Does Smoking Cause Lower Educational Attainment and General Cognitive Ability? Triangulation of causal evidence using multiple study designs

Objectives Observational epidemiological studies have found associations between smoking and both poorer cognitive ability and lower educational attainment; however, evaluating causality is more challenging. We used two complementary methods to attempt to ascertain whether smoking causes poorer cognitive ability and lower educational attainment. Design A cohort study (Study One) and a two-sample Mendelian randomization study using publicly-available summary statistics (Study Two). Setting The Avon Longitudinal Study of Parents and Children (ALSPAC), a birth-cohort study based in Bristol, United Kingdom, and general population samples from published genome-wide association studies (GWAS). Participants Up to 12,004 young people in ALSPAC (complete case analysis N = 2,107) (Study One and Study Two), and summary statistics from three previously published GWAS (not individual-level data) (Study Two). Main outcome measures Cognitive ability at age 15 (assessed via the Wechsler Abbreviated Scale of Intelligence) and educational attainment at age 16 (assessed via school records) (Study One), and educational attainment (measured as years in education) and fluid intelligence from previously published GWAS (Study Two). Results In Study One, heaviness of smoking at age 15 was associated with lower cognitive ability at age 15 and lower educational attainment at age 16. Adjustment for potential confounders and earlier cognitive ability or educational attainment attenuated findings although evidence of an association remained (e.g., fully adjusted cognitive ability beta - 0.736, 95% CI −1.238 to −0.233, P = 0.004; fully adjusted educational attainment beta −1.254, 95% CI −1.597 to −0.911, P < 0.001). Comparable results were found in sensitivity analyses of multiply imputed data. In Study Two, two-sample Mendelian randomization indicated that both smoking initiation and lifetime smoking lower educational attainment and cognitive ability (e.g., smoking initiation to educational attainment inverse-variance weighted MR beta −0.197, 95% CI −0.223, −0.171, P = 1.78 × 10 −49 ). Educational attainment results were robust to various sensitivity analyses, while cognition analyses were less so. Conclusions Our results provide evidence consistent with a causal effect of smoking on lower educational attainment, although were less consistent for cognitive ability. The triangulation of evidence from observational and Mendelian randomisation methods is an important strength for causal inference. Summary boxes What is already known on this topic Associations are seen between smoking and both educational attainment and cognition. These is some evidence that educational attainment might causally influence smoking, but causality in the opposite direction has not been assessed. What this study adds Using multiple methodologies, we found evidence consistent with a causal effect of smoking on lower educational attainment. An exploration of potential mechanisms could inform the development of interventions to mitigate this risk.