The present prospective test study of hypotheses addressed the question whether psychological factors are predictive of survival time in lung cancer patients. The hypotheses were: Emotional distress, depression and depressive coping are associated with shorter survival; hope and active coping with longer survival. The study was based on a sample of n = 103 patients who were investigated post-diagnosis and before the beginning of primary treatment. Emotional distress and hope were assessed by clinical scales (self-reports and interviewer ratings), depression by the Depression Scale of von Zerssen, depressive coping and active coping by the Freiburg Questionnaire on Coping with illness by Muthny. At follow-up, which took place three to five years later, n = 74 patients had died, for n = 29 patients the survival data are censored. The prediction of the survival time was performed applying multivariate analyses (Kaplan-Meier-method, Cox-Regression), adjusting for biological risk factors (histological classification, stage of the disease, type and amount of treatment, Karnofsky performance status, age). Results were as follows: Active coping and hope were associated with longer survival, emotional distress, depression and depressive coping with shorter survival, respectively. These associations were found consistently across assessment methods. The predictive effects of coping and distress were statistically independent of the influence of the somatic risk factors. The best psychological predictor was the interviewer rating of active coping. Its predictive power equalled that of the Karnofsky performance status. However, there was evidence that the effects of the psychological factors varied somewhat in interaction with treatment modalities. The findings are discussed from a methodological perspective. Possible causal models and mechanisms are presented which could account for interactions of psychological measures and the course of the disease: Thus, it can be conceived that psychological effects were mediated by patients' compliance with medical treatment. In addition, it cannot be ruled out that psychological factors themselves were influenced by the physical status of the patients at the time of entry to the study.