The validity and reliability of clinic-based and telephone-based versions of a 4 word delayed recall test were evaluated in nondemented elderly individuals (n=644) participating in a simulated primary prevention clinical trial. There was no significant difference in the average scores achieved by participants tested in clinic (mean=3.40) or by telephone (mean=3.47) and the 2 groups had similar distributions of scores. Delayed recall scores were significantly, but weakly, correlated with scores on a rigorous verbal memory task, were lower in participants in Clinical Dementia Rating stage 0.5 than in those in Clinical Dementia Rating stage 0, and were lower in those with subjective memory complaints than in those without complaints. There was only fair correspondence between scores achieved at initial testing and 3 months later for both versions of the test. There were no differences in the average scores achieved by men or women, those older (age 80 to 93) or younger (age 75 to 79) than age 80, or those with white or nonwhite ethnicity. Participants with low education scored significantly lower than those with high education. Results suggest that clinic-based and telephone-based versions of the Four Word Delayed Recall Test are valid and reliable and can be used to screen for possible memory deficits in elderly individuals. However, the psychometric properties of the test are relatively weak and do not support the general use of the test for clinical and research purposes if the use of a more rigorous memory test with a wider range of possible scores is feasible.