Abstract
OBJECTIVE: This study aims to explore the association between coffee consumption and the prevalence of hearing loss in American adults based on a national population-based survey. DESIGN: Cross-sectional analysis of reported audiometric status and coffee intake from the 2003-2006 National Health and Nutrition Examination Survey (NHANES). Multivariate logistic regression, forest plots and restricted cubic spline (RCS) analyses were used to explore the associations and dose-response relationships between coffee consumption frequency and hearing loss. SETTING: The USA. PARTICIPANT: This study included 1894 individuals aged ≥ 20 from the 2003-2006 NHANES. RESULTS: In this study, the prevalence of speech-frequency hearing loss (SFHL) and high-frequency hearing loss (HFHL) among the participants was 35·90 % and 51·54 %, respectively. Compared with those who no consumed coffee, non-Hispanic White who consumed ≥ 4 cups/d had higher prevalence of SFHL (OR: 1·87; 95 % CI: 1·003. 3·47). And a positive trend of coffee consumption frequency with the prevalence of HFHL was found (P(trend) = 0·001). This association of HFHL was similar for participants aged 20-64 (P(trend) = 0·001), non-Hispanic White (P(trend) = 0·002), non-noise exposure participants (P(trend) = 0·03) and noise-exposed participants (P(trend) = 0·003). The forest plots analysis found that the association between 1 cup-increment of daily coffee consumption and the prevalence of HFHL was statistically significant in males. RCS model supported a positive linear association of coffee consumption with SFHL (P for overall association = 0·02, P for nonlinearity = 0·48) and a positive non-linear association of coffee consumption with HFHL (P for overall association = 0·001, P for nonlinearity = 0·001). CONCLUSION: Our findings suggested that coffee consumption was associated with higher prevalence of hearing loss. Further cohort studies in larger population are needed to investigate these findings.