Nonlinearity association of blood cobalt with the risk of anemia among middle-aged and older adults: National Health and Nutrition Examination Survey

Abstract

ABSTRACT Objective The use of cobalt alloys in medical implants poses a high risk of cobalt exposure, yet there is a lack of evidence regarding the association between blood cobalt levels and anemia. This study aimed to explore the link between blood cobalt levels and the onset of anemia and to identify potential threshold levels of blood cobalt that could affect anemia. Methods The US National Health and Nutrition Examination Survey data from 2017 to 2020 were analyzed for this cross-sectional study. This study primarily employed multivariate logistic regression, stratified interaction analysis, restricted cubic splines (RCS), and threshold effect analysis to explore the relationship between blood cobalt concentration and anemia. Results The study included 5510 participants and among them 12.2% were diagnosed with anemia. Logistic regression model indicates a positive correlation between blood cobalt levels and the risk of anemia. RCS shows that the relationship between ln cobalt concentration and anemia was non-linear (J-shaped). The ln cobalt inflection point was approximately 0.81. The odds ratio of anemia with ln cobalt ≥ 0.81 was 4.00 (95% CI: 2.95–5.43, p < 0.001), the odds ratio of anemia with ln cobalt < 0.81 was 0.73 (95% CI: 0.45–1.18, p = 0.201). Conclusions The analysis unveiled a non-linear relationship, indicating that elevated blood cobalt levels were linked to a heightened likelihood of developing anemia in middle-aged and older adults; the cut-off value of ln cobalt was approximately 0.81. The findings of this study warrant further investigation.