Validating the time and change test to screen for dementia in an older Hispanic population.

Abstract

To the Editor: Dementia is an underdiagnosed and undertreated health problem, especially in older minority people, emphasizing the need for screening tests for dementia that are valid, accurate, and easy to administer across ethnic groups. Many current tests require a significant amount of time to administer and contain ethnic/cultural/language biases and an overemphasis on verbal/reading abilities.1-4 In response to these difficulties, the Time and Change (T&C) test has been developed. It is a simple, rapid performance-based test of a subject's ability to tell time and make change and has been validated for the detection of dementia in inpatient and outpatient populations.5,6 Although African-American and Caucasian populations were well represented in these samples, they were only 1% Hispanic. We administered the T&C test in a convenience sample of 222 community-living residents of Chicago, Illinois, aged 55 and older and self-identified as Hispanic. “Hispanic,” referring to people who immigrated to the United States from Spanish-speaking countries of Latin America, or to their descendants, is a vague term that encompasses a variety of cultural and racial groups. Nevertheless, the sharing of a common language, a history of recent immigration, and cultural similarities make “Hispanic” a relevant category to use when studying screening tools that inevitably are influenced by the language and culture in which they are designed. Two bilingual medical student interviewers administered an interview in the subject's language of preference (Spanish or English). The Mini-Mental State Examination (MMSE)7,8 and the T&C test were included as part of the interview. The T&C test was performed as described below; see Inouye et al. for further details and to obtain training materials.5 Time. The participant was shown a large clock-face diagram with the hands set at 11:10. The participant was cued: “Please tell me what time it says on this clock.” Change. Three quarters, seven nickels, and seven dimes were placed in front of the participant, who was cued: “Please give me a dollar in change.” If the subject had been living in the United States for less than 5 years and admitted confusion as to the values of the coins, the interviewer explained the values and then resumed the test. The sample was predominantly Mexican (49%), of Mexican ancestry born in the United States (20%), or of Puerto Rican (19%) origin. Eighty-three percent of participants chose to be tested in Spanish; 17% were tested in English. The mean age ± standard deviation of the sample was 70.3 ± 7.6 (range 55–92), and mean years of formal education was 5.6 ± 4.5 (range 0–18). A total of 11.3% of subjects failed the T&C, and 10% scored less than 18 on the MMSE. (A MMSE score of <18 was chosen as a score low enough to be indicative of some degree of cognitive impairment regardless of the subject's educational/literacy level.) Low T&C scores were correlated significantly with low MMSE scores. Figure 1 demonstrates the concordance of T&C and MMSE scores. Using an MMSE score of less than 18 as a proxy for cognitive impairment, the T&C has a sensitivity of 78.3% and a specificity of 93.0%, giving a negative predictive value of 97.4%. Score (pass or fail) of the Time and Change (T&C) test by Mini-Mental State Examination (MMSE) score. The T&C test was less influenced than the MMSE by the subject's length of time living in the United States and by years of formal education, although very low educational levels did adversely affect T&C performance. This could signify that the T&C test reduces test bias while preserving other correlations between educational level and cognitive function. The T&C test appeared to be a useful test for screening for cognitive impairment in a community setting. The high negative predictive value of the T&C test makes it especially useful in excluding dementia. The T&C test examines cognitive areas such as conceptualization and visuospatial functions, which are less emphasized by the MMSE.5 By incorporating skills that are important in daily living, the T&C test may also give more information about a patient's ability to live independently. Additional advantages of the T&C test were ease of administration, high acceptability to participants, and relatively infrequent educational, cultural, or language bias compared with the MMSE. Resource limitations prevented identification of impaired cognitive function with true criterion standard comprehensive testing. However, the MMSE was used for comparison in this study because it has been extensively validated and is widely utilized for the assessment of cognitive impairment.7 Future studies with improved sampling procedures and a complete diagnostic evaluation for dementia are likely to further validate the T&C test for clinical use in Hispanic populations.