Relationship of race/ethnicity and blood pressure to change in cognitive function.

OBJECTIVES: To determine whether there are racial/ethnic differences regarding the relationship of level of blood pressure to change in cognitive function in older people. DESIGN: Longitudinal data 1986 to 1989 on representative, older, community-residing African Americans and whites. Blood pressure levels were assessed and a brief screen of cognitive functioning, the Short Portable Mental Status Questionnaire (SPMSQ), was performed at baseline and 3 years later. SETTING: Five contiguous counties in the Piedmont area of North Carolina. PARTICIPANTS: African-American (n = 2,260) and white(n = 1,876) participants in the Duke Established Populations for Epidemiologic Studies of the Elderly, aged 65 to 105 at baseline. MEASUREMENTS: The outcome measure was change in SPMSQ score over 3 years. Covariates included age; education; gender; self-reported diabetes mellitus, stroke, heart attack, current smoking, and depressive symptomatology;and use of antihypertensive medication. The primary independent variable was measured blood pressure. RESULTS: In unadjusted analyses, a statistically significant U-shaped relationship was found between systolic (but not diastolic) blood pressure levels and change in SPMSQ score over a 3-year period for older white men and women. No such relationships were found between these blood pressure measurements and change in SPMSQ score in older African Americans. These findings remained after adjustment for initial SPMSQ score, demographic characteristics, and use of antihypertensive medication. There were no significant interactions between race and blood pressure on change in cognitive function. CONCLUSION: Decline in cognitive function was associated with extremes of systolic blood pressure in older white people. Although a similar but muted nonsignificant association was found in older African Americans, the curves for the two groups were not significantly different. Further studies in older African Americans are needed.

Duke Scholars

Author Carl F. Pieper Biostatistics & Bioinformatics, Division of Biostatistics

Author Gerda G. Fillenbaum Psychiatry & Behavioral Sciences, Adult Psychiatry & Psychol ...

Author Daniel German Blazer Psychiatry & Behavioral Sciences, Behavioral Medicine & Neur ...