The zero offset changes with temperature, but any long-term changes (barring global warming <g>) are due to changes in the state of the battery in the crank. As it runs runs down (over years - our track SRM from 2002 is still going strong on its original battery) the zero offset will gradually decrease.
What Tom A. and I were discussing were changes in the slope itself. At least at one point, this would drift steadily for a couple of years, then begin to stabilize (and as I mentioned, neither of ours have seemingly changed in 5+ y). SRM blamed this on a bad batch of epoxy that required a long time to cure, but I've always wondered if perhaps it weren't due to "work hardening" of the crank spider to which the strain gages are bonded (any engineers out there, feel free to chime in and correct me if this isn't possible/logical).