It was postulated that bile acids can interact with esophageal epithelia of rats, either by the gastro-duodeno-esophageal anastomosis operation, or by adding bile concentrate to the diet. These interactions would cause reflux diseases and eventually esophageal cancer. Rats gastrectomized and jejunostomized to allow bile acids to reflux into the esophagus developed many carcinomas in 50 weeks, while other modifications that kept bile out of the esophagus did not produce any carcinogenesis, thus supporting the hypothesis. Therefore, we conclude that bile acids refluxing into the esophagus of humans should also cause cancers, especially in Westernized societies with their high fat diets, which provide an abundant supply of bile. Bile acids can enter the model OE33 cells and activate the oncogene c-myc at pH 4, the gene complex NF-kappaB at pH 6.5, and start proliferation at neutral pH. 50% of Barrett's metaplasia contained activated c-myc, and 40% of Barrett's Esophagus patients contain activated NF-kappaB. Since normal human esophageal epithelia contained neither activated c-myc nor NF-kappaB, these activations must also occur in Barrett's patients. Acid suppression therapy is used to treat these patients, and will solubilize free bile acids and some of the glycine conjugates, allowing them to enter the epithelia. Taurine conjugates (20% of bile) will also enter the epithelia unaffected by acid suppression therapy. All these internalized bile acids will start carcinogenesis. Therefore, techniques to keep bile acids out of the stomach, or prevent them from reacting, must be developed, but until then, acid suppression therapy should be restricted, not promoted.