A conference abstract is a 350-word summary of a study presented at a meeting; the full paper is the peer-reviewed article that follows, often much later or never. What goes missing in between is the participant flow, the subgroup tables, the adverse-event listings, and the statistical analysis plan — the detail that decides whether a headline number means anything.
For Alzheimer's specifically, the picture is better than in most fields for pivotal drug trials and worse than it looks everywhere else. The big readouts now publish in a journal the same day they are presented. The smaller studies, the biomarker analyses, and the corporate press releases are where the gap still lives.
Medical information disclaimer: This article is for general educational purposes only and does not provide medical advice, diagnosis, or treatment. Always consult a physician or other qualified health professional about symptoms, medications, tests, or treatment decisions.
Table of Contents
- What an abstract cannot physically hold
- How often the full paper never arrives
- When the numbers change between abstract and paper
- Why Alzheimer's drug trials mostly closed this gap
- The corporate topline is where the gap survives
- How to read a conference finding before the paper exists
- Frequently Asked Questions
What an abstract cannot physically hold
The Alzheimer's Association caps abstracts for its annual conference at 350 words in a fixed Background/Methods/Results/Conclusions structure, according to its AAIC 2026 abstract submission guide. At that length, a study cannot report how many people were screened, how many dropped out and why, or how the outcomes were defined before the data came in. That means the absence is structural, not evasive.
A researcher presenting honest work still cannot tell you in 350 words whether the 12% benefit came from the whole cohort or from one subgroup that happened to move. The timing compounds it. AAIC abstracts appear in an online supplement to *Alzheimer's & Dementia* months after the meeting — the association says AAIC 2026 abstracts will post by the end of December 2026. The news coverage runs in July.
How often the full paper never arrives
The Cochrane methodology review "Full publication of results initially presented in abstracts," updated in November 2018 across 425 reports, found that fewer than half of abstracted studies reached full journal publication within ten years, and about two-thirds of randomised trials did. Roughly a third of trials shown at meetings never produce a paper at all. The selection is not random.
The same review found positive results were published in full more often than non-positive ones, so the studies that stay permanently at abstract length are disproportionately the null and negative ones. For a reader following dementia research, that is the part worth internalising. The abstracts you see cited years later skew toward things that worked, and the quiet failures are underrepresented in the checkable record.
When the numbers change between abstract and paper
Even when the paper does arrive, it often does not say the same thing. A retrospective review in *Plastic and Reconstructive Surgery – Global Open* compared abstracts with their eventual manuscripts and found minor discrepancies in 91% of pairs and major ones in 76%, with results changing in 63% of the major cases and 19% showing a double-digit change in a quantitative result. Three studies reversed direction outright.
That study is surgical, not neurological, and the rates will differ by field. But the mechanism is universal: abstracts are submitted months before a meeting, often on incomplete data, and the analysis keeps moving. The practical reading is that an abstract number is provisional. Treat a percentage from a conference slide as a rough indication, not as a figure to plan care around.
Why Alzheimer's drug trials mostly closed this gap
The field's pivotal readouts now publish simultaneously. Eisai presented full Clarity AD results at CTAD on 29 November 2022 — 1,795 participants across 235 sites, 27% slowing of decline, a −0.45 difference on CDR-SB at p=0.00005 — and the company's announcement notes the *New England Journal of Medicine* paper appeared the same day. Lilly repeated the pattern with TRAILBLAZER-ALZ 2, presenting results in 1,736 participants with confirmed amyloid and tau at AAIC Amsterdam on 17 July 2023 alongside simultaneous publication in JAMA.
Two consecutive pivotal trials doing this established it as the field's expectation. So if you are reading about a major anti-amyloid trial, the full paper usually exists on day one. Look for it; it will answer the questions the press summary does not.
The corporate topline is where the gap survives
The remaining hole is the press release issued weeks or months before any presentation. Novo Nordisk announced on 24 November 2025 that its evoke and evoke+ trials of oral semaglutide missed significance on CDR-SB, and the Alzheimer's Association responded the same day that it "looks forward to hearing a deeper dive into the data" at CTAD in December — a plain statement that a topline is not evidence. The December presentation showed why that distinction matters.
Per Alzheimer Europe's account, the trials covered 3,808 adults aged 55–85 with biomarker-confirmed mild cognitive impairment or mild Alzheimer's dementia, and reported biomarker changes without clinical benefit. A dissociation like that is uninterpretable from a topline. Which biomarkers moved, by how much, and in whom are exactly the details a press release cannot carry and a full paper must.
How to read a conference finding before the paper exists
A few checks separate a result you can weigh from one you should park: None of this requires reading statistics. It requires noticing which document you are actually looking at.
- Ask whether a journal article was published the same day. For pivotal Alzheimer's trials, it usually was.
- Check the participant count and who they were — biomarker-confirmed diagnosis, disease stage, age range. An abstract has room for this much.
- Treat effect sizes as provisional until the paper lands, given how often quantitative results shift.
- Be sceptical of a company topline with no presentation date attached.
- If a finding is biomarker-only with no cognitive or functional outcome reported, wait. That is the pattern where full-paper detail changes the interpretation most.
Frequently Asked Questions
Is a conference presentation peer-reviewed?
Abstracts are screened by a scientific committee, not peer-reviewed the way a journal article is. The review that catches statistical and reporting problems happens at the paper stage.
Should I bring a conference finding to my doctor?
You can, but say where it came from. A clinician will weigh a presented abstract differently from a published trial, and that difference is the point.
Why would a negative result never get published?
Journals have historically favoured positive findings and authors invest less effort in writing up null results. Cochrane's 2018 review measured the imbalance directly in what reaches full publication.
