We compare first scattered point dose spread arrays generated by Monte Carl
o and an analytic method. The analytic method models energy deposition usin
g Klein-Nishina cross sections for Compton scatter and approximations for e
lectron transport. Assumptions in the analytic method are shown to be valid
within a region of the point dose spread kernel in which meaningful compar
isons can be made. Differences between the models are less than 10% for the
forward scatter directions for radii greater than the electron range assoc
iated with the first scattered Compton photon. Differences in the backscatt
er region are discussed and indicate that the analytic model is useful for
identifying large errors that might be present in numerically generated fir
st scatter point dose spread arrays. The analytic method is simple and usef
ul for validating first scatter kernels. (C) 1999 American Association of P
hysicists in Medicine. [S0094-2405 (99)02305-6].