Abstract
BACKGROUND: The relationship between exposure to ethylene oxide (EO) and the risk of developing osteoarthritis (OA) remains unclear. We aimed to explore the association between EO exposure and OA risk among young and middle-aged adults. METHODS: We utilized data from the 2013-2018 National Health and Nutrition Examination Survey, involving 2,380 individuals aged 20-60 years. Weighted multivariable regression models, smooth curve fitting (SCF), subgroup analysis and interaction tests were employed to examine the association between EO exposure and OA risk. Furthermore, we performed variable selection via least absolute shrinkage and selection operator regression and multivariable regression analyses to construct a prediction model. RESULTS: Increased EO exposure was associated with increased OA risk. After full adjustment, individuals in the highest tertile of EO exposure had a significantly greater OA risk (224% increase) than did those in the lowest tertile of EO exposure (OR = 3.24; 95% CI: 1.61-6.52; p for trend = 0.002). SCF did not indicate any nonlinear associations. There was no statistically significant interaction observed in any of the subgroups (all p > 0.05). We built a prediction model visualized with a nomogram. This prediction model demonstrated good discriminatory power, excellent precision, and potential clinical benefits. CONCLUSION: The findings of our research demonstrated that among middle-aged and young adults, EO exposure was positively associated with OA risk. A prediction model was developed by integrating EO exposure with other factors readily acquired from users to assist in the evaluation and management of high-risk OA groups.