MPOWER policy intensity and the global burden of smoking-attributable respiratory cancers: lag-aligned inequality regressions with Bayesian age–period–cohort projections (1990–2050)

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Background Smoking-related respiratory cancers, including lung, laryngeal and nasopharyngeal cancers, remain a major global public health challenge. The WHO MPOWER framework has guided international tobacco control, but most evaluations have focused on smoking prevalence and cessation rather than disease-specific burden. The present study introduces an ecological, lag-aligned inequality regression framework, linking MPOWER policy scores with age-standardised disability-adjusted life-years (DALYs) of smoking-related respiratory diseases derived from the Global Burden of Disease (GBD) study. Materials and methods We quantified the global burden of smoking-attributable respiratory cancers from 1990 to 2021 using GBD 2021 estimates, the WHO MPOWER policy database and World Bank ageing statistics. We analysed age-standardised mortality rates (ASMRs) and age-standardised DALY rates (ASDRs) by sex and region and projected trends to 2050 with a BAPC model. For the policy analysis, we ran associational, ecological, lag-aligned global inequality regressions, regressing the annual cross-country 90–10 ASDR gap (Δq_t) on the global-mean MPOWER score and global-mean ageing aligned by lags. Results Smoking-attributable ASDRs and ASMRs declined from 2008 to 2021, although disparities by sex and region persisted. BAPC projections suggest continued decline through 2050, particularly among women. Countries with higher MPOWER scores, especially for Monitor, Protect and Enforce measures, were associated with narrower ASDR gaps with slower progress among men. Conclusion Our findings suggest that robust MPOWER implementation is associated with reduced inequality in smoking-related cancer burden. Prioritising monitoring, protective measures and enforcement, while tailoring strategies by sex and age, may further accelerate the reductions. Future studies with unit-level data are needed to strengthen causal inference.