Given the dynamic characteristic of an individual’s drinking behaviours, comprehensive consideration of alcohol consumption variation using repeated measures may improve insight into the nature of its association with blood pressure (BP) change. We examined the association between longitudinal alcohol consumption (trajectory and quantity) and changes in BP and pulse pressure (PP) among Korean aged ≥ 40 years living in rural areas. Totally, 1682 hypertension-free participants who completed all three health examinations (median, 5·3 years) were included. All three visits were used to determine the cumulative trajectory of and quantity of alcohol consumption and the latest two visits and the last visit were used for the recent trajectory and the most recent quantity of alcohol consumption, respectively. Changes in BP and PP from the baseline to the third visit were used as outcome. In men, ≥30 ml/d cumulative average alcohol consumption was associated with the greatest increase in systolic BP (SBP) in both baseline outcome-unadjusted (2·9 mmHg, P = 0·032) and -adjusted models (3·6 mmHg, P = 0·001), and the given association for the most recent alcohol consumption was observed in the baseline outcome-adjusted model (3·9 mmHg, P = 0·003). For PP, similar associations were observed only in the baseline outcome-adjusted model. No meaningful associations in diastolic BP in men and any BP or PP in women existed. The quantity of alcohol consumption rather than the trajectory may be significantly related to raised SBP, and a possible short-term influence of the most recent alcohol consumption may exist when baseline SBP is adjusted in men.