Our objective was to test, in tube-fed patients whether treatment with antibiotics, the presence of hypoalbuminemia, or the use of hypertonic tube feeding is associated with a higher incidence of diarrhea; how often tube feeding actually causes diarrhea; and whether administration of a Lactobacillus preparation reduces the incidence of diarrhea. Our study design included a randomized, double-blind, placebo-controlled trial of patients on tube feeding for at least 5 days. Stool weights and clinical assessment of bowel function were used as outcome measures. Diarrhea was defined as > 200 g of stool, or three or more liquid stools, in any 24-h period. The tube feeding was considered responsible for diarrhea only when the latter resolved on discontinuation of the feeding. When diarrhea did not resolve, other causes were sought. Of 62 patients enrolled, 41 reached a trial end point. Of these, 34 completed 5 days of feeding without diarrhea, and 7 experienced diarrhea. Although diarrhea was associated with hypoalbuminemia and with protracted treatment with antibiotics, in only 1 subject who had a history of gastric surgery was it caused by tube feeding. The other 6 cases of diarrhea were caused by factors other than tube feeding, mainly drugs administered through the tube. Lactobacillus treatment did not alter the risk of diarrhea. Diarrhea occurs more commonly in tube-fed patients who have low serum albumin levels and have been treated with antibiotics for long periods, but these associations are generally not causal. Hypertonic feeding formulas are not associated with increased risk of diarrhea. Most cases of diarrhea in tube-fed patients are caused by factors extraneous to the tube feeding.