Levels of care, real costs, insurance, safety, and how to distinguish a serious clinical program from a beautiful sales pitch.

How to use this guide

You are probably reading this because someone you love is in trouble, the clock feels loud, and the internet has been no help at all. Every treatment website looks the same: an infinity pool, a golden retriever, a sunset, and the word “healing” in a serif font. You cannot tell the excellent programs from the dangerous ones, because they use the same photographer.

The wrong program does more than waste money. It can waste the window when someone is finally willing to accept help, convince a family that “treatment does not work,” and return a vulnerable person to the same life with less hope than before. The right program is not necessarily the prettiest, the most expensive, or the one with a bed available tonight. It is the one matched to the actual person, with a real plan for what happens inside and what happens after.

This is the guide I wish families had before they called me. It is long, so use the table of contents to jump. Read the one-minute triage box below first, read the checklist at the end if you have five minutes, and read the whole thing if you have a weekend and a decision to make.

One thing to hold onto: the brochure is not the program. The marble lobby is not the treatment. The treatment is the people doing the work, the structure around that work, the medical oversight, and the plan for what happens when the doors open again. And the case running through this guide (“Marcus”) is a fictional composite built from patterns I see constantly, not a real patient.

If you or someone you love is in immediate danger right now, this guide is not the move. Call or text 988 (the Suicide and Crisis Lifeline) or call 911.

One-minute triage: where are you actually?

• Immediate danger, or medically unstable (active suicidal intent, an overdose, confusion, seizures, chest pain). This is an emergency. 988 or 911, or the nearest ER. The rest can wait.
• Possible alcohol or benzodiazepine withdrawal (heavy daily drinking or daily benzo use, now stopping). Withdrawal from these can be fatal. Do not do this at home unsupervised. Go to Section 2.
• Safe for the moment, but cannot function at home. You are choosing a level of care. Go to Sections 2 through 4.
• Repeated treatment failure, or a diagnosis that never quite fit. The question has shifted from “which program” to “what is actually driving this.” Go to Sections 5 and 11.
• The person refuses treatment, or does not believe there is a problem. Do not start calling luxury programs as if admission is already agreed. Start with a qualified intervention professional, an addiction psychiatrist, or a family-based approach like CRAFT. Go to Section 10.

Los Angeles resources (free or low-cost)

• LA County 24/7 Help Line, mental health and substance use, one number: (800) 854-7771. Press 1 for crisis and mental health, 2 for substance use disorder services, 3 for veterans.
• LA County Substance Abuse Service Helpline: (844) 804-7500. 24/7 screening and referral straight to a treatment provider.
• 211 LA County: dial 2-1-1, 24/7, for any health or social service.
• SAMHSA National Helpline: 1-800-662-4357, free and confidential, 24/7.
• Verify a California facility’s license: the Department of Health Care Services facility licensing directory (dhcs.ca.gov), which also lists every county’s access line.

Table of contents

1. Meet Marcus
2. First decision: emergency, withdrawal, or treatment?
3. Levels of care, and when each one fits
4. The plan: duration, active treatment, and the staircase down
5. Choosing the route: addiction, psychiatric, or integrated
6. How to vet a program before you admit
7. Money, insurance, and getting care covered
8. The supports that hold the plan together
9. Adjuncts and emerging treatments
10. Family involvement without control or abandonment
11. When diagnostic clarification and neuropsychology help
12. Marcus, illustrated
13. A family’s quick-start checklist
14. The decision in front of you

Looking for specific programs? After reading this guide, use the companion Los Angeles Mental Health & Addiction Treatment Directory — programs by level of care, with approximate cost, insurance, specialties, and who each is best for. This guide teaches you how to choose; the directory shows you what is out there.

1. Meet Marcus

Marcus is thirty-five. He is gay, he is funny, and he is drowning.

He grew up in Nashville, in a family with real money and a specific relationship to appearances. He did fine in school, got diagnosed with ADHD somewhere along the way, and never quite figured out what he wanted to do with his life. He moved to Los Angeles partly for work and mostly to get away from a family dynamic that felt like a held breath. In LA, the cocaine and the chemsex found a home, because in certain circles both are so normalized that nobody around him thought anything was wrong. He had a good therapist for years who tried hard to help.

He has been to treatment three times. He does not know how to help himself, and neither do his parents, who are paying for all of it and are frightened, because a while back Marcus scared them badly with an attempt to hurt himself. They do not talk about that.

One night Marcus fell off his balcony. It was a first-floor balcony, so the fall was short, but he hit his head and wondered if he had a concussion. That question, “did I hurt my brain,” is what finally landed him in my office. Not the cocaine. Not the three prior programs. A short fall and a headache.

Hold onto Marcus. Almost every decision in this guide is one his family had to make, badly, three times, before someone slowed it down and asked a better question.

2. First decision: emergency, withdrawal, or treatment?

Before you shop for a program, answer a safety question, because the wrong first move here is the one that hurts people.

Is this an emergency? If someone is in acute danger, actively suicidal with intent, medically unstable, overdosing, then hospitalization is the right call and emergency care is not optional. Call 988 or 911 or go to the ER. Nothing in this guide, and no program’s marketing, should talk you out of emergency care for someone in immediate danger. That floor does not move.

A word on psychiatric hospitalization, because families fear it. Inpatient psychiatric care, especially an involuntary hold (a 5150 in California), can be a hard experience, and quality varies enormously between facilities. Two things are true at once. When someone is genuinely at acute risk, that level of care is correct, and the goal is finding a good facility, not avoiding care. And a lot of avoidable hospitalizations happen because someone was under-treated or mismatched earlier, so getting the level of care right the first time prevents crises. If a hospitalization is needed, you are not powerless in choosing where, and you can advocate for a better unit even under pressure.

Is withdrawal dangerous? This is the one non-negotiable medical fact in the whole guide, so I will say it once and plainly: withdrawal from alcohol and from benzodiazepines can kill you. Seizures and delirium tremens are real. Unsupervised detox from either at home is genuinely dangerous, not just uncomfortable. Opioid withdrawal is miserable and rarely fatal in an otherwise healthy adult, though complications and dehydration carry their own risks. People do attempt “home detox,” and physicians sometimes manage outpatient withdrawal for appropriate candidates, but that requires medical supervision and the right candidacy. This is not a place to improvise. Assessment by a physician, not a guide, decides who can safely withdraw where.

Detox is not treatment. Medically managed withdrawal gets the substance out of the body over a few days and does almost nothing to change the life that produced the use. Leaving after detox and going home is one of the most dangerous things a person can do, and Section 4 shows you why. Detox is the door, not the room.

So the sequence is: stabilize safely first, then treat. If you are past the safety questions, you are choosing a level of care.

3. Levels of care, and when each one fits

The field uses these terms loosely and sales reps use them dishonestly, so here is the plain-English version, roughly most intensive to least. Treat the descriptions as what good care should include, not as guarantees the label delivers. Ask.

Medically managed withdrawal (detox). Short-term, supervised care through the acute physical part of stopping, usually five to ten days. Intensity varies, and acute psychiatric or medically managed inpatient care can be more intensive than ordinary residential. Not treatment on its own.

Residential / RTC (“rehab”). Living at a licensed facility with structure and clinical programming. This is what most people picture. Do not assume the label guarantees daily individual therapy, continuous psychiatric oversight, or high clinical intensity. Those are exactly the things to verify (Section 6). Length varies from the insurance-driven twenty-eight days to ninety days or more.

PHP (Partial Hospitalization Program). “Day treatment.” The person lives in sober living or at home and comes in for programming most of the day, most days, then goes back at night. A real step down that is still a serious daily commitment.

IOP (Intensive Outpatient Program). Programming several days a week for a few hours, built around returning to work, school, or life. The workhorse of the step-down phase.

OP (Outpatient). Weekly or twice-weekly therapy and psychiatry. Maintenance, not acute care.

Sober living / recovery residence. Housing, not treatment. A structured, substance-free place to live, with rules, curfews, testing, and peer accountability, usually paired with PHP or IOP. Good sober living is scaffolding while someone rebuilds a life. Bad sober living is a flophouse with a logo, and there is a lot of it (Section 6).

Dual diagnosis / co-occurring care. Care built to treat a psychiatric condition and a substance condition at the same time, in an integrated way, because in real people they are tangled. A program that treats the addiction and hands the trauma a pamphlet is not a dual-diagnosis program regardless of the website.

Which rung fits (orientation, not a placement decision)

Situation	Usually points toward
Dangerous withdrawal or medical instability	Medically managed withdrawal, or a hospital
Acute suicide or psychiatric risk	Emergency or inpatient psychiatric care
Cannot stay safe or abstinent outside 24-hour structure	Residential
Stable overnight but needs most-day programming	PHP
Functioning at home but needs several structured sessions weekly	IOP
Stable and maintaining gains	Outpatient
Home environment threatens recovery	Recovery residence alongside PHP or IOP

Read this as orientation, not an individualized placement determination. A clinician assessing the specific person makes the actual call.

Reasons to go up in level of care, and reasons not to

Go up when the person cannot stop despite real consequences and real effort, when the home environment is part of the problem, when a co-occurring condition has outrun outpatient care, or when safety is the question. Addiction is not a willpower problem, and treating it like one is why families stay stuck for years.

Do not go up when a person is stable and progressing in outpatient care and a shiny program just called, when the “problem” is defined entirely by an anxious family and not by clinical reality, or when residential is being used to avoid the slower work of building a life. And when someone has been to residential several times and walked out or been discharged each time, the question stops being “which residential” and becomes “what is actually driving this, and is there a different door.” That question is Section 11.

The right answer is almost never the most intensive option or the cheapest. It is the appropriate level for this person’s actual needs, followed by the step down, followed by the step down again.

4. The plan: duration, active treatment, and the staircase down

A treatment plan has to answer three questions: how long care should last, what actually happens during that time, and how the person steps down safely afterward.

Enough treatment matters

NIDA, the federal research agency, is direct that for many people, treatment engagement of less than 90 days is of limited effectiveness, and longer engagement generally produces better outcomes. Ninety days is not a marketing number invented by ninety-day programs. It is roughly how long the post-acute stuff (sleep, mood, cravings, concentration) takes to move from fragile to durable, and how long new habits need to hold.

What that does not mean is a single universal calendar for every case. The right amount of residential care inside that window depends on medical risk, psychiatric complexity, the home environment, prior treatment history, and whether the person can function safely at a lower level. Twenty-eight days treated as a completed course of care is where a lot of people get set up to fail.

Here is my position, and I will hold it: in severe, recurrent, or highly unstable cases, roughly three months of residential care followed by roughly three months of structured step-down (IOP plus sober living) is often the most defensible plan. Not because six months is a magic number, but because it usually gets a person across the engagement threshold and keeps a hand on their back through the most dangerous stretch, which is the next idea. This is a recommendation, not settled science, and it should flex to the person.

The calendar in plain English: For many people, plan for at least 90 days of continuous treatment engagement across appropriate levels of care. For severe, recurrent, or highly unstable cases, my practical recommendation is often closer to three months of residential treatment followed by three months of structured step-down. The person’s risk and functioning determine how much of that time belongs at each level.

For context on why this is a chronic-illness problem and not a moral failing: relapse rates for substance use disorders run about forty to sixty percent, in line with hypertension and asthma. Nobody says a diabetic “failed” because their blood sugar spiked and the plan needed adjusting. Same category of problem, same need for a long-term, adjustable plan.

Rule number one: occupied bed-days are not treatment

From the first stabilized day, there has to be a real, individualized plan that gets worked every day the person is in care. The exception is the front end, when someone is still detoxing or stabilizing and the only job is safety. Once that window closes, every day without an active plan is a day burned.

Time in a bed is not the same as treatment, and the industry blurs the two on purpose. A person can spend ninety days and a quarter of a million dollars in a beautiful facility and come out no better, because “treatment” was a loose rotation of groups and meals and equine afternoons with no driving clinical plan underneath, no defined targets, no one accountable for the specific things wrong with this specific person. That is expensive containment, and it is common. The duration point only holds if the days are used.

The failure mode families do not see coming: a person who drifts through treatment learns they can pass the time without doing the work, and then everyone concludes they “tried treatment and it didn’t work,” when what failed was the absence of a plan. A person cannot fail a plan that never existed. And the plan has to include the aftercare and the step-down from day one, not as a discharge-week scramble from whatever beds are open. That is the whole subject of a separate piece I have written: aftercare is not a phase you bolt on at the end, it is where treatment either becomes a life or stays an episode, and the assessment that drives it has to start at admission. Build the staircase down while the person is still standing at the top of it.

Why the weeks right after opioid treatment are especially dangerous

You may have asked how many people die at these places. The honest answer is that people rarely die inside a real program. The danger spikes the moment they leave, and this is where I have to be precise about which drug, because the evidence is overwhelmingly about opioids.

Getting sober, or even completing a detox, lowers opioid tolerance. If a person then relapses and uses the amount they used to use, that amount can now be a lethal dose, and in the fentanyl era the margin is almost gone. A Norwegian cohort found mortality in the first four weeks after medication-free inpatient treatment running roughly fifteen times higher than later, and every early death was an opioid overdose in someone who had dropped out. After opioid detox specifically, a large share of people relapse within the first weeks. The same lost-tolerance dynamic is why the two weeks after release from jail carry sharply elevated overdose risk.

Do not silently extend this to cocaine, alcohol, eating disorders, or depression, where the risks are real but different. For opioids, though, the point is stark and it drives the whole continuum: the moment a person exits structure with a reset tolerance and no scaffolding is the most lethal moment in the process. When a family says “he did his thirty days, he’s coming home,” for an opioid case that is walking someone up to the edge of the most dangerous window and letting go. The step-down is not padding. It is the hand on the back during the drop.

And for opioid use disorder, medication changes the odds. Staying on buprenorphine or methadone through and after treatment measurably lowers overdose and all-cause mortality. A program whose ideology forbids medication for opioid use disorder is not offering a philosophy, it is offering a liability.

What this means for you: aim for enough treatment plus a real step-down, insist that an actual plan gets worked daily, and for opioids, do not skip the transition or the medication.

Marcus’s family had already paid for treatment three times. What they had never been given was a plan that extended beyond the bed.

5. Choosing the route: addiction, psychiatric, or integrated

People arrive from different doors, and the door shapes the path. A cleaner way to sort it than “addiction versus mental health” is by what is actually driving the case:

• Primarily substance-driven.
• Primarily psychiatric (trauma, severe depression, OCD, an eating disorder), with substance use riding on top.
• Truly co-occurring, the two braided together.
• Unclear or repeatedly misidentified, which is Section 11’s territory.

If it is primarily substance-driven

Addiction treatment is broader than any single philosophy. It includes medication for opioid or alcohol use disorder, contingency management, cognitive and behavioral therapies, harm reduction, recovery housing, and mutual-help fellowship. The free on-ramps are real and everywhere:

• AA / NA, and their secular cousins.
• A sponsor, someone further along who walks you through the steps, a relationship that does real work no app replicates.
• A home group and service, because belonging and responsibility to other people is load-bearing.
• CODA / SLAA for the compulsive sexual behavior and love-addiction piece (relevant to Marcus), free.

One honest caveat, because “just go to AA” sounds simple and is not. AA is thousands of separate meetings that range widely in tone, format, and religiosity, from heavily Christian speaker meetings to secular discussion groups, from old-timer rooms to young people’s, LGBTQ, women’s, and profession-specific meetings, and the feel changes block to block. A meeting that saves one person is the same meeting the next person walks out of, and a bad first meeting drives more people away for years than almost anything else. So the task is not “find a meeting,” it is audition meetings until you find your people, which asks for persistence at the moment a person has the least of it. Two things worth knowing. Mainstream AA is a documented poor cultural fit for a lot of people, including many from racial, ethnic, and immigrant communities, and the culturally adapted meetings that would fit better are not reliably available. And AA is not the only game: secular alternatives like SMART Recovery, Women for Sobriety, and LifeRing exist, and in an observational comparison of active participants, people reported cohesion and satisfaction at least as high as in 12-step groups (a comparison of engaged members, not a randomized proof one approach beats another). If AA did not take the first time, that is usually a fit problem, not a verdict on recovery.

The escalation when the free tier is not enough: stabilize, then residential, then structured step-down and sober living, with medication where indicated. Abstinence and fellowship are one legitimate spine. They are not the only one.

If it is primarily psychiatric

For the person whose substance use rides on top of a psychiatric condition, and especially for the person who has “failed” addiction treatment repeatedly because nobody treated the thing underneath, start with a real psychiatric evaluation and a psychiatrist and therapist. If outpatient cannot hold it, the path is psychiatric residential, then PHP, then IOP, then supportive living. This route is often the more coverable one, because “medically necessary psychiatric care” is a category insurers understand. And the modality has to match the condition: trauma needs trauma-focused care, OCD needs exposure and response prevention rather than generic supportive talk (Section 6).

If it is truly co-occurring

Most of the hard cases I see are both, braided together. The trauma drives the using, the using deepens the depression, the depression feeds the trauma. NIDA is explicit that co-occurring disorders must be treated together. The path is dual-diagnosis residential (real integration, not addiction-with-a-side-of-feelings), then IOP, then supportive living, with case management through all of it (Section 8) and the approach matched to the actual driver.

Abstinence versus harm reduction is a case-by-case call, not a war

This gets treated like a holy war and should not be. Abstinence-based, 12-step approaches save lives. Harm reduction (meeting people where they are, reducing damage, keeping people alive long enough to change, including medication and naloxone) also saves lives. Both have evidence. The substance matters most of all: opioids in the fentanyl era shift the calculus hard toward keeping people alive by any means, including medication. The history matters, and the person’s own values matter. Someone with a decade of failed abstinence attempts and a near-fatal overdose may need harm reduction and medication to survive to the point where deeper change is possible. Someone else thrives in the structure of a fellowship. A program that treats its single philosophy as the only real recovery is telling you it will fail the patients who do not fit its mold, which is a lot of patients.

6. How to vet a program before you admit

Here is the part the marketing is designed to keep you from doing. Lead with the questions. The reasons they matter come right after.

Eight questions to ask before paying a deposit

Ask these and watch how they answer, because good programs answer easily and bad ones get cagey.

1. Licensing, for the exact address. Is the specific house and level of care where your family member will actually stay licensed by the state (in California, DHCS for residential SUD), and accredited by The Joint Commission or CARF? A polished brand can run several houses under different arrangements, so confirm the license covers the exact address and level of care, not just the company name.
2. Medical and psychiatric coverage, and who responds overnight. Not just “is there a medical director,” which is too easy to answer yes to. Ask who provides medical and psychiatric care, how often they are physically present, how quickly they can evaluate a patient who is struggling, and who responds in the middle of the night.
3. Who runs the groups, and the staffing. Licensed clinicians, or coaches and recent alumni? What is the clinician-to-client ratio, and the overnight staffing? Ask about the behavioral health technicians and overnight staff too, the people who sit with clients in the ordinary hours, because they often matter more to whether someone makes it than the director you met on the tour. I know, because I did that work. High turnover in the people who actually sit with your kid tells you more than any brochure.
4. The active plan. Is there an individualized plan worked every day with defined targets and someone accountable, or just a schedule of groups (Section 4)?
5. Referral conflicts. If a “consultant” or “interventionist” pushed one facility hard and fast, ask directly whether they are paid for the referral. Anything but a clean no, walk.
6. Current leadership. Programs change. The excellent program from two years ago may have lost its clinical director and gutted its staff since. Ask who runs it now, how long they have been there, and what changed.
7. Aftercare, in writing. What is the discharge and step-down plan, specifically, and who coordinates it? A program that cannot describe day 91 has not thought about the day people relapse.
8. The photos-to-substance ratio. The more the pitch is about the property and the less about the clinical model, medical oversight, and outcomes, the more nervous you should be.

And if a program advertises a success rate, ask how success was defined, who was counted, how long patients were followed, and whether an independent party collected the data. A percentage without those answers is marketing, not an outcome.

What to say when you call. Screenshot this. “Before we discuss admission, I need to understand the clinical plan, staffing, total cost, and discharge pathway. Who owns the program? Who will be the treating clinician, and how often will individual therapy happen? What specialty treatment is actually delivered? What is the projected length of stay, and what could insurance change? What is the proposed next level of care? And does anyone involved in this referral receive compensation? Please send the answers, the complete fee schedule, and the proposed treatment plan in writing.”

Compare three programs on paper

Ask every program the same questions and put the answers side by side. A program that refuses to answer basic clinical, staffing, ownership, or pricing questions in writing should not receive your deposit.

Question	Program A	Program B	Program C
Current state license (exact address)?
Current accreditation?
Who owns the program?
Medical/psychiatric coverage and availability?
Individual therapy per week?
Who runs groups?
Overnight staffing ratio?
Specialty treatment actually delivered?
Projected length of stay?
Total cash price, and what it excludes?
Insurance authorization process?
Written step-down plan?
Referral compensation disclosed?

Why these questions matter

The reason to be this rigorous is that the industry has a fraud problem measured in billions. In Arizona, ProPublica and the Arizona Center for Investigative Reporting exposed a Medicaid scheme, eventually pegged by the state at as much as $2.5 billion, that deliberately targeted Native Americans, warehousing people in sober homes where operators often let them keep using because sick, using clients kept the billing going. Medical examiner records showed at least forty deaths in those homes between 2022 and 2024. The engine is “body brokering,” paying kickbacks for referrals and treating people with addictions as revenue units to be bought and cycled, documented across Florida, Arizona, California, and beyond. A 2024 New Jersey State Commission of Investigation report found that state’s rehab industry “rife with abuses,” including facilities falsifying patients’ urine tests to manufacture a relapse and keep the money flowing, and operators have been criminally convicted for sexually exploiting the vulnerable people in their care. When you hand a frightened person to a facility, you are handing over someone with almost no power inside that building. The questions above are how you tell a real program from a beautiful sales pitch.

Match the specialty, not just the level of care

Fit is where a placement saves someone or wastes a year. Ask specifically:

Need	What competent care should include	What to ask
OCD	Actual exposure and response prevention (ERP)	Who delivers ERP, and how often?
Eating disorder	A defined nutritional and therapeutic model	How do you decide between DBT, RO-DBT, or another approach for a given patient?
LGBTQ / chemsex	Affirming clinicians and chemsex-specific relapse work	Is chemsex treated directly or just acknowledged?
Trauma	An evidence-based trauma model, with appropriate timing	Which trauma methods, and when in the stay?

Two deserve a note. OCD treated with reassurance instead of ERP can get worse, because reassurance is a compulsion. And eating disorder programs sound alike and are philosophically miles apart: standard DBT targets emotional chaos and under-control, while Radically Open DBT (RO-DBT) targets the rigid, over-controlled, perfectionistic profile that drives a lot of restrictive eating, and putting a patient through the wrong one can stall recovery. Choosing an eating disorder program without knowing which model it uses, and whether it fits the person, is choosing blind.

7. Money, insurance, and getting care covered

This is where families feel trapped, or guilty, or both, so let me be straight, including about numbers the industry works to keep vague. Most centers will spend an hour on your insurance benefits and go quiet when you ask the price. Treat a program that will not give you a number as a program telling you something.

What it costs (2026 market rates)

Level of care	Typical cost (before “extras”)	Notes
Detox	~$1,000–$2,500 per day (~$3,000–$25,000 per stay)	Medical stabilization only, usually 3–10 days
Residential, mainstream (national)	~$15,000–$60,000+ per 30 days	Wide range
Residential, LA / Malibu luxury	commonly ~$50,000–$100,000+ per month; top of the local market ~$135,000–$165,000	Sticker prices are often the floor
PHP (day treatment)	~$8,000–$15,000 per month	Lives in sober living or at home
IOP	~$5,000–$10,000 per month (coastal LA ~$8,000–$16,000)	The step-down workhorse
OP	~$1,000–$3,000 per month	Maintenance
Sober living	standard ~$800–$3,000; luxury coastal ~$5,000–$15,000; ultra-high-end estates ~$20,000–$30,000+	Housing, not treatment; usually on top of the PHP/IOP fee

Two programs quoting the same monthly price may be selling completely different packages. Before you compare numbers, ask for a written list of what is included and what is billed separately: detox, psychiatric visits, medication management, individual therapy, lab testing, transportation, family work, sober living, case management, outside specialist appointments, and private-room surcharges.

What programs actually cost

Costs vary enormously by level of care and setting. As rough Los Angeles anchors (2026): medical detox runs about $1,000 to $2,000 a day; residential runs roughly $6,000 to $30,000 a month at mainstream programs and $40,000 to $135,000 or more at luxury ones; PHP runs about $7,000 to $13,500 a month; IOP about $5,000 to $10,000; sober living about $800 to $5,000. In-network care can cost far less after your deductible, and Medi-Cal covers detox, residential, and outpatient at no cost for those who qualify.

For named programs with their specific costs, insurance, levels of care, and specialties, see the companion Los Angeles Mental Health & Addiction Treatment Directory. Two programs quoting the same price can be selling completely different packages, so before you compare numbers, ask each for a written list of what is included and what is billed separately.

Looking for actual programs? This guide teaches you how to evaluate treatment. The companion Los Angeles Mental Health & Addiction Treatment Directory lists programs by level of care, with approximate cost, insurance accepted, specialties, and the population each one serves.

Here is the thing the price tag will never tell you: more expensive does not mean better. I have worked inside programs at the top of this market and left them, not over money, but because the staffing changed or because I did not like how they were being run. A beautiful facility with a famous name can be coasting on reputation while the clinical team that made it good has quietly turned over. The rate buys real estate, privacy, and amenities. It does not guarantee the staff, the medical coverage, the clinical model, or the way the place is actually run on the day your family member walks in. Use the eight questions in Section 6 and the comparison worksheet, not the price, to tell a real program from an expensive one.

Where to find the actual programs

This guide deliberately names no programs, so it stays evergreen and gives you a framework rather than a sales pitch. The current, organized list of Los Angeles programs, by level of care, with approximate cost, insurance accepted, specialties, and what each is best for, lives in the companion Los Angeles Mental Health & Addiction Treatment Directory. Read this guide first so you know what to ask; then use the directory to shortlist, and run every finalist through the eight questions in Section 6 and the comparison worksheet in Appendix A.

8. The supports that hold the plan together

The clinical program gets the attention. The supportive layer determines whether the plan survives contact with real life, and families under-fund it.

Case management. A good case manager is the connective tissue: they coordinate the psychiatrist, therapist, program, and family so no two are working from a different story, they do the boots-on-the-ground work of getting a person to the right meeting and showing up when the wheels come off, and they catch people in the cracks where people vanish. It works because the failure mode in this world is fragmentation, and case management is the job of preventing it. Cost is mostly private pay: hourly rates run roughly $100 to $250, higher on the Westside and for licensed or RN-level managers, with monthly retainers from a few hundred dollars for basic monitoring to a few thousand for crisis-level coordination, and specialized recovery firms running to about $10,000 a month. One correction worth making: insurance may cover defined care-management delivered inside a medical system (Medicare, for instance, covers chronic care management and behavioral-health integration), but independent, high-touch recovery case management is commonly private pay. And at the top tier, price does not guarantee credentials, boundaries, or availability, so make staffing another thing you verify, not assume. The real vetting questions: real experience with this population rather than a generic “life coach,” a caseload small enough to actually give your person time, references from clinicians they have worked alongside, and, most important, boundaries rather than enabling. A good case manager supports the person and refuses to do the family’s enabling. A bad one is an expensive concierge. Ask directly how they handle a client trying to triangulate them against the team or family, and listen for a clear, immediate answer.

Sober coaches and the “rent a house, hire coaches” model. Some families rent a house and hire coaches to manage a person day to day, a bespoke one-person program. It is usually not ideal, because it can be under-structured and can quietly become an expensive form of enabling. But it has a use: for the person who has been discharged from residential or refuses a higher level of care, it can be a bridge to get some scaffolding around someone who has none. It is expensive, it lives or dies on the quality of the people, and it should be a bridge, not a destination.

Faith communities. If the person carries a genuine faith, their church, synagogue, temple, or mosque can be a powerful support, when it offers belonging and accountability rather than shame, and when it complements clinical care instead of replacing it. Faith and evidence-based treatment are not rivals, and the people who have both tend to do well.

9. Adjuncts and emerging treatments

A growing menu of treatments gets marketed hard, sometimes well ahead of the evidence. The through-line: structured care is the spine, and these help when the structure is there and distract when it is not. Two things people call “adjuncts” are not adjuncts and were handled earlier: medication for opioid use disorder is first-line, mortality-reducing treatment (Section 5), and home detox is a medical-safety question, not a modality (Section 2).

Ketamine and esketamine (Spravato). Esketamine, the nasal spray sold as Spravato, is FDA-approved for treatment-resistant depression in adults and for depressive symptoms in adults with major depressive disorder with acute suicidal ideation or behavior, given in-office under monitoring. Its own label is explicit that it has not been shown to prevent suicide or reduce suicidal ideation, and that it does not replace hospitalization when hospitalization is warranted. Racemic ketamine (IV, IM, sublingual) is used off-label, widely, via clinics and telehealth. Where it fits: the treatment-resistant depression that often sits under a substance problem. It is not an addiction cure, and it warrants caution in someone with an active substance use disorder, since ketamine itself is a dissociative with abuse potential.

TMS (transcranial magnetic stimulation). Noninvasive magnetic pulses to specific brain regions, no anesthesia, no drug, FDA-cleared for treatment-resistant depression and for OCD. Newer accelerated and theta-burst protocols exist, and specific accelerated systems have their own FDA clearances, so ask about the particular protocol and its clearance rather than assuming every intensive “Stanford-style” course carries the same standing. It treats the mood or OCD engine, which can be exactly what an assessment identifies as the driver.

Psilocybin. Precision matters here because the hype is enormous. Psilocybin is not FDA-approved for anything and remains federally illegal. Supervised adult access is legal only in Oregon and Colorado, out of pocket, outside a medical-diagnosis model. The research is genuinely promising for treatment-resistant depression, and a synthetic formulation (COMPASS’s COMP360) is in late-stage development, with the company projecting a final application to the FDA around late 2026 and a possible launch in 2027, on its own timeline, not an FDA guarantee. Be wary of any clinic overselling this as available and proven for a complex psychiatric and substance history. (For contrast, MDMA-assisted therapy for PTSD was declined by the FDA in 2024 and sent back for more data.)

Cognitive rehabilitation. Underused and squarely in my wheelhouse. Chronic substance use, and any real head injury, can leave measurable deficits in attention, processing speed, and executive function. Cognitive rehabilitation uses restorative exercises, compensatory strategies, and functional retraining to address them, and for the person whose thinking is genuinely impaired, it can make the rest of the work more possible. The evidence varies by presentation, so it is a serious tool to consider, not a guaranteed fix.

Hyperbaric oxygen (HBOT). Included because unrecognized brain injuries are common in this population and families ask. Be clear-eyed: HBOT is FDA-approved for a specific list of conditions, but for traumatic brain injury and cognitive recovery the evidence is mixed and it is investigational, oversold by plenty of clinics. It is not a factor that should decide your choice of program, and it is not a substitute for structured care.

10. Family involvement without control or abandonment

This is the chapter families want least and need most. I will be direct, because directness helps here where softness has not.

Many of the complex cases that reach me have been, in part, enabled into the room, not out of malice but out of love and fear: the apartment that made the using possible, the consequences smoothed away, the rescues that each quietly removed a reason to stop. That is the most human thing in the world, and it can also be part of what keeps the problem alive. There is often a saving-face layer too, especially in families with status, where the energy that should go into getting the person well goes into managing the story, and the secrecy isolates the person inside the shame that fuels the using. And there is triangulation, where the person plays the team, the family, and the providers against each other so no two hold the same information, which is exactly the gap case management exists to close.

There is also a spectrum problem, and a real research base behind it. On one side is the family that is over-involved in the wrong way: intrusive, enmeshed, calling constantly, demanding to run the clinical plan, and harassing the staff trying to do their jobs. On the other is the family that checks out entirely and treats treatment as a place to store a problem. Both predict worse outcomes. The construct is “expressed emotion,” measured as criticism, hostility, and emotional over-involvement, and high expressed emotion is one of the most reliable predictors of relapse identified. The 1998 Butzlaff and Hooley meta-analysis established this robustly across schizophrenia and, in its analysis, mood and eating disorders (it was not itself a meta-analysis of substance-use relapse). In substance use specifically, separate studies find that high perceived criticism at treatment intake predicts greater relapse, while family support predicts better participation. Both the critical-hostile pole and the over-involved pole are on the harmful side. Warmth is not the problem. Enmeshment, criticism, and control are. And none of this is about blaming families: expressed emotion is not a character verdict, it is a modifiable feature of the emotional atmosphere, and good family work moves it.

So here is what actually helps, more useful than another paragraph of theory:

• Pick one family spokesperson to communicate with the team, so the person cannot triangulate five relatives.
• Authorize appropriate communication among the providers so the team can hold a consistent line.
• Do not make side deals with the person that contradict the plan.
• Show up for scheduled family work without trying to run daily treatment.
• Set financial and housing boundaries before discharge, not during the next crisis.
• When the person is ambivalent or refusing care, consider an evidence-based family approach like CRAFT (Community Reinforcement and Family Training), which teaches families how to move a resistant loved one toward help without coercion or enabling.

And the hard part that is not the family’s job to do for the person: the person has to walk through the door, listen to the case manager even when it is uncomfortable, and stop triangulating. A perfect plan does not survive continued enabling, and I have watched it fail against enabling more times than I can count. The clinical work and the family work have to happen together.

11. When diagnostic clarification and neuropsychology help

Look back at Marcus. He arrives with an ADHD diagnosis, a decade of chemsex and cocaine, a self-harm history, a possible concussion, and three failed programs. Which of those is the driver? Because if you treat the wrong one, you fail, expensively, again.

This is where a comprehensive neuropsychological evaluation earns its place, but I want to be precise about when, because that makes the case stronger, not weaker. Neuropsychological evaluation is especially valuable when repeated treatment has failed, when cognitive functioning is genuinely in question, when diagnoses conflict, when developmental and psychiatric explanations stay tangled, or when the right treatment setting cannot be determined from ordinary clinical assessment. It is not the first move for everyone. Plenty of people need an addiction-medicine assessment, a psychiatric evaluation, or urgent stabilization first, and active intoxication, withdrawal, insufficient abstinence, or acute psychiatric instability can all limit what testing can even interpret. Timing matters.

When it is indicated, a good evaluation untangles the drivers: it distinguishes trauma from ADHD from OCD from substance-related cognitive impairment from a genuine head injury from a spectrum profile, and from the common reality of several at once. Knowing what you are actually dealing with is what takes the noise out of the decision, the difference between shopping blind and shopping with a map of the terrain.

Where I think my field falls short, and where I try not to, is the report that ends in generic recommendations the family then has to execute alone, in exactly the confusing market this guide describes. Better than a list is a real handoff: specific referrals based on current knowledge of the programs, direct coordination when appropriate, and a concrete plan for the next level of care, matched to the actual driver. Done that way, the evaluation does not just describe the problem, it saves a family a great deal of the learning-by-trial-and-error that this whole guide exists to shortcut. In a complex case that has already failed treatment more than once, a well-timed evaluation can be one of the highest-leverage expenditures in the process. It is a fraction of a single month of luxury residential and a small fraction of the cost of guessing wrong and cycling again.

And no evaluation does the walking for anyone. The person still has to walk through the door, stay consistent, listen to the right case manager, and stop triangulating, and the family still has to stop enabling. The eval builds the right plan. Living it is a separate job, and both are required.

12. Marcus, illustrated

Marcus is a composite, so let me be honest about what that means: I am going to walk the pathway a case like his tends to take, not report a real outcome. Treat this as the illustration it is.

He came in about his head, so the workup started there. In the composite, a careful evaluation would look for a persistent neurocognitive disorder attributable to the fall and, in a case like this, not find one, which is worth establishing rather than guessing. But the thinking would not read as clean either: enough substance-related cognitive slowing, in attention and processing, to name a mild, likely substance-related cognitive problem that needs a targeted intervention, because you cannot do deep therapeutic work on a foggy engine. The developmental history and findings would not support ADHD as the primary explanation, and would point more toward trauma, which changes the whole plan, since you treat those very differently and stimulants are not the answer to trauma. There would be OCD traits, real but more likely passenger than driver. And there would be a previously unflagged, unspecified eating disorder hiding in plain sight.

From there the plan writes itself: a gay-affirming residential and sober living program with a chemsex-specific track, using a trauma-focused approach paired with chemsex relapse-prevention work, because that targets what is actually driving the case rather than the label he walked in with. Targeted cognitive support for the thinking. A dietitian and eating disorder support for the piece nobody caught. And a case manager to hold the whole thing together and, frankly, to interrupt the family dynamics.

A family like Marcus’s often does not like a plan like this, because it is specific and uncomfortable and names things a decade of not-naming worked hard to avoid, and does not let anyone save face. In the version I am illustrating, they might even fire off a one-star review on the way out. I will admit the joke is mine. I will not pretend the review proves anything about them, because reaching for a family’s anger as evidence of their pathology is exactly the kind of thing a good evaluation is supposed to make unnecessary.

Where a case like this can go, when the plan fits and the person stays: the affirming care that actually matches, the trauma treated as trauma, the eating piece addressed, a community that does not require hiding, and a case manager who does something no report can, like helping someone find a technical school and a craft, and with it an answer to the question that has haunted him since Nashville, what he wants to do with his life. That is the point of getting the driver right. The noise comes out of the plan, the plan fits the person, and then the person and the family still have to live it.

13. A family’s quick-start checklist

If you read nothing else, read this.

First, safety (Section 2) – Immediate danger or medical instability: 988 or 911 now, not after you finish researching. – Heavy alcohol or benzodiazepine use and stopping: do not detox at home unsupervised, it can be fatal.

Choosing care (Sections 3–5) – Match the level of care to the actual need, and match the specialty too. – Sort by the actual driver: substance-primary, psychiatric-primary, or co-occurring. – Plan the whole staircase up front, residential to PHP or IOP to sober living, and aim for enough engagement to clear the roughly 90-day threshold, with more residential time in severe, recurrent, or unstable cases. Do not treat 28 days as a finished course. – For opioids, do not skip the transition or the medication, that is where people die.

Vetting any program (Section 6) – State licensed? Joint Commission or CARF accredited? – Named, board-certified medical director actually on site? – Licensed clinicians, not just coaches? Ratios and overnight staffing? Tech turnover? – An individualized plan worked daily with real targets, or just a schedule of groups? – Does anyone steering you get paid for the referral? (Not a clean no, walk.) – Who runs it now, and what has changed? – What is the written discharge and step-down plan, and who coordinates it?

Money and insurance (Section 7) – Ask the price, in dollars, and treat a refusal as a flag. – Get projected length of stay and discharge criteria in writing before admitting. – Ask the utilization team, point-blank, how good they are at concurrent review and appeals. – Line up the step-down from week one, because a cut can come sooner than promised. – Out of reach? Start with the LA and California resources at the top of this guide.

Red flags – “A bed is available today, but only today.” – Won’t tell you the price, or won’t put length of stay and discharge criteria in writing. – Won’t discuss medication for opioid use disorder. – Treats its single philosophy as the only real recovery. – Can’t describe what happens on day 91.

The family’s own work (Section 10) – One spokesperson, no side deals, boundaries set before discharge, and support the case manager instead of triangulating. A perfect plan does not survive continued enabling.

When to get an assessment (Section 11) – Repeated treatment failure, genuine questions about cognition, or conflicting diagnoses. Get the driver identified before paying for another placement built for the wrong one.

14. The decision in front of you

You do not need to become an expert in the entire treatment industry tonight. You need to avoid the dangerous mistake, ask the next honest question, and refuse to let genuine urgency be turned into sales pressure.

No program can promise recovery. But the right people, working the right plan at the right level of care, can give someone a real chance to begin it. That is what you are choosing: not a building, not a brand, and not thirty days away from home, but the conditions under which change might finally hold.

Choose the people, not the property. Choose the plan, not the promise. And build the staircase down before treatment begins.