Nonlinear thresholds and bidirectional effects: Paternal smoking’s dose-response impact on mental health in 21,212 Chinese preschoolers

J Affect Disord. 2025 May 1:S0165-0327(25)00756-6. doi: 10.1016/j.jad.2025.04.169. Online ahead of print.

ABSTRACT

BACKGROUND: Paternal smoking is a modifiable risk factor implicated in child mental health, yet its nonlinear dose-response relationship and residual risks post-cessation remain unclear. This study evaluates the association between paternal smoking exposure, cessation, and mental health risks in Chinese preschoolers.

METHODS: This cross-sectional study, conducted from February to March 2025, enrolled 21,212 parent-child dyads from 189 kindergartens in a western city, China. Children aged 3-6 years were assessed for mental health using the validated Chinese Strengths and Difficulties Questionnaire (SDQ), yielding a Total Difficulties Score (TDS) and prosocial behavior subscale scores. Paternal smoking history was self-reported and categorized as never, former, or current smokers, with current smokers’ cumulative exposure quantified in pack-months (PM: mild, <15; moderate, 15-30; heavy, >30). Multivariable regression models, adjusted for child factors, for socioeconomic and parental confounders, examined associations between paternal smoking exposure and SDQ outcomes. Restricted cubic splines assessed nonlinearity, with thresholds identified via derivative analyses.

RESULTS: Among 21,212 children (mean age 4.81 ± 0.89 years; 48.2 % female), 18.5 % had a TDS exceeding the SDQ clinical threshold (>14).. At-risk children exhibited socioeconomic disparities (lower parental education, rural residence, poverty) and higher paternal heavy smoking. Heavy paternal smoking (>30 PM) elevated TDS risk by 23 % (Adj. OR = 1.23, p < 0.001), while low exposure (<15 PM) unexpectedly reduced risk (Adj. OR = 0.87, p = 0.004). Former smokers’ children had near-significant TDS elevation (Adj. OR = 1.26, p = 0.052). Nonlinear analysis revealed thresholds at 15.0 PM (rapid risk accumulation) and 19.7 PM (saturation phase), delineating distinct risk trajectories. Prosocial behavior showed bidirectional effects-improved with low exposure (Adj. OR = 1.14, p < 0.001) but impaired with heavy smoking (Adj. OR = 0.89, p = 0.008) and former smoking (Adj. OR = 0.72, p < 0.001). Heavy smoking increased TDS by 0.61 points (95 % CI: 0.41-0.81, p < 0.001).

CONCLUSIONS: Paternal smoking demonstrates a nonlinear, threshold-driven association with child mental health, with residual risks persisting post-cessation. The 19.7 PM threshold marks critical saturation in neurodevelopmental risk, underscoring the need for stratified preconception interventions targeting heavy smokers. Public health policies must integrate paternal cessation counseling, anchored at this threshold, to disrupt intergenerational mental health disparities.

PMID:40318793 | DOI:10.1016/j.jad.2025.04.169