STATISTICS HELP!!!!!!!!! Please help me?

Heres the question:
Nurse practitioners are nurses with advanced qualifications who often act much like primary care physicians. Medical researchers want to design a study to compare effectiveness of nurse practioners and physicians. They recruit 1200 volunteer subjects for the study, each of whom has been diagnosed with asthma diabetes or high blood pressure.

1. Describe a completely randomized design for this experiment
2. Describe a design for this experiment that uses blocking to improve on the design in (1).. Explain carefully how blocking improves the study design.
3. Can your experiment in (2) be carried out in a double- blind manner? Justify your answer.

Im sorry, I am on break right now and we got these set of questions, and I mistakenly left my book in my locker and cannot find any sites to help.
please help!

I took Stats two years ago....so my memory's a bit fuzzy, but maybe it will jog yours some?

Okay so you could randomly assign the subjects numbers 1-1200. Numbers 1-600 are sent to nurse practitioners and numbers 601-1200 are sent to physicians.

For blocking...you could block each group as in those with asthma, those with diabetes, and those with high blood pressure. Then you could compare the effectiveness of the treatments on each condition between the nurse practitioners and the physicians.

I know that double-blind means that the nurse practitioners and physicians wouldn't know which patient had what diagnosis and the subjects wouldn't know if they were seeing a nurse practitioner or a physician, so that would help eliminate bias on the subjects' part, since most people think physicians would be more effective than nurse practitioners.


I hope this helps!