Responsiveness to a self-administered diet history questionnaire in a work-site dietary intervention trial for mildly hypercholesterolemic Japanese subjects: correlation between change in dietary habits and serum cholesterol levels.

Modification of lifestyle, especially of diet, is considered important for prevention of cardiovascular disease. Dietary assessment is generally too troublesome to use in a large number of subjects for prevention. We have therefore developed a self-administered diet history questionnaire (DHQ), an easier dietary assessment method than conventional methods, with reasonable validity for use in dietary intervention studies. Responsiveness, i.e., sensitivity to a change in a target variable, is one type of validity required for a dietary assessment method which is used for the evaluation of the effect of dietary interventions. We examined the responsiveness of the DHQ using the data from a 12-week work-site dietary intervention trial including 63(54 men and 9 women) mildly hypercholesterolemic Japanese (age range: 22-59 years, serum cholesterol > or = 200 mg/dl). Dietary habits were assessed by the DHQ before and after the trial. Pearson's correlation coefficients between the change in serum cholesterol and Keys score calculated from the dietary data were 0.33 and 0.32 (p < 0.01) with and without adjustment for possible confounding factors, respectively. Forty-two percent of the total variation of serum cholesterol change was explained by the initial serum cholesterol level, the change in body mass index, and the Keys score. The results suggest that the DHQ showed adequate responsiveness to the serum cholesterol change resulting from dietary intervention.