Skip to main content
. Author manuscript; available in PMC: 2023 Mar 24.
Published in final edited form as: Sci Total Environ. 2022 Aug 6;850:157830. doi: 10.1016/j.scitotenv.2022.157830

Figure 2:

Figure 2:

Results from 100 repeated holdouts testing the association between the phthalate index and latency to the first growth spurt, adjusting for sex, SGA, NICU morbidity score, and year of birth (N=101; training n~40, validation n~61). A. shows the distribution of WQS and WQS2 estimates and B. shows the distribution of weights for each phthalate metabolite. For both A. and B., data points represent the estimate for each holdout and the closed diamond shows the mean effect estimate. For A., the boxplot shows the 25th, 50th, and 75th percentiles and the whiskers show the 2.5th and 97.5th percentiles and for B., the boxplot shows the 25th, 50th, and 75th percentiles and the whiskers show the 10th and 90th percentiles.