CMS Just Wrote “Not Sick Enough Yet” Into Federal Medicaid Policy

Justin Bell, PhD, was sitting in a doctor’s office, escalating on painkillers, when the physician looked at him and called it “a bump in the road” — then sent him out the door. No referral. No follow-up question. No curiosity about what “escalating” meant or what Bell might have needed to stop. Bell tells this story himself, today, in an essay on Recovery Review describing remarks he wrote for medical students — because two decades later, he’s a psychologist and a person in long-term recovery, and he still remembers exactly what it felt like to be waved off by someone whose job was to see him.

Here’s the sentence you should screenshot:

The same instinct that made one doctor decide Bell wasn’t sick enough to help is now written into federal Medicaid rules that take effect in ten days.

On June 1, 2026, the Centers for Medicare & Medicaid Services issued an interim final rule — the regulatory text agencies publish when a law tells them to move fast — implementing the new 80-hour-a-month Medicaid work requirement created by this year’s federal tax and spending law. States have to start enforcing it by January 1, 2027. The rule is open for public comment until July 31, 2026 — ten days from today. Buried inside it is a definition of who counts as too sick to be required to work, and that definition treats a substance use disorder (the clinical term for what most people just call addiction) almost exactly the way that doctor treated Bell: as something you have to prove is serious enough, permanent enough, bad enough, before anyone in power will take it seriously.

This is not a coincidence of timing. It’s the same logic wearing two different uniforms — one white coat, one federal regulation — and it’s worth understanding both halves of the story, because the second half is moving fast and most people haven’t heard about it yet.

Bell’s Whole Argument Is That Recovery Is the Normal Outcome, Not the Miracle One

Bell wrote to medical students because he wants them to walk into exam rooms with a different assumption than the one his doctor had. His central data point: according to SAMHSA’s 2024 National Survey on Drug Use and Health, 31.7 million American adults have, at some point, considered themselves to have had a problem with alcohol or drugs. Of those 31.7 million, 74.3% — 23.5 million people — now consider themselves in recovery or fully recovered. Bell’s framing, almost word for word from SAMHSA’s own data: recovery isn’t a long shot. It’s the most common outcome there is.

That statistic is a direct rebuke to the doctor in his story, and to a lot of clinical training more broadly. If three out of four people who’ve had a serious problem eventually resolve it, then treating a patient’s substance use history as a life sentence — as evidence they’re fundamentally unreliable, untreatable, a bad bet — isn’t clinical caution. It’s just wrong, and it’s wrong in a way that actively makes the outcome worse, because how a doctor treats you in that appointment is itself part of what determines whether you get better.

Bell’s second point is about what actually works, and it’s where he pushes back on a narrower kind of medical thinking — the idea that a prescription is the whole treatment. “The roots of addiction,” he writes, “are like a layered web of knots — isolation, trauma, distrust — and each must be untangled by human connection.” Medication for opioid use disorder (drugs like buprenorphine that reduce cravings and withdrawal, sometimes called MOUD) works. Bell isn’t arguing against it. He’s arguing that a pill without a relationship attached to it treats the chemistry and ignores the wound — the isolation, the trauma, the reasons a person stopped trusting people in the first place. You cannot prescribe your way out of distrust. Someone has to rebuild it, one interaction at a time.

That’s why Bell spends real space on peer recovery specialists — people with their own lived experience of addiction who are trained and often certified to support others going through it. He tells physicians directly not to treat peers as “ex-patients” or hand them the scut work nobody else wants. “Their ability to build trust in your patients,” he writes, “may be the most valuable asset you have in healing SUDs.” A peer specialist can sit with a patient’s shame in a way a fifteen-minute appointment can’t. Dismissing them wastes the one relationship most likely to actually reach someone.

Dismissing them wastes the one relationship most likely to actually reach someone.

And his last instruction is the simplest, and the one that circles back to his own story: be curious. “The less you assume and more you ask, the more you will be seen as an ally.” His doctor assumed. That’s the whole essay in one sentence — a physician who assumed instead of asked, and a patient who walked out of that office with a diagnosis he could feel but nothing he could use.

If you’re reading this in recovery yourself, you already know this story before Bell finishes telling it. You’ve had the version where someone in a position to help decided, on sight, that you weren’t worth the extra two minutes. Bell’s argument isn’t really about medical school curricula. It’s about whether the people with power over your care are willing to find out who you actually are before they decide what you deserve.

The CMS Rule Draws the Same Line Bell’s Doctor Drew — Just With a Federal Seal On It

Now to the policy part of this story that’s about to have consequences for millions of people, most of whom have never read a CMS document in their life.

H.R. 1, the reconciliation bill Congress passed earlier this year, requires most Medicaid-expansion adults to log 80 hours a month of work, school, or community service to keep their coverage. The law carved out an exemption for people who are “medically frail” — a category that Congress itself said should include anyone with a substance use disorder. That was the deal states thought they were getting: an SUD diagnosis, like a disability or a serious mental illness, would exempt you.

CMS’s June 1 rule doesn’t honor that deal. According to the agency’s own fact sheet and reporting from KFF, the rule says a diagnosis alone can’t be used to prove medical frailty — states now also have to determine that the condition “significantly impairs” a person’s ability to meet the work requirement. And Georgetown University’s Center for Children and Families reports the rule goes further still for people with SUD specifically: it excludes anyone CMS considers to be in “stable recovery” for five or more years from the exemption entirely — on the theory that if you’ve made it that long, you no longer need the protection.

Sit with that for a second, because it is exactly Bell’s doctor’s logic dressed up in regulatory language. “A bump in the road” was one clinician deciding a person’s addiction wasn’t serious enough to warrant real attention. CMS’s five-year cutoff is the federal government deciding that a person’s addiction — once they’ve cleared an arbitrary number of years — is no longer serious enough to warrant protection. Both judgments run on the same false premise: that addiction is a phase you graduate out of, provable and disprovable like a broken bone that either has or hasn’t healed, instead of what the clinical and recovery research communities actually understand it to be — a chronic, relapsing condition that people manage, often successfully, for the rest of their lives, the same way someone manages diabetes or hypertension. Georgetown’s analysts ask the obvious question CMS never answers: who decides someone is in “stable recovery,” using what clinical criteria, and how does a fixed cutoff square with the reality that this is a condition people live with, not one they finish?

It gets worse on the process side. The rule tightens self-attestation — meaning your own signed statement that you have a condition — so that starting in 2028, most people can only self-attest to medical frailty one time during their entire enrollment. Get sick again, or relapse, after that one window closes, and you may need new paperwork, a new determination, a new fight, at the exact moment you have the least capacity to fight for anything. A person managing early recovery from active addiction is now expected to navigate a bureaucratic proof-of-injury process on top of everything else — proof that has to satisfy a caseworker who has never met them, working from a state’s implementation of a federal standard that even KFF’s health policy analysts say is murkier than the underlying statute intended.

Here’s why this isn’t an abstraction. Georgetown’s analysis puts the number at nearly 30% of adults enrolled in Medicaid through the expansion pathway who have a substance use disorder or a mental health condition. KFF’s research on Medicaid and the opioid epidemic found the program covers close to four in ten nonelderly adults with opioid use disorder nationally — and every state Medicaid program is required to cover buprenorphine, the medication Bell’s essay treats as one half of what actually works, alongside human connection. Losing that coverage doesn’t just mean losing a card in your wallet. It means losing access to the medication that keeps withdrawal and cravings from running your life, at the exact moment a new bureaucratic hurdle is asking you to prove, in writing, that you’re still sick enough to deserve it.

Losing that coverage doesn’t just mean losing a card in your wallet.

This Is an Old Fight With a New Form

Strip away the acronyms and CMS’s five-year cutoff is just the latest draft of a much older American argument: that there are worthy poor and unworthy poor, and addiction is the line examiners use to sort people into one pile or the other. It shows up whenever a program has to decide who deserves help, because someone, somewhere, has to write the rule that separates “sick” from “sick enough.” That job has gone badly, over and over, whenever it’s been handed to people who’ve never had to live on the other side of the form.

None of this is the first time American policy has decided some people struggling with addiction are worth helping and others aren’t. In 1996, Congress imposed a lifetime ban on cash welfare and food stamps for anyone with a felony drug conviction, sober or not, employed or not, decades later or not. The provision got two minutes of debate on the Senate floor before passing by voice vote. Senator Phil Gramm’s justification, as documented by The Sentencing Project, was blunt: people “violating the Nation’s drug laws” shouldn’t get welfare. It took nearly thirty years, state by state, before all but one finally opted out. The pattern repeats in the CMS rule: a policymaker deciding, from a distance, how much someone must prove before they’ve earned help.

The felony ban and the five-year “stable recovery” cutoff share a design flaw, not just a spirit: both assume addiction is something a person did, once, that a sufficiently long clean record can outrun — instead of a chronic condition that can require ongoing support the same way a manageable but real heart condition does. Nobody asks a person with hypertension to prove they’ve gone five consecutive years without a single elevated reading before their medication stays covered. The ask only shows up for conditions our culture has decided are moral failures dressed up as diagnoses. That’s the through-line connecting a Senate floor vote in 1996 to a Federal Register filing in 2026: different decade, different bureaucracy, same instinct to make people re-earn the help they’re already entitled to.

What’s Still True, and What You Can Actually Do About It

Nothing in this rule has taken your coverage yet. As of today, July 21, 2026, Medicaid’s substance use disorder benefits are unchanged, buprenorphine and methadone coverage rules haven’t shifted, and the work requirement itself doesn’t take legal effect in most states until January 1, 2027. The rule is still in its comment period — it closes July 31, 2026, which is the one lever ordinary people, clinicians, and advocates still have before implementation guidance hardens into practice.

If you’re a case manager, peer specialist, or clinician reading this: you have ten days. CMS is required to review public comments on interim final rules, and a comment from someone who actually treats or works alongside people with SUD carries weight that a think-tank brief doesn’t — describe, specifically, what a five-year “stable recovery” cutoff misunderstands about relapse and chronic illness, and submit it before July 31 through the docket referenced in the CMS fact sheet. Separately, start now — this week — identifying which of your patients or clients might fall into the newly narrowed medically frail category, so you’re not doing crisis documentation in December for a January 1 deadline. And read Bell’s essay before your next intake appointment. His instruction to physicians — be curious, ask instead of assume — is free, it works today, and it’s the opposite of what a five-year cutoff asks a caseworker to do.

Bell’s whole essay is a plea for individual clinicians to look at one person and ask before they judge. CMS’s rule is the inverse: it looks at 23.5 million people who’ve resolved a substance use problem and 31.7 million who’ve faced one, and it asks a caseworker to judge from a distance, on a timeline, without ever asking. Bell walked out of that exam room decades ago and became the psychologist who writes to the next generation of doctors so they don’t repeat his doctor’s mistake. The federal government doesn’t get that same do-over quietly. It gets one until July 31 — and after that, the mistake becomes policy.

It gets one until July 31 — and after that, the mistake becomes policy.