Wait, I'm missing something. Isn't the patient number in col B the baby's ID? I mean, the patient is the baby, not the mother, right? And isn't col C the date the patient entered the NICU? Apparently one or both of those assumptions are wrong, if you can't always go by the admit date.

About desired output: What I'm hearing is that it worked alright to do this manually on a daily and weekly basis, but a) to do the comparison for a whole year's worth you need Excel to do a lot of the work for you, and b) since you're doing that you may as well make it work for you on the weekly stats too. So we're talking about two kinds of output: A comparison of 2013 with 2012, and (if it can be done conveniently) a better way of doing the weekly summary. And for that matter you have to collect daily numbers too. But first things first: The whole-year comparison is the one that would cost you the most effort.

Now, I'm hearing two different things about that:

1. "....manual stats that had to be reported once a week, which included # of admits, days, and now our administrators want it compared to last year as well". That sounds like a week-by-week count of 2013's admissions and number of days' stay, compared week-by-week with 2012.
2. "Our final output wouldn't include patient account #'s - just totals by month." Not by week, after all, then?

And you want "a count, by month, for every day they are housed in the SCN/NICU". But since this isn't by patient, I guess what you want is the total patient-days. For example, if patient A stayed 12 days and patient B stayed 5, that's 17 patient-days and that's all you want for the "days" part of this?)