Skip to main content
NUTRITION 2026 Main banner

(OR11-04-26) Development and Validation of the Diet-Related Small Habits (DISH) Survey and Assessment of the DISH Score

Sunday, July 26, 2026
3:06 PM - 3:18 PM ET
Location: National Harbor 10-11

    Presenting Author(s)

  • Hemangi B. Mavadiya, PhD, MPH, RD, NBC-HWC (she/her/hers)

    Assistant Professor of Nutrition and Dietetics
    University of Tennessee at Chattanooga
    Chattanooga, Tennessee, United States

    Disclosure(s): No relevant financial relationship(s) with ineligible companies to disclose.

    • Co-Author(s)

  • YL

    Yunxia Lu, Msc PhD

    Joe C Wen School of Population Health and Public Health, University of California, Irvine, California, United States

Disclosure(s):
Hemangi B. Mavadiya, PhD, MPH, RD, NBC-HWC: No relevant financial relationship(s) with ineligible companies to disclose.
Objectives: Diet-related Small Habits (DISH) are simple, day-to-day, evidence-based behaviors that may help increase adherence to dietary recommendations, as the literature indicates low compliance in the population. DISH includes direct behaviors such as eating vegetables and protein before starchy foods or indirect behaviors such as making a grocery list before shopping. Since DISH are not measured in dietary questionnaires, this study aimed to develop and validate a novel DISH survey, which would enable the measurement of DISH score in a sample of US adults.

Methods: A 46-item survey measured DISH awareness and behavior. Survey development included conceptualization, an extensive literature review, and identification of key mechanisms underlying the five DISH domains. Fourteen experts, mostly registered dietitians, evaluated content validity. Data collected from 30 participants were analyzed for the test-retest reliability and the DISH score. Factor analysis was conducted using 342 responses from the DISH online survey. Convergent validity was assessed using the Average Variance Extracted (AVE) and Composite reliability (CR). Discriminant validity was assessed using the Fornell–Larcker criterion.

Results: The average Spearman’s correlation coefficient for time 1 versus time 2 was 0.5 and 0.6 for awareness and behavior items, respectively. The average awareness score was higher than the behavior score at both time points. Confirmatory factor analysis (CFA) supported the a priori five-domain structure, with good model fit statistics. For both awareness (CR = 0.69–0.82) and behavior (0.55–0.80), CR was acceptable for most factors. AVE values were low (awareness = 0.29–0.36; behavior =0.22–0.33). Factor pairs did not meet the Fornell–Larcker criterion.

Conclusions: Overall, while the CFA supported the a priori 5-domain model, high factor correlations and low AVEs suggest a more parsimonious factor structure may better represent DISH items. Higher awareness compared to the lower score for behavior indicated that having awareness about DISH did not translate into healthy actions, suggesting the need for DISH-based interventions.

Funding Sources: Susan Samueli Integrative Health Institute, University of California, Irvine (UCI), and the Cancer Research Coordinating Committee, UCI, Office of the President.