Primary literature
researchThe original published reports of research — the papers in which a study's own methods and results first appear — as distinct from reviews, textbooks, and databases that summarise them. Reference values are traced back to it wherever possible, and where a value cannot be located there, the absence is recorded rather than filled in from a secondary account.
Primary literature consists of the original journal articles, study reports, and conference papers in which researchers first describe what they did and what they found — before any review article, textbook, or database has summarised or reframed the work.
It sits at the source layer of the evidence hierarchy. Reviews and meta-analyses draw on it; textbooks distil it; databases index it. When a dose range, a mechanism claim, or an interaction severity appears on these pages, the provenance trail leads back to a primary paper. Where it cannot be traced there, that gap is recorded rather than filled in from a secondary account that may have compressed, averaged, or shifted the original finding.
What this design can establish
A primary study can establish what its authors measured, in their specific sample, under their specific conditions. A pharmacokinetic paper documents how a compound behaved in the participants or animals studied; a clinical trial documents whether an outcome differed between groups; a case report documents that a specific event occurred in one patient.
Because methods, participant details, and raw results are reported directly, a careful reader can assess what the study actually asked — and whether its conclusions follow from what was done. That transparency is the defining feature of primary literature, and it is why a well-conducted review cites primary papers rather than other reviews: the authority travels through the chain back to the source.
What it cannot
Reading a primary study establishes what happened in that study. It does not establish that the finding generalises — to different populations, doses, routes, durations, or combinations. A single positive result is not a settled claim; a single negative result does not close a question.
Primary literature is also not a quality guarantee. Peer review catches some errors but not all. Small samples, uncontrolled conditions, undisclosed conflicts of interest, and publication bias — where null or negative results go unpublished — can all distort the picture without disqualifying a paper from the record. The distinction a reader needs is between this was published and this is established: primary literature is the former; whether it is also the latter requires reading what was actually done.
AI-generated · not yet verified by a human reviewer
Harm-reduction reference — not medical advice.