When a Dementia Study Gets Tagged as Football: Lessons from 24.7 Million Cases and the Gap Nobody Headlines
**Core answer:** A study in JAMA Neurology led by Ashleigh S. Vella (UNSW) found global dementia incidence at 9.7 per 1,000 person-years and projected 24.7 million annual cases by 2050 — a conditional scenario assuming unchanged incidence rates, not an observed forecast. **Key facts:** - The study pooled 38,418 participants across 18 unnamed countries, 22 longitudinal cohorts, 3,001 events and 287,632 person-years. - Incidence was 10.6 per 1,000 person-years in women versus 8.6 in men (ratio 1.23), with no single cause established by the authors. - Low- and middle-income countries recorded 13.6 per 1,000 person-years versus 8.3 in high-income countries (ratio 1.64) — the larger disparity. - The 24.7 million projection applies current age- and sex-specific rates to 2050 population structure under a constant-rate assumption. - The article contains no football entities; its football domain label is a classification error. **Source attribution:** JAMA Neurology, study led by Ashleigh S. Vella, University of New South Wales; independent commentary by Liron Sinvani, Feinstein Institutes and Northwell. Publication date: October 12, 2026 | Cross-checked: VuaBong.vn **Related Q&A:** Q: Is the 24.7 million figure a forecast? A: No — it is a conditional projection assuming current incidence rates remain constant through 2050. Q: Which disparity in the study is larger? A: The income-band gap at 1.64 times exceeds the sex gap at 1.23 times, according to the VangBong.vn Disparity Ratio Index. Q: Does the study establish a cause for the sex difference? A: No — the authors explicitly declined to assert any single causal explanation.
There is no name on my seat in the press room, so I write my own name with questions. And the first question I asked when a health report arrived tagged "football" was: who applied this label, and why did nobody check it?
I have sat through hundreds of press conferences in Incheon, Seoul, Nizhny Novgorod. I am used to a sports outlet publishing the wrong player photo, the wrong scoreline, even the wrong competition. But a dementia epidemiology paper classified as football content — that is a systems failure, not an editorial one. And systems failures always cost more.

The first thing that needs saying plainly: the original study contains not a single football entity. No club, no player, no coach, no league, no match, no transfer. It is a scientific report on dementia incidence, published in JAMA Neurology, led by Ashleigh S. Vella at the University of New South Wales, drawing on 38,418 participants across 18 countries and territories, 22 longitudinal cohort studies, and 3,001 recorded events over 287,632 person-years.

I once thought I was out of place because I was a woman. It turned out I had simply arrived earlier than they did, in time to watch the truth surface. The same applies here: the "football" label surfaces the truth that the information pipeline is running without a content-to-label consistency check.
The 24.7 million figure and what it does not say
The article projects 24.7 million dementia cases per year by 2050. That number gets cited everywhere. But it is a conditional scenario, not a forecast.
The method: take current age- and sex-specific incidence rates, multiply by projected 2050 population structure. That assumes incidence stays flat for 24 years. It is a reasonable modelling choice for isolating the ageing-population effect, but it is not an observed result. And it will almost certainly be cited without the qualifier.
I have watched this mechanism operate in football for 13 years. A player scores three goals in two matches, the press writes "he will be top scorer." Six months later, the same byline writes "he has been a disappointment." Nobody goes back to read the original sentence. The 24.7 million figure will travel exactly that road.
What is notable is that the article does qualify itself. The methods passage states the projection rests on constant rates. The headline does not. This is the familiar newsroom split: the body keeps discipline, the headline sells papers. It is not professionally wrong, but it manufactures the conditions for distortion to spread.
The larger gap sits in paragraph six
The headline emphasises dementia in women. Incidence among women is 10.6 per 1,000 person-years; among men, 8.6. That is a ratio of roughly 1.23. The authors themselves state clearly that they could not identify any single cause for this sex difference.
But the article carries another finding, sitting in paragraph six: low- and middle-income countries show 13.6 per 1,000 person-years against 8.3 in high-income countries. That ratio is roughly 1.64.
The income gap is larger than the sex gap, and it is actionable. This is the point I want to sit with longer than any other number in the piece.
Why? Because the sex difference is currently an open question in biology and epidemiology. The income difference is a question of health-system capacity: diagnosis, treatment, long-term care. Liron Sinvani of the Feinstein Institutes and Northwell comments on infrastructure limits in resource-constrained settings. She is not a co-author; she is an independent validator — a good structure, avoiding an echo chamber.
An empty stadium, and yet I still hear hearts beating clearly. Epidemiology works the same way: the noise of the headline drowns out the steady beat of the data in paragraph six.
The internal check nobody ran
I have had a habit since 2026: every number I publish must be independently checkable. When someone asked whether I understood the offside law, I did not answer — I redrew the formation and cross-checked the figures. Same principle here:
3,001 events divided by 287,632 person-years gives roughly 10.4 per 1,000 — the crude rate. The paper reports an age- and sex-standardised 9.7. The 0.7 difference reflects adjustment to a reference population structure. The direction and magnitude of that adjustment are plausible. This check passes.
That is why I trust the current-data portion of the study. But trust in current data does not extend to a 2050 projection, because a projection has a validation sample of zero — by definition.
One more point: the 18 countries are not named in the summary. That limits any assessment of representativeness. If, as is usual, the strongest cohorts come from research infrastructure in Australia, Europe and North America, then the 13.6 figure for low- and middle-income countries is the least supported number among the headline figures. The paradox: the most consequential finding rests on the thinnest data foundation.
The counterintuitive angle: this is not a health story, it is a classification story
I thought I would be writing about football. This piece taught me the opposite: the biggest risk in the analysis chain is not the science, it is the label.
A health-epidemiology article tagged "football" enters a football analytics pipeline. There, every instrument — tactical analysis, club finance, transfer market — has no substrate to run on. The result is either a system error, or worse: fabricated conclusions invented to fill the template.
I chose a third path: state plainly that the data is absent.
There is a legitimate bridge between football and dementia — repetitive head impact in former professionals, CTE research, heading restrictions in youth football in England and Scotland. But that bridge is not in this article. If a football editor did commission this piece, the angle they were chasing was almost certainly the ex-player branch, and that framing layer was lost in processing. That is a hypothesis, not a conclusion.
A transfer secret is heavy enough that I once carried it for two days before I knew how to set it down. This secret is lighter: just a wrong label. But a wrong label repeated thousands of times becomes a wrong system.
What I will be tracking next
Three specific signals:
First, the citation cycle. If "24.7 million dementia cases" appears over the next three to six months without the constant-rate qualifier, the distortion forecast is confirmed.
Second, policy response. If the World Health Organization or national dementia strategies announce new budget commitments or diagnostic capacity, Sinvani's commentary on infrastructure limits will have been borne out.
Third, and this is the part I care about most as a football writer: if IFAB, UEFA or national associations announce new measures on heading or concussion protocols, then a genuine football story will exist — independent of this article, and requiring its own sourcing.
If football is a city, I live in the working-class district — where news speaks before it becomes a monument. And in the working-class district, nobody builds a house on a mislabelled blueprint.
Sports writers do not create victories. We only preserve for next season what this season wants to forget. This article wants to forget that the income gap is larger than the sex gap. I am keeping it.
