However, I'm still quite confused about what this study really shows. I was initially blaming the Mail article for mangling the stats: the stats they've thrown in from other sources are very hard to understand without context - eg the vagina construction bar chart adds up to 100% - does this mean that every single patient was in pain 2 years later? Or does it just show the proportion of the patients in pain who needed medical treatment compared with those who didn't (not very informative if the latter).
However, having looked at the original study, I'm still confused.
The abstract says firstly:
In early adolescence, 11% of participants reported gender non-contentedness. The prevalence decreased with age and was 4% at the last follow-up (around age 26).
(So, less than the 70-80% desistance rates seen in gender clinic studies, but I think that comes out as over a 60% reduction.)
However, it also says,
Three developmental trajectories of gender non-contentedness were identified: no gender non-contentedness (78%), decreasing gender non-contentedness (19%), and increasing gender non-contentedness (2%).
I think this is where the confusion upthread has arisen, with the suggestion that 80% didn't decrease. In fact most of those 80% didn't decrease because they didn't express any discontentedness (what a word!) in the first place.
I'm struggling to bring the two statements together though. How can 19% of people having a decreasing trajectory account for a drop of over 60%. Is it to do with the 11 > 4% change being measured between two points, and the 'trajectory' figure taking into account multiple intermediate points? Or something else? I have no maths or science beyond GCSE, so I'd be really grateful if anyone with more knowledge in this area could explain.