Association between non-essential metal exposure and cognitive function in rural older adults: effect modification by essential metals
- VernacularTitle:非必需金属暴露与农村老年人认知功能的关联：必需金属的修饰效应
- Author:
Yong ZENG
1
;
Yang YANG
1
;
Xu WANG
1
;
Pengfei LI
1
;
Yong LYU
1
;
Linsheng YANG
2
Author Information
- Publication Type:Investigation
- Keywords: essential metal; non-essential metal; cognitive function; older adult; modification
- From: Journal of Environmental and Occupational Medicine 2026;43(8):979-987
- CountryChina
- Language:Chinese
- Abstract: Background China has the largest elderly population in the world, and age-related cognitive impairment has become an increasingly important public health issue. Objective To investigate the associations of exposure to selected essential metals (EMs) and non-essential metals (NEMs) with cognitive function among rural older adults and to examine whether EMs modified the association between NEW exposure and cognitive functions. Methods A total of 1045 adults aged 60 years and above were enrolled using baseline survey data from a cohort study. Inductively coupled plasma mass spectrometry (ICP-MS) was used to determine urinary concentrations of five EMs (selenium, cobalt, vanadium, manganese, and boron) and five NEMs (arsenic, chromium, cadmium, lithium, and tin). The Mini-Mental State Examination (MMSE) scale was used to assess cognitive function. Generalized linear models (GLMs) and restricted cubic spline (RCS) models were used to examine the associations of individual EMs and NEMs with MMSE scores. Quantile g-calculation (QGC) and Bayesian kernel machine regression (BKMR) models were used to evaluate the associations of EM and NEM mixtures with MMSE scores. Results The mean age of the participants was (69.72±6.63) years, and 511 participants (48.90%) were male. The mean MMSE score was 21.06±5.79. After adjustment for covariates, GLM analysis showed that cobalt (β=0.575, 95%CI: 0.233, 0.918), selenium (β=0.440, 95%CI: 0.045, 0.836), and vanadium (β=0.633, 95%CI: 0.251, 1.016) were positively associated with MMSE scores, whereas tin (β=−0.379, 95%CI: −0.728, −0.029) was negatively associated with MMSE scores. After categorizing metal concentrations into quartiles and including them in the GLM, cobalt (βQ4 vs. Q1=1.697, 95%CI: 0.806, 2.589) and vanadium (βQ4 vs. Q1=1.598, 95%CI: 0.719, 2.477) remained positively associated with MMSE scores, whereas tin (βQ4 vs. Q1=−0.890, 95%CI: −1.771, −0.008) and lithium (βQ4 vs. Q1=−0.794, 95%CI: −1.689, −0.007) were negatively associated with MMSE scores. RCS model showed linear positive associations of cobalt and vanadium with MMSE scores, a linear negative association of tin with MMSE scores, and an inverted U-shaped association between lithium and MMSE scores. Both QGC and BKMR models indicated that the EMs mixture was positively associated with MMSE scores, with cobalt contributing the most. The NEM mixture showed a negative but nonsignificant association with MMSE scores, with lithium contributing the most. When EMs and NEMs were jointly modeled, no significant association was observed between the overall metal mixture and MMSE scores. Interaction analysis suggested a positive interaction between EMs and NEMs in relation to MMSE scores. Stratified analysis showed that, compared with participants with high cobalt or EMs exposure, those with low cobalt or EMs exposure had a more pronounced and statistically significant negative association between NEMs exposure and MMSE scores. Conclusion The EM mixture is significantly and positively associated with MMSE scores, whereas the NEMs mixture shows a negative but nonsignificant association. EMs and NEMs have a positive interaction on MMSE scores; EMs attenuate the association between NEMs and MMSE scores.
