Determinants and pathways of smoking cessation among 147,691 adults living with disability

Abstract

OBJECTIVES: Adults living with disability face systemic and psychosocial barriers that make quitting smoking harder, yet the pathways driving this inequality remain poorly understood. This study identified determinants of smoking cessation and examined pathways linking disability severity with cessation among adults with disability. STUDY DESIGN: Cross-sectional study. METHODS: Data were drawn from the Annual Population Survey between 2017 and 2023, comprising 147,691 individuals who had ever smoked with self-reported disability. Multivariable logistic regression estimated demographic, socioeconomic, and psychosocial correlates of being an ex-smoker versus a current smoker. A generalised structural equation model was then used to describe path-specific associations from severe activity limitation to smoking cessation through employment status, housing tenure, and life satisfaction, using robust (Huber-White) variance estimators. A Karlson-Holm-Breen (KHB) decomposition was performed as a sensitivity analysis to adjust for rescaling in the logistic models. RESULTS: Overall, 65.7% of participants were ex-smokers. In adjusted models, older age, being married or cohabiting, and having a degree were associated with higher odds of cessation, whereas unemployment, economic inactivity, renting (housing tenure), and lower life satisfaction were associated with lower odds. Severe activity limitation was associated with reduced cessation (adjusted odds ratio = 0.88, 95% CI 0.86 to 0.90, p < 0.001). In the path model, this association was largely captured by path-specific coefficients for renting (log-odds = -0.55, 95% CI -0.58 to -0.53, p < 0.001), economic inactivity (-0.37, 95% CI -0.43 to -0.31, p < 0.001), unemployment (-0.15, 95% CI -0.19 to -0.11, p < 0.001), and lower life satisfaction (probit coefficient = -0.08, 95% CI -0.09 to -0.06, p < 0.001). The summed indirect association (-1.15, 95% CI -1.23 to -1.07, p < 0.001) and smaller residual direct component (-0.13, 95% CI -0.16 to -0.10, p < 0.001) indicated that most of the overall relationship was captured by socioeconomic and psychosocial pathways. The KHB sensitivity analysis produced directionally consistent total, indirect, and direct log-odds components (total = -0.37, indirect = -0.13, direct = -0.24; all p < 0.001), supporting the robustness of the decomposition. CONCLUSIONS: Among adults with disability, cessation is strongly patterned by socioeconomic position and wellbeing. Although cross-sectional, these findings support cessation strategies that combine individual support with action on material insecurity and psychosocial wellbeing.