BACKGROUND: Discharge against medical advice may be associated with more readmissions. OBJECTIVE: To evaluate DAMA in patients with acute ischemic stroke (AIS) and identify the relationship between DAMA and 30-day unplanned readmissions. DESIGN: A retrospective cohort study. PARTICIPANTS: The National Readmission Database was used to identify inpatients with a primary diagnosis of AIS who were either discharged home or DAMA between 2010 and 2017 in the USA. MEASURES: Demographic features, hospital type, comorbidities, stroke risk factors, severity indices, and treatments were compared between patients discharged routinely and DAMA. Multivariable logistic regression was used to evaluate predictors of DAMA, and a double robust inverse probability of treatment weighting method was used to assess the association between DAMA and 30day unplanned readmissions. KEY RESULTS: Overall, 1,335,484 patients with AIS were included, of whom 2.09% (n = 27,892) were DAMA. The prevalence of DAMA in AIS patients increased from 1.65 in 2010 to 2.57% in 2017. The rates of 30-day unplanned readmissions for DAMA and non-DAMA patients were 16.81% and 7.78%, respectively. Patients with drug abuse, alcohol abuse, smoking, prior stroke, psychoses, and intravenous thrombolysis had greater odds of DAMA. DAMA was associated with all-cause readmissions (OR, 2.04; 95% CI, 2.01-2.07) and remained a strong predictor for transient ischemic attack/stroke-specific and cardiacspecific causes of readmissions. CONCLUSIONS: Although the DAMA rate is low in AIS patients, DAMA is a risk factor for all-cause and recurrent stroke-specific readmissions. Future studies are needed to address issues around compliance and engagement with health care to reduce DAMA.