Background: Acute liver injury of uncertain aetiology is often drug related and quantitative information about the associated risk is scarce. Aim: To estimate the risk of acute liver injury associated with the use of drugs. Methods: In a population survey study, 126 cases of acute liver injury were prospectively assembled from January 1993 to December 1999, in patients over 15 years of age, in 12 hospitals in Barcelona (Spain). We estimated the relative risk for each drug as the ratio between the incidence of acute liver injury among the exposed population to the drug and the incidence of acute liver injury among those not exposed to it. Drug consumption data were used to estimate the exposed population. Results: Isoniazid, pyrazinamide, rifampicin, amoxicillin with clavulanic acid, erythromicin, chlorpromazine, nimesulide, and ticlopidine presented the highest risk (point relative risk > 25). Amoxicillin, metoclopramide, captopril and enalapril, furosemide, hydrochlorothiazide, fluoxetine, paroxetine, diazepam, alprazolam, lorazepam, metamizole, low-dose acetylsalicylic acid and salbutamol showed the lowest risk (point relative risk < 5). Conclusions: This study provides a risk estimation of serious liver disease for various drugs that will be useful in its diagnosis and management, and when comparing with the drug therapeutic benefit in each indication. Some observed associations would be worth specific studies. © 2007 The Authors.