This review of the medical literature from 1994 to 2003 summarizes the relationship between raw and cooked vegetables and cancer risk and examines whether they may affect cancer risk differently. Twenty-eight studies examined the relationship between raw and cooked vegetables and risk for various cancers. Twenty-one studies assessed raw, but not cooked, vegetables and cancer risk. The majority of these assessed risk of oral, pharyngeal, laryngeal, esophageal, lung, gastric, and colorectal cancers. Most showed that vegetables, raw or cooked, were inversely related to these cancers. However, more consistent results were found for oral, pharyngeal, laryngeal, esophageal, and gastric cancers. Nine of the 11 studies of raw and cooked vegetables showed statistically significant inverse relationships of these cancers with raw vegetables, but only 4 with cooked vegetables. The few studies of breast, lung, and colorectal cancers also suggested an inverse relationship with both raw and cooked vegetables, but these results were less consistent. In the two studies of prostate cancer, there was no association with either raw or cooked vegetables. One of two bladder cancer studies found an inverse relationship with cooked, but not raw, vegetables. Possible mechanisms by which cooking affects the relationship between vegetables and cancer risk include changes in availability of some nutrients, destruction of digestive enzymes, and alteration of the structure and digestibility of food. Both raw and cooked vegetable consumption are inversely related to epithelial cancers, particularly those of the upper gastrointestinal tract, and possibly breast cancer; however, these relationships may be stronger for raw vegetables than cooked vegetables.