There are two concerns, coolant degradation as it acidifies with age, and inhibitor package depletion.
You can check how the glycol is aging with a pH meter (strips won't give you enough resolution). You should make sure your pH meter is calibrated against pH buffer standards because the probes quickly drift out of calibration, especially when they are not stored correctly. The pH should be around 8.0 in a healthy glycol solution for automotive use. And yes.... it ages quicker with heat. The degradation is a chemical reaction and ethylene glycol primarily degrades into glycolic acid (some oxalic, some formic) and all chemical reactions proceed quicker with energy input. Thus ethylene degrades into the acids quicker at higher temps than it would at lower.
I've also seen studies that show inhibitor degradation over time. The azoles, in particular, drop with time. I'm not sure of the mechanism whether it is bonded to copper/brass as the inhibiting layer or if it is lost in the non-sealed system. Either way, you could just start at a higher level. The assumption is that it will be run at 50/50 so run it at 65/35 so you have more to start with.
It is so cheap to dump it on a regular basis that I err on the side of caution.