Chronic Candida vulvovaginitis cannot, at present, be treated effectively with local or oral antimycotic medication. Göttlicher & Meyer [Vulvovaginalmykose. Klinische Ergebnisse einer epidemiologischen Langzeitstudie, Thieme, Stuttgart (1998); Mycoses41 (1998) 49] formulated a hypothesis to explain this fact. They say that stress is the main cause of vulvovaginal mycosis, and that frequently cited predisposing somatic risk factors are not causing the illness. Two research projects were carried out to test this hypothesis. The first project was conceived to provide direct evidence for two empirical statements derived from the Göttlicher-Meyer hypothesis: (i) stress as a predictor of vulvovaginitis is least valid as the commonly assumed somatic factors and (ii) combinations of factors that reliably discriminate index women from symptom-free women point to aspects of psychosocial development, particularly stress, and not to somatic factors. Between March and November 1999, 309 successive patients were randomly chosen from those successively treated in one in-patient and two out-patient settings. Of those chosen, 117 had had at least one episode of vaginal candidosis within the 2 years prior to enlistment in the study. The remaining 192 patients had experienced no such illness. For the 117 index patients both of the above empirical statements were confirmed by statistical analyses. The second project was designed to test predictions concerning the ineffectiveness of traditional antimycotic treatment derived from the Göttlicher-Meyer hypothesis. Three different empirical statements were derived from the hypothesis--each based on the assumption of a Poisson distribution of relapses that remains uninfluenced by treatment attempts. Each of the statements was tested and confirmed in independent samples--the first in the sample used in project one, the second in a sample of 206 women insured by public Healths Management Organizations (HMOs) and the third in a sample of 179 women insured by private HMOs. Each patient had evidenced Candida albicans vulvovaginitis at least once in the time interval between 1996 and 2000. Statistical analyses confirmed each of the empirical statements, thereby substantiating the Göttlicher-Meyer hypothesis. The authors conclude that (i) psychosocial factors, particularly stress, are the primary causes of Candida albicans vulvovaginitis. Accompanying somatic factors are of little statistical significance in explaining occurrence and relapse. (ii) Traditional antimycotic treatment influences only the symptoms of the illness, not its causes, and has no effect on the probability of a relapse. Increased treatment effectiveness can be achieved only at the cost of a complete re-orientation concerning the causes of vaginal mycosis. Interventions designed to strengthen the patients' immune response are among the most promising.