*2.5. Data Abstraction*

The following information was abstracted from the articles identified in the full-text review: author, publication year, sample size, sampling method, time between assessment tool and outcome, assessment tool(s) used, age at which assessment tool was used, outcome(s), age at which outcome was evaluated, and measure(s) of association. For studies that used an assessment tool at multiple time points without separate effect measures, we only reported the greater duration from child development assessment to outcome (e.g., for tool applied at age 11 and age 16 and outcome measured at age 19, we only reported association between age 11 to age 19). In some studies, more than one assessment tool was used, including some child development assessment tools that were not on our pre-specified list. In these cases, only the associations between the tools on the list and our outcomes of interest were reported. Effect measures are presented adjusted for standard sociodemographic factors (i.e., parent education, sex, and wealth) unless otherwise stated. Regression coefficients and odds ratios are presented with statistical significance, and confidence intervals when measures of statistical significance were unavailable.

**Figure 1.** A Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flowchart of the selection process of published studies.
