Incidence and prevalence of cleft lip and palate: what we think we know.

The descriptive epidemiology of the spectrum of orofacial cleft disorders has many methodologic problems, including (1) casefinding using data sources such as birth certificates, fetal death certificates, and hospital records that often produce ascertainment bias, selection bias, or both and (2) the multiple comparisons problem (i.e., the chance occurrence of statistically significant findings). The resultant incidence and prevalence rates from studies with inadequate designs or inadequate data are limited and may be misleading. A variety of reasons is advanced to explain the wide discrepancies in reported statistics on orofacial clefting from different geographic areas, ethnic groups, and time periods. Specific recommendations are offered for producing better epidemiologic data. An example of how higher quality descriptive statistics can be used for future hypothesis testing is also provided.