[{"data":1,"prerenderedAt":92},["ShallowReactive",2],{"topic-research-science":3},{"articles":4,"guides":91},[5],{"id":6,"title":7,"author":8,"authorTitle":9,"body":10,"date":77,"description":78,"draft":79,"extension":80,"featured":81,"heroCredit":82,"heroImage":83,"meta":84,"navigation":81,"path":85,"readingTime":86,"seo":87,"stem":88,"topic":89,"__hash__":90},"articles\u002Farticles\u002Fwhat-we-mean-when-a-drug-works.md","What we mean when we say a drug \"works\"","Dr Alani Reyes","psychiatrist and clinical researcher",{"type":11,"value":12,"toc":69},"minimark",[13,17,25,30,33,39,42,46,49,53,66],[14,15,16],"p",{},"Every few months a headline announces that a psychedelic compound has produced\n\"unprecedented\" results for depression, addiction or trauma. The effect sizes\nare often genuinely large. They are also frequently misunderstood — by readers,\nby reporters, and sometimes by the people running the studies.",[14,18,19,20,24],{},"This is not an argument that the research is weak. Much of it is careful and\npromising. It is an argument that ",[21,22,23],"em",{},"promising"," is a specific, limited claim, and\nthat learning to hear it precisely is the most useful skill a curious person can\nbring to this field.",[26,27,29],"h2",{"id":28},"a-result-is-a-sentence-with-hidden-clauses","A result is a sentence with hidden clauses",[14,31,32],{},"When a trial reports that a single dose \"reduced depression scores by 50 per\ncent at six weeks\", several quieter facts travel underneath that sentence: how\nmany people were studied, who they were, what the comparison group received,\nhow long anyone was followed, and how many dropped out before the end.",[34,35,36],"blockquote",{},[14,37,38],{},"A number without its denominator is a rumour with a decimal point.",[14,40,41],{},"Small early-phase studies exist to detect whether an effect is plausible, not\nto measure it accurately. The honest reading of an exciting Phase 2 result is\nnot \"this works\" but \"this is worth the expense and risk of finding out\nproperly\".",[26,43,45],{"id":44},"the-unblinding-problem","The unblinding problem",[14,47,48],{},"Psychedelics present researchers with an awkward fact: people generally know\nwhether they have taken one. That breaks the blind that ordinary drug trials\nrely on, and expectation is a powerful medicine in its own right. Good teams now\ndesign around this — active placebos, independent raters, careful measurement of\nwhat participants believed they received — but a study that ignores it should be\nread with that gap in mind.",[26,50,52],{"id":51},"what-good-evidence-will-look-like","What good evidence will look like",[54,55,56,60,63],"ul",{},[57,58,59],"li",{},"Larger samples that include the people clinicians actually treat, not only\nthe unusually well and unusually motivated.",[57,61,62],{},"Longer follow-up, because durability is the whole question for a one- or\ntwo-dose treatment.",[57,64,65],{},"Honest accounting of harms, including the difficult or destabilising\nexperiences that averages tend to hide.",[14,67,68],{},"None of this is a reason for despair, and none of it is a reason for hype. The\nfield is young. The most respectful thing we can do for it — and for the people\nwaiting on it — is to describe it accurately while it grows up.",{"title":70,"searchDepth":71,"depth":71,"links":72},"",3,[73,75,76],{"id":28,"depth":74,"text":29},2,{"id":44,"depth":74,"text":45},{"id":51,"depth":74,"text":52},"2026-05-12","Psychedelic trials report dramatic numbers. Understanding what those numbers measure — and what they quietly leave out — is the first step to reading them honestly.",false,"md",true,"Higher Place — original artwork","\u002Fimg\u002Fheroes\u002Fresearch-science.svg",{},"\u002Farticles\u002Fwhat-we-mean-when-a-drug-works",9,{"title":7,"description":78},"articles\u002Fwhat-we-mean-when-a-drug-works","research-science","Tkyz07EdnWjT1kxk7cDMpvnKlOZB3vOj7Jk_bpKGxbo",[],1779224636277]