In this longitudinal study using the LSAC cohort, the researchers examined the relationship between early childhood IQ and adolescent health behaviors at age 16-17. The findings are compelling: higher childhood IQ predicted both lower risk-taking behaviors and better health habits even after adjusting for an extensive set of covariates including family SES, parental health behaviors, non-cognitive traits, and regional disadvantage. The authors employed Gelbach decomposition and Cinelli & Hazlett sensitivity analyses, and the relationships held up remarkably well. The effect sizes were modest but statistically significant and comparable to other established predictors, which is noteworthy given that these behaviors compound across the lifespan.
The decomposition analyses revealed fascinating differences between the two behavioral domains. For risk-taking, family health behaviors and ethnocultural background were the primary confounders, while peer characteristics at age 14-15 accounted for roughly two-thirds of the IQ-risk-taking association (this suggests that higher-IQ adolescents may selectively associate with peers who reinforce prosocial norms). In contrast, the IQ-health habits relationship was more strongly confounded by socioeconomic disadvantage, implying that structural factors play a larger role in self-regulatory behaviors. This divergence is conceptually interesting and aligns with dual-process models of adolescent development, where risk-taking is more socially mediated and health habits more environmentally constrained.
While the study strengthens the case for IQ as an upstream determinant of health inequalities, I’m somewhat critical of the causal framing. The authors acknowledge they cannot make definitive causal claims, yet the discussion leans heavily toward interpreting IQ as a flexible resource that enables better decision-making. However, the peer selection findings could just as easily reflect gene-environment correlation rather than a cognitive advantage in navigating social environments. I’m also curious about the policy implications they propose. Interventions targeting health literacy are reasonable, but the suggestion that “improving cognitive ability” could reduce health disparities feels both vague and potentially troubling without more nuance about what that entails. That said, the study is methodologically solid and adds valuable evidence to the literature on cognitive ability and life outcomes.
Link to full article: https://doi.org/10.1016/j.ssmph.2025.101887
