Is reading for pleasure in adolescence good for mental health? A counterfactual and within-person analysis in a large longitudinal study

Abstract Purpose: Reading has been proposed to be a protective factor in mental health; however, testing this is made challenging by the vulnerability of the reading-mental health association to confounding. Methods: We used two complementary approaches to address this: propensity score matching and random intercepts cross-lagged panel models (RI-CLPM) in a large longitudinal study: the Zurich Project on Social Development from Childhood to Adulthood (z-proso). For the counterfactual analyses, mental health outcomes of anxiety, depression, and psychosis-like symptoms were measured at ages 17 and 20 and reading engagement was measured at ages 15 and 17. Matching variables included a wide range of factors potentially related to reading and mental health outcomes. Results: There was no consistent effect of reading for pleasure on anxiety, depression, and psychosis. The only significant effects were detrimental effects of reading on anxiety and psychosis at age 20 when using non-bi-partite matching. Conclusion: Reading for pleasure without considering content and style of engagement may not be a protective factor in adolescent mental health.