CONTEXT: The role of alcohol intake in influencing longitudinal trajectories of adiponectin is unclear. OBJECTIVE: The objective of the study was to examine the association between alcohol intake and changes in the circulating levels of adiponectin over repeat measures. DESIGN, SETTING, AND PARTICIPANTS: A prospective cohort study of 2855 men and women (74% men with a mean age of 50 y at baseline) drawn from the Whitehall II study. Data from study phases 3 (1991-1993), 5 (1997-1999), and 7 (2002-2004) were used. MAIN OUTCOME MEASURE: Adiponectin serum concentrations (nanograms per milliliter) were measured, and alcohol intake was defined in terms of number of UK units (1 U = 8 g ethanol) consumed in the previous 7 days on three occasions. Cross-sectional associations between alcohol and adiponectin levels were calculated using linear regression. A bivariate dual-change score model was used to estimate the effect of alcohol intake on upcoming change in adiponectin. Models were adjusted for age, sex, ethnicity, and smoking status. RESULTS: Alcohol consumption was cross-sectionally associated with (log transformed) adiponectin levels (β ranging from .001 to .004, depending on phase and level of adjustment) but was not associated with changes in adiponectin levels over time [γ = -0.002 (SE 0.002), P = 0.246]. CONCLUSION: Alcohol intake is not associated with changes in circulating adiponectin levels in this cohort. This finding provides evidence that adiponectin levels are unlikely to mediate the relationship between moderate alcohol consumption and reduced risk of type 2 diabetes. It is important to consider dynamic longitudinal relationships rather than cross-sectional associations.