I wanted to forward a link which
skibbley posted today, as it's actually rather impressive. The National Health Service trust for a relatively small area of the UK (200,000 people) actively tried to find out how many transgendered people were living in the area so that they could improve their services. It's surprisingly well-designed, and clearly acknowledges all the limitations, yet manages to draw several useful conclusions and recommendations nonetheless.
I especially like the bits about hitting GPs with a cluebat, and making sure all of the healthcare workers in the Trust are using the same guidelines and criteria. The points about how the majority of trans people become "stealth" post-transition and the increase in the number of children with gender dysphoria presenting for treatment are particularly worth a provider of healthcare noticing. It is important that trans people who retain some of their original body parts (e.g. prostates & cervices), are offered the appropriate screening in a sensitive manner, and it's important that a coherent policy exists for dealing with children too young to consent to surgery that is more useful than "send them away until they're 18".
Obviously, the document isn't going to be directly relevant to people who live in countries without state-sponsored healthcare, but I'm sure some of these conclusions and recommendations would be worth forwarding to insurance companies if you have enough energy to do so.
Calderdale NHS Survey of the Trans Population (pdf).
I'm not screening comments because I don't have enough time to unscreen the useful ones, but if you want to say something privately, post anonymously. Ideally signed with your name or lj name so I know who you are :)