Approved medications now cover most of the major substance use disorders, but the strength of evidence varies sharply by indication
A decade-spanning umbrella review has pulled together the best available evidence on approved pharmacologic treatment for substance use disorders, and the picture is broad rather than uniform. Across alcohol, opioid, tobacco, sedative-hypnotic, and overdose-related care, the review identified 41 approved medications represented by 25 distinct molecules, along with four completed phase 3 randomized trials with posted results. The dataset is large enough to matter at the bedside: 371 randomized trials, 62 observational studies, and more than 360,000 combined participants and patients across the reviewed evidence base.
The practical value of this synthesis lies in how clearly it separates disorders with mature medication options from those where treatment remains thin. For alcohol use disorder, opioid use disorder, and tobacco use disorder, the therapeutic arsenal is no longer hypothetical. For stimulant use disorder, the pipeline is still small enough that even one completed phase 3 signal deserves careful attention.
Alcohol use disorder shows consistent benefit, but the effect size remains modest
In alcohol use disorder, 86 randomized trials involving 18,643 participants examined seven medications, along with three studies of alcohol withdrawal or intoxication. The main signal was a reduction in the risk of returning to any drinking, with relative risk reductions ranging from 7% to 17% versus placebo. That is a clinically meaningful direction of effect, especially in a disorder where relapse is common and sustained abstinence is difficult to maintain, but it is not a dramatic cure signal.
The medications act through different neurobiologic pathways, which helps explain why the class is not interchangeable. Some target reward circuitry, others dampen craving, and some blunt the reinforcing pull of alcohol exposure. In practice, the modest effect size matters less than the fact that several approved options now offer a pharmacologic foothold for patients who are ready to reduce heavy drinking or sustain abstinence after detoxification.
A single percentage point can change the trajectory of repeated relapse prevention when the alternative is no medication at all.
Opioid use disorder remains the clearest example of medication-driven recovery support
The evidence base for opioid use disorder was smaller than the alcohol and tobacco literature but clinically powerful. Forty-nine randomized trials with 7,804 participants, plus 21 observational studies involving 155,111 patients, covered 12 medications. The highest abstinence rates ranged from 41% to 43% compared with 5% in placebo. Those numbers describe a therapeutic gap that is hard to ignore: pharmacologic treatment does not merely trim symptoms, it can separate sustained abstinence from near-complete failure in placebo-controlled settings.
The mechanism is straightforward and clinically important. Opioid agonist or partial agonist strategies stabilize receptor activity, reduce withdrawal, suppress craving, and lower the volatility that drives return to use. In overdose-prone populations, that stabilization often functions as a bridge from repeated crisis to long-term engagement. The observational evidence also fits what clinicians see in routine care: when treatment is retained, outcomes improve, and survival can improve with it.
Overdose reversal has the strongest immediate survival signal in the review
Forty-one observational studies involving 74,114 patients evaluated nine medications for opioid overdose reversal. Survival ranged from 92% to 98%. In a clinical setting where minutes determine outcome, a survival range in that band is not a statistical curiosity; it is the difference between a rescue intervention and a fatal event. The review also included one medication for opioid withdrawal, reinforcing how much of the opioid literature still centers on acute stabilization rather than long-term remission alone.
The biology here is direct. Overdose reversal works by rapidly countering opioid receptor-mediated respiratory suppression. When administration is timely, it restores ventilation before hypoxic injury becomes irreversible. The data make clear why overdose reversal sits apart from most other addiction treatments: the endpoint is immediate survival, not only reduced use over months or years.
Tobacco cessation remains one of the most mature pharmacotherapy areas in addiction medicine
The tobacco literature was the largest in the review. Two hundred twenty-three randomized trials involving 104,524 participants assessed eight medications for tobacco use disorder. These treatments produced 1.3 to 2.3 times greater likelihood of abstinence compared with placebo. Among substance use conditions, tobacco has one of the deepest and most reproducible pharmacotherapy pipelines, and this review confirms that the effect is not fragile.
Here, too, the clinical logic is grounded in neurobiology. Nicotine replacement and other cessation agents reduce withdrawal, attenuate cue-driven craving, and blunt the reinforcing cycle that keeps smoking in place. The magnitude of benefit may look smaller than the opioid abstinence figures, but tobacco cessation trials are notoriously difficult, and the observed relative gains are meaningful when translated into population health. In a disorder with high cumulative morbidity, even incremental increases in abstinence compound over time.
Sedative-hypnotic overdose reversal remains narrowly represented
For sedative-hypnotic or anxiolytic use disorder, the review identified 13 randomized trials with 994 participants covering one medication used for benzodiazepine overdose reversal. The evidence base is far thinner than the one for alcohol, opioids, or tobacco. That thinness itself is clinically revealing. It signals how unevenly pharmacologic progress has reached the major addiction categories and how much of routine practice still depends on supportive care, careful tapering, and risk mitigation rather than a broad medication menu.
Phase 3 pipeline data point to future expansion, not immediate replacement
Four completed phase 3 randomized trials with posted results were identified across the treatment pipeline, involving 1,026 participants. Three trials, enrolling 623 participants, examined three medications for alcohol use disorder. One trial, with 403 participants, evaluated a combination medication treatment for stimulant use disorder. The presence of these trials matters because it shows the field is still moving beyond the current approved set, but the list remains short relative to the burden of disease.
Stimulant use disorder remains the most obvious gap in the approved-treatment framework. Unlike alcohol or opioid use disorders, where multiple agents already have demonstrable benefit, stimulant pharmacotherapy is still working toward a durable standard of care. A single phase 3 program cannot close that gap, but it does mark a direction of travel.
Methodologic rigor supports the overall signal, while heterogeneity limits direct comparison
This review was registered in PROSPERO and followed PRISMA methods. Investigators searched PubMed, MEDLINE, and the Cochrane Library for systematic reviews published from 1 January 2016 to 1 January 2026, and also searched trial registries for completed phase 3 medications with posted results. Systematic reviews were appraised with ROBIS and AMSTAR-2, and phase 3 trials with RoB2. Two independent reviewers handled study selection and resolved disagreements by discussion and third-reviewer consensus.
The review’s structure strengthens confidence in the overall conclusions, but the evidence is not perfectly comparable across disorders. Different endpoints, different populations, and different intervention classes make direct cross-condition ranking difficult. Even so, the clinical hierarchy is hard to miss: tobacco, opioid, and alcohol use disorders already have approved medications with measurable benefit, overdose reversal saves lives acutely, and stimulant treatment remains the area where the pipeline is still trying to catch up with need.