Skip to main content
. 2018 Oct 10;126(10):107001. doi: 10.1289/EHP3546

Figure 2.

The first row in the graphical representation consists of three scatter plots with regression lines plotting MUF T2 (y-axis) across MUF T1 (x-axis) (Pearson r equals 0.41), MUF T3 (y-axis) across MUF T1 (x-axis) (Pearson r equals 0.31), MUF T3 (y-axis) across MUF T2 (x-axis) (Pearson r equals 0.36), respectively, for MUF unadjusted where ICC is 0.37 (95 percent confidence intervals 0.34 to 0.4). The second row in the graphical representation consists of three scatter plots with regression lines plotting MUF T2 (y-axis) across MUF T1 (x-axis) (Pearson r equals 0.40), MUF T3 (y-axis) across MUF T1 (x-axis) (Pearson r equals 0.35), MUF T3 (y-axis) across MUF T2 (x-axis) (Pearson r equals 0.51), respectively, for MUF subscript SG where ICC is 0.39 (95 percent confidence intervals 0.36 to 0.42). The third row in the graphical representation consists of three scatter plots with regression lines plotting MUF T2 (y-axis) across MUF T1 (x-axis) (Pearson r equals 0.37), MUF T3 (y-axis) across MUF T1 (x-axis) (Pearson r equals 0.34), MUF T3 (y-axis) across MUF T2 (x-axis) (Pearson r equals 0.51), respectively, for MUF subscript CRE underscore 1 where ICC is 0.38 (95 percent confidence intervals 0.36 to 0.43). The fourth row in the graphical representation consists of three scatter plots with regression lines plotting MUF T2 (y-axis) across MUF T1 (x-axis) (Pearson r equals 0.38), MUF T3 (y-axis) across MUF T1 (x-axis) (Pearson r equals 0.35), MUF T3 (y-axis) across MUF T2 (x-axis) (Pearson r equals 0.52), respectively, for MUF subscript CRE underscore 2 where ICC is 0.40 (95 percent confidence intervals 0.35 to 0.42).

Pearson r correlations between pairs of trimesters (T1, T2, T3) and intraclass correlation coefficients (ICCs) across trimesters [with 95% confidence interval (CI)] for log10-transformed maternal urinary fluoride (MUF) levels without adjustment, with adjustment for specific gravity (MUFSG), and with adjustment for creatinine using two methods of adjustment (MUFCRE_1 and MUFCRE_2). Individual data points represent individual observations, solid lines represent regression lines.