Tom Yates
@tomayates.bsky.social
Epidemiology, infection
https://profiles.ucl.ac.uk/37299-tom-yates/about
https://profiles.ucl.ac.uk/37299-tom-yates/about
Clinical pharmacology friends, how likely is it that patients would develop sulfamethoxazole crystals in urine if given prophylactic doses of co-trim PLUS methenamine? The case reports I have seen are all patients on high dose co-trim
#IDSky #ClinPharm
#IDSky #ClinPharm
July 16, 2025 at 5:11 PM
Clinical pharmacology friends, how likely is it that patients would develop sulfamethoxazole crystals in urine if given prophylactic doses of co-trim PLUS methenamine? The case reports I have seen are all patients on high dose co-trim
#IDSky #ClinPharm
#IDSky #ClinPharm
How is @groupe-sncf.com still using this terrible infographic, which massively understates the environmental benefits of train travel
#dataviz
#dataviz
April 13, 2025 at 7:14 PM
How is @groupe-sncf.com still using this terrible infographic, which massively understates the environmental benefits of train travel
#dataviz
#dataviz
Here is the OBR's distributional impact assessment of Rachel Reeves' changes so far
Hard to understand why richest households haven't been asked to contribute more
Doesn't include funding additional defence spending with cuts to DFID, which is spectacularly regressive
@stellacreasy.bsky.social
Hard to understand why richest households haven't been asked to contribute more
Doesn't include funding additional defence spending with cuts to DFID, which is spectacularly regressive
@stellacreasy.bsky.social
March 26, 2025 at 7:42 PM
Here is the OBR's distributional impact assessment of Rachel Reeves' changes so far
Hard to understand why richest households haven't been asked to contribute more
Doesn't include funding additional defence spending with cuts to DFID, which is spectacularly regressive
@stellacreasy.bsky.social
Hard to understand why richest households haven't been asked to contribute more
Doesn't include funding additional defence spending with cuts to DFID, which is spectacularly regressive
@stellacreasy.bsky.social
Here is the SAPIT trial
ART initiation within 4 weeks of TB treatment start vs after completion of intensive phase in people with TB
Maybe the trial should be in people with CD4<50 with CNS OIs, where ART initiation is typically delayed for longer
ART initiation within 4 weeks of TB treatment start vs after completion of intensive phase in people with TB
Maybe the trial should be in people with CD4<50 with CNS OIs, where ART initiation is typically delayed for longer
March 20, 2025 at 9:23 PM
Here is the SAPIT trial
ART initiation within 4 weeks of TB treatment start vs after completion of intensive phase in people with TB
Maybe the trial should be in people with CD4<50 with CNS OIs, where ART initiation is typically delayed for longer
ART initiation within 4 weeks of TB treatment start vs after completion of intensive phase in people with TB
Maybe the trial should be in people with CD4<50 with CNS OIs, where ART initiation is typically delayed for longer
We tried various assumptions about time to intervention and time to event
In all scenarios, IMMORTOOL predicted substantial ITB
Whilst it seems likely the benefits of IVIG have been exaggerated, we cannot tell the true effect as other biases, e.g. confounding by indication, may be present
[10/n]
In all scenarios, IMMORTOOL predicted substantial ITB
Whilst it seems likely the benefits of IVIG have been exaggerated, we cannot tell the true effect as other biases, e.g. confounding by indication, may be present
[10/n]
January 10, 2025 at 9:29 AM
We tried various assumptions about time to intervention and time to event
In all scenarios, IMMORTOOL predicted substantial ITB
Whilst it seems likely the benefits of IVIG have been exaggerated, we cannot tell the true effect as other biases, e.g. confounding by indication, may be present
[10/n]
In all scenarios, IMMORTOOL predicted substantial ITB
Whilst it seems likely the benefits of IVIG have been exaggerated, we cannot tell the true effect as other biases, e.g. confounding by indication, may be present
[10/n]
NEW PREPRINT!
Enjoyed working with @tom-parks.bsky.social and @petedodd24.bsky.social on immortal time bias
In its simplest form, ITB results from misallocation of person time, i.e. pre treatment time in intervention group being counted as time on treatment
www.medrxiv.org/content/10.1...
[1/n]
Enjoyed working with @tom-parks.bsky.social and @petedodd24.bsky.social on immortal time bias
In its simplest form, ITB results from misallocation of person time, i.e. pre treatment time in intervention group being counted as time on treatment
www.medrxiv.org/content/10.1...
[1/n]
January 10, 2025 at 9:29 AM
NEW PREPRINT!
Enjoyed working with @tom-parks.bsky.social and @petedodd24.bsky.social on immortal time bias
In its simplest form, ITB results from misallocation of person time, i.e. pre treatment time in intervention group being counted as time on treatment
www.medrxiv.org/content/10.1...
[1/n]
Enjoyed working with @tom-parks.bsky.social and @petedodd24.bsky.social on immortal time bias
In its simplest form, ITB results from misallocation of person time, i.e. pre treatment time in intervention group being counted as time on treatment
www.medrxiv.org/content/10.1...
[1/n]
If you want to quantify harm - presumably the purpose here - I think you need some comparator
How many of these patients would have died anyway?
Approaching death leads to AMR (healthcare contact, antibiotic exposure), and you need to unpick this from the deaths caused by AMR
How many of these patients would have died anyway?
Approaching death leads to AMR (healthcare contact, antibiotic exposure), and you need to unpick this from the deaths caused by AMR
January 9, 2025 at 3:50 PM
If you want to quantify harm - presumably the purpose here - I think you need some comparator
How many of these patients would have died anyway?
Approaching death leads to AMR (healthcare contact, antibiotic exposure), and you need to unpick this from the deaths caused by AMR
How many of these patients would have died anyway?
Approaching death leads to AMR (healthcare contact, antibiotic exposure), and you need to unpick this from the deaths caused by AMR
But in the solid organ transplant population? The data there looked pretty convincing
December 26, 2024 at 1:27 PM
But in the solid organ transplant population? The data there looked pretty convincing
This is an interesting proposition
I've got several articles that at 85% complete
If I submit here, I can plagiarise the remaining 15%!
I've got several articles that at 85% complete
If I submit here, I can plagiarise the remaining 15%!
November 21, 2024 at 1:43 PM
This is an interesting proposition
I've got several articles that at 85% complete
If I submit here, I can plagiarise the remaining 15%!
I've got several articles that at 85% complete
If I submit here, I can plagiarise the remaining 15%!
They are oversimplifications but @resfoundation.bsky.social's distributional assessment of Rachel Reeves' 1st budget suggests it is broadly progressive
Here's hoping the next one will be bolder
Super disappointing that the two child limit remains, and the petrol duty vs bus fare thing was rubbish
Here's hoping the next one will be bolder
Super disappointing that the two child limit remains, and the petrol duty vs bus fare thing was rubbish
November 2, 2024 at 12:47 PM
They are oversimplifications but @resfoundation.bsky.social's distributional assessment of Rachel Reeves' 1st budget suggests it is broadly progressive
Here's hoping the next one will be bolder
Super disappointing that the two child limit remains, and the petrol duty vs bus fare thing was rubbish
Here's hoping the next one will be bolder
Super disappointing that the two child limit remains, and the petrol duty vs bus fare thing was rubbish
This is my favourite
November 1, 2024 at 7:56 AM
This is my favourite
October 22, 2024 at 2:34 PM
Naive question about hazard ratios
Does HR for all-cause mortality go all the way to 1 with sufficient duration follow up, as suggested by @miguelhernan.bsky.social here, or will relative reductions in person time denominator due to harmful exposure mean it doesn't quite get to 1?
#statsky #episky
Does HR for all-cause mortality go all the way to 1 with sufficient duration follow up, as suggested by @miguelhernan.bsky.social here, or will relative reductions in person time denominator due to harmful exposure mean it doesn't quite get to 1?
#statsky #episky
October 15, 2024 at 4:37 PM
Naive question about hazard ratios
Does HR for all-cause mortality go all the way to 1 with sufficient duration follow up, as suggested by @miguelhernan.bsky.social here, or will relative reductions in person time denominator due to harmful exposure mean it doesn't quite get to 1?
#statsky #episky
Does HR for all-cause mortality go all the way to 1 with sufficient duration follow up, as suggested by @miguelhernan.bsky.social here, or will relative reductions in person time denominator due to harmful exposure mean it doesn't quite get to 1?
#statsky #episky
Green energy folk!
Triple glazed Velux are really expensive
If you have two 114 x 118cm loft windows, how much saving on gas/electricity bill are you likely to get if you opt for the various triple glazed options vs double glazing (the default)?
Hope to switch to heat pump soon
Triple glazed Velux are really expensive
If you have two 114 x 118cm loft windows, how much saving on gas/electricity bill are you likely to get if you opt for the various triple glazed options vs double glazing (the default)?
Hope to switch to heat pump soon
September 25, 2024 at 6:39 PM
Green energy folk!
Triple glazed Velux are really expensive
If you have two 114 x 118cm loft windows, how much saving on gas/electricity bill are you likely to get if you opt for the various triple glazed options vs double glazing (the default)?
Hope to switch to heat pump soon
Triple glazed Velux are really expensive
If you have two 114 x 118cm loft windows, how much saving on gas/electricity bill are you likely to get if you opt for the various triple glazed options vs double glazing (the default)?
Hope to switch to heat pump soon
There must be code available to simulate M bias? Any pointers? My DAG below. Exposure and outcome are binary. Disease severity is a bundle of different binary things - probably easiest to model as severe vs not. Open to collaborating. Important clinical question!
#episky @lucystats.bsky.social
#episky @lucystats.bsky.social
September 5, 2024 at 12:37 PM
There must be code available to simulate M bias? Any pointers? My DAG below. Exposure and outcome are binary. Disease severity is a bundle of different binary things - probably easiest to model as severe vs not. Open to collaborating. Important clinical question!
#episky @lucystats.bsky.social
#episky @lucystats.bsky.social
Sorry, works for me. Screenshot below.
August 24, 2024 at 6:38 PM
Sorry, works for me. Screenshot below.
Me and Helen Payne are putting on an event at the Royal Society - come along!
'Indirect effects of cytomegalovirus infection: mechanisms and consequences'
14 and 15 October - hybrid, in London and online
KEYNOTE FROM STANLEY PLOTKIN!
Register here royalsociety.org/science-even...
#IDsky #EpiSky
'Indirect effects of cytomegalovirus infection: mechanisms and consequences'
14 and 15 October - hybrid, in London and online
KEYNOTE FROM STANLEY PLOTKIN!
Register here royalsociety.org/science-even...
#IDsky #EpiSky
August 22, 2024 at 11:07 AM
Me and Helen Payne are putting on an event at the Royal Society - come along!
'Indirect effects of cytomegalovirus infection: mechanisms and consequences'
14 and 15 October - hybrid, in London and online
KEYNOTE FROM STANLEY PLOTKIN!
Register here royalsociety.org/science-even...
#IDsky #EpiSky
'Indirect effects of cytomegalovirus infection: mechanisms and consequences'
14 and 15 October - hybrid, in London and online
KEYNOTE FROM STANLEY PLOTKIN!
Register here royalsociety.org/science-even...
#IDsky #EpiSky
Thanks, yes, spotted that
I am told the BEAT-TB results not yet shared publicly
Here are the randomised data from the other trials (table from our paper)
Composite primary outcomes dominated by early cessation and treatment switches rather than microbiological end points
I am told the BEAT-TB results not yet shared publicly
Here are the randomised data from the other trials (table from our paper)
Composite primary outcomes dominated by early cessation and treatment switches rather than microbiological end points
August 21, 2024 at 6:29 AM
Thanks, yes, spotted that
I am told the BEAT-TB results not yet shared publicly
Here are the randomised data from the other trials (table from our paper)
Composite primary outcomes dominated by early cessation and treatment switches rather than microbiological end points
I am told the BEAT-TB results not yet shared publicly
Here are the randomised data from the other trials (table from our paper)
Composite primary outcomes dominated by early cessation and treatment switches rather than microbiological end points