Abstract
BACKGROUND: Limited evidence links paternal drinking (41.5% prevalence) to preschoolers' behavioral difficulties in Western China, necessitating Strengths and Difficulties Questionnaire-based studies to assess psychosocial outcomes. METHOD: From February 28 to March 5, 2025, a cross-sectional study enrolled 25,017 parent-child dyads across 189 public kindergartens in a western city, China, using stratified sampling. After excluding 2,408 non-parents and 1,397 dyads with invalid age data, 21,212 remained. Paternal alcohol intake was categorized as nondrinkers, ex-drinkers, or current drinkers (light [≤ 2.86 g/day], moderate [>2.86-20 g/day], heavy [>20-40 g/day]. Child mental health was assessed using the Strengths and Difficulties Questionnaire (SDQ), with total difficulties (TDS >14 vs. ≤14) and prosocial behavior (PB <6 vs. ≥6) as categorical outcomes. Multivariable logistic regression, including a combined model of drinking status and daily intake, examined associations with these outcomes, adjusting for child, parental, and household factors. Continuous scores (TDS, PB, and SDQ subscales) were analyzed using multivariable linear regression, with false discovery rate correction applied to subscale analyses. Stratified analyses tested effect modification by covariates. RESULTS: Paternal alcohol consumption was consistently associated with adverse child mental health outcomes. Ex-drinkers had elevated odds for both total difficulties (TDS >14; adjusted OR = 1.54, P <0.001) and reduced prosocial behavior (PB <6; adjusted OR = 1.43, P <0.001). Among current drinkers, dose-response patterns were evident: for TDS, odds rose from 1.21 for light drinking to 1.73 for heavy drinking, with each additional 100 g/week associated with 20% higher odds (P <0.001); For PB, the association with paternal alcohol intake was weak but statistically significant (OR = 1.01, p = 0.01), reflecting a flat trend without clear dose-response. Stratified analyses revealed stronger associations for TDS among boys, higher-income households, and children of non-working fathers; paternal effects were attenuated when mothers were current drinkers; for PB, maternal employment status was the sole significant modifier. CONCLUSIONS: Paternal current and ex-drinking were associated with adverse child mental health outcomes. Future longitudinal studies with interaction models and larger samples are needed to confirm mechanisms and refine high-risk subgroup identification.