Took Princess into the vet this morning to review her prednisone schedule (one per week being marginal at best) and trim the rest of her claws that Husband and I couldn't manage. A few days ago she broke her skin and ripped out large chunks of fur in very short order, so we had to act quickly.
For reference her steroid use has now been declared "long term" and will be managed as such. Dosage for the immediate future to be one every three days, then gradually increasing the interval once more to somewhere around the 5-6 day mark. And of course, keep the claws trimmed, which is easier said than done.
Have you ever trimmed the claws of an uncooperative (and especially foot-shy) fluffy cat? Her name isn't Princess Twinkle Toes for nothing - from the very beginning it was clear that she was fussy about her feet. With the fluff between her toes it takes that much longer to get a clear view of the target, and claws can only be trimmed so much. Go too far and you get blood and pain. Princess would have us believe that there is excruciating pain involved even in safe trimming, so the screaming (hers) adds a certain flair to the proceedings.
So, clearly the task is to keep her claws trimmed so she can't break the skin and give her the minimum medication to prevent the scratching in the first place. But the amount of regrowth required for safe trimming is about the same as that required for self-harm. It's a race.
It would be interesting to build a mathematical model of the various processes to predict the ideal schedule. Claw growth should be reasonably straightforward to plot (including potential variables for the effects of medication and the weather on growth rates). It should be possible to identify minimum trimmable length and minimum damage length. Over that there's the cyclic function of the drug schedule, where the probability of irritation/scratching increases towards the end of the cycle. The drug cycle has to be as long as possible and must therefore include at least some irritation at the end.
Easy.