I'm curious about that too. We use a RO system, and so far I've only replaced the carbon sediment filters. I'm a year or two away from replacing the RO membrane cartridge. I cut them open when I replace them, I always wanted to do some kind of chemical analyses of the filters. Visibly they're just brown and saturated with rust from old galvanized pipes our building uses.
I think it's actually pretty hard to tell much from the membrane itself, my understanding is that most of the process is based in the membrane preferring the smaller water-only molecules and allowing the waste run off to carry away anything else. So they shouldn't really be catching much in the element.
They fail either from the chlorine degrading the membrane film which will start allowing larger dissolved solids through (your TDS will increase) or by having the pores scale and plug over time which will slowly decrease the output.
