BACKGROUND: Diabetes is an independent risk factor for cognitive impairment. We aimed to investigate the association between the glucagon-like peptide-1 (GLP-1) receptor agonist dulaglutide and cognitive impairment as an exploratory analysis within the Researching Cardiovascular Events With a Weekly Incretin in Diabetes (REWIND) trial. METHODS: REWIND is a randomised, double-blind placebo-controlled trial at 371 sites in 24 countries. We included men and women (aged ≥50 years) with either established or newly diagnosed type 2 diabetes and additional cardiovascular risk factors, glycated haemoglobin of up to 9·5% (80 mmol/mol) on a maximum of two oral glucose-lowering drugs with or without basal insulin, and a body-mass index of at least 23 kg/m. Participants were randomly assigned (1:1) subcutaneous injections once a week of either dulaglutide (1·5 mg) or an equal volume of matching placebo. Randomisation was done using a computer-generated code with stratification by site. Participants and all study personnel were masked to treatment allocation until the database was locked. Participants were followed up at least every 6 months for the composite primary outcome of stroke, myocardial infarction, or death from cardiovascular or unknown causes. Cognitive function was assessed at baseline and during follow-up using the Montreal Cognitive Assessment (MoCA) and Digit Symbol Substitution Test (DSST). We present here the exploratory primary cognitive outcome, which was the first occurrence of a follow-up score on MoCA or DSST that was 1·5 SDs or more below the baseline mean score in the participant's country. All analyses were done using an intention-to-treat approach. The REWIND trial is registered with ClinicalTrials.gov, NCT01394952. FINDINGS: Between Aug 18, 2011, and Aug 14, 2013, 9901 participants were randomly assigned to either dulaglutide (n=4949) or placebo (n=4952). During median follow-up of 5·4 (IQR 5·1-5·9) years, 8828 participants provided a baseline and one or more follow-up MoCA or DSST scores, of whom 4456 were assigned dulaglutide and 4372 were assigned placebo. The cognitive outcome occurred in 4·05 per 100 patient-years in participants assigned dulaglutide and 4·35 per 100 patient-years in people assigned placebo (hazard ratio [HR] 0·93, 95% CI 0·85-1·02; p=0·11). After post-hoc adjustment for individual standardised baseline scores, the hazard of substantive cognitive impairment was reduced by 14% in those assigned dulaglutide (HR 0·86, 95% CI 0·79-0·95; p=0·0018). INTERPRETATION: Long-term treatment with dulaglutide might reduce cognitive impairment in people with type 2 diabetes. Further studies of this drug focused on brain health and cognitive function are clearly indicated. FUNDING: Eli Lilly and Company.