What Does DSM-5 Say About Dual Diagnosis?

When a clinician works through the DSM-5, they are not looking for a single “co-occurring” code to drop onto your chart. They are building two or more separate diagnoses, side by side. One line might read major depressive disorder. Another might read alcohol use disorder, moderate. A third might name post-traumatic stress disorder. Each one has to meet its own criteria on its own terms ¹.

That structure is why SAMHSA defines co-occurring disorders broadly: any combination of a DSM-5-TR mental disorder and a substance use disorder, with no required pairing ¹. Depression plus alcohol use counts. Bipolar disorder plus stimulant use counts. PTSD plus opioid use counts. The combinations are not pre-packaged, which means your care should not be either. Commonly associated conditions include anxiety, depression, PTSD, bipolar disorder, ADHD, psychotic disorders, and personality disorders ⁵.

If you have ever left an evaluation with only one of your conditions written down, this is where things went wrong. A thorough DSM-5 workup treats each disorder as a real diagnosis that deserves its own name, its own severity rating, and its own treatment plan. You are not a label. You are a person with two or more clinical pictures that need to be seen clearly and treated together ⁴.

If you were diagnosed before 2013, you may remember the old language: substance abuse or substance dependence. The DSM-5 retired that split. In its place is a single diagnosis — substance use disorder — rated on a continuum based on how many symptoms you meet from the DSM-5 criteria list.

• Two to three symptoms is mild.
• Four to five is moderate.
• Six or more is severe ³.

That shift matters more than it might look on paper. The old system treated “abuse” and “dependence” as different conditions, which encouraged clinicians (and insurers) to decide you were one or the other. The continuum reflects what addiction actually looks like in real life. People move along it. A moderate alcohol use disorder today can become severe after a stressful year, or ease back toward mild with sustained treatment and support.

The continuum is also a reminder that progress is real and measurable. Dropping from severe to moderate is not a failure to “fully recover.” It is a documented clinical change, and it is worth noticing. Every symptom you no longer meet is a step your treatment team can see on paper ⁸.

Here is the question that quietly shapes almost every co-occurring diagnosis: did the mental health symptoms come from the substance, or were they already there?

The DSM-5 treats this as a fork in the road. On one path is a substance-induced mental disorder — symptoms that showed up during intoxication, withdrawal, or soon after, and that will likely fade once the substance is out of the picture. On the other path is an independent mental disorder — a condition that exists on its own, with its own trajectory, that happens to be co-occurring with substance use. Same symptoms on the surface. Very different diagnoses underneath ².

Timing is how clinicians tell them apart. SAMHSA’s guidance is direct about it: some mental disorders are really substance-induced mental disorders, and an abstinent period of weeks or sometimes months may be needed before an independent co-occurring disorder can be diagnosed confidently ². That is not a bureaucratic delay. It is the clock the brain needs to show what it looks like when the substance is no longer driving the picture.

Think about what this means in practice. If you entered an evaluation while still in acute withdrawal and walked out with a diagnosis of major depressive disorder, that diagnosis is provisional at best. Your depression might be real and independent — or it might be the chemistry of withdrawal talking. A careful DSM-5 clinician will tell you which one they suspect, and when they plan to look again ⁴.

Behind the fork between substance-induced and independent sits a short list of questions a trained clinician has to answer, in order. None of them are optional, and none of them can be answered in a fifteen-minute intake.

1. When did the symptoms start, relative to the substance use? Did the hallucinations, the panic, the depression, the mood swings appear before you ever used — or only during intoxication and withdrawal? The DSM-5 is clear that symptoms arising only during or shortly after substance exposure point toward a substance-induced disorder until proven otherwise ².

2. Do the symptoms persist after you stop? This is where the one-month benchmark becomes useful. In the psychosis literature, a primary psychotic disorder is considered when symptoms predate substance use, occur independently of exposure, or persist for at least one month after intoxication or acute withdrawal has ended ¹⁰. Persistence is evidence. Fading is also evidence.

3. What does collateral history and toxicology say? A single self-report in a crisis is not enough. Family accounts, old medical records, prior psychiatric notes, and current toxicology all help the clinician see a longer timeline than the person in front of them can usually recall ¹⁰. For substance-induced psychotic disorder specifically, the DSM-5 requires evidence that the substance is etiologically related and that a primary psychotic disorder does not better explain the symptoms ⁹.

4. Have other causes been ruled out? Medical conditions, medications, head injury, sleep deprivation, and other psychiatric conditions all have to be considered before a diagnosis is locked in ².

5. Is this diagnosis provisional, and when will it be reassessed? SAMHSA’s guidance emphasizes provisional diagnosis and repeated reassessment, because intoxication and withdrawal can mimic or obscure psychiatric illness ². A thorough assessment names what it does not yet know, and sets a date to look again.

If your evaluation moved through these questions — even imperfectly — you were given a real DSM-5 workup. If it skipped past them, you were given a label.

The gap between substance-induced and independent disorders is not evenly distributed. There are specific presentations where misdiagnosis happens again and again, and knowing them can help you recognize when a second opinion is reasonable.

• Stimulant and cannabis psychosis. Methamphetamine, cocaine, and heavy cannabis use can produce hallucinations and delusions that look identical to a first episode of schizophrenia. The DSM-5 asks whether the psychosis started during or shortly after use, and whether it persists for at least a month after the substance has cleared ¹⁰. In the real world, people sometimes get a schizophrenia diagnosis in the ER before anyone has watched them through a full abstinent period. Sometimes it holds. Sometimes it quietly resolves ⁹.

• Depression in early sobriety. The first weeks after stopping alcohol or opioids can look like severe depression — low mood, anhedonia, poor sleep, hopelessness. For some people, those symptoms lift as the nervous system recalibrates. For others, the depression stays, and it becomes clear it was always there underneath. Treating only one possibility ignores the other ².

• Anxiety during withdrawal. Panic, racing thoughts, and intense worry are hallmarks of withdrawal from alcohol, benzodiazepines, and stimulants. They are also hallmarks of an independent anxiety disorder. Timing and persistence are the only way to tell the difference reliably ².

If any of this sounds like the evaluation you or your loved one received, you are not being paranoid for asking questions. Overlap between psychiatric symptoms and substance effects is exactly why NIMH stresses that accurate diagnosis is key and why comprehensive assessment by clinicians trained in both domains matters so much ⁴. A second look is not doubting your care team. It is doing what the DSM-5 itself asks of them.

The DSM-5’s method — diagnose each condition separately, decide what is substance-induced versus independent, reassess over time — only works if someone is actually looking at both pictures at once. Sequential care, where you get sober first and then address the mental illness, or treat the depression first and then “deal with” the drinking, breaks that method before it starts.

Here is why that matters for complex cases specifically. If you have treatment-resistant schizophrenia alongside stimulant use, a program that detoxes you and then refers you out for psychiatric care is making a bet that your psychosis will stay stable without antipsychotic adjustments during the exact weeks your brain is most unstable. If you have rapid-cycling bipolar disorder with alcohol use, a mental health clinic that stabilizes mood without touching the drinking is treating around the condition that is destabilizing the mood in the first place. If trauma is driving opioid use, treating the opioid use disorder without the trauma leaves the engine running ¹¹.

Each handoff between systems is also a handoff of information — and information tends to get lost. The timeline a careful DSM-5 workup depends on, the collateral history, the toxicology, the reassessment plan — those pieces rarely travel intact between a detox unit, an outpatient psychiatrist, and a therapist who were never in the same room ^{3, 8}.

“Integrated treatment” gets used loosely in marketing. The DSM-5 gives it a more specific meaning. If a diagnosis depends on watching symptoms across substance use, intoxication, withdrawal, and abstinence — and on reassessing provisional diagnoses as the picture clarifies — then the people doing the watching have to share a chart, a treatment plan, and a conversation. That is what integrated means in a clinical sense: one team, both conditions, from the first day ⁸.

In practice that looks like a psychiatrist and a therapist who talk about the same person each week, medication decisions that account for withdrawal timelines and psychiatric targets at the same time, and a treatment plan that names each DSM-5 diagnosis with its severity rating and a reassessment date. SAMHSA’s TIP 42 is built around this coordination for exactly this reason — co-occurring disorders are the audience it was written for ⁸.

Honest caveat: the research on what works best in these populations is still catching up. The 2024 review of the field describes a multimodal approach — contingency management, CBT, DBT, motivational interviewing, family interventions — as the practical foundation, while noting that pharmacologic evidence in people with active co-occurring SUD is often limited by small samples and trials that excluded them in the first place ¹¹. Integrated care is not a guarantee of a clean outcome. It is the setting where DSM-5 diagnosis can actually be done well, and where the next adjustment to your treatment plan is a shared decision rather than a referral. That is the standard worth asking for.