Personality test data and patterns of life changes over a 3-year period were collected on 73 male medical students to determine the influence of these mediating variables on the occurrence of illness. The results showed that life change units tended to be stable over time within groups which had high and low scores and psychological attributes were significantly different in the two groups. It was found that the subjects who had sustained high life change in all 3 years (the Sustained Group) were associated with a significantly greater number of health changes of more serious kinds than the subjects who had high life change in year 1 but low life change in the subsequent 2 years (the Short Term Group) and that the Short Term Group had greater psychological strengths than the Sustained Group. These findings are congruent with the data from a slightly different analysis in which healthy subjects in the original High Life Change Group obtained lower scores in subsequent years while ill subjects maintained high scores and the healthy subjects showed greater psychological strengths than ill subjects. The data were interpreted to show evidence that psychological characteristics do influence the occurrence of illness and that health is more likely to be impaired by sustained stress than by a short term major crisis.