Cervical cancer mortality trends and inequities, 1999 to 2023: Joinpoint and spatial analyses integrated with...

Cervical cancer is preventable and screen-detectable, yet it continues to cause significant mortality, with inequities across populations and regions. This study aims to identify modifiable causal exposures contributing to cervical cancer and explore how they vary across different populations. The objective is to inform targeted prevention strategies and optimize resource allocation to address these disparities. We used a two-part complementary design. First, cervical cancer mortality data (1999-2023) were extracted from Centers for Disease Control and Prevention Wide-ranging Online Data for Epidemiologic Research and stratified by age, sex, race/ethnicity, census region, urbanization level, and state. Death counts and age-adjusted mortality rates were calculated. Joinpoint regression estimated average annual percent change (AAPC) and segment-specific annual percent change, and spatial patterns were visualized. Second, we conducted two-sample Mendelian randomization (MR) to evaluate genetically predicted trace element/nutrition-related exposures in relation to cervical cancer risk. Analyses included allele harmonization and instrument selection; inverse-variance weighted estimation was the primary approach (fixed vs random effects based on heterogeneity). Robustness was assessed using MR-Egger, weighted median, MR-Pleiotropy RESidual Sum and Outlier, and sensitivity analyses (leave-one-out, funnel, and scatter plots). Reverse MR was performed to examine the causal direction. Mortality declines differed markedly across strata. Age-adjusted mortality rate decreased significantly among those aged 65 to 74 and ≥85 years (AAPC = -1.81 and -2.81, respectively), whereas the decline in the 25 to 34-year group was smaller and not statistically significant (AAPC = -2.10, 95% confidence interval crossed 0). Heterogeneous declines were also observed by race/ethnicity, region, and urbanization level. In MR, genetically predicted vitamin C (7 single-nucleotide polymorphism instruments) showed a modest inverse association with cervical cancer risk in inverse-variance weighted fixed-effects analysis (odds ratio = 0.9962, 95% confidence interval = 0.9924-1.0000; P = .048). Weighted median and MR-Egger were directionally consistent but nonsignificant. No strong evidence of pleiotropy or heterogeneity was detected, and sensitivity analyses were generally robust. Reverse MR yielded no consistent evidence. Cervical cancer mortality has declined overall, but inequities persist, underscoring the need to pair vaccination and screening with more equitable access and continuity of care. MR provides suggestive genetic support for a small protective association of vitamin C, warranting validation with stronger instruments and larger datasets.