Dietary intake tools are used in epidemiological and interventional studies to estimate nutritional intake. The past-month Canadian Diet History Questionnaire II (CDHQII) has not yet been validated. This study aimed to assess the validity of the CDHQII in adults by comparing dietary results from the CDHQII to the same participants’ 24-h recalls consisting of two weekdays and one weekend day. The recalls were collected using the validated multiple-pass method. Participants were asked to complete both tools at baseline, and again at 3-month follow-up. The study further aimed to determine which dietary intake tool was preferred by study participants by comparing completion rates. Data collection occurred at baseline (pre-intervention) and 3-month follow-up (post-intervention). Paired sample t-tests were conducted to compare means for the following nutrients (grams and %kcal): calories, protein, carbohydrates, total fat, saturated fat, unsaturated fat and sodium. Intraclass correlation coefficients of agreement and coefficients of variation were further calculated. Chi-square tests were used to determine the dietary assessment method with the greatest participant completion rate. At baseline (n = 104), there were no significant differences between the results of the CDHQII and three 24-h recalls (averaged), with overall moderate correlation coefficients. At 3-months (n = 53), there were significant differences (p < 0.05) between dietary intake collection methods for all nutrients assessed in this study, except for saturated fat (%kcal), unsaturated fat (%kcal), protein (%kcal) and sodium (mg). Correlation coefficients were moderate. A significantly greater proportion of participants completed the three 24-h recalls compared to the CDHQII after 3 months (completion rates of 67.2% vs. 50.8% of the sample, respectively). The CDHQII provided estimates of mean nutritional intake (calories, macronutrients and sodium) that were comparable to mean intake established from three 24-h recalls, at baseline and was validated in a sample of primarily middle-aged, college-educated, Caucasian female adults with overweight and obesity for mean baseline or cross-sectional measurement only but not for assessing individual/patient dietary intake in clinical practice (r = 0.30–0.68). This tool was not validated at 3-month follow-up. Additionally, participants preferred the three 24-h recalls to the online, past-month CDHQII.
Available at: http://works.bepress.com/jason-gilliland/34/