Background: The study of the spatial variation in disease incidence is a vital component of descriptive epidemiology. The production of attractive and informative disease maps complements any formal statistical analyses of spatial variations and often their visual impact will influence the recipient of the information much more than the accompanying statistics. Like any other graphical display, however, a map can both inform and mislead.
Methods: We describe methodological alternatives and pitfalls for the production of disease maps using data on recent suicides in England and Wales as an illustrative example.
Results: Decisions concerning the methods of mapping of suicide rates, including smoothing, choice of grouping criteria and colouring scheme dramatically influence the resulting map and how it might be interpreted by the reader.
Conclusions: This study provides clinicians and other non-specialist research workers an insight into the methodological pitfalls of disease mapping and atlas production and should also act as a methodological framework for a critical appraisal of published maps and atlases.