Poll: How do you handle subjects having very different patterns?

#1
Good day everyone - I am curious as to what you all do in these situations.

If you are running a sleep study, and you are simply using two groups. What do you do if you know people in these 2 groups (and in the real world) have VERY different sleep patterns?
 
#2
Good day everyone - I am curious as to what you all do in these situations.

If you are running a sleep study, and you are simply using two groups. What do you do if you know people in these 2 groups (and in the real world) have VERY different sleep patterns?
Can you give a little bit more information about your study? Are the two groups randomized? Is grouping your main exposure (independent variable) of interest? What is your outcome measure?
 
#3
No problem.

Two groups are randomized
This is based on us testing two separate treatments

There is two dependent variables (two different questionnaires), based on self reported measures.