Current techniques for measuring dental fluorosis: issues in data analysis.

Abstract

The currently popular scoring systems used to diagnosis fluorosis use different measurement units, evaluate variable numbers of sites per person, and involve non-comparable groupings of clinical symptoms. Although none of these factors is related to the level of fluoride exposure in the examined population, their combined effect produces fluorosis prevalence values for a population which vary considerably among and within these scoring systems. Intrinsic factors for a scoring system include the inclusion of a questionable category, the minimal level of fluorotic involvement, and the number of affected sites within a subject required for case definition. Thus, a case definition of fluorosis for each scoring system, although not mandatory, would certainly be desirable so that dental epidemiologists and clinical investigators can interpret fluorosis scores relative to risk assessment. On the other hand, ratios of fluorosis prevalence magnitudes, as evidenced by odds ratios, can be more stable between scoring systems when groups with different fluoride exposure levels are compared. There is a strong correlation between extent and specific measures of fluorosis severity for Dean's Index (DI) and the Tooth Surface Index of Fluorosis (TSIF) scoring system, as well as within each scoring system separately. Parallel patterns in fluorosis severity were found among groups with different fluoride exposure on severity levels of fluorosis may be better understood by using relative measures rather than by using differences in severity levels.