An empirically supported treatment (EST) is a psychological therapy that has been tested in controlled research studies and shown to produce meaningful improvement for a specific mental health condition. The concept was formalized in the 1990s by the Society of Clinical Psychology, a division of the American Psychological Association, to help clinicians and patients distinguish therapies backed by rigorous evidence from those based only on tradition or personal preference.
The designation matters because hundreds of therapy approaches exist, and not all of them have been tested with the same level of scientific scrutiny. When a treatment earns the “empirically supported” label, it means independent researchers have put it through structured trials, compared it against control conditions, and found it produces real, measurable benefits.
How a Treatment Earns the Designation
The original framework for classifying these treatments, known as the Chambless Criteria after psychologist Dianne Chambless, established two tiers. A therapy could be labeled “well-established” if it had positive results from at least two rigorous, controlled studies conducted by different research teams. A therapy rated “probably efficacious” needed positive results from two or more controlled studies by the same research team, or at least one well-controlled study plus supporting evidence from less rigorous designs or single-case experiments.
In both cases, the studies needed to use a treatment manual, a detailed, step-by-step guide spelling out exactly what the therapist should do in each session. This requirement exists so that the therapy being tested is consistent and replicable. Without a manual, researchers can’t be sure every patient in the study received the same treatment, which makes it impossible to know what actually produced the results.
In 2015, the Society of Clinical Psychology adopted updated criteria, sometimes called the Tolin Criteria, which replaced the original Chambless framework. The newer system uses a grading method borrowed from medical research called GRADE (Grading of Recommendations Assessment, Development, and Evaluation). Instead of simply counting studies, the updated approach evaluates the overall quality of the evidence and weighs benefits against potential harms. Treatments receive a recommendation of “very strong,” “strong,” or “weak” based on several factors: how consistent the findings are across studies, whether the studies have major design flaws, and whether benefits hold up at least three months after treatment ends. A “very strong” recommendation also requires that at least one study has demonstrated the treatment works outside of a research setting, in real-world clinical practice.
What This Looks Like in Practice
Cognitive behavioral therapy (CBT) for depression is one of the best-known examples. Across dozens of trials, CBT produces a medium-sized treatment effect compared to control conditions, with larger effects when compared to no treatment at all. In practical terms, about three out of every four people who would not improve on their own will improve with CBT. The therapy performs comparably to antidepressant medication for most patients, including those with more severe depression, provided the therapist is well trained and experienced in the approach.
What sets CBT apart from medication, though, is what happens after treatment stops. Relapse rates are consistently lower for people who completed CBT than for those who discontinue antidepressants. The therapy appears to give patients lasting skills that continue to protect against future episodes, something medication alone does not provide. Combining CBT with medication produces somewhat better results than medication by itself, though the added benefit is modest.
Prolonged Exposure therapy for PTSD is another prominent example. It carries the strongest recommendation as a first-line PTSD treatment in every major clinical practice guideline worldwide, including those from the APA, the VA/Department of Defense, and the UK’s National Institute for Health and Care Excellence. Based on intent-to-treat data (which counts everyone who started treatment, not just those who finished), 53% of people who begin Prolonged Exposure no longer meet diagnostic criteria for PTSD. Among those who complete the full course, that number rises to 68%. Long-term follow-up data shows 83% of treated patients remained free of a PTSD diagnosis six years after their initial treatment.
EST vs. Evidence-Based Practice
These two terms sound interchangeable but refer to different things. An empirically supported treatment is a specific therapy for a specific disorder that has passed a defined research threshold. Evidence-based practice is a broader philosophy for making clinical decisions. It rests on three pillars: the best available research evidence (which includes ESTs), the clinician’s own expertise and judgment, and the patient’s individual characteristics, values, and preferences.
In other words, an EST is one ingredient in evidence-based practice, not the whole recipe. A clinician practicing in an evidence-based way would consider the research support for a given therapy but also factor in what they know about the patient sitting in front of them, including cultural background, personal goals, and prior treatment experiences.
Why Treatment Manuals Are Central
The requirement that empirically supported treatments follow a manual is both a strength and a source of ongoing debate. On the research side, manualization is essential. When therapists in a study follow the same protocol, researchers can be confident that the results reflect the treatment itself rather than differences in how individual therapists happened to deliver it. Studies with high treatment fidelity, meaning the therapist closely followed the protocol, are dramatically more likely to produce clear, statistically significant results. One review found that studies with high fidelity to the treatment plan were nine times more likely to find significant outcomes than those with low fidelity.
This matters for patients too. If a therapy was tested with a specific structure and sequence of techniques, deviating significantly from that structure means you may not be getting the treatment that was actually shown to work.
Common Criticisms
Not everyone in the field agrees that the EST framework captures what makes therapy effective. One recurring concern is that structured manuals might interfere with the therapeutic relationship, the trust and connection between therapist and patient that many clinicians view as a core driver of change. Surveys of therapists have revealed worries that following a manual could make them appear rigid or too agenda-driven to genuinely respond to what a client brings into the room.
Research on this question has been mixed. Some studies find no meaningful difference in the quality of the therapeutic relationship when manuals are used. Others suggest that less experienced therapists may struggle to balance following a protocol with being attuned to the person in front of them. The concern is not unfounded, but it may reflect more about training and flexibility than about manuals themselves.
A broader criticism targets the research model underlying ESTs. Controlled trials typically study people with a single, clearly defined diagnosis, but real patients often present with multiple overlapping conditions, complicated life circumstances, and backgrounds that may not match the demographics of the original study participants. Critics argue that the controlled conditions that make a study rigorous also make its results harder to generalize to everyday clinical settings. The 2015 update to the classification criteria addressed this in part by requiring real-world effectiveness data for the highest recommendation tier.
Conditions With Established Treatments
The APA has developed or is developing clinical practice guidelines for several major conditions, including PTSD, depression across different age groups, chronic musculoskeletal pain, and obesity in children and adolescents. Beyond these formal guidelines, the Society of Clinical Psychology maintains a broader listing of treatments that have been evaluated against the EST criteria, spanning anxiety disorders, obsessive-compulsive disorder, substance use disorders, eating disorders, insomnia, and more.
The landscape continues to evolve as new therapies are tested and existing ones accumulate more evidence. A treatment that currently has a “weak” recommendation could move to “strong” as additional high-quality studies are published, and treatments that were once considered well-established can be re-evaluated if newer evidence raises questions about their effectiveness or safety relative to alternatives.