That's a very interesting question Charlie. I have pondered it quite a bit because I have had the privilege of having had many bad objectives pass through my hands. Currently, I have a box full of them, awaiting their fate of either a respectable repair or a dis-respectable disposal.
From my experience, which covers a large amount of Spencer and AO optics, relatively less Bausch & Lomb or Leitz , less Zeiss( due to avoidance), plus a few scattered other brands; the problem can be classified according to the way it manifests itself and therefore one can either avoid or mitigate it somewhat. It manifests itself differently from manufacturer to manufacturer, from cementing material to cementing material and tends to occur in batches.
The first thing is, it helps to identify what the cementing material is because depending on that, the problem will be different. Up to a certain point in time, and it obviously varies according to the manufacturer, all of the cement was Canada Balsam. Most makers had some problems with Canada Balsam but the problems vary in their severity and form. I think there are 3 basic problems with it and doubt if there is anything one can now do to halt the most common one, drying out. First, there is contamination. This is what causes the previously referenced batching of de-lamination , where objectives of the same type, suffer similar de-lamination in the same element or elements, while other elements or objectives in the series do not . Usually there is a milky swirly type of film throughout the cement, probably due to some chemical reaction taking place over time. If that hasn't happened by now, then it probably isn't going to. Second, there is the effect of airborne chemicals, or mild airborne solvents , that enter from the exterior and cause a kind of circumferential breakdown, from the outside in. Again, it probably happened in the past , when many facilities did not have the best ventilation conditions or non at all. The problem is probably not going to get any worse. This can mimic the third problem, the outright drying out of the cement. In drying, the balsam will develop anything from air bubbles, to cracks and crystals and yellowing.
I have had some success in re-dissolving the balsam with ethanol, and healing most of the above problems, even to the point of getting a perfect result but how long it will last is another question. If it lasts as long as the original...50 years or so, then I guess it is a good enough job. The good thing is, that most of the objectives cemented with balsam and currently in good condition will probably stay so, for a considerable period of time. Keeping them in a cannister when not in use helps reduce drying because balsam can overdry and crack but if the lenses are in good shape now, after 50 years, then probably they are sealed into their location sufficiently, to resist drying. Some makers edged cemented lenses with a sealant.
With synthetic cements, which mostly arrived from the 50's on, the problem is more complex. The problems seem similar in appearance, though , to those associated with balsam minus the yellowing. Bausch & Lomb had some problems , which appear on the surface to mimic the problems with balsam quite closely but the solvents required to heal the problem are pretty heavy duty; methyl iodide and the like. I have also had some success with mild heat( boiling water), and have evolved some pretty low tech , novel methods of clamping lens doublets. B & L's problems seem to occur in batches, so my guess is that they got into some cement that was mis-manufactured. Either that, or their preparation and curing at the production level went off the rails. It does seem that in this case, the problem showed up quickly, so an objective unaffected, is unlikely to be affected any time soon. I have duplicates of B & L objectives, with some de-laminated and others, not. The good ones show no sign of deterioration over 15 years or so. Generally, most of the synthetic cements used have been pretty good. They don't seem to suffer from drying as much as balsam but then , there are fewer years since their adoption too.
Then there is the problem of the effect of light and heat. If you took 20 objectives from a production that later showed to have a disproportionately high rate of failure and put 10 into use, up to the point of where some failed but had another 10 sitting on a shelf wrapped in tissue paper, would the failure rate be statistically similar? I don't think so. Firstly, some objectives, probably have a tendency to concentrate heat in certain elements. It may be a small amount but it is there. Secondly, there must be an effect of certain wavelengths of light upon the chemistry of the cement. There does seem to be some evidence that an unused objective has less of a chance of having cement problems, than one that has been used.
So; the general rules might be; avoid conditions that promote drying and don't leave an objective in position over a light source if not being used and avoid excessive intensity where possible. I don't think there is too much else, one can do to avoid the problem, which ultimately seems to be mostly a manufacturing defect: crappy cement, poor homogenization and or curing. I have objectives that are over 100 years old, that have been used to the point of the knurling being worn off and barely visible, and they have impeccable glass in them.