Age and illness duration in Chronic Fatigue Syndrome

[Dolphin]

The data is all basically in table 1.

 

[Dolphin]

The way they dichotomised this means that you don't know whether it would have held for different thresholds. It would have been interesting if they looked at treating age and illness duration as continuous variables and seeing if a relationship existed.

 

[Dolphin]

It would have been a little interesting if they had also looked at collapsing the data into above 55 and below 55; also above 10 years illness duration and below 10 years illness duration to see if differences arose. One could still do this with the existing data I suppose but I don't feel inclined to do it myself.

 

[Kati]

The way they dichotomised this means that you don't know whether it would have held for different thresholds. It would have been interesting if they looked at treating age and illness duration as continuous variables and seeing if a relationship existed.

'Treating age', what treatment? There are no approved treatments for ME.

Personally after 8 years I am still going through the different stages of grief. i believe it was @JenB who tweeted about grief, and how it never ended with a chronic disease, let alone an untreated chronic disease.

 

[Dolphin]

The way they dichotomised this means that you don't know whether it would have held for different thresholds. It would have been interesting if they looked at treating age and illness duration as continuous variables and seeing if a relationship existed.

'Treating age', what treatment? There are no approved treatments for ME.

Personally after 8 years I am still going through the different stages of grief. i believe it was @JenB who tweeted about grief, and how it never ended with a chronic disease, let alone an untreated chronic disease.

I mean "treating" in the statistical sense: plotting a graph and looking for a pattern e.g. can you create a regression line that fits the data.

 

[Kati]

I mean "treating" in the statistical sense: plotting a graph and looking for a pattern e.g. can you create a regression line that fits the data.

Thick fog here !