OBJECTIVE : A window period of approximately 3-6 months is usually adopted in studies that evaluate hepatic encephalopathy (HE) risk in proton pump inhibitor (PPI) users. However, HE risk after short-term PPI exposure remains unclear. We explored the effect of short-term PPI exposure using a case-crossover study design. METHODS : Records of patients with decompensated cirrhosis who had received an HE diagnosis were retrieved from the National Health Insurance Research Database. PPI use rates were compared for case and control with window periods of 7, 14, and 28 days. The adjusted self-matched odds ratio (OR) and 95% confidence interval (CI) from a conditional logistic regression model were used to determine the association between PPI use and HE risk. RESULTS : Overall, 13 195 patients were analyzed. The adjusted OR for HE risk after PPI exposure was 3.13 (95% CI = 2.33-4.20) for the 7-day window, 4.77 (95% CI = 3.81-5.98) for the 14-day window, and 5.60 (95% CI = 4.63-6.78) for the 28-day window. All PPI categories, except omeprazole and pantoprazole, were associated with an increased HE risk. Irrespective of other precipitating factors, such as recent gastrointestinal bleeding or infection, PPI significantly increased HE risk. CONCLUSIONS : Short-term PPI use is significantly associated with HE in patients with decompensated cirrhosis. Physicians should use PPI in these patients for appropriate indications, and carefully monitor signs of HE even after short-term exposure. Owing to the limitations of retrospective design in the current study, further study is warranted to confirm our findings.