Defining and measuring stages of change for dietary behaviors: readiness to meet fruit, vegetable, and grain guidelines among Chinese Singaporeans.

OBJECTIVE: To assess the ability of 2 algorithms to classify people by stage of change for consuming the recommended servings of grains (cereal foods) and total fruit and vegetables.
DESIGN: Assessment of stage involved an objective behavioral measure in the form of a self-administered food frequency questionnaire, followed by a brief telephone interview to assess intentions of subjects to increase intake to meet the recommended servings. Validity of the stage classification was assessed by comparison with three 24-hour dietary recalls.
SUBJECTS: One hundred and one Singaporean Chinese subjects (mean age = 38.7; 51% men) were recruited from 716 respondents who had taken part in a survey investigating factors influencing consumption of grains, fruit, and vegetables. STATISTICAL ANALYSES PERFORMED: Differences in mean intake by diet recalls across the stages were investigated using analysis of variance. Sensitivity, specificity, and predictive values of the algorithms were also determined.
RESULTS: There were significant increases across the stages in mean intake of grains (men: F(2,48) = 20.30, P < .001; women: F(2,47) = 23.39, P < .0001), and total fruit and vegetables (men: F(2,48) = 30.29, P < .005; women: F(2,47) = 37.29, P < .0001). Based on diet recalls for grains intake, the algorithms classified 89% of subjects having inadequate intakes into the preaction stages, and 75% of those having adequate intakes into the action or maintenance stages. For fruit and vegetables, 93% of subjects having inadequate intakes were classified into the preaction stages, and 76% of those having adequate intakes were classified into the action or maintenance stages.
CONCLUSION: Algorithms developed to assess stages of change for food-based rather than nutrient goals, and which include an objective assessment of intake, appear to improve the accuracy of stage classifications.