The original paper is quite readable and interesting in how the researcher works through a number of alternative hypotheses.

https://www.nber.org/system/files/working_papers/w35387/w353...

An tangential part of the paper is partnered men:

  Among baseline partnered men (which is where partnership dissolution is 
  well-defined), I find a 6.8pp increase in men leaving their baseline partners 
  (s.e. 2.9), which rises to 12.1pp after 1.5+ years (s.e. 6.1). Unlike women, 
  partnered men leave existing relationships with weight loss.