methods for selection a subset of population to measure biomarkers

#1
Hi! In a quite large time-to-event clinical trial project, the population is very large (more than 1,000 people). I want to measure the subjects' bio-markers. But the issue is: the cost of measuring everyone's bio-marker is super high. So I probably would like to choose a subset of these patients to have their blood-sample measured. But I don't know how to choose this subset so it won't bias the conclusion of these markers on the outcome. Does anyone has a good idea or can offer a hint, please?
Look forward to your professional advice. Thank you!
 

hlsmith

Omega Contributor
#2
We need more info!!


Will they have to come in to get an EXTRA measurement? Do they find out the results of the measurement? Primarily, random sampling is your friend, however if this is an extra measurement you need to make sure a bias (systematic error) does not affect who opts to come in.