Why this page exists
People reading about a medication deserve to know what kind of evidence sits behind each statement. This page sets out the evidence hierarchy PsychiatryRx uses when writing and reviewing content, and explains what it means when a page says something is "well established" versus "preliminary."
The hierarchy we use
When the medical editor reviews a page, sources are weighed roughly in this order, with stronger evidence preferred over weaker:
- Meta-analyses. Statistical syntheses of multiple randomized trials. These give the most reliable picture of the average effect of a medication.
- Systematic reviews. Structured reviews of the available evidence on a question, using transparent inclusion criteria.
- Clinical practice guidelines. Recommendations from recognized bodies, such as the American Psychiatric Association, the U.S. Preventive Services Task Force, and the National Institute for Health and Care Excellence.
- Randomized controlled trials. Individual trials that randomize patients between treatments. Strong, but a single trial rarely settles a question on its own.
- Cohort and observational studies. Useful for real-world effectiveness, safety signals, and questions that cannot ethically be randomized.
- Expert opinion and clinical experience. Used to interpret the evidence above, and to fill gaps where good evidence does not yet exist. Identified as such when used.
The regulatory sources we treat as primary
Alongside the research hierarchy, PsychiatryRx relies on a small set of primary regulatory and government sources for medication-specific facts:
- U.S. Food and Drug Administration prescribing information, including boxed warnings, indications, dosing, and safety information.
- MedlinePlus, from the U.S. National Library of Medicine.
- The National Institute of Mental Health.
When the FDA label and an outside source disagree on a U.S.-specific fact, such as an approved indication or a warning, the label wins.
How uncertainty is described
Where the evidence is strong, pages say so plainly. Where it is mixed or preliminary, pages say that too. Phrases like "there is good evidence that," "studies are mixed," and "this is not well established" are used deliberately to signal where a statement sits in the hierarchy above.
What evidence can't do
Even the best evidence describes averages across groups of people. It can't tell you how a specific medication will work for you, how you'll tolerate it, or whether it's the right choice given your other conditions and medications. That's what your prescriber is for. Pages on PsychiatryRx are written to help you have a better conversation with that prescriber, not to replace it.
Recommendations may change
Medicine moves. Guidelines are revised, new safety signals appear, and what looked like the right answer five years ago may not look the same today. We update pages as evidence changes, and the last-reviewed date on each page shows when it was last checked against current sources.