagn. 2015;11(1):75-82. PMID 25531670.
• Herbell K, Graaf G. “Parents' Perspectives in Accessing Psychiatric Residential Treatment for Children and Youth: Differential Experiences by Funding Source.” Child Youth Serv Rev. 2023;154:107148. PMID 37841201.
• American Association for Community Psychiatry. Level of Care Utilization System (LOCUS): Patient and Family Guide. 2021. Six dimensions; scores change over time; use by clinicians and insurers
• American Society of Addiction Medicine. About The ASAM Criteria. Placement, continued service and transfer standards
• 29 CFR 2560.503-1. Concurrent care decisions, urgent care timeframes, 180-day appeal window, right to the criterion relied on
• 45 CFR 147.136 and 29 CFR 2590.715-2719. Internal claims and appeals and external review for non-grandfathered plans; four-month filing window; 45-day and 72-hour external review timeframes; binding effect
• New Jersey Department of Banking and Insurance. Independent Health Care Appeals Program. Eligibility and exclusions, carrier-paid cost, 45-day and 48-hour decision timeframes, binding effect, Maximus effective January 1, 2022, program phone numbers
• Maximus. New Jersey Independent Health Care Appeals Program: Provider Reference Guide. May 2022. Four-month filing window; three appeal stages; NJ FamilyCare eligibility; expedited grounds including continued stay
• Medicaid and CHIP Payment and Access Commission. Payment for services in institutions for mental diseases (IMDs). 16-bed definition; ages 21 to 64; 15-day managed care option
• 42 CFR 441.155. Individual plan of care for inpatient psychiatric services for individuals under age 21; 30-day review
• 988 Suicide & Crisis Lifeline