Methods Longitudinal data collection included self-administered questionnaires and medical data (aspartate aminotransferase (AST) and alanine aminotransferase (ALT) liver enzymes). Two analyses were performed to assess the association between coffee (⩾3 cups a day) and daily chocolate intake and abnormal values of AST and ALT (AST or ALT >2.5 × upper normal limit (UNL)) (N = 990) over time, after adjustment for known correlates. Logistic regression models based on generalized estimating equations were used to take into account the correlations between repeated measures and estimate adjusted odds ratio. Results After adjustment, patients reporting elevated coffee consumption and daily chocolate intake were less likely to present abnormal ALT (OR = 0.65; p = 0.04 and OR = 0.57; p = 0.04, for coffee and chocolate respectively), while only patients reporting elevated coffee consumption were less likely to have abnormal AST values (p = 0.05). Nevertheless, the combined indicator of coffee and chocolate intake was most significantly associated with approximately 40% reduced risk of abnormal liver enzymes (p = 0.003 for AST; p = 0.002 for ALT). Conclusions Elevated coffee consumption and daily chocolate intake appear to be associated with reduced levels of liver enzymes in HIV-HCV co-infected patients. Further experimental and observational research is needed to better understand the role that polyphenol intake or supplementation can play on liver disease and liver injury.