Diagnosis Just Got Even Less Rigorous

Last week's post discussed problems that arise when one particular experience of addiction is universalized to stand for all– a common problem in policy arguments. Public policy deals in population-scale solutions that approach people in categories. But not only do people have widely varying experiences of addiction; we do not even have a reliable way to determine who is addicted. Substance use that is legal and normalized (drinking, cannabis use) can become quite destructive without attracting outside attention, while even the most limited use of stigmatized illicit substances (heroin, meth) may automatically be labeled addiction, triggering high-stakes intervention. Racism and other types of bias also frequently figure into whose substance use is presumed to be a problem.

When I was poking around last year's DSM text revisions for the post on moral injury becoming a (yawn) Z-code, I found another update with much greater potential impact. What I believe is the single most important step in diagnosing substance use disorders (SUDs) was edited out last year, and from what I can tell, not even patient advocacy groups have taken note.

Sign up for A Cure For Addiction

Let's get realistic about substance use treatment.

Subscribe

Email sent! Check your inbox to complete your signup.

No spam. Unsubscribe anytime.

I'm no fan of the DSM overall, but I have found a lot of utility in the current (DSM-V) diagnostic criteria for substance use disorders (PDF, start at page 1122). We have various different definitions of addiction and questionnaires about substance use, but here's what I like about the DSM criteria:

• They are identical across substances. What constitutes a use disorder for amphetamines, for example, will also constitute an alcohol or cannabis use disorder. This means assumptions based solely on what drug is being used cannot hold up clinically.
• They make no reference to legal involvement, policing or punishment. Criminalized activities can quickly bring labels of behavioral pathology. But which substance use attracts authorities' attention– and which patients this more often happens to– are not clinical measures and cannot masquerade as symptoms.
• They are diverse and capacious. There are 11 criteria, representing a combination of what used to be called substance "abuse" and, separately, "dependence" in the DSM-IV. This allows patients to present very differently from one another with everyone still qualifying for care.
• They are subjective. The criteria do not identify any amount or frequency of use as disordered. Instead, it is the qualitative effect of substance use on one's life and relationships that is being assessed. This is a much more person-centered foundation for clinical work because it identifies where and how substance use is causing problems for this particular person, as determined by themselves.

So what is this porentous editing-out in 2025's text revision? That last bit above.

Previously, criteria or symptoms had to be "endorsed" by the patient. In practice, this means that when I do assessments with the DSM criteria, I read each one out loud and ask my patient if it applies to them or not (and talk through any ambiguities). For example: "Spending a lot of time getting the substance, using it, or recovering from its effects," or "Continuing to use even though it is making a medical or mental health issue worse."

You might assume everyone simply says "no" across the board to be done with it– but they don't. Regardless of what brought them to the diagnostic process, and even whether they had a choice in it, most patients are curious about what is officially (clinically) a substance use problem, and whether they are considered to have one. This does not mean they will agree with the finding, much less what to do about it. But being able to show them the symptoms of substance use disorder, including which ones and how many they have just identified in themselves, is often a start toward their accepting treatment. This is as opposed to strong-arming them in based on my own (or someone else's) insistence that they need to change.

But now the term "endorsed" has been removed from diagnostic instructions (see page 10), and criteria can be selected purely by the clinician's judgment and/or input from a third party. I know some clinicians were already diagnosing SUDs in this manner, either overruling patients' input on the criteria or not even bothering to ask them. But this is now considered correct procedure.

The general context, of course, is that people frequently deny having substance use problems even when others believe they do. I don't recommend treatment in this situation, because treatment by definition addresses an identified problem. A patient who doesn't believe they have a problem, accordingly, will not engage– and why would they? Harm reduction is the appropriate voluntary service for such a person. A diagnostic process that simply bypasses the patient's understanding of themselves and allows the clinician unilaterally to pronounce them addicted is a setup for even more forced treatment.

The more immediate context is a generalized rise in coercive treatment practices for people labeled as addicted and/or mentally ill. An excellent newsletter on this trend is PsychForce Report by Rob Wipond- think legal conservatorships, involuntary commitments, and forced medication. See also this earlier post on the July 2025 executive order (US) encouraging states to remove people from public spaces into indefinite "treatment," i.e. institutional confinement:

On “Dying With Their Rights On”

Does this look like someone with too many rights? The notion of people “dying with their rights on” comes from arguments for forced psychiatric hospitalization, dating back to the 1970s. As the rhetoric goes: a desperately sick person who refuses services, and doesn’t meet criteria for involuntary admission, will

A Cure For AddictionRosalie Genova

To sum up: "behavioral health" diagnostics have always been subject to bias and manipulation. Highly stigmatized diagnoses are especially sensitive in that providers and patients often disagree, yet diagnosis can trigger clinical, legal, or occupational interventions that the patient does not want. And now the greatest strength of the professional diagnostic process for SUD- patients' subjective endorsements of symptoms- has been dismantled.

This a step backward in our standard of care, reflecting a bipartisan repressive move back toward brute-forcing "treatment." Responsible, person-centered providers should ignore the DSM language change and continue relying on patient endorsements. People are going to continue being rounded up and locked away- on all kinds of pretenses- and SUD diagnostic instructions now stand to facilitate this effort. You don't have to personally contribute your signature and credentials.