The value of serum beta-carotene concentration as an indicator of steatorrhoea was investigated in 50 patients with steatorrhoea (fecal fat greater than 7 g/day), 53 controls, and 22 patients with gastrointestinal disease without steatorrhoea. In the control group, beta-carotene concentrations were normally distributed when plotted logarithmically. The mean value was 131 micrograms/dl. The lower limit of normal, based on a 2-SD confidence interval, was 47 micrograms/dl. beta-Carotene concentrations and fecal fat excretion were correlated in a reciprocal, hyperbolic function (r = -0.66). Twenty-nine of the 50 patients with steatorrhoea had beta-carotene concentrations less than 47 micrograms/dl (sensitivity 58%; specificity 93%). Referring to an additional cut-off point of 100 micrograms/dl, beta-carotene concentration had a sensitivity of 88%. These data show that low plasma beta-carotene concentrations (less than 47 micrograms/dl) can be regarded a specific and useful indicator of steatorrhoea and thus obviate fecal fat analysis. Values greater than 47 micrograms/dl, however do not exclude steatorrhoea. Based on a two-step interpretation beta-carotene is thus both a useful screening test for steatorrhoea (with a cut-off point of 100 micrograms/dl) and, more important, a valid, simple, and clinically practical alternative for fecal fat analysis (if values are less than 47 micrograms/dl).