On three weeks of sobriety, Thursday afternoon insurance denials, and the uncomfortable amount of influence a coverage decision can have over someone's treatment.
I've spent a significant portion of my career working in behavioral health, and there are certain things about the industry that, no matter how many times I experience them, still manage to make me question whether the system is functioning the way anyone actually intended.
Insurance utilization reviews are pretty high on that list.
I understand why they exist. Residential treatment is expensive, and there needs to be accountability surrounding the level of care being provided, the services being reimbursed, and whether patients are receiving treatment that's appropriate for their needs. I've worked on the financial and operational sides of healthcare long enough to recognize that approving every requested service without question would create an entirely different set of problems.
But I've also spent years working alongside clinical teams, watching patients move through different levels of care, and trying to reconcile what clinicians believe is appropriate with what insurance companies are willing to authorize. And there are moments when that disconnect feels less like responsible utilization management and more like a system that's become extraordinarily comfortable making consequential decisions about people it doesn't really know.
Consider a scenario that I suspect will sound familiar to anyone who's worked in residential behavioral health.
A patient has been in residential treatment for approximately three weeks. They've maintained sobriety throughout that time, which may be one of the longest periods they've gone without using substances in years. They're beginning to develop a therapeutic relationship with their clinician, recognize some of the patterns contributing to their substance use, and understand how their psychiatric symptoms, environment, relationships, and coping mechanisms have influenced their behavior.
They're making meaningful progress, but they're also nowhere near finished.
Perhaps they're still experiencing passive suicidal ideation. They aren't reporting an immediate plan or intent to harm themselves, but the thoughts haven't disappeared, and the treatment team remains concerned about their ability to maintain stability outside the residential environment. Their coping skills are improving, although they haven't consistently demonstrated that they can use them independently, and the recovery supports they'll need after discharge are still being established.
The therapist believes another couple of weeks would allow the team to continue addressing those vulnerabilities while developing a more thoughtful transition plan. The objective isn't to keep the patient in residential treatment indefinitely. It's to prepare them to move to a lower level of care with a reasonable expectation that they'll be able to maintain the progress they've made.
Then Thursday arrives, and the insurance company determines that continued residential treatment is no longer medically necessary.
The patient is no longer experiencing acute withdrawal, they're participating in programming, and they aren't expressing active suicidal intent. From the reviewer's perspective, the remaining concerns appear manageable through partial hospitalization or intensive outpatient treatment.
The clinical team disagrees and requests a peer-to-peer review, hoping to clarify why it believes the patient still requires residential care. Unfortunately, the conversation doesn't change the decision.
Now the team has a new problem, because even if everyone agreed that stepping down was appropriate, the patient doesn't have an established plan to begin treatment at the next level. Discharging someone into the weekend without confirmed services, adequate support, or a workable safety plan introduces additional concerns, so the program asks the insurer for transitional days.
The insurance company agrees to authorize the weekend.
How generous.
On paper, this gives the team Friday, Saturday, and Sunday to arrange an appropriate transition, with discharge anticipated on Monday. In practice, Friday is essentially the only remaining business day to coordinate with a receiving program, verify insurance benefits, arrange medication management, communicate with family or recovery supports, address transportation or housing concerns, and prepare the patient for a significant change in treatment.
The weekend may provide a few additional days in the residential environment, but it doesn't magically make outpatient admissions teams, insurance representatives, or community providers available to finalize everything that's still outstanding.
And somewhere in the middle of all this, the therapist is trying to explain to a patient who has spent three weeks beginning to trust the treatment process why the plan has suddenly changed.
I realize this is a hypothetical scenario, but it's built around a dynamic I've encountered repeatedly throughout my career. What frustrates me isn't simply that the insurer and the treatment team disagree. Qualified professionals can reasonably interpret clinical information differently, and the treating team isn't automatically correct simply because it knows the patient.
What bothers me is how quickly a complicated clinical decision can become an administrative deadline, leaving the people responsible for the patient's care to manage the consequences.
And when the patient is dependent on insurance coverage to remain in treatment, the practical difference between an insurer making a coverage decision and an insurer determining the course of treatment becomes increasingly difficult to appreciate.
When Getting Better Becomes the Argument for Why You No Longer Need Treatment
There's an uncomfortable irony in how continued-stay reviews sometimes unfold, particularly when a patient begins demonstrating improvement.
Someone enters residential treatment because their substance use, psychiatric symptoms, or circumstances have reached a point where the treatment team believes they require a structured, intensive environment. The program provides clinical support, medication management when appropriate, therapeutic interventions, and a setting designed to reduce exposure to some of the immediate risks and stressors contributing to their instability.
Ideally, those interventions begin working. The patient's symptoms become more manageable, they start participating more meaningfully in therapy, and their behavior suggests they're developing skills they didn't have when they arrived. That's exactly what we want to see, and it's the progress clinicians have been working to achieve.
But then the progress is documented during a continued-stay review, and suddenly it can become part of the justification for reducing coverage.
I understand the reasoning behind that determination. Residential treatment isn't intended to be a permanent living arrangement, and a patient shouldn't remain in a more restrictive setting when their needs can be safely and effectively addressed elsewhere. Moving people through the continuum of care as their condition improves is a fundamental part of appropriate treatment.
What I question is whether we're always distinguishing between a patient who's stable because of the treatment environment and a patient who's prepared to maintain that stability without it.
If someone has achieved three weeks of sobriety while living in a structured residential program, that progress is absolutely meaningful. But during those three weeks, they've also been removed from many of the environmental circumstances that contributed to their substance use. Their days have been organized around treatment, they've had access to clinical support, and they haven't had to navigate the same level of independent decision-making they'll encounter after discharge.
Their progress matters, but so does the environment in which it occurred.
The more useful question is whether they've developed the ability to maintain that progress when the structure changes. Can they manage cravings independently? Are they able to recognize when psychiatric symptoms are escalating and use the appropriate supports? Is their living environment conducive to recovery? Do they have relationships and resources outside treatment that can provide meaningful support?
And when we're talking about someone who continues to experience passive suicidal ideation, how are those thoughts being understood within the context of their history, current functioning, protective factors, and the circumstances they'll be returning to?
Passive suicidal ideation doesn't automatically justify residential treatment, and its presence shouldn't be treated as a substitute for a comprehensive risk assessment. But neither should the absence of an active plan or intent become an overly convenient reason to conclude that significant vulnerabilities have resolved.
I think we sometimes become so focused on identifying whether the most acute symptoms are still present that we overlook the broader question of whether the patient can sustain improvement with less intensive support.
The American Society of Addiction Medicine's ASAM Criteria recognize the importance of that distinction. Their framework emphasizes individualized, multidimensional assessment and movement through levels of care based on clinical needs, progress, and the patient's ability to receive appropriate services at a different level, rather than an arbitrary number of treatment days.
That makes sense to me, particularly because recovery doesn't occur according to a universal timeline. Two patients with similar diagnoses and similar lengths of sobriety may have very different needs depending on their histories, symptoms, environment, and available supports.
Which is why I find the phrase medical necessity both incredibly important and occasionally infuriating.
From an insurance company's perspective, medical necessity is the standard used to determine whether a particular service or level of care qualifies for reimbursement under the applicable clinical criteria and plan requirements. From the treatment team's perspective, the question is often broader and more individualized: What does this person need right now, and what level of care provides the most appropriate opportunity to address those needs?
Ideally, those questions should lead to similar conclusions. But when they don't, the resulting disagreement can have enormous consequences.
An insurer may acknowledge that a patient still needs treatment while determining that residential care is no longer justified. That's an important distinction, because denying continued residential coverage isn't necessarily the same thing as denying all behavioral health treatment.
But if the patient can't afford to remain in residential care without insurance, the decision about what will be reimbursed can effectively determine which treatment options remain available.
Technically, the insurer hasn't discharged the patient. The treating clinician can continue recommending residential care, and the patient can theoretically choose to remain.
It's just that somebody has to pay for it.
I suppose that's a comforting distinction if you're the person responsible for explaining why the clinical recommendation hasn't changed, but the patient is still expected to leave on Monday.
What Happens in Residential Treatment Doesn't Always Fit Neatly Into One Therapy Note
There's another part of this conversation that I think is worth examining before we even get to the peer-to-peer review, and it has to do with how we're representing a patient's treatment experience in the documentation being evaluated.
I've worked in several residential programs where patients are told upon admission that they'll receive an individual therapy session each week, and generally speaking, they do. But anyone who's worked in a residential setting knows that the rest of the patient's treatment experience doesn't simply disappear between those appointments.
Patients are participating in groups, interacting with peers, engaging with residential staff, attending psychiatric appointments when indicated, working on treatment assignments, and attempting to apply new coping skills in an environment where difficult emotions and interpersonal situations don't always wait for a scheduled therapy session.
And frankly, some of the most clinically meaningful information about a patient's functioning may emerge outside the therapist's office.
Someone might demonstrate impressive insight during their individual session but struggle to manage conflict with another resident later that evening. They may be able to explain their relapse-prevention plan beautifully in therapy while continuing to have difficulty recognizing escalating cravings or using coping skills independently.
They might participate appropriately in groups throughout the week but experience significant emotional dysregulation when confronted with a situation that resembles something they'll encounter after discharge.
Those observations don't necessarily mean the patient requires continued residential care, but they can be relevant when evaluating how much support the person needs and whether they're prepared to function with less structure.
And this is where documentation becomes particularly important.
I have some fairly strong opinions about group notes, which is probably enough material for an entirely separate article. Anyone who's spent time reviewing behavioral health documentation knows that there's an enormous difference between documenting that a patient attended a group on emotional regulation and documenting something clinically meaningful about how they participated, responded, or demonstrated difficulty applying the material.
A note stating that someone attended group, participated appropriately, and appeared receptive to the intervention may accurately describe what happened during that particular session. It doesn't necessarily tell us how the patient functions in other situations or whether the skills being discussed are translating into meaningful behavioral change.
I'm certainly not advocating for turning every group interaction into a dissertation, because the last thing most residential clinicians need is another creative way to spend their evenings documenting. But when we're discussing medical necessity, the quality and relevance of the clinical information available can make an enormous difference.
And then there's the reality of the therapist's workload.
I've worked in programs where clinicians were expected to manage caseloads of eight to ten highly complex patients while facilitating groups, responding to crises, communicating with families, coordinating referrals, attending clinical meetings, and helping overwhelmed residential staff navigate situations requiring additional clinical support.
Somewhere in there, they're also expected to complete accurate and timely documentation, understand payer requirements, participate in utilization reviews, and occasionally enjoy the luxury of eating lunch.
I'm not going to turn this into an article about clinician workload, although I could certainly write one. My point is that the expectations we place on residential therapists directly affect how clinical information is gathered, documented, and communicated.
A therapist who's managing multiple complex cases and responding to competing responsibilities may understand a patient's needs quite well while struggling to ensure that the entire clinical picture is consistently reflected in the record.
And when the documentation doesn't adequately capture what's happening, the insurance reviewer may be evaluating a considerably narrower picture of the patient than the treatment team is using to make its recommendations.
To be fair, reviewers may have access to psychiatric evaluations, treatment plans, interdisciplinary progress notes, utilization-review summaries, and other clinical information beyond the weekly individual therapy note. The documentation reviewed varies by payer and process, and it would be inaccurate to suggest that every insurer makes its determination based on one session.
But the problem remains when the clinical record is fragmented or doesn't clearly explain the relationship between the patient's symptoms, functioning, treatment response, and continued need for a particular level of care.
If the individual therapy note reflects improvement, the group notes provide little meaningful information, and important residential observations aren't clearly incorporated into the clinical assessment, the reviewer may see a patient who appears substantially more stable than the treatment team believes them to be.
Which brings me to something I've spent a considerable amount of time trying to teach my own clinical teams.
What Are the Barriers to Step-Down?
I've held many trainings with clinicians throughout my career, and there's one question I've repeated so often that I'm fairly certain some of them could recite it in their sleep.
WHAT ARE THE BARRIERS TO STEPDOWN?
I've told therapists to tattoo it on themselves, frame it above their desks, write it on their mirrors, or doodle it in little hearts if that's what helps them remember. I don't particularly care how they choose to immortalize the phrase, although the heart doodles would be a nice touch. I just want them to understand how important the question is and be able to answer it without hesitation.
Not with a vague explanation about how the patient could benefit from more treatment or isn't quite ready to leave. I want clinicians to be able to identify exactly what's preventing the patient from safely and appropriately transitioning to a lower level of care, how those barriers are affecting their functioning, and what interventions are being used to address them.
And I want that clinical reasoning reflected throughout the documentation, not introduced for the first time during a concurrent review because someone suddenly realized the authorization was about to expire.
Let's say a patient is maintaining sobriety within the residential environment but continues to experience significant cravings and difficulty using coping skills independently. Those observations are clinically relevant, but they don't necessarily establish why the patient requires residential treatment instead of PHP or IOP.
The documentation needs to connect those concerns to the patient's actual functioning and explain why the support available at a lower level of care may not yet be sufficient. It should also describe what the treatment team is doing to address those concerns and how the patient's response is informing the plan.
The same principle applies when there are unresolved psychiatric symptoms, concerns about the discharge environment, difficulties with emotional regulation, or other factors affecting the transition. Those issues deserve to be assessed and documented accurately, including whether they can be managed through outpatient treatment, recovery housing, or other community supports.
Because not every barrier requires another week in residential care, and clinicians should be able to recognize when a less intensive setting is appropriate.
What matters is that the record explains the clinical reasoning behind the recommendation rather than relying on the assumption that the recommendation should speak for itself.
And when I say the barriers should be reflected throughout the documentation, I don't mean copying the same paragraph into every note for three weeks. That would be a spectacular way to produce a lot of documentation without necessarily communicating anything useful.
I mean ensuring that the treatment plan, individual therapy notes, meaningful group observations, case management documentation, and discharge planning collectively tell an accurate, evolving clinical story. If a patient is struggling with a particular issue, the record should reflect how it's affecting them and what the team is doing about it. When the concern improves or resolves, the documentation should acknowledge that progress.
The objective is to help the patient overcome those barriers, not to preserve them on paper as an ongoing justification for reimbursement.
I think that's an important distinction because I've never believed the purpose of clinical documentation should be to manufacture a stronger insurance argument. Documentation should accurately support patient care, communication, treatment planning, and clinical decision-making.
But if we know that an insurer may ultimately rely on that same record to determine whether treatment can continue, shouldn't we make sure the documentation meaningfully reflects the patient's needs?
And shouldn't we be preparing clinicians to understand how their clinical observations and recommendations will be interpreted?
That brings me to another problem I've encountered throughout my career, one that has less to do with the patients themselves and considerably more to do with the clinicians being asked to defend their treatment.
We're Training Clinicians to Provide Care, but Who's Teaching Them to Fight for It?
I've had multiple therapists come into my office in tears after a peer-to-peer review.
And that has stayed with me, because these weren't clinicians who didn't care about their patients or lacked the ability to provide meaningful treatment. They were people who knew their patients, believed they were advocating for appropriate care, and walked away from a conversation with an insurance company's medical reviewer feeling belittled, overwhelmed, or completely defeated.
The way they described being spoken to was, frankly, awful. Dismissive, intimidating, and at times unnecessarily adversarial. What frustrated me even more was recognizing that many of these clinicians hadn't been adequately prepared to navigate that kind of conversation.
They weren't necessarily unprepared to provide treatment. They were unprepared to defend that treatment within an insurance system operating under an entirely different set of expectations.
And there's something else about this process that I've always found troubling. In all my years working with insurance reviews, I have never encountered an insurer's medical reviewer who had actually met the patient whose treatment they were evaluating.
Not once.
I recognize that's my experience, not proof that every insurer conducts reviews in exactly the same way. And I'm not suggesting that every independent reviewer must personally meet every patient to make a legitimate determination. Evaluating clinical records is a normal part of healthcare oversight, and independent reviews can serve an important purpose.
But I think it's worth considering the dynamic involved when a clinician who's spent hours working directly with a patient is expected to defend their recommendation to someone whose understanding of that individual comes from documentation and a relatively brief conversation.
That imbalance becomes particularly noticeable when the treating clinician is newer to the profession.
Many residential programs employ therapists who have recently completed their master's degrees, are working toward independent licensure, or are gaining their first significant experience with complex psychiatric and substance use populations. I've worked with newer clinicians who demonstrate exceptional judgment, compassion, and clinical ability, and I don't believe their relative inexperience with insurance processes should be mistaken for a lack of competence.
But providing excellent clinical care and navigating utilization management are not the same skill.
Think about what these professionals have spent years studying. They've learned about diagnosis, therapeutic interventions, crisis assessment, trauma, treatment planning, ethics, and the complexities of human behavior. They're learning how to establish therapeutic relationships with people who may be experiencing some of the most difficult circumstances of their lives.
Then they enter residential treatment and discover that another part of their job involves understanding concurrent reviews, continued-stay criteria, authorization deadlines, and how to communicate medical necessity in terms that satisfy a payer.
And in some cases, they're expected to learn all of that while managing a full caseload.
Graduate education isn't entirely silent on insurance and healthcare systems. The 2024 CACREP standards include third-party reimbursement and practice-management concepts, while social work education addresses advocacy and the broader systems influencing care.
But familiarity with those concepts isn't the same as practical preparation for presenting a complex case during a continued-stay review, responding to a denial, or navigating an intimidating peer-to-peer discussion.
Those skills often develop after graduation, and the quality of that education can depend heavily on the organization employing the clinician.
Which is why I believe leadership has an enormous responsibility here.
If we're going to expect therapists to communicate why a patient continues to require residential care, we should be teaching them how medical-necessity criteria operate, how clinical documentation supports level-of-care decisions, and how to advocate effectively when a payer disagrees.
We also need to ensure they have access to experienced clinical supervisors, medical leadership, and utilization-review support, particularly when a review becomes complicated or adversarial.
A peer-to-peer conversation should provide an opportunity for a meaningful discussion between qualified professionals. Depending on the payer's process, that may involve physician-level participation, even when the patient's primary therapist has the most extensive ongoing therapeutic relationship.
I understand why medical leadership may be involved, but I think organizations need to make sure the treating therapist's knowledge is adequately represented, regardless of who formally conducts the review.
And when a review doesn't go well, the response shouldn't simply be to tell the therapist they needed to document better.
Sometimes the documentation genuinely is insufficient, and that's worth addressing. But there is a difference between holding a clinician accountable for accurate clinical documentation and leaving them to navigate a specialized, potentially intimidating process without the preparation or support necessary to do it well.
That's why my obsession with barriers to step-down has never been about teaching clinicians how to win arguments with insurance companies.
It's about making sure they can clearly articulate their clinical reasoning, that the record accurately represents the patient's needs, and that the organization has done its part to support them when those recommendations are questioned.
Because there's something deeply frustrating about watching someone who's been doing meaningful therapeutic work with a patient walk into your office in tears after an insurance review, particularly when you recognize that the distress isn't simply about losing an authorization.
It's about being made to feel that their judgment doesn't matter, even though they'll remain responsible for helping the patient navigate whatever happens next.
And when the review results in a Thursday denial with transitional coverage through Sunday, those professional and emotional pressures don't disappear. The same clinician may now be responsible for helping coordinate an accelerated discharge while supporting a patient who's understandably concerned about the sudden change.
I don't think we can separate those administrative experiences from the broader conversation about quality care, because they directly affect the people responsible for delivering it.
What Does the Research Actually Tell Us About Recovery and Treatment Duration?
As much as I have strong feelings about this process, I think it's important to distinguish what I've personally witnessed from what the research can demonstrate.
And the research offers meaningful reasons to take treatment continuity and discharge planning seriously, although it doesn't support every assumption someone might make about residential length of stay.
The U.S. Surgeon General's report Facing Addiction in America describes substance use disorders as conditions that often involve recurrence, citing commonly reported relapse rates of approximately 40% to 60%. Those figures are broadly comparable to recurrence rates for other chronic illnesses, including diabetes, hypertension, and asthma.
That's important context because it reinforces that recovery isn't necessarily achieved simply because someone has experienced several weeks of improvement. Substance use disorders frequently require ongoing care, support, and monitoring, particularly when the individual has complex clinical needs.
But the statistic doesn't tell us that 40% to 60% of people who leave residential treatment after three weeks will relapse, and it doesn't establish that an insurance denial causes relapse. The estimate reflects recurrence across substance use disorders more broadly, and outcomes vary considerably depending on the individual, the substance involved, the treatment provided, and the follow-up period.
I think those distinctions matter because we weaken the argument when we use legitimate research to support claims it wasn't designed to establish.
The National Institute on Drug Abuse has also emphasized that adequate treatment duration is important, historically noting that participation in addiction treatment for less than ninety days is generally associated with limited effectiveness.
That's certainly worth considering when we're discussing someone who's only been sober for three weeks, but it doesn't mean every patient requires ninety days in residential care.
The broader concept is sustained engagement in appropriate addiction treatment, which may involve residential services followed by PHP, IOP, outpatient therapy, medication treatment, recovery housing, and other supports.
And when researchers have examined residential treatment duration specifically, the findings haven't always been consistent. Some studies have associated longer stays with better outcomes, while research involving Veterans Health Administration residential programs has found that longer average stays don't necessarily produce greater improvements across every patient population.
I don't think that undermines the importance of adequate treatment. It reinforces why the right answer depends on the person's needs and the quality and continuity of the care being provided.
The same applies to transitions involving psychiatric symptoms and suicide risk.
A 2017 systematic review and meta-analysis published in JAMA Psychiatry examined 100 studies involving more than 17,000 suicides following discharge from psychiatric facilities. Researchers found particularly elevated suicide rates during the first three months after discharge, with a pooled annualized rate of approximately 1,132 suicides per 100,000 person-years during that period.
These findings concern psychiatric facility discharges, not specifically insurance-driven residential substance use treatment denials. They also don't establish that a particular Thursday denial or Monday transition increases someone's risk by a quantifiable amount.
But they do reinforce the importance of carefully assessing risk during transitions between levels of care.
The National Action Alliance for Suicide Prevention and SAMHSA have emphasized the importance of coordinated discharge planning, safety planning, timely follow-up, and meaningful connections between inpatient and outpatient providers.
Those recommendations make sense because a discharge plan should represent an actual connection to continuing care, not simply a document stating that an individual has been referred somewhere.
And that's what brings me back to the weekend authorization.
If the clinical team believes that a safe transition requires coordination with another program, medication follow-up, appropriate housing, transportation, or additional recovery supports, those needs don't automatically become easier to address because the insurer has authorized three transitional days.
Some patients may already have those arrangements in place and be able to move safely to PHP or IOP with relatively little delay. Others may have significant unresolved needs that require more time and coordination.
The important question isn't whether every patient should remain in residential care for a particular number of days. It's whether the proposed transition is clinically appropriate and realistically achievable for the person involved.
And I think that's a much more useful standard than assuming the calendar reflects readiness.
Insurance Companies Have a Point, but That Doesn't Settle the Argument
I don't want to pretend that insurers are always wrong or that treatment programs are somehow immune to financial incentives.
Residential care is expensive, and organizations reimbursed for treatment days have a financial interest in maintaining census. Some programs undoubtedly provide services that aren't necessary, fail to transition patients when appropriate, or document clinical acuity in ways that don't accurately reflect the services being delivered.
That deserves oversight, and I think behavioral health organizations should be able to explain why the level of care they're providing is appropriate and what meaningful benefit patients are receiving from it.
If someone can safely continue treatment through PHP or IOP, I don't believe they should remain in residential care simply because another bed day can be billed. More intensive treatment isn't automatically better treatment, and remaining in a restrictive environment longer than necessary can have its own drawbacks.
But I also don't believe the existence of a utilization-review process automatically means the decisions it produces are clinically appropriate.
Independent oversight should involve meaningful examination of the patient's needs, the services being provided, and whether the proposed level of care can actually address the identified concerns. If the review focuses too narrowly on the absence of acute symptoms without adequately considering functioning, environment, treatment response, or realistic continuing-care options, it's fair to question whether the determination reflects the full clinical picture.
That responsibility also extends to the organizations providing care. If the treating team believes residential treatment remains necessary, the documentation should clearly explain why the patient's needs cannot currently be managed at a lower level, rather than relying on generalized statements that more treatment would be beneficial.
This is exactly why I spend so much time emphasizing barriers to step-down in my trainings. The objective is to make the clinical reasoning clear enough that another qualified professional can understand it, evaluate it, and respond meaningfully.
But I also think we need to acknowledge that even excellent documentation doesn't guarantee agreement, and an unfavorable decision doesn't automatically establish that the treating clinician's assessment was wrong.
There are legitimate differences in clinical judgment, and we should have processes capable of addressing those differences respectfully and transparently.
There's also a larger legal and regulatory conversation involving mental health parity. The Mental Health Parity and Addiction Equity Act establishes requirements intended to prevent covered health plans from applying more restrictive treatment limitations to behavioral health benefits than to comparable medical and surgical benefits.
That includes certain prior-authorization and medical-management requirements. The law doesn't guarantee approval of every residential treatment request, and an individual denial isn't automatically evidence of a parity violation. But the existence of those protections reflects a broader recognition that insurance practices can create barriers to mental health and substance use treatment.
Which makes me particularly interested in whether we're measuring the outcomes associated with utilization-management decisions.
I went looking for comprehensive national information showing how often residential or PHP continued-stay requests are denied based on medical necessity, how frequently those decisions are overturned, and what happens to patients afterward.
And the available data is disappointingly limited.
A March 2026 KFF analysis of 2024 Marketplace claims data found that insurers on HealthCare.gov denied approximately 19% of in-network claims. Among reported reasons for denial, 5% involved medical necessity, while other denials were associated with administrative issues, missing authorizations, exclusions, or reasons that weren't clearly specified.
But those numbers describe claims for services, not the full universe of prospective and concurrent authorization decisions. They can't tell us how frequently residential treatment authorizations are discontinued or how often clinicians disagree with those decisions.
They also can't establish what happens to patients after a disputed continued-stay denial.
So I can't responsibly argue that most residential authorizations are denied because of medical necessity, even though that's certainly a familiar reason for disagreement in the environments where I've worked.
The publicly available research simply doesn't support a reliable national estimate.
And I think that lack of transparency should concern both insurers and treatment providers, because without more meaningful information, it's difficult to evaluate whether these decisions are improving care or merely controlling immediate expenditures.
I'd be interested in knowing how often patients whose continued residential coverage was denied successfully transition into the recommended lower level of care, and how frequently those transitions are followed by treatment disengagement, emergency department visits, psychiatric hospitalization, or readmission.
I'd also want to understand how those outcomes compare with clinically similar patients who received additional treatment, recognizing that those comparisons would require careful research rather than simply assuming one decision caused a particular outcome.
That seems like a reasonable expectation for an industry that spends an extraordinary amount of time discussing evidence-based treatment.
If we're going to use clinical criteria to determine which services are medically necessary, shouldn't we also be interested in whether the decisions made using those criteria are producing the outcomes we intended?
If the Patient Comes Back, Was the First Decision Actually Cost-Effective?
There's another part of this conversation that I find particularly interesting from an operational and financial perspective.
When an insurer declines to authorize another two weeks of residential treatment, the immediate financial implication is fairly straightforward. The plan avoids paying for those additional days, and if the patient can safely transition to less intensive care without compromising treatment effectiveness, that may be an entirely appropriate use of healthcare resources.
But the calculation becomes more complicated when the transition doesn't go as planned.
If a patient disengages from continuing care, experiences a recurrence of substance use, requires emergency psychiatric services, or returns to residential treatment, the resulting costs can extend well beyond the initial episode of care.
And those costs aren't exclusively financial. The patient may experience additional distress, relationships may be disrupted, families may find themselves navigating another crisis, and the individual may become less willing to seek treatment again.
None of those outcomes is inevitable, and I wouldn't suggest that every relapse or readmission can be traced back to an insurance decision. Recovery is complicated, and people can experience setbacks even when their care has been coordinated thoughtfully.
But I do think these possibilities raise legitimate questions about what we're measuring when we describe utilization management as cost-effective.
If success is defined primarily by reducing the cost of the current treatment episode, shortening a residential stay may accomplish that. But if the objective is to support sustained recovery while avoiding unnecessary future healthcare utilization, we need to consider a much broader picture.
The problem becomes even more complicated when financial responsibility is distributed across different organizations and systems. A patient may change insurance plans, receive emergency services through another provider, or eventually access treatment funded through a different source. The organization benefiting from the immediate reduction in spending may not be the same one absorbing the cost of subsequent care.
That doesn't prove that longer residential treatment would have prevented a setback, but it does demonstrate why looking only at the cost of the current authorization can be misleading.
And I think this is where both insurers and treatment providers could benefit from more meaningful transparency and shared accountability.
Treatment programs should be able to demonstrate that the additional services they're requesting are clinically appropriate, that their interventions are addressing identifiable needs, and that patients are making progress toward the least intensive level of care capable of supporting them. Insurers, meanwhile, should be able to explain how their medical-necessity criteria are applied and demonstrate that those processes are supporting appropriate treatment rather than functioning primarily as mechanisms for reducing spending.
When a treating team and an insurer disagree, I would also like to see greater attention paid to what happens after the determination. A patient who transitions successfully to PHP after three weeks of residential care may provide evidence that the step-down was appropriate. Another who experiences significant treatment disruption may raise questions about whether the timing, support, or clinical assessment could have been different.
Neither outcome, on its own, proves that the original authorization decision was right or wrong. But examining patterns across patients and programs could tell us considerably more about the effectiveness of utilization management than simply counting how many days were approved or denied.
And considering how frequently behavioral health organizations are expected to demonstrate measurable treatment outcomes, I don't think it's unreasonable to ask the systems influencing access to that treatment to do the same.
The Part I Keep Coming Back To
What stays with me isn't the idea that insurance companies and clinical teams sometimes disagree. That's inevitable in a system where different professionals have different responsibilities, financial considerations, and interpretations of appropriate care.
It's the way those disagreements can unfold, and the amount of responsibility that remains with the people providing treatment after the coverage decision has been made.
I think about the therapists who've come into my office after difficult insurance reviews, sometimes in tears, describing how they'd tried to explain their patients' needs and left the conversation feeling dismissed or professionally diminished. Some were relatively new clinicians who hadn't yet received adequate preparation for the process, while others had considerable experience and were still frustrated by how little room there seemed to be for meaningful clinical discussion.
I think about how much work goes into residential treatment beyond the weekly individual therapy session, and how easily that work can become fragmented across documentation systems, group notes, crisis interventions, and interdisciplinary conversations. When the record doesn't adequately communicate the patient's functioning and remaining needs, it's harder for someone reviewing the case from a distance to understand why the treatment team is recommending continued care.
And I think about the trainings I've held, where I've repeatedly asked clinicians to identify the barriers to step-down because I genuinely believe that understanding those barriers is essential to appropriate treatment planning. I've never wanted teams to document concerns simply to extend authorizations. I want them to understand what the patient needs, what the team is doing to address those needs, and what has to change before a less intensive level of care becomes appropriate.
That's a clinical responsibility, regardless of whether an insurance company is involved.
But when a coverage decision changes the timeline for treatment, I think insurers have a responsibility to consider that reasoning carefully, too. A review process should be capable of distinguishing between a patient who's improving within a structured environment and one who's prepared to continue improving with substantially less support. It should also account for whether the proposed services are actually available and whether the transition can be coordinated appropriately.
That doesn't mean approving every request for additional residential days. Nor does it mean assuming that a clinician's recommendation should automatically take precedence over an independent review.
It means treating the disagreement as something worth understanding rather than simply resolving through an administrative deadline.
And perhaps that's what I find most troubling about the Thursday denial scenario.
We have a patient who's made meaningful progress during three weeks of residential treatment, a clinical team that believes additional time is necessary, and an insurance reviewer who concludes that treatment can continue at a lower level of care. The disagreement itself is not unreasonable, and either conclusion could potentially be supported depending on the patient's circumstances.
But once the insurer declines continued coverage and grants only the weekend for transition, the conversation changes. Now the team has to coordinate a complicated discharge within a window that may provide only one functional business day, while the patient is trying to process an unexpected change and the treating clinician is attempting to preserve as much continuity as possible.
I can appreciate that the insurer has provided transitional coverage rather than simply ending payment immediately. In some situations, that may be sufficient, particularly when continuing-care services are already arranged.
But when the identified barriers remain unresolved and the necessary supports aren't yet available, I don't think a weekend extension automatically addresses the clinical concerns that prompted the request for additional treatment in the first place.
The calendar doesn't tell us whether the patient can manage cravings independently, whether their psychiatric symptoms are sufficiently stable, or whether their recovery environment will support continued progress. And it certainly doesn't tell us whether the receiving program has an opening on Monday.
Those things require assessment, coordination, and clinical judgment.
I believe behavioral health organizations should be held accountable for providing appropriate care, documenting their clinical reasoning, and helping patients move through treatment without unnecessary restrictions or delays. I've spent too much time on the operational side of healthcare to believe that financial accountability is somehow incompatible with quality treatment.
But I also believe accountability needs to extend to the systems that influence whether patients can access that care. If insurance companies are going to make determinations that substantially affect treatment duration and transitions, those determinations should reflect an accurate understanding of the patient's needs, and we should be interested in evaluating the outcomes that follow.
And if we're going to expect therapists to advocate effectively through those processes, particularly clinicians who are relatively new to the profession, then we need to prepare and support them rather than leaving them to learn through difficult, sometimes demoralizing experiences.
Ultimately, I don't think the most important question is whether a patient has spent enough days in residential treatment to satisfy a particular authorization standard. What matters is whether the current level of care remains appropriate, whether the next level can realistically meet the patient's needs, and whether the transition is being driven by clinical readiness rather than primarily by the expiration of coverage.
Which brings me right back to the question I've spent years asking my clinical teams, occasionally with suggestions that they tattoo it somewhere visible.
What are the barriers to step-down?
If those barriers have been sufficiently addressed and a safe, appropriate lower level of care is available, then let's help the patient move forward. That's what good treatment is supposed to accomplish, and keeping someone in residential care unnecessarily doesn't serve them.
But if significant barriers remain, the clinical team should be able to explain them clearly, and the insurance reviewer should be willing to meaningfully consider that explanation. And when the decision is that the patient can step down anyway, I think we should be able to explain why that transition is appropriate and how it can realistically be supported.
Because the person receiving treatment doesn't experience medical necessity as a definition in an insurance policy or a set of criteria being discussed during a peer-to-peer review. They experience it as a decision that changes what happens to them next, sometimes with very little warning and without fully understanding why.
For the patient who's been sober for three weeks and is finally beginning to believe that recovery might be possible, Monday morning is not an abstract coverage determination. It's the moment they leave the environment that's helped them stabilize and begin trying to maintain that progress somewhere else.
Maybe they're ready. Maybe the insurer's assessment was entirely appropriate, and continuing treatment through PHP or IOP is exactly what they need.
But if we're going to make that determination, shouldn't we be able to explain it in terms of the patient's actual clinical needs rather than simply pointing to the authorization?
And shouldn't we care enough to find out whether we were right?
Because telling someone they're ready to step down doesn't make them ready. And a weekend extension, however generous it might look on paper, doesn't automatically make Monday a safe day to leave.
