However, our study participants had very similar dietary patterns because all of them had their three daily meals in the same school central canteen. Therefore, potential confounding effect from diet, if any, would be likely to alter the observed associations modestly rather than materially. Finally, although we used a repeated-measure study protocol to increase the study efficiency, our sample size is modest and the study results should be interpreted with caution. Future studies with larger sample sizes are needed to confirm our study findings.