Introduction: Studies on adult bipolar patients have demonstrated a disability associated with the bipolar disorder, even in euthymic patients, but there is a lack of data in the elderly population. Material and method: A cross-sectional, multicentre study on a consecutive sample of ambulatory bipolar patients (DSM-IV-TR criteria), aged 65 years or over. Retrospective and cross-sectional sociodemographic and clinical data were collected, as well as the Clinical Global Impression for Bipolar Modified scale (CGI-BP-M) and the level of disability using the World Health Organisation Disability Assessment Schedule (WHO/DAS). The disability was assessed globally and by areas. The presence of a moderate to maximum disability compared to a mild to no disability was considered a dependent variable. Results: A moderate to maximum global disability was present in 43.6% of the sample. By areas, occupational functioning was the area most frequently affected, and personal care the least affected. The only variables which were associated with disability were the presence of medical comorbidity (P = .01), increased age (P = .005) global clinical severity (P = .0001) and in the depressive pole (P = .03). There was no relationship between clinical subtype, duration of the disease, number of previous episodes, number of hospitalisations, or other clinical variables and the degree of disability. Conclusions: These data underline the need to establish specific therapeutic strategies in the approach to depressive symptoms and medical comorbidity, with the aim of minimising the disability in elderly bipolar patients. Given the lack of current data, new studies are needed with larger samples and control groups. © 2010 SEP y SEPB. Published by Elsevier España, S.L. All rights reserved.