Adrian Esterman | Epidemiology @profesterman

This is an intriguing body of evidence. Multiple observational studies report associations between routine vaccination and lower dementia risk. Whether that's a direct biological effect or reflects other differences between vaccinated and unvaccinated people remains uncertain. As always, association ≠ causation. But when consistent associations are seen across multiple vaccines and populations, they deserve serious investigation.

Worth remembering that healthcare-associated COVID is still a patient safety issue. Victorian data found at least 6,212 patients acquired COVID in hospital during 2022–23, with 586 subsequent deaths. Infection prevention shouldn't be something we only think about during pandemics.

Denis - The COVID info guy - @BigBadDenis

a day ago

"Hundreds of patients died after catching COVID in Victorian hospitals, new data shows" 6 May 2024 At least 6,212 patients caught COVID in hospital in 24 months (2022-2023) and 586 died — that's about 60 infections and 6 deaths per week, on average. abc.net.au/news/2024-05-0…

5 148 305 6K 51

Six cancer patients relapsed, and all of them had a COVID booster." Sounds damning. Now try: "Six people crashed their cars today, and all of them had eaten breakfast." Also true, and also proves nothing, because nearly everyone eats breakfast. And nearly everyone got the booster.

And I am one of them Staff weren't even wearing masks in the transplant unit.

Tricia Dearborn @TriciaDearborn

3 days ago

"medically vulnerable patients reported having to ask hospital staff to wear masks. 'We'd never expect people to have to individually ask their surgeon to wash their hands, we just expect that high standards are followed when we go into a health facility, that you will be safe'"

1 9 13 1K 1

Check out this article: cnn.com/2026/06/24/wor…

It had to happen eventually. It is a major concern, not just for our poultry industry, but also for wild birds and marsupials. indailysa.com.au/news/just-in/2…

@RealAnitaWhite @7NewsSydney The main concern is elderly and vulnerable people, where COVID booster uptake is miserable, and flu uptake is going down.

Excellent article about the current Ebola outbreak.

Atul Gawande @Atul_Gawande

3 weeks ago

Speed of response is everything, and it's still lagging far behind. Essential explainer on the Congo Ebola outbreak from @JeremyKonyndyk. nytimes.com/interactive/20…

6 54 139 99K 16

Latest estimate from the USA, 16% of adults infected with COVID developing Long COVID. However, that is from hospital data, not a population sample: an older (average age >50), sicker cohort. Direct surveys put current cases nearer 1 in 10 of those infected. jamanetwork.com/journals/jaman…

Friday methods post A statistically significant result is not necessarily an important one. With a large enough sample size, even tiny differences can produce very small p-values. That's why it's important to consider: • Effect sizes • Confidence intervals • Clinical or public health relevance • Potential sources of bias The key question isn't "Is the effect different from zero?" but "Is the effect large enough to matter?" Statistical significance tells us something happened. Effect size tells us whether anyone should care!

This is great news. Cambridge Uni group have successfully completed a Phase I trial of a new pan-sarbeco virus family vaccine, which includes SARS-CoV-2. Now recruiting for Phase II. New AI-designed vaccine ‘could protect against whole families of viruses’ edp24.co.uk/news/national/…

@anchinya Agreed, fair point. hMPV is one to watch. I focused on the three respiratory viruses with vaccines and solid ABS data, but you've flagged a genuine gap. Might be worth its own article!

Is flu now deadlier than COVID? Read my new article. theconversation.com/how-flu-overto…

Good points. The article focused on mortality data and why influenza is currently causing more deaths than COVID. Long COVID and the possible effects of repeated COVID infections on the immune and vascular systems are important concerns, but they're separate questions from the mortality trends discussed in the piece. The evidence that these factors are driving this year's flu season is still evolving.

It's an empirical estimate from outbreak investigations rather than a particular modelling framework. Some studies use mean-field approaches, some include spatial structure. The main point is that Ebola transmission depends heavily on close-contact behaviours (especially caregiving and funerals), which are much less common and more controllable in developed-country settings.

An Ebola case travelling internationally is always taken seriously, but it's important to keep the risk in perspective. Ebola is not spread like COVID, influenza or measles. Transmission generally requires close contact with infected bodily fluids, which is why rapid case identification and contact tracing are so effective. The key question is not whether someone travelled, but whether public health authorities can identify and monitor contacts quickly.

Two viruses are in the headlines: Ebola in the DRC, hantavirus on a cruise ship. Here's what almost nobody mentions: about 8% of your DNA was put there by viruses. One captured viral gene is the reason mammals can grow a placenta. #Ebola New piece: medium.com/@adrian.esterm…

@DrNeilStone Ebola is one of those diseases that reminds us how important strong public health systems are. The virus itself is terrifying enough. Add delayed detection, conflict, mistrust and weak healthcare infrastructure, and outbreaks can become extraordinarily difficult to control.

I have just been asked by a journalist if the general public should be worried about another pandemic. I told him that we should be more concerned about the drop in vaccine coverage rates. We can't spend our lives like hermits, worried sick about another pandemic.