Dietary patterns of older adults in a rural, tri-ethnic community: a factor analysis approach

Health promotion interventions for older adults designed to alter nutrient intake patterns often fail C, because they do not consider their total dietary patterns. Factor analysis can be used to determine which foods or food groups cluster together. Food patterns of 130 African American (34%), Native American (30%), and white (36%) adults greater than or equal to70 years in two rural communities are reported here. A modified Block food frequency questionnaire was administered orally by trained interviewers to elicit usual diet. The 110 individual foods from the questionnaire were combined into 38 groups. Principal components factor analysis with promax rotation was used to identify 7 patterns that accounted for 45% of the variance of dietary intake frequency. These factors represent a variety of patterns, such as "American Diet", "traditional breakfast," and "meats." With the exception of gender, few differences in patterns were observed for selected demographic and health characteristics. This approach can summarize complex dietary data into meaningful patterns and assist health promotion efforts among older adults. (C) 2003 Elsevier Inc. All rights reserved.