Tomáš Formánek
@tformanek.bsky.social
MSCA Fellow @ Aarhus University and Visiting Researcher @ University of Cambridge studying the mental-physical health interface.
I focus on the aetiology of co-occurring mental and physical health conditions and the consequences of such co-occurrences.
I focus on the aetiology of co-occurring mental and physical health conditions and the consequences of such co-occurrences.
Any contexts in which it *can be* beneficial? 🥸
October 8, 2025 at 1:55 PM
Any contexts in which it *can be* beneficial? 🥸
His tempo at the end was furious! Will probably not run that fast in any context, not to mention after 50km in the Alps 😄
August 28, 2025 at 7:08 PM
His tempo at the end was furious! Will probably not run that fast in any context, not to mention after 50km in the Alps 😄
"Protective causal factor" feels like the one I was looking for! It is succinct and feels intuitive. Thanks a lot for sharing it 🤓
August 27, 2025 at 11:04 AM
"Protective causal factor" feels like the one I was looking for! It is succinct and feels intuitive. Thanks a lot for sharing it 🤓
It's challenging to be technically 100% correct but also succinct 😬 My problem is that saying "negative association" or "inverse association" is neither 100% correct but also not really casually understandable 🥲
August 26, 2025 at 5:06 PM
It's challenging to be technically 100% correct but also succinct 😬 My problem is that saying "negative association" or "inverse association" is neither 100% correct but also not really casually understandable 🥲
It always depends 😄
Your example would work nicely for reporting the actual results. However, imagine that you want to succinctly describe the character of the association. Positive and null results are fairly clear labels for standard purposes. Inverse association seems difficult to grasp, though.
Your example would work nicely for reporting the actual results. However, imagine that you want to succinctly describe the character of the association. Positive and null results are fairly clear labels for standard purposes. Inverse association seems difficult to grasp, though.
August 26, 2025 at 5:47 AM
It always depends 😄
Your example would work nicely for reporting the actual results. However, imagine that you want to succinctly describe the character of the association. Positive and null results are fairly clear labels for standard purposes. Inverse association seems difficult to grasp, though.
Your example would work nicely for reporting the actual results. However, imagine that you want to succinctly describe the character of the association. Positive and null results are fairly clear labels for standard purposes. Inverse association seems difficult to grasp, though.
I recently used this meme for illustration 🤡
August 12, 2025 at 11:59 AM
I recently used this meme for illustration 🤡
No RStudio either!
August 7, 2025 at 5:33 PM
No RStudio either!
Brilliant – I can hear the swearing when making typos trying to call the package 😆
June 1, 2025 at 10:11 AM
Brilliant – I can hear the swearing when making typos trying to call the package 😆
indiscrete? But cheekycont sounds pretty good 😄
May 31, 2025 at 3:17 PM
indiscrete? But cheekycont sounds pretty good 😄
ConTreat? Or ContinueR? 🤡
May 31, 2025 at 3:08 PM
ConTreat? Or ContinueR? 🤡
Ah, that one is on hazard ratios, here are two specifically on odds ratios: academic.oup.com/esr/article-... academic.oup.com/aje/article/...!
Logistic Regression: Why We Cannot Do What We Think We Can Do, and What We Can Do About It
Abstract. Logistic regression estimates do not behave like linear regression estimates in one important respect: They are affected by omitted variables, ev
academic.oup.com
April 2, 2025 at 4:59 PM
Ah, that one is on hazard ratios, here are two specifically on odds ratios: academic.oup.com/esr/article-... academic.oup.com/aje/article/...!
I'm not a statistician, but I think interpreting odds ratio magnitude is particularly tricky. Here is a nice discussion: www.sciencedirect.com/science/arti....
Relative rates not relative risks: addressing a widespread misinterpretation of hazard ratios
The use of the Cox proportional hazards model is ubiquitous in modern medical research. Despite the widespread implementation of this model, the termi…
www.sciencedirect.com
April 2, 2025 at 4:54 PM
I'm not a statistician, but I think interpreting odds ratio magnitude is particularly tricky. Here is a nice discussion: www.sciencedirect.com/science/arti....