Terrible decision on the face of it.

Read this.

“Urgent health warning as most contagious disease on earth is detected at Mall of America” . sure sounds scarier than . “Child visiting shopping mall tests positive for measles”.

This kid could go far in EBM. Ask questions. Test hypotheses.

P = 0.06 . ". Utility company investigated for sparking the worse fire in Los Angeles opted against de-energizing their transmission lines because the strength of the wind fell just shy of the threshold at 59mph at Edison's Eaton Canyon monitoring site - just 1 mph short".

Assisted dying and predicting end of life. A short commentary.

The discussion on prostate cancer screening followed the usual pattern of being led by the advocate whose life (he claims) was saved by screening. Of course, the opposing viewpoint (of the disadvocate) is never given as they’re not around to give it.

4 new prisons a year. The number required to keep up with “demand”, according to Shabana Mahmood on R4 this morning.

Welsh times table.

Work led by a very talented and hard working MSc student. Glad to be a part of this work.

Not wishing to detract from the gravity of the situation but I was wondering why the yacht was apparently called Bayesian?

micro stats #8 . Ventile: . Any of the nineteen points that divide an ordered distribution into twenty parts, each containing one twentieth of the population.

According to R4 this evening violent crime rates has never been lower. Anyone believe that? . Where’s the figures?.

According to the National Audit Office, spending on the pandemic totalled £376 billion. 16 times the cost of bailing out the banks in 2007-09.

Lecturing about screening, when the fire alarm goes off and we all evacuate the building - . it was a false alarm - . A teaching opportunity presents itself.

Dutifully sharing this; . AI can help radiologists detect pneumothoraces.

On R4 this morning. “The growing good of the world is partly dependent on unhistoric acts; and that things are not so ill with you and me as they might have been, is half owing to the number who lived faithfully a hidden life, and rest in unvisited tombs.”. –  Eliot, G.

Non-informative censoring ;. occurs if the survival times provide no information about the censorship.

Random: . subjects who are censored at.time t should be representative of all who remained at risk with respect to their survival experience. Independent: . is random censoring conditional on each level of covariates.

micro stats #7 . Censoring:. Non-informative, independent and random. Confident you know the difference?.