Measures of life crisis, manifest distress and maladaptive coping were used to predict future illness behavior in male college students. A group of 92 Ss was screened by an internist's examination as being free from disease, and was administered a series of self-ratings measuring the above dimensions. One year later, 79 of these students were available to be recontacted and were asked to describe their health records during the intervening time; 65 (82%) replied. Of the total number, 23 reported being ill and seeking medical treatment, whereas the remaining 42 either indicated no symptoms or else treated themselves with proprietary medication or rest in bed. Analysis of the 2 groups' premorbid reports indicated that the treatment-seeking Ss had scored significantly higher on each dimension, using t test comparisons. Secondly, independently established criterion scores were applied to this sample. Chi square analysis indicated that 71% of the cases could be correctly designated using this method-ie, before becoming ill, Ss who later sought care for their symptoms were more likely to have scored above the cut-off points than those who did not eventually seek care. We also scaled degree of incapacitation along a 0-10 continuum and correlated such illness behavior with the premorbid scores. Each measure correlated significantly with the criterion and with each other. A multiple R of 0.43 was obtained when four scale scores were used in a regression equation to estimate degree of incapacitation. These results suggest that premorbid indicators of unresolved life stress accurately predict who will seek care for illness.