Journals, researchers, and even funding structures all favor positive, novel findings — a study that finds a clear effect is more publishable, more citable, and more career-advancing than one that finds no effect at all. The result is that studies showing “no significant difference” are disproportionately likely to end up unpublished, sitting in a metaphorical file drawer rather than in the literature. This is called publication bias, or the file-drawer problem, and it means the published research on any given compound tends to look more consistently positive than the full body of research that was actually conducted.

If you only ever encounter positive-sounding studies for a given compound, that isn’t necessarily because every study on it succeeded — it may partly reflect which studies made it into print. This is one more reason the evidence-hierarchy thinking from the previous lesson matters: a single positive in-vitro study sitting alongside an absence of any negative results doesn’t mean no negative results exist, it may just mean they were never published. Reading widely, checking trial registries where they exist, and staying appropriately skeptical of a uniformly positive literature are all reasonable responses to this problem.