+Note different x-axis. *Stratified exposure totals were not given for the following studies: Phiri et al. (N = 273) [56], Mihrshahi et al. (N = 366) [57], Kurup et al. (N = 554) [58], and Gamboa et al. (N = 194) [24], Aimpun et al. (N = 553) [59], and Lello et al. (N = 330) [60]. *A 95% confidence interval/standard error (SE) was not available so it was not included in the forest plot: Lello et al. (odds ratio: 0.81) [60]. *Adjusted effect estimate: Phiri et al. [56]and Mihrshahi et al. [57]. *Inverted Log [odds ratio] and SE from effect estimate of barefoot exposure: Phiri et al. [56], Modjarrad et al. [25], Mihrshahi et al. [57], Kurup et al. [58], Gunawardena et al. [23], Gamboa et al. [24]. *Log [odds ratio] and SE calculated from raw data: Martinez et al. [61], Liabsuetrakul et al. [62], and Khan et al. [63].