Do you think that a cohort incidence study would be worth while? This question should be addressed from the standpoint of statistical power, of public health, and of scientific/biologic interest.