with bilaterally symmetrical and diffuse disturbances in enamel mineralization that might be mistaken for fluorosis. More recently, Rweneyonyi et al. (1999) reported higher prevalences of severe enamel fluorosis at higher altitudes than at lower altitudes in Ugandan populations with the same water fluoride levels.
Some evidence from animal studies indicates that genetics might contribute to susceptibility to enamel fluorosis (Everett et al. 2002). It has also been proposed that use of the antibiotic amoxicillin during infancy might contribute to the development of enamel fluorosis of the primary teeth (Hong et al. 2004).
A number of review articles evaluate the strengths and deficiencies of the various indexes used to diagnose and characterize the degree of enamel fluorosis (Clarkson 1989; Ellwood et al. 1994; Kingman 1994; Rozier 1994). In general, the following observations may be made:
The various indexes use different examination techniques, classification criteria, and ways of reporting data. All indexes are based on subjective assessment, and little information is available on their validity or comparability. Prevalence data obtained from these indexes also can vary considerably because of differences in study protocols and case definitions. Nevertheless, the American Dental Association (2005) considers severe and even moderate fluorosis “typically easy to detect.”
Examiner reliability is an important consideration in evaluation studies. Systematic interexaminer variability has been reported (Burt et al. 2003). Rozier (1994) noted that only about half the studies available in 1994 provided evidence that examiner reliability was evaluated. Although almost all of those assessments were conducted in populations in which severe enamel fluorosis was very rare, they showed an acceptable level of agreement.
Agreement among examiners tends to be lower when enamel fluorosis is recorded at the level of the tooth or tooth surface than when it is recorded at the person level.
In many reviews and individual studies, all levels of enamel fluorosis severity are grouped together. This approach is less problematic at comparatively low levels of fluoride intake, where all or almost all of the cases are mild or moderate in severity. At higher intake levels, such as those typically found in communities with water fluoride concentrations at the current MCLG of 4 mg/L or the current SMCL of 2 mg/L, it is more informative to report results for the different levels of fluorosis severity. Those reviews in