Long-term stability of food patterns identified by use of factor analysis among Swedish women.

Limited data exist on the reproducibility of food patterns measured by factor analysis as well as on the use of stability of patterns over time. Our primary objective was to explore the long-term stability of food patterns derived by confirmatory factor analysis among 33,840 women participating in the Swedish Mammography Cohort. Our secondary objective was to compare factor solutions from confirmatory factor analysis with those derived by use of exploratory factor analysis. Diet was assessed by a food frequency questionnaire in 1987 and 1997, and food patterns were derived by the use of exploratory and confirmatory factor analysis. We observed four major food patterns (Healthy, Western/Swedish, Alcohol, and Sweets) at both time points. Correlations between confirmed food patterns in 1987 and 1997 were 0.37 for the Healthy pattern, 0.27 for the Western/Swedish pattern, 0.54 for the Alcohol pattern, and 0.44 for the Sweets pattern (P < 0.0001 for all associations). Patterns derived by the use of exploratory factor analysis were strongly associated with those derived by the use of confirmatory factor analysis (r >/= 0.90, P < 0.0001, for all associations). The patterns derived in this study were similar to those derived in other studies, indicating reproducibility of food patterns across populations. Our study suggests that food patterns measured by the use of confirmatory factor analysis are reproducible over time, and weaker correlations observed may reflect natural changes in eating behavior, the food supply, and/or perceptions of what is considered healthy. Testing whether patterns measured over a long time period can be used in analytic studies is the next step in assessing the validity of this method.