Table 2. Unadjusted and Adjusted Associations Estimated From Linear Regression Models of Fluoride Exposure Variables and FSIQ Scores.
| Variable | Difference (95% CI) | |||
|---|---|---|---|---|
| Unadjusted | Adjusted Estimates, Regression Coefficients Indicate Change in Outcome pera | |||
| 1 mg | 25th to 75th Percentiles | 10th to 90th Percentiles | ||
| MUFSGb,c | −2.60 (−5.80 to 0.60) | −1.95 (−5.19 to 1.28) | −0.64 (−1.69 to 0.42) | −1.36 (−3.58 to 0.90) |
| Boys | −5.01 (−9.06 to −0.97) | −4.49 (−8.38 to −0.60) | −1.48 (−2.76 to −0.19) | −3.14 (−5.86 to −0.42) |
| Girls | 2.23 (−2.77 to 7.23) | 2.40 (−2.53 to 7.33) | 0.79 (−0.83 to 2.42) | 1.68 (−1.77 to 5.13) |
| Fluoride intaked,e | −3.19 (−5.94 to −0.44) | −3.66 (−7.16 to −0.15) | −2.26 (−4.45 to −0.09) | −3.80 (−7.46 to −0.16) |
Abbreviations: FSIQ, Full Scale IQ; HOME, Home Observation for Measurement of the Environment; MUFSG, maternal urinary fluoride adjusted for specific gravity.
a
Adjusted estimates pertain to predicted FSIQ difference for a value spanning the interquartile range (25th to 75th percentiles) and 80th central range (10th to 90th percentiles): (1) MUFSG: 0.33 mg/L, 0.70 mg/L, respectively; (2) fluoride intake: 0.62 mg, 1.04 mg, respectively.
b
n = 512.
c
Adjusted for city, HOME score, maternal education, race/ethnicity, and including child sex interaction.
d
n = 400.
e
Adjusted for city, HOME score, maternal education, race/ethnicity, child sex, and prenatal secondhand smoke exposure.