The aim of three-dimensional conformal therapy (3DCRT) is to treat the Planning Target Volume (PTV) to the prescribed dose while reducing doses to normal tissues and critical structures, in order to increase local control and reduce toxicity. The evaluation tools used for optimizing treatment techniques are three-dimensional visualization of dose distributions, dose-volume histograms, tumor control probabilities (TCP) and normal tissue complication probabilities (NTCP). These tools, however, do not fully quantify the conformity of dose distributions to the PTV. Specific tools were introduced to measure this conformity for a given dose level. We have extended those definitions to different dose levels, using a conformity index (CI). CI is based on the relative volumes of PTV and outside the PTV receiving more than a given dose. This parameter has been evaluated by a clinical study including 82 patients treated for lung cancer and 82 patients treated for prostate cancer. The CI was low for lung dosimetric studies (0.35 at the prescribed dose 66 Gy) due to build-up around the GTV and to spinal cord sparing. For prostate dosimetric studies, the CI was higher (0.57 at the prescribed dose 70 Gy). The CI has been used to compare treatment plans for lung 3DCRT (2 vs 3 beams) and prostate 3DCRT (4 vs 7 beams). The variation of CI with dose can be used to optimize dose prescription.