Skip to main content
. 2022 Jul 11;130(7):076001. doi: 10.1289/EHP10197

Figure 4.

Figure 4 is a forest plot, plotting Study or subgroup and odds ratio inverse variance, random, 95 percent confidence intervals year, including (bottom to top) Road: Non noise specific: Total (95 percent confidence intervals), 1.80 [1.50, 2.17]; Test for subgroup differences: Chi squared equals 43.62, degrees of freedom equals 1 (uppercase p equals 0.00001); I squared equals 97.7 precent. Test for overall effect: uppercase z equals 6.26 (uppercase p equals 0.00001); Heterogeneity: Tau squared equals 0.16; Chi squared equals 466.45, degrees of freedom equals 21 (uppercase p equals 0.00001); I squared equals 95 precent. Test for overall effect: uppercase z equals 3.14 (uppercase p equals 0.002); Heterogeneity: Tau squared equals 0.01; Chi squared equals 17.44, degrees of freedom equals 6 (uppercase p equals 0.0008); I squared equals 66 precent; Subtotal (95 percent confidence intervals), 1.13 [1.05, 1.22]; Bartels and others [47], 1.01 [0.88, 1.15] 2021; Martens and others [34], 1.10 [1.02, 1.18] 2018; Evandt and others [37], 1.20 [1.11, 1.30] 2017; Bodin and others [33], 1.12 [0.96, 1.31] 2015; Frei and others [63], 1.22 [1.03, 1.44] 2014; Halonen and others [64], 0.99 [0.88, 1.12] 2012; Brink [56], 1.43 [1.18, 1.72] 2011; Road: Noise specific: Test for overall effect: uppercase z equals 8.28 (uppercase p equals 0.00001); Heterogeneity: Tau squared equals 0.11; Chi squared equals 95.40, degrees of freedom equals 14 (uppercase p less than 0.00001); I squared equals 85 precent; Subtotal (95 percent confidence intervals), 2.32 [1.90, 2.84]; Brink and others [46], 2.56 [2.27, 2.90] 2019; Evandt and others [37], 3.19 [2.68, 3.78] 2017; NORAH [42], 1.92 [1.63, 2.26] 2015; Brown and others [60], 2.55 [2.15, 3.02] 2015; Bodin and others [33], 2.44 [1.84, 3.24] 2015; Phan and others Thai Nguyen [57], 19.93 [8.27, 48.02] 2010; Phan and others Hue [57], 1.49 [0.98, 2.26] 2010; Phan and others Ho Chi Minh City [57] 1.13 [0.76, 1.70] 2010; Phan and others Hanoi [57], 1.28 [0.86, 1.90] 2010; Phan and others Da Nang [57], 12.49 [4.64, 33.60] 2010; Hong and others [61], 1.26 [0.74, 2.14] 2010; Ristovska and others [62], 2.45 [1.71, 3.50] 2009; Sato and others Kumamoto [59], 1.41 [0.94, 2.12] 2002; Sato and others Gothenburg [57], 3.50 [2.34, 5.22] 2002; and Sato and others Sapporo [59], 2.66 [1.20, 5.89] 2002 (y-axis) across Less disturbed, ranging from 0.02 to 0.1 in increments of 0.1 and 0.1 to 1 in increments of 0.9, and 1 to 10 in increments of 9 and 10 to 50 in increments of 40 (x-axis) for risk of bias, including selection bias, exposure assessment bias, bias due to confounding, and reporting bias.

Forest plot for the odds of being highly sleep disturbed by road noise per 10-dB increase in Lnight (combined estimate derived from all relevant outcomes within studies). Subgroups are presented for questions that mentioned noise as the source of the disturbance, and questions that did not specify noise as the source of the disturbance. Risk of bias: A: selection bias; B: exposure assessment; C: confounding; D: reporting bias. Green (+) denotes low risk of bias, red (–) denotes high risk of bias, yellow (?) denotes unclear risk of bias. Plots were generated using an inverse-variance (IV) random effects method across the full noise range for each individual study (not restricted to 40–65 dB Lnight). Note: CI, confidence interval; df, degrees of freedom; Lnight, nighttime noise; NORAH, Noise-Related Annoyance, Cognition and Health.