Objectives To examine the longitudinal association between objective measure of sleep duration and subsequent recurrence of peptic ulcer disease (PUD) in older patients with mild cognitive impairment (MCI) and the potential influence of change in the severity of depression on this association. Method Older Chinese patients with MCI and Helicobacter pylori (H. pylori)-infected PUD (N = 2,208) were recruited from hospitals in the People's Republic of China between 2010 and 2014. H. pylori was eradicated and PUD cleared in 2,015 patients by the end of 2014; 1,866 of these were followed for up to 48 months. Sleep duration was measured using an accelerometer. The Kaplan-Meier method was used to assess how PUD recurrence in older patients, as confirmed with esophagogastroduodenoscopy, varied with the levels of sleep duration and change in depression. Multivariate Cox-proportional hazards models were calculated to examine the associations between sleep duration, depression, and PUD recurrence. Result The present results are based on no change, a decrease, or an increase in caregiver-assessed depressive symptoms. Multivariate analyses showed that short sleep duration was associated with PUD recurrence during the 48-month follow-up period (hazard ration [HR] = 2.685; 95% confidence interval [CI: 1.622, 4.538]). Sleep duration did not affect recurrence in patients without depression (HR = 1.325; 95% CI [.878, 1.998]) or with reduced depression (HR = 1.048; 95% CI [.695, 1.581]). However in patients with unchanged or increased depression, a higher risk of PUD recurrence was found in short sleepers (HR = 1.598; 95% CI [1.042, 2.451]; HR:HR = 2.668; 95% CI [1.720, 4.083], respectively) than in medium and long sleepers. Discussion Short sleep duration is associated with a greater risk of PUD recurrence, but decrease in depression or absence of depression may alter this relationship. (PsycINFO Database Record (c) 2019 APA, all rights reserved).