One of the most important parameters determining the sensitivity of X-ray telescopes is their effective area as a function
of the X-ray energy. The computation of the effective area of a Wolter-I mirror, with either a single layer or multilayer
coating, is a very simple task for a source on-axis at astronomical distance. Indeed, when the source moves off-axis the
calculation is more complicated, in particular for new hard X-ray imaging telescopes (NuSTAR, ASTRO-H, NHXM,
IXO) beyond 10 keV, that will make use of multilayer coatings to extend the reflectivity band in grazing incidence.
Unlike traditional single-layer coatings (in Ir or Au), graded multilayer coatings exhibit an oscillating reflectivity as a
function of the incidence angle, which makes the effective area not immediately predictable for a source placed off-axis
within the field of view. For this reason, the computation of the off-axis effective area has been so far demanded to raytracing
codes, able to sample the incidence of photons onto the mirror assembly. Even if this approach should not be
disdained, it would be interesting to approach the same problem from an analytical viewpoint. This would speed up and
simplify the computation of the effective area as a function of the off-axis angle, a considerable advantage especially
whenever the mirror parameters are still to be optimized. In this work we present the application of a novel, analytical
formalism to the computation of the off-axis effective area and the grasp of the NHXM optical modules, requiring only
the standard routines for the multilayer reflectivity computation.