OBJECTIVES: To determine the association of co-medications and co-morbidities and the risk of dementia, in patients with chronic heart failure (CHF). METHODS: A national retrospective cohort of Veterans Affairs (VA) patients, with at least one outpatient claim of CHF from 1st October 1, 1996 to September 30, 2000, was identified. Drug exposure/co-morbidity data was analyzed from October 1, 1998 to September 30, 2000 and the outcome of dementia (yes/no) was identified based on ICD-9-CM codes from October 1, 2000 to September 30, 2002. Patients with non-utilization of VA services for two years and prevalent cases of dementia were excluded. Multivariate logistic regression was conducted to determine outcome estimates, adjusted for socio-demographics, co-morbidities and co-medications. RESULTS: Out of 242,734 patients, 2.77% met the definition for dementia. About 75% were elderly and 79% had hypertension. Females, African-Americans had significantly higher risk of dementia than males, whites, respectively. Risk of dementia increased progressively with age. In adjusted analyses, hypertension (OR: 1.172, 95% CI: 1.092-1.259), diabetes (OR: 1.085, 95% CI: 1.031-1.142) and seizure (OR: 1.786, 95% CI: 1.598-1.997) significantly increased the risk of dementia. Adjusted analyses of one-year exposure to Angiotensin II Receptor Blockers (ARBs) (OR: 0.785, 95% CI: 0.633-0.974), digitalis (OR: 0.915, 95% CI: 0.838-0.998) and two-year exposure to ARBs (OR: 0.715, 95% CI: 0.595-0.859), statins (OR: 0.909, 95% CI: 0.852-0.971), diuretics (OR: 0.877, 95% CI: 0.819-0.938), digitalis (OR: 0.925, 95% CI: 0.868-0.986) had protective effect on dementia. Adjusted analyses of one-year exposure to beta-blockers (OR: 1.077, 95% CI: 1.001-1.159), had increased risk of dementia. CONCLUSIONS: This exploratory study highlights that major cardiovascular and cerebrovascular co-morbidities are significant risk factors for dementia. It indicates the beneficial effect of drug classes like ARBs, digitalis, diuretics, and statins. It provides direction for further evidence based research on patients with multiple co-morbidities and co-medications.