The anecdotally reported putative influence of psychosomatic factors on the clinical manifestation of cancer was examined in a prospective study on the relationship of typical attitudes and subsequent cancer incidence. The concept included an assessment of the typical reaction towards an important event and its duration rather than counting life events. The hypothesis - as derived from a preceding case-control study - was that adverse events causing a chronic conflict eventually leading to hopelessness and (masked) depression increase the susceptibility for malignant growths. In the cohort of 1,353 persons interviewed in 1966 (with a questionnaire containing 109 items), 204 cases of cancer were identified until 1976. Persons reporting hopelessness showed a steadily increasing risk ratio, depending on the intensity of this reaction. No risk elevation was found in those reporting anger and excitement after an important event. The duration had an influence in that already after 3 years of an uninterrupted hopelessness and depression the risk ratio was elevated and remained increased. Other diseases did not show the same consistent pattern. A gradual increase of cancer became evident more than 3 years after an event causing unresolved depression and hopelessness of more than 1 year's duration. These results have to be seen in the context of individual exposure to carcinogens as well as important host factors. They seem, however, to indicate that increased susceptibility to cancer and a timely relationship to its manifestation will have to be considered in a defined proportion of the population, determinable by psychosocial methods.