I would do whatever I can to use non-pharmacologic methods to comfort the patient and help him sleep
I would explain him that what he was taking home may or may not be safe for him right now. He is in hospital because he is sick, isn't he? He takes many other meds now for that reason. Now, ALL meds are working in the same human body at once. Some of them work together, and it can be good or not. Some counteract each other, same story. I am a nurse, I cannot determine if that 1 mg of Xanax safe right now. Doctor can, would you like me to call?
If I call, I will start from fresh set of vitals taken by me personally. And I will tell the whole story, not only "hi, this is KatieMI from X, about your patient ftom 1234, can he get something for sleep?". In the case I described sbove, I would sure mention loading amio, drip and quickly escalating opioids.
I would not tell about "home dose Xanax for sleep" at all, unless it was verified or I can verify the dose. Most conscientious providers would not order it "for sleep" anyway.
If a provider will indeed order Xanax or something equally borderline (say, Versed - I had such orders in ICU), I would politely doubt it right then and there. I would mention my discomfort re. possibility of respiratory depression and then "just suggest" Valium, low dose Restoril, eszopiclone, Remeron or Benadryl.
If the provider still wants Xanax, I would call to whoever manages the amiodarone and cardizem and ask that person directly if they think it would be okay. Cardiologists are usually better verced in complicated pharmacology of their own drugs.
I have strong pharmacology background, so doctors usually listened to what I said.
About "1 hour apart" rule:
For example, you work with IV dilaudid to be given q2h.
1) check renal and liver functions.
2) if they are more or less normal, proceed on your own; if clearly not, call pharmacy and speak with PharmD, not tech
3) go Google and search for "dilaudid half life". Or use free app like Epocrates or drugbook.
You do not need a peer-reviewed article. The first link Goodle has is from something named "therecoveryvillage.com". It is good enough, as the number will be the same everywhere.
Dilaudid half life is about 2 hours. So, if you gave dose at 10 AM and the next at 12, your dose #2 will be "catching tail" of the first one as 1/2 of #1 will still be there. After #3 at 2 PM, you will have 1/4 of #1, 1/2 of #2 and whole #3. After #4, you will be close to "doubling" (draw it if it is difficult to understand). Which can be fine if you treat acute postop pain or cancer pain, but not for chronic.
Usually showing Higher Ups that you understand that works. But I must tell you - it sucks to be more than average intelligent person in general, and especially if you are a nurse.