Frederik Feys

Vrije Universiteit Brussel | VUB · Family Medicine and Chronic Care

Hi everyone !

I have a challenging time to crack the code in the following analysis (https://www.fhi.no/globalassets/dokumenterfiler/rapporter/2020/should-individuals-in-the-community-without-respiratory-symptoms-wear-facemasks-to-reduce-the-spread-of-covid-19-report-2020.pdf).

Norway population is 5,432,580 I am confused how to implement the assumption "20% of people infectious with SARS-CoV-2 do not have symptoms".

What is the formula used to get the number of masks needed to wear to prevent 1 infection?

Thank you for your time

Frederik

Hello everyone

Who is expert in discovering confounding? We just wrapped up preliminary work on the possible association between psychological well-being and covid-19 survival. We found a weak link.

However, what confounding do you think might be at play?

Happy to discuss

Frederik

Hi everyone

I modeled using a uni-variate regression.This is a plot of fitted versus residual values. Clearly this model violates the assumption of normality. What could be going on?

Thank you

Frederik

Hi everyone,

The world is in crisis; scientific communities work to their best to find answers. From our side, we worked on a risk assessment for the lockdown strategy, focusing on psychological well-being.

We submitted to medrxiv and this was the shocking reality:

medrxivpreprint censored our manuscript because "the work challenges a widely accepted public health measure".

What do you think? Isn't challenging work what drives science forward? Questioning existing paradigms; admit knowledge gaps and improve to make progress ...

Lets discuss this,

I get this weird error. A rather basic action to add your data (spreadsheet file) to a preprint manuscript...? How to do this?

Thanks

Frederik

Hello everyone!

A standard approach after the diagnosis of leukemia is immunosuppression and later on transplantation of donor cells.

A doctor may state that "if we do not start therapy immediately, you (the patient) will die within weeks."

What evidence is there for this statement? What research exists that studies no intervention (watchful waiting)?

Thank you

Hi Everyone,

I want to analyse how well two scales agree for detecting changes in wellbeing. I look at the Bland-Altman Method. One scale is a simple score from 0 - 10 for wellbeing. The other scale is the well established Beck’s Depression Inventory-II (goes from 0 - 63) . I wonder how to analyse the agreement?

Thanks!

Fred

Hi everyone!

I have a 2 repeated measures that was asked at 3 moments in time (within-subject factor). We doubled the measure to control for response bias. So the measure (likert scale) probed the same topic but one time positively framed ("did this help you?" and another negatively framed ("did this hinder you?").

I use a GLM repeated measures strategy to analyse. How to put the 2 measures into the same model? Do I add the hindered measures as a covariate? Please see screenshot for more.

Thanks for helping me out!!

Fred

Hi everyone!

I am stuck with a seemingly very simple logistic regression analysis. What is the impact of diabetes(nominal var, values 0-1) on preterm birth (nominal var, values 0-1)? In the output, the Nagelkerk is < .00. Weird, since the expected coefficient (Exp (B) is 1.77. To wich I conclude, that diabetes increases the risk with 77% for preterm birth. I would expect the Nagelkerk to be around .30 or so? Am I doing this analysis correctly? Any input very much welcome!

Thanks

Fred