The Revolving Door: Why Treating Mental Illness Properly Would Cost Less

Inspired by: “The Revolving Door” by Adam Iscoe (The New Yorker, May 22, 2023)

Introduction – The Door That Never Stops Turning

Every year in America, hundreds of thousands of people with serious mental illness circle through the same institutions: police stations, emergency rooms, county jails, short-term psychiatric wards, and then back again to the streets. The system calls this “recidivism,” but it might more accurately be called systemic design—a feedback loop built on neglect.

Adam Iscoe’s “The Revolving Door” follows one such life in New York City, where the state’s psychiatric facilities have become civic hinges, swinging open and shut for the same patients dozens of times a year. His story is local, but the pattern is national. From Los Angeles’s Skid Row to Chicago’s West Side, to rural counties where the jail doubles as the mental-health unit, the U.S. has turned short-term containment into its primary form of treatment.

The consequences are measured not only in human suffering but in dollars. The current architecture—fragmented, punitive, and crisis-driven—costs billions more than humane, continuous care would. Yet we keep building more doors instead of more rooms.

The Landscape of Neglect

To understand how this happened, we have to look backward. In the 1960s and ’70s, deinstitutionalization was supposed to replace massive state hospitals with robust community clinics. Instead, the first part happened and the second never did. Public hospitals closed, budgets shrank, and the promised network of local supports materialized only in pilot programs and grant proposals.

The result: emergency departments and jails became the default institutions of mental-health care. A sheriff in Ohio recently observed that he now oversees the largest psychiatric facility in his state—his own county jail. Nationwide, roughly two million people with serious mental illness are booked into U.S. jails each year, and about 40 percent of those released will return within twelve months.

For the individuals inside the loop, “treatment” often means sedation, a plastic tray of food, and a paper prescription at release. There is no follow-up visit, no housing, no therapist, no continuity of medication. When the symptoms return—as they inevitably do—the person re-enters the cycle through whichever institution happens to open first.

This landscape isn’t accidental; it’s a budgetary mirage. The spending appears lean because the costs are scattered—police departments pay part, hospitals another, counties a third. No single ledger captures the whole price of dysfunction. But when you assemble the full picture, the math becomes indefensible: the United States spends more per capita managing mental-health crises than it would providing sustained, coordinated care.

Beneath the statistics lies a moral geometry. Systems bend toward what they measure, and ours measures throughput—how fast a patient can be stabilized, documented, and discharged—not recovery. The revolving door is efficient only at spinning.

The Economics of Failure

If moral arguments can’t stop the revolving door, perhaps economics can. The U.S. mental health system is not just broken—it is expensive in precisely the ways that yield the worst outcomes. Each cycle through jail or an emergency room costs far more than an entire year of coordinated, community-based treatment. But because those costs fall on different agencies—police, hospitals, courts—no one entity sees the full bill.

The numbers tell the story better than any ideology. Consider a single repeat patient in a major city: multiple 911 calls, transport by two officers, several ER visits, a three-day psychiatric hold, then release with a bus token. Within a month, the cycle repeats. Add up every jurisdiction’s share—municipal, county, state, federal—and the same person may cost $70,000 a year while never receiving sustained therapy, housing, or stability.

Contrast that with programs like Housing First in Utah and Houston, or Assertive Community Treatment (ACT) teams in New York and North Carolina. These models replace episodic interventions with continuous engagement: a stable apartment, a clinician who knows the patient, medication management, and job or education support. Their total annual cost per person runs about $25,000—one-third of the status quo. In other words, the humane option is the fiscally conservative one.

Sidebar: By the Numbers – The Economics of the Revolving Door (These are illustrative, order‑of‑magnitude ranges drawn from multiple program evaluations)

Interpretation: For every dollar spent sustaining the revolving door, the humane alternative yields roughly three dollars in measurable savings within five years—not counting the invisible dividends of restored families, reduced trauma, and safer communities. The most successful pilots share one feature: continuity. When care follows the person instead of the crisis, the revolving door finally slows.

Beyond the Ledger

The numbers are only the floor of the argument. Economists can quantify savings, but they struggle to price despair, stigma, or the erosion of public trust that occurs when citizens watch loved ones fail in plain sight. Each year of delay magnifies secondary costs: the child removed from a parent’s custody, the eviction that leads to chronic homelessness, the untreated psychosis that becomes a violent headline.

In this light, the revolving door is not just a policy failure but a civic accounting error—one that hides its interest charges in every other social budget line. The alternative—continuous, compassionate, evidence-based care—is less a reform than a correction.

Systems That Work (and Why They’re Rare)

Across the country, a handful of programs show that the revolving door can, in fact, be stopped. They differ in geography and design, but each shares one essential trait: they treat continuity as medicine.

The Clubhouse Model – Belonging as Treatment

In midtown Manhattan, Fountain House pioneered what is now called the Clubhouse model—a community built on membership rather than diagnosis. Participants help run the daily operations: staffing the kitchen, publishing a newsletter, connecting peers to jobs, and mentoring newcomers. Clinical care happens quietly alongside purpose. Members who once cycled through hospitals every few months now maintain housing and employment for years at a time. Independent studies find that Clubhouse participation substantially reduces psychiatric hospitalization and significantly increases competitive employment compared with usual care.

What’s radical here isn’t therapy; it’s relationship. When a person is missed if they don’t show up, treatment becomes human again.

Assertive Community Treatment – Care That Finds You

The ACT model flips the traditional outpatient structure on its head. Instead of asking the patient to find care, it sends a team—psychiatrist, nurse, social worker, and peer specialist—to the patient. Services are mobile, flexible, and available 24/7. An ACT team might meet a client at a park bench, help refill a prescription, resolve a landlord dispute, or coordinate medical appointments. These teams have been operating for decades, and the data are consistent: often cutting hospital use roughly in half or more for the highest‑need clients, arrests decline, and adherence to medication rises sharply. The average caseload per team is about 100 individuals, meaning a single ACT team can stabilize an entire neighborhood’s worth of high-need residents.

Housing First – Shelter as the Foundation of Sanity

Utah reported dramatic reductions in chronic homelessness—on the order of 70–90% in some measures—during its initial Housing First push, alongside substantial drops in ER use and jail stays among participants. Houston and other cities that have adopted similar models have also documented large declines in chronic homelessness.

CAHOOTS – Rewriting First Response

In Eugene, Oregon, the CAHOOTS program replaces police cruisers with vans staffed by a medic and a crisis worker. They respond to 911 calls involving mental illness, homelessness, or addiction—situations that rarely require handcuffs. In a recent year, CAHOOTS handled tens of thousands of calls, and only about 1% required police backup. City estimates put the annual savings from diverted police and ER responses in the several‑million‑dollar range.

CAHOOTS and its offspring (like Denver’s STAR program) demonstrate that compassionate response is not only safer but cheaper. When a crisis is treated as a medical event rather than a criminal one, the outcomes shift dramatically.

Why Success Doesn’t Scale

Despite clear evidence, these models remain islands in a storm. The barriers are systemic:

• Fragmented funding. Medicaid covers medical care but not housing; HUD funds housing but not psychiatric services. Each agency guards its silo.
• Political optics. Spending on jails reads as “public safety.” Spending on treatment reads as “welfare.”
• Workforce shortages. The U.S. has fewer than one psychiatrist per 10,000 residents, and burnout drives turnover in nursing and social work.
• Liability fears. Hospitals discharge quickly to avoid lawsuits for holding patients against their will, even when that means returning them to the street.

These disincentives create what systems theorists call a reinforcing feedback loop: neglect breeds crisis, crisis inflates costs, inflated costs justify further austerity. To break the loop, someone has to own the whole system—funding, outcomes, and accountability. Until then, the success stories remain exceptions proving a tragic rule.

The Feedback Loop of Failure

When a system produces the same results everywhere, it is no longer accidental. America’s revolving door isn’t a collection of mistakes; it’s a structure reinforced by the very feedback loops that should be healing it.

Funding, for instance, flows to visible crises rather than invisible prevention. Cities get reimbursed for police overtime and hospital admissions, not for long-term stability. Each ambulance trip generates a line item; each avoided trip generates nothing. Hospital administrators are judged on turnover speed, not continuity of recovery. Legislators can point to new psychiatric beds as proof of progress even if the same patients occupy them in endless rotation.

Insurance reimbursement deepens the loop. Short-term stabilization is billable; sustained case management often is not. Private hospitals discharge early because longer stays reduce profitability. Public hospitals discharge early because beds are scarce. Both decisions look efficient in isolation—but together, they guarantee return visits.

Meanwhile, the cultural narrative still frames serious mental illness as moral failure or personal weakness. “Noncompliance” becomes a clinical label, not a symptom of an overwhelmed person navigating a labyrinth. The stigma that once kept people institutionalized now keeps them untreated.

The result is a national system optimized for throughput rather than healing—a machine that spins faster the more it fails.

What “Doing It Right” Would Look Like

Imagine the opposite system—one that measures its success by absence: fewer crises, fewer police calls, fewer hospital readmissions. Its basic architecture is simple:

• Continuity instead of fragmentation. Every patient has a single care coordinator with authority across health, housing, and justice domains. The goal isn’t to manage paperwork but to manage lives coherently.
• Housing as infrastructure. Housing becomes part of healthcare funding. A person stabilized under a roof costs a fraction of what they cost on the street.
• Integrated crisis response. 911 diverts all behavioral-health calls to dedicated teams—nurse, clinician, peer counselor—backed by on-call psychiatrists. Police respond only when safety truly requires it.
• Outcome-based funding. Federal and state reimbursement ties directly to reduction in crisis episodes, not just number of patients “served.” Systems are rewarded for keeping people out of hospitals and jails.
• Peer and family partnerships. Every regional network employs individuals with lived experience of mental illness as staff and mentors. Family caregivers are trained and compensated as part of the continuum, not sidelined.
• Long-term accountability. National dashboards track metrics—recidivism, employment, stable housing—over five and ten years, not quarterly budget cycles.

If this model sounds idealistic, it isn’t. Versions already exist elsewhere. In regions of Finland that adopted the Open Dialogue approach, long‑term psychosis outcomes—relapse rates, hospital days, and disability—have been markedly better than typical U.S. figures reported in comparable cohorts. The difference lies not in miracle drugs but in continuity, compassion, and design.

A Rough Cost of Wholeness

As a rough thought experiment, if we invested on the order of tens of thousands of dollars per year per person in comprehensive care, the total might be comparable to what we already spend indirectly on crisis services, lost earnings, and family burden. Evaluations of existing programs suggest that, over several years, many of those investments would likely pay for themselves in reduced crisis and incarceration costs.

• $630 billion in crisis and incarceration costs,
• $193 billion in lost earnings,
• $400 billion in family and social costs (caregiving, evictions, legal fees).

Doing it right would pay for itself in five years and generate long-term dividends in safety, productivity, and human dignity. Yet politically, the payoff falls outside election cycles. As one policy analyst put it, “We fund disasters, not defenses.”

The paradox is that reform is not a matter of invention but of integration. The pieces already exist; they simply live in separate budgets. To heal the system, we must first teach it to see itself whole.

The Moral Accounting

Behind every policy debate lies an unspoken question: What kind of society do we believe ourselves to be? When the revolving door turns, it’s not just patients who pass through—it’s our collective conscience.

The moral cost of the current system is not abstract. It appears in the face of a mother waiting through a 72-hour psychiatric hold, knowing her child will be released with no plan. It shows up in police officers trained for violence rather than empathy, in doctors forced to discharge people they know will return. Each broken link becomes someone else’s problem, until the chain itself collapses.

Economists can calculate savings, but only ethics can measure neglect. The real loss is not money but potential—the talents and relationships never realized, the innovations and communities that vanish into untreated illness. Every person stabilized is not just a cost avoided but a story restored.

This is the deeper logic of interconnectedness that Humboldt’s Home seeks to illuminate: systems of care mirror systems of thought. When we see people as problems to be processed, we create machinery that processes endlessly. When we see them as part of a living whole, we design systems that heal.

Conclusion – Stopping the Spin

If the revolving door is the symbol of our failure, it can also become the blueprint for our repair. Doors, after all, are made to open and close with intention. We could choose to slow the mechanism—to replace rotation with refuge.

Doing so requires that we act as if wholeness were measurable: fewer crises, longer intervals of wellness, stable housing, steady work, restored dignity. These are not miracles; they are outcomes that thousands of local pilots have already achieved.

The task is national integration, not national invention. We already know what works; we simply fail to scale it. The choice is stark but simple: continue funding a cycle that multiplies suffering, or invest in continuity that compounds healing.

To stop the door, we must first decide who belongs inside.

Classroom Prompts

• Why do societies invest so much in crisis and so little in prevention?
• What structural incentives—political, financial, or cultural—keep the revolving door spinning?
• How might the design principles of successful models (continuity, housing, community) apply to other broken systems like foster care or criminal justice?
• If you were tasked with redesigning the U.S. mental-health system from scratch, what would your top three priorities be—and why?

Annotated Sources

• Iscoe, Adam. “The Revolving Door.” The New Yorker, May 22, 2023. The narrative foundation for this essay, chronicling Houston’s psychiatric emergency center and the recurring patients it serves.
• National Alliance on Mental Illness (NAMI). “Mental Health by the Numbers,” 2024. Provides up-to-date national data on prevalence, incarceration rates, and economic cost of untreated mental illness.
• Substance Abuse and Mental Health Services Administration (SAMHSA). “Behavioral Health Barometer,” 2023. Offers comparative data by state on treatment access, hospitalization, and outcomes.
• Treatment Advocacy Center. “The Shortage of Psychiatric Beds in the U.S.,” 2023. Explains the historical and policy context of deinstitutionalization and its consequences.
• Bureau of Justice Statistics (DOJ). “Indicators of Mental Health Problems Among Jail Inmates,” 2022. Key evidence for the scale of incarceration as de facto treatment.
• Fountain House. “Impact Report,” 2022. Documents success of the Clubhouse model in reducing hospitalizations and improving employment outcomes.
• City of Eugene. “CAHOOTS Annual Report,” 2023. Provides operational data and cost savings for mobile crisis intervention models.
• HUD Office of Policy Development. “Evaluation of Housing First Programs,” 2023. Quantifies cost savings and long-term outcomes for supportive housing interventions.

© 2025 Michael A. Pink