@AliceSitch

Statistician, Test and Prediction Research Team at University of Birmingham

 Birmingham, England
Joined March 2009
  • Tweets90
  • Following172
  • Followers160
15Media
This is my view on P-values and multiplicity pubmed.ncbi.nlm.nih.gov/2845…
10
30
9
145
🐐🙌
Lewis Hamilton
@LewisHamilton5 Feb 2022
I’ve been gone. Now I’m back!
💔
3
Massive thanks to @Tashamaguire72 @FerranteLavinia and Simon for organising the BESaTE virtual Christmas event. @BarneyUoB needs to listen to more music in preparation for next year!
3
6
>330 healthcare workers are participating in @CopeWestMids🎉. Thanks to @Luce0903Lucy @BWC_NHS @__Barley @wmasjoshmiller @OFFICIALWMAS @annabelgrins @UHBResearch @uhbtrust for their hard work! Interested in our study? Find out more: birmingham.ac.uk/cope-study
2
2
3
5
Calling ALL employees working for @OFFICIALWMAS @uhbtrust and @BWC_NHS. We are exploring the impact of #COVID19 in healthcare settings. If you have had a PCR test, join our study! @Peymane_Adab @DrKiranRai @AlexRichter3 @e_sapey
6
5
9
We’ve started recruiting at one of our sites. If you work @uhbtrust and have had a swab test for COVID-19, we want to hear from you! Visit birmingham.ac.uk/COPE-study to find out more @Peymane_Adab @e_sapey @AlexRichter3 @DrKiranRai
7
2
6
✍️| 'Too many corners are being cut in the race to find a Covid-19 antibody test' says Professor Jon Deeks (@deeksj) in a new article for @guardian⬇️ theguardian.com/commentisfre…

Too many corners are being cut in the race to find a Covid-19 antibody test | Jon Deeks

Despite optimistic coverage, the results from a new test seem too good to be true, says professor of biostatistics Jon Deeks

theguardian.com
5
7
I recently asked the followers of my academic Instagram account (instagram.com/psyconogy/) what their number one piece of advice would be to future PhD students. This is what they said... Please feel free to RT and share with any PhD students who are starting soon!
1
19
9
34
😐
Richard Zybert@richardzybert3 May 2020
Shapps says we have best statisticians and that is why our death rate is so high. I am trying to understand - do statisticians kill people?
2
An appeal to Biomarker companies in the Oxford COVID Serology study doi.org/10.1101/2020.04.15.2… For the benefit of all, please tell us if your test was evaluated in this study, and which number test it is. It is the morally right thing to do. Please RT to get to the companies
53
4
50
3/3 Anonymity agreements can only create publication bias, and reduce the value of the study. The COVID-19 epidemic must be the time to stop tolerating such corporate protectionism. Lives are at stake.
2
9
1
39
2/3 ... the authors can’t tell us what the 9 tests were as manufacturers insisted on anonymity. How should we interpret the results? What about the other 100+ tests? Study was funded by NIHR, MRC and UK Government. When are such corporate anonymity restrictions unethical?
1
5
1
28
1/3 The UK Oxford COVID-19 serology study now available here doi.org/10.1101/2020.04.15.2… - 9 serology tests assessed in 31-40 COVID-19 patients and 60-142 controls. Well done study to the research team BUT (sorry ... ) ... #COVIDー19 #nihr @NDMOxford
2
23
5
59
Please RT & shout it out loud: If you are worried your baby or child is sick then the #nhs is here to help you now: to diagnose and give appropriate treatment. This may just be reasurance BUT it may also be life saving care. A pandemic doesn’t stop every other disease #pedsicu
This Post is from an account that no longer exists. Learn more
2
28
3
36
The evidence is *woeful* for AI apps that "screen for skin cancer". And yet we give them CE marks and money. This is risky. It risks misleading the public and professionals. It risks rewarding failure, and thereby failing to incentivise true innovation. Our editorial in BMJ
The BMJ
@bmj_latest11 Feb 2020
A new study throws general concerns about the use of data analytics and machine learning in healthcare into stark relief with its close examination of the evidence on diagnostic apps for skin cancer. @jessRmorley @Floridi @bengoldacre look at what it found ow.ly/SQUZ50yiqyx
3
57
4
102
We’ve produced some guidance to help design studies that compare tests #diagnostics #accuracy
Jon Deeks FMedSci@deeksj10 Feb 2020
How good are we at comparing tests? Not very was our answer, but there were some really good examples and we provide some guidance to help @Richard_D_Riley @ExeterTestGroup @UoB_IAHR @UoB_TERG @deeksj see doi.org/10.1016/j.jclinepi.2…
2
2
Stats by Sue Mallett and Laura Quinn, members of the test evaluation research group @UoB_IAHR
National Institute for Health and Care Research
@NIHRresearch10 May 2019
‘One scan could replace multiple scans to streamline #cancer assessment’, NIHR-funded research by @UCL and @UCLH published today in @theLancet . Read more: bit.ly/2LyEaEU
5
Thank you all for the excellent talks. @NihrG
Dr Kristian Brock@CatchTwentyToo10 May 2019
Real pleasure to speak at the NIHR Precision Medicine meeting last week with @JMSWason @joyfeed and @CindyBillingham . Many thanks to @AliceSitch for organising (and my mum for looking after my girls so I could go for a pint afterwards)
1
Thank you all for the lovely comments. Many thanks to @deeksj, @jacdinnes and Sue Mallett for supervising and to @hstmcleod for chairing.
Jon Deeks FMedSci@deeksj27 Mar 2019
CONGRATULATIONS to @AliceSitch on defending her PhD thesis today on Design and Analysis of Biological Variability Studies and Optimal Monitoring Strategies. Thx to @rafaoxford and Chris Price for their rigorous examination! @NIHRBhamBRC @UoB_IAHR.
2
17
Load more