OBJECTIVE Palliative care clinicians and researchers often seek information about patients from informants. This research examines the extent of agreement between information from patients and family caregivers who were asked to serve as collateral sources of information about the patient. METHODS Sixty-six patients with advanced cancer and their family caregivers participated in the study. Two measurement contexts were examined: Direct observation of patients' cognitive performance (Mini-Mental State Examination) was compared with carers' subjective reports about patients' everyday cognition (Cognitive Decline subscale of the Psychogeriatric Assessment Scale), and subjective reports about patient depression were compared between patients and carers who completed parallel forms of the same scale (Geriatric Depression Scale and Geriatric Depression Scale-Collateral Source, respectively). The relationship between patient-completed and carer-completed measures was examined in four ways: the correlation between total scores in the sample, agreement about the prevalence of impaired functioning and of specific symptoms in the sample, agreement concerning whether particular patients showed impaired functioning, and agreement about the presence or absence of a specific symptom for a particular patient. RESULTS Although most measures of agreement showed that information provided by patients and carers was related, the magnitude of discrepancies was substantial. CONCLUSIONS There was no empirical justification for substituting information from a collateral source for information provided by the patient for any of the measures of agreement for either cognition or depression. The use of information from collateral sources is most appropriate when information from the patient is likely to be incomplete or inaccurate, when insight into caregivers' needs or understanding is sought, and when using a multi-informant approach to assessment.