ABSTRACT
Epidemiology is the study of the distribution of disorders in defined populations. The populations studied are commonly defined in terms of age, sex or geographical location, but any measurable characteristic could be used. Comparisons of the frequency of conditions across cultures can be very misleading if the samples are not based on defined populations. In order to appreciate the potential contribution of epidemiology to psychiatry it is necessary to distinguish between the terms 'incidence' and 'prevalence'. There is a way of avoiding this ethnocentric bias, namely to develop psychiatric interviews in non-Western countries based on the complaints of people attending traditional healers and taking account of folk categories of illness. Some patients with an insidious onset of schizophrenia may not even be taken to a traditional healer, but might live out their lives disabled but sheltered by their family.
