Dear Frederic, 

The samples the two researchers used are not identical.  From your reference: The present study aimed to assess the validity of the EPDS as a screening instrument for major depressive episode (MDE) among adults from the general population. Both sexes were thus included.  The scale was originally developed for women. Furthermore, cultural and age differences between samples are at hand. Calculate the mean in your own sample, start from there, look at the normal curve.  See where the lowest third falls and where the highest third falls. You can look at my hysterectomy and vasectomy groups and see how I used BDI there.

 Thank you for this excellent answer, Béatrice. Unfortunately, I don't think that I am allowed to do like this : indeed, our purpose is to determinate the ratio of MDE in young fathers. So, I can't make the measurements and then choose the cut-off in a way the results will match wih our hypothesis, it would be bad research ;-)

But you are right, I can't make a comparison between these studies because they studied differents things in different groups. I need to chose the closest one from our population and object of study.

I will read with great interest your work about depression and anxiety among hysterectomised women.

Actually in qualitative research it is always done in this order: You gather your results and the closely on them, they look like ashes. Gray and without logic.  Then all of a sudden the Phoenix bird raises up in all his glory and dusts off the ashes!

You see what surprised you and then you build up your introduction and discussion. It is not at all bad research, why do we need research then if we would know already in our "working hypotheses" how things look like in real life! 

Although used widely in perinatal/postpartum depression research, remember that the Edinburgh is intended to have clinical utility as a screening instrument to identify women who are experiencing symptoms of depression (not to verify the diagnosis of major depressive episode). A major depressive episode requires a certain number of symptoms occurring at a certain severity over a specified period of time, not just the presence of "some" depressive symptoms. ... Cut-off points are arbitrary and are guaranteed NOT to capture some men who are depressed, and TO capture some who are not depressed. I think you will want to frame your introduction as well as your discussion of the results in such a way that explains thoroughly why you chose your particular cut-off, and how this choice impacts your findings. If your project is strictly about identifying rates of major depressive episodes in this particular population, then you are going to need more ways of measuring it in addition to the EPDS. If you do use multiple methods, then in addition to producing results on the rate of  major depression in your particular population, you can also add to the body of knowledge about what EPDS cut-off score was optimal in identifying as many cases of major depression as possible and screening out non-cases. Be sure to provide information on demographics such as culture and age, since as was noted above this will affect your results. 

Hello,

What a fascinating question to pose Frederic. I would echo some of Kimberly and Beatrice's remarks in terms of the Edinburg scale being design and used for PPD in women. I wonder why are you (and your research group) considering this tool in the first place? Why not using a well validated depression scale such HAMD or MADRS and apply it to your sample of men in the post-natal period?

In any case, I look forward to learning more about your research in men experiencing depression in the post-natal period.

Best regards,

Fidel

Thank You Fidel,

Our choice to use the EDS, rather than conventionnal scales for this study, stand on litteracy on this subject. The EDS has been conceived to be more specific for young adults after the birth of a child. Of course, this has been conceived for mothers, but it seems logical to use the same scale for young fathers.