Association between cannabis consumption and serum Klotho levels in middle-aged U.S. adults: NHANES cross-sectional analysis

BACKGROUND: With expanding U.S. cannabis legalization, clarifying its health impacts is critical. This study explores the association between cannabis use frequency and serum Klotho—a biomarker of aging and metabolic health. METHODS: We analyzed 5,827 adults aged 40–59 years from National Health and Nutrition Examination Survey (NHANES) 2007–2016. Cannabis use was self-reported by Computer Assisted Self Interview (ACASI) system and validated questionnaires and serum Klotho concentrations were quantified by ELISA kits. Weighted multivariable-adjusted generalized linear models evaluated associations, with subgroup analyses and interaction tests exploring potential effect modification by covariates. Sensitivity tests (restricted to non-smokers) were performed to investigate potential confounding by tobacco use. Additionally, E-value (assessing unmeasured confounding) was calculated to evaluate the robustness of primary findings. The population attributable fraction for Klotho reduction was calculated to quantify the population-level relevance of the observed association. RESULTS: Cannabis use frequency demonstrated an inverse dose-dependent association with Klotho. Frequent users had significantly lower Klotho than never users (adjusted Estimate = -57.94, P = 0.027), with a significant trend (P for trend = 0.022). This association persisted in non-smokers (P = 0.047), confirming a cannabis-specific effect. E-value analysis showed the primary association was robust to unmeasured confounding (E-value = 4.77, 95% CI: 1.37, 8.57), exceeding the typical threshold for strong confounding (E-value > 1.5). Subgroup analyses revealed amplified Klotho reductions in frequent users with prolonged sedentary activity (> 240 min/day), higher vitamin D (serum 25(OH)D ≥ 72.5 nmol/L), and cancer (P-interaction < 0.05). The population attributable fraction was 3.13% (95%CI: 0.47%, 6.54%), indicating that approximately 3.1% of Klotho reduction in the study population was potentially attributable to frequent cannabis use. CONCLUSIONS: Frequent cannabis use is associated with reduced serum Klotho concentrations, a finding that may relate to biological aging-related mechanisms—particularly in metabolically healthier subgroups and individuals with prolonged sedentary behavior or cancer. This association remained robust after adjustment for numerous confounders and exclusion of smokers, highlighting the need for tailored monitoring of frequent cannabis users, especially vulnerable subgroups.