Ying-Jen Chang, Chia‐Li Kao, Kuo‐Mao Lan
Background Low magnesium status has been linked to cerebrovascular conditions that overlap with pathways implicated in dementia, yet prior studies have relied on single magnesium measurements obtained in earlier population-based cohorts. Whether repeatedly documented low magnesium status is associated with incident dementia in real-world clinical populations is unclear. Methods This retrospective propensity score-matched cohort study used data from the TriNetX Global Collaborative Network (2016–2023). Adults aged ≥50 years with two serum magnesium values <1.7 mg/dL within a 1-year window were compared with matched controls with two normal values (1.70–2.20 mg/dL). A 1-year landmark period was applied before the outcome ascertainment. The primary outcome was incident all-cause dementia. The secondary outcomes included dementia subtypes, stroke, and mortality. Prespecified positive and negative control outcomes, subgroup and sensitivity analyses, sequential multivariable Cox regression, and an exploratory analysis of high magnesium status were conducted. Results After propensity score matching, 162,936 patients were included in each group. Low magnesium status was associated with incident all-cause dementia (HR 1.33; 95% CI, 1.26–1.40, p < 0.001), with the strongest association observed for vascular dementia (HR 1.63; 95% CI, 1.42–1.88, p < 0.001). Associations were also observed for stroke (HR 1.37, p < 0.001) and mortality (HR 1.52, p < 0.001). The results remained consistent across the sensitivity and multivariable analyses. High magnesium status was associated with all-cause dementia (HR 1.25, p < 0.001) and vascular dementia (HR 1.40, p < 0.001), but not Alzheimer’s disease. Conclusion Repeatedly documented low magnesium status was associated with an increased risk of incident all-cause dementia, particularly vascular dementia, suggesting a potential role of magnesium homeostasis in long-term cognitive function.