OBJECTIVE To determine the difference in the incidence of diarrhea among subjects given one of three formulas with varying fiber concentrations administered by nasogastric (NG) tube, variables affecting incidence of diarrhea, discomforts other than diarrhea associated with NG tube feedings, and effects of changing from continuous to interval feedings on incidence of diarrhea and discomforts. METHODS Prospective, double-blind, randomized study. METHODS Midwestern tertiary care center otolaryngology nursing unit. METHODS Eighty randomized subjects who were 18 years or older, English-speaking, and undergoing head and neck cancer surgery that required an NG tube postoperatively and who had no gastrointestinal (GI) illness within two weeks prior to surgery. METHODS Subjects received continuous administration of formula containing no fiber, 7 gms/L of fiber, or 14 gms/L of fiber until they reached the caloric intake goal and then were advanced to interval feedings. Patients' medical records provided past medical history and information on medication administration. A bedside flow sheet was used for documenting incidence of diarrhea and other GI discomforts. METHODS Amount of fiber in the formula administered, patient's genders and prior food aversions, and antibiotics' effect on diarrhea and other GI discomforts. RESULTS Multiple logistic regression showed significant odds ratios (ORs) for developing diarrhea in female subjects (OR = 7.96), subjects who had prior food aversions (OR = 2.67), and subjects receiving broad spectrum antibiotics (OR = 3.22). Diarrhea was four times more likely to occur in males who received fiber-free formula. Of all subjects, 70% experienced GI discomforts with continuous feedings, and 50% experienced discomforts when advanced to interval feedings. CONCLUSIONS Fiber formulas reduced the incidence of diarrhea in male subjects but not in female subjects. Antibiotics' effect on diarrhea paralleled the findings of other studies. CONCLUSIONS Use formulas with fiber for males. Liquid stools do not require interruption of tube feeding; GI discomforts warrant interruption. Interval feeding schedules require monitoring similar to continuous feeding schedules.