Methods

In this cross-sectional epidemiologic investigation, we merged county-level fluoridation data from the Centers for Disease Control and Prevention’s Water Fluoridation Reporting System with dental caries data from 1999-2004 and 2011-2014 cycles of the National Health and Nutrition Examination Survey (NHANES).³ Counties were classified as predominantly fluoridated where 75% or more of the population was served by fluoridated water and otherwise as less fluoridated. Dental caries was quantified as the sum of decayed and filled primary (ie, deciduous) tooth surfaces (dfs) for children aged 2 to 10 years (n=5835), and the sum of decayed, missing, and filled permanent tooth surfaces (DMFS) for those aged 6 to 17 years (n=8384). The University of North Carolina’s Office of Human Research Ethics determined that this study does not constitute human subjects research. We regressed dental caries on family income to poverty ratio (modeled as a continuous variable), county fluoridation status (dichotomous variable), and the interaction of those variables, adjusting for survey cycle, age, sex, race/ethnicity, rural-urban location, and time since last dental visit. Analysis took account of the NHANES complex survey design, producing estimates for the US child and adolescent populations. Data analysis was conducted between July 2017 and September 2018; SAS, version 9.3 (SAS Institute Inc), software was used. Significance was determined at P<.05.

Results

Children in families with low income—but not those living in poverty—were less likely than other income groups to live in a predominantly fluoridated county (Table). Inverse income gradients in dental caries were most pronounced in the primary dentition.

In predominantly fluoridated counties, the income gradient in dfs was attenuated by 41% (estimated slope [B]=–0.62; SE, 0.13) compared with less-fluoridated counties (B–1.06; SE, 0.18) and the interaction was statistically significant (P=.03) (Figure, A). Absolute and relative fluoridation-related caries reductions were most pronounced for the lowest income level. For permanent teeth, there was 18% attenuation of the income gradient in DMFS in predominantly fluoridated counties (B=–0.32; SE, 0.06) compared with less-fluoridated counties (B=–0.39; SE, 0.08), although the interaction was not significant (P=.49) (Figure, B). The interaction with income was significant (P=.002) under a more restrictive assumption that the percentage of population living in a fluoridated area, when modeled as a continuous variable, has a linear effect on the income gradient in DMFS.

Discussion

To our knowledge, this is the first US study to show evidence that water fluoridation attenuates income-related inequalities in dental caries. The degree of attenuation was less pronounced in the permanent dentition, possibly because the level of decay was about half that of primary teeth. Greater attenuation in the permanent dentition might be seen in early adulthood, as the burden of DMFS doubles between adolescence and early adulthood.⁴

One limitation of this study is its cross-sectional study design, which restricts causal interpretations. Also, in the absence of information about individuals’ consumption of water, county of residence was used as a proxy for exposure to fluoridation. The resulting misclassification of exposure likely biases measures of association towards the null.

These findings have public health relevance, as dental caries in the primary dentition affects one-half of children aged 6 to 8 years.³ In 2013, the United States spent $26.9 billion on child and adolescent oral health, exceeding the combined spending on asthma, upper respiratory tract infections, other infectious diseases, and anxiety.⁵ Economic modeling suggests that fluoridation of currently unfluoridated US communities with populations of 1000 or more would annually save $2.5 billion in costs related to dental caries.⁶ The current findings provide additional support for water fluoridation as a means toward the Healthy People 2020 goals of achieving health equity, eliminating disparities, and improving the health of all groups.

References