Lightnews — Scholar-powered news

Thiago @csthiago.bsky.social · 20h

The BBC article about the paper. God... "hormone reboot"... From a survey paper....

Thiago @csthiago.bsky.social · 5d

It looks that there is a problem with the statistics as well. The propensity score only uses baseline variables and treats them as time fixed. But, a second infection is a time-varying exposure. So, the "unexposed" group has been defined conditionally on the future. Right, @pwgtennant.bsky.social ?

1

Thiago @csthiago.bsky.social · 12d

Also mediation with counts

Thiago @csthiago.bsky.social · 12d

Help!
Where can I find resources about causal mediation of ordinal outcomes? (using SEM)

Thanks 😄
#rstats @dingdingpeng.the100.ci

2 1 2

Thiago @csthiago.bsky.social · 17d

Updates about it. The CIs are likelihood ratios CIs, which can be very asymmetrical.
About the cumulative incidence, they have reinforced me that the all lower exposure curves initiate before the higher doses one.
(I still feel a bit uneasy with almost all low CIs being 1)

Reposted by Thiago

Fjorida Llaha @fllaha.bsky.social · 20d

I’ve had the same impression while looking at academic epidemiology postdoc offers. So much focus on AI, machine learning, LLM experience. Are these the core skills we need now? What about subject matter knowledge, study design, bias, statistics, population health knowledge? Outdated?

Lorenzo Fabbri @epilorenzofabbri.com · 21d

Looking at job offers (both academia and industry) and it’s basically all “AI” & Co. 😔
And WTF is “experience with LLM” for an academic post in epidemiology??

1 2 10

Thiago @csthiago.bsky.social · 22d

I haven't contact them / or post on pubpeer. Someone shared this paper with me today and I found it those bits very strange.
I will email them tomorrow

Thiago @csthiago.bsky.social · 22d

Oh. I haven't paid close attention to this figure. There are even participants starting not at 0%.
About the "rr" is complicated to know, once they have used time varying exposure and idk how to interpret cum incidence in this context.

1 1

Thiago @csthiago.bsky.social · 23d

About the RR (relative risk in the paper, not risk ratio), which is the HR is... but to me looks misleading

1 1

Thiago @csthiago.bsky.social · 23d

Sleuths,
Can someone explain to me how these CIs are possible?
www.nejm.org/doi/full/10....
It is not clear how they got RR from the Cox PH, as it would require choosing a time point (looks like an interpretation of HR as RR)

@lonnibesancon.bsky.social @sophieehill.bsky.social @retractionwatch.com

2 1

Thiago @csthiago.bsky.social · Sep 5

Thanks! I will take a look

1

Thiago @csthiago.bsky.social · Sep 5

2/2
The outcome model has multiple categorical variables; I only found binary and continuous methods in MICE (for mixed models). Any ideas?
#rstats #episky

1 1

Thiago @csthiago.bsky.social · Sep 5

1/2
Does anyone know how to proceed for multiple imputations if one of the variables (fully observed) in the model has more than 5000 levels (in the outcome model, it is a mixed effect)?

#rstats #episky

1 1 3

Thiago @csthiago.bsky.social · Aug 25

It's also possible to use data.table speed through tidytable (tidyverse syntax)

1

Thiago @csthiago.bsky.social · Aug 24

It is very interesting book. The story behind "cleaning" in surgery

Thiago @csthiago.bsky.social · Aug 21

📌

Thiago @csthiago.bsky.social · Aug 18

If you look in the WoS www.webofscience.com/wos/author/r...

www.webofscience.com

1

Thiago @csthiago.bsky.social · Aug 18

Almost all self citations in open repositories

1 1

Thiago @csthiago.bsky.social · Aug 17

guides.lib.uiowa.edu/c.php?g=5437...

University Libraries | The University of Iowa

Guides: ORCID @UI: CrossRef

guides.lib.uiowa.edu

1 1

Thiago @csthiago.bsky.social · Aug 17

I think you can also link the orcid to crossref. www.crossref.org/community/or... (I think that is the best one)

ORCID auto-update - Crossref

An ORCID iD is a persistent identifier for individual researchers and scholarly contributors. It allows everyone (authors, publishers, funders, and research institutions) to uniquely identify the work...

www.crossref.org

1 1

Thiago @csthiago.bsky.social · Aug 17

You can sync WoS to Orcid and let it do the udpates

1 1

Thiago @csthiago.bsky.social · Aug 16

📌

Thiago @csthiago.bsky.social · Aug 12

I don't know if that will change in the future (pessimistic hat). Even "traditional" methods still suffer from inadequate peer review, just see the infinitude of papers with time varying exposure/treatment being treated as time fixed and being published in "prestigious" journals

2

Thiago @csthiago.bsky.social · Aug 8

Check out our last paper about the adverse consequences of arbovirus infection during pregnancy:
www.nature.com/articles/s41...

We have use the exact same approach for dengue, zika and chikunugnya. (including a negative control exposure!)
#dengue, #zika, #chikungunya, #epidemiology

Perinatal outcomes of symptomatic chikungunya, dengue and Zika infection during pregnancy in Brazil: a registry-based cohort study - Nature Communications

In this study of ~7 million Brazilian births, the authors show that chikungunya, dengue, or Zika infection during pregnancy is associated with an increased risk of adverse perinatal outcomes, with the...

www.nature.com

1 2

Thiago @csthiago.bsky.social · Jul 3

I would say that there is nothing special (in terms of wrong things) compared to others fields. Equally wrong everywhere

1 1