Robin Blythe

What can the Clinical Frailty Scale tell us about planning for death? By Dr Robin Blythe and Prof. Adrian Barnett. Blog post for palliAGED and CareSearch, the Palliative Care network.

Contributed a blog to CareSearch/palliAGED on our external validation of the #clinicalfrailtyscale. The CFS is very widely used but we showed limited prognostic value. Better to think of it as a cue for discussing ageing and death with older pts than as a marker to assign treatment
#statsky #ageing

Maybe this is a tangent, but I see these tools as highly useful because they lead to better science, not because they replace peer review. I genuinely do not know what the purpose of peer review is, or perhaps, is meant to be.

The dumbest part of 'back to the land' fantasy and cottagecore is people forgetting that subsistence farming basically gave you a pretty high chance of starving to death every year

I reckon the hard part of that job doesn't manifest until you're in your 50s and basically struggling to get out of bed every morning, but you have to or else the work just doesn't happen, you don't get paid, and everyone's miserable.

This is where I am at the moment (supervising research scientists rather than students, but still) and articulating your request is hard. You have to broach the topic somewhat carefully or it looks like you're accusing them of not putting in the work. Which... maybe they aren't.

Something I see as emblematic of their system - they asked me to look through a list of hundreds of editors and asked me to find one who understood my paper. That process took 3 months.

There's some ok stuff in SR but I kind of agree. I find it hard to look at it as anything other than a glorified medrXiv

Haha you don't need to tell me, that was my first analyst job, for a magazine distributor. OLS in Excel, but manual tweaks to regularise. Model must have been 10 years old and untouched for fear of breaking it.

I know you were joking, but you never know how the qual people will take it. You might say they're a reflexive bunch

I have been tossing up whether to do a course, but think I prefer the self-directed stuff, so these are great!

Bold to assume they made this chart using actual coding

Got it. I'll have to read in more depth to understand, as I'm myself doing a DCE for the first time only now. It's surprisingly fun

Worth asking a qual scientist but I think there actually is a big push for transparency wrt explicitly linking themes with the (qualitative) data! In fact I think many qual researchers are more aware of researcher degrees of freedom issues than quant people are -- so your decree still holds

Data, sometimes I understand (privacy, IRB etc). But refusing to share code is inexcusable and I've recommended rejection outright for it multiple times when the study has hinged on the model.

I was thinking more along the lines of potential political attitudes towards govt vs third parties. But I take your point that the 3 countries probably have socioeconomic drivers that are consistent across each country

No just in Documents. I do import some data files from Dropbox at the start of each script (institutional requirement) but I do this on my home laptop too. Maybe windows is automatically treating Documents as OneDrive? Something I should check.

One thing the Stats NZ bureau are not are hypocrites, so I thankfully haven't yet had to listen to that particular brand of cognitive dissonance

Good idea, will give it a try!

Congrats! How did you end up deciding on addressing potential heterogeneity between your 3 countries in the sample? Grappling with a similar issue myself for distinct ethnic heterogeneity in a DCE.

We tried to get a local mirror and were politely told where we could stick it

In the NZ integrated data infrastructure environment, you need an individual ticket for each package INCLUDING dependencies. Tidyverse install for a single project could be weeks of admin for you and some poor sysadmin, lol

Sometimes the case -- but the issue is actually worst without a VPN at all, just on my office wifi.

How do you mean? I don't use a proxy when I'm connected to office internet.

Good idea for the former. I've asked about the latter before (mostly because I want a machine with more RAM and more threads) and didn't get a satisfactory answer. Prayer would have helped!

Out of curiosity, why? For your own sanity? To make them reconsider being rude?

Anyone work within annoying firewalls at work? RStudio usually takes 3-4 minutes to open, and it can take 30-40 minutes to download all the CRAN hosted packages I need to start a project. On the same network with my personal machine, the whole process is over in around 90 seconds. #rstats

I bet this will be interpreted by some racists as proof that migration is out of control -- "they don't even put English on the signs anymore!!1"

I lost around 15kg between end of my PhD and starting new job (due to the magic of stress) and I am legit swimming in my old clothes. I spent a few weeks looking like a street orphan fallen on hard times. 4 extra holes in my belt 😂

On slowing down A blog about statistics, meta-research, metascience and academia

Some notes on slow science medianwatch.netlify.app/post/slow_sc...

In fact the whole paper is hilarious