If it is difficult to predict how many participants will be eligible or be attracted to a study, the optimum number should be specified. Responses such as "don't know" or "as many as we can recruit" to questions about the number of participants are not acceptable. A discussion of the problems faced in recruiting participants should be included.
Multicenter studies, in which data will be pooled and recruitment may
vary, present a special problem for investigators. The application should
provide information about the total picture, including both the number
of participants to be studied at the University or by University researchers
and information on overall recruiting.