Objective: To develop and validate the Time and Change (T&C) test, a simple, standardized method for detecting dementia in a diverse older outpatient population with varying levels of education.
Design: A prospective cohort validation study.
Setting: Two outpatient clinics at an urban teaching hospital.
Participants: The concurrent validation sample consisted of 100 consecutive outpatients 70 years of age or older who were 58% non-white and had a 16% dementia prevalence rate and educational levels ranging from 0 to 17+ years. Reliability was tested in a sample of 42 consecutive outpatients 75 years of age or older with a 36% dementia prevalence rate.
Measurements: T&C ratings were validated against a reference standard based on the Blessed Dementia Rating Scale and the Mini-Mental State Examination. Reliability, contribution to physician recognition of dementia, and ease of use were assessed.
Results: In the outpatient setting, the T&C had a sensitivity of 63%, specificity of 96%, a negative predictive value of 93%, a positive predictive value of 77%, and test-retest and inter-observer reliability agreement rates of 95% and 100%, respectively. When T&C results were added to the physician's documentation of dementia, the number of missed cases decreased from 44% to 19%, and the number of overcalled cases decreased by 100%. When timed cut points were added, the T&C test had a sensitivity of 94 to 100%, specificity of 37 to 46%, negative predictive value of 98 to 100%, positive predictive value of 23 to 25%, and test-retest and inter-observer agreement rates of 82% and 70 to 75%, respectively.
Conclusion: The T&C test is a simple, accurate, reliable, performance-based tool that can improve physician ability to recognize dementia in diverse outpatient populations.