628-P: Factor Structure and Validity of a Dietary Screener among Diverse Adults with Suboptimally Controlled Type 2 Diabetes (T2D)

Abstract

Objective: Although following a healthy diet is a key aspect of diabetes self-care, few brief, validated dietary measures exist. The 8-item Starting the Conversation (STC) questionnaire shows promise but has limited validation, particularly among Latinos, who are disproportionately impacted by T2D. Methods: We assessed the STC’s factor structure in English and Spanish, convergent and divergent validity, and internal reliability at baseline using data from 812 predominantly Latino adults with T2D who participated in a trial of telephonic self-management support (Female = 56.9%; Latino = 85.8%; Mean (SD) Age and A1C = 59.2 (10.8) , 9.3 (1.8) , respectively) . We also examined the STC’s sensitivity to this intervention over 12 months. Results: Exploratory factor analysis yielded 2 factors: Junk Food and Healthy Food. When 3 items were excluded due to low loadings, a single Junk Food factor was retained, explaining 40% of variance in STC scores. Cronbach’s alphas for the 5-item Junk Food scale were similar in Spanish (α = .61) and English (α = .57) at baseline. The Junk Food scale demonstrated convergent validity with BMI (r = .13, p < .001) and self-reported diabetes self-care (r = -.33, p < .001) , but not with A1C (r = .04, p = .328) , systolic BP (r = -.09, p = .011) , or diastolic BP (r = -.04, p = .312) . Demonstrating divergent validity, the Junk Food scale did not correlate with 4 scales assessing beliefs about medication (rs ≤ .06; ps ≥ .103) . There was no significant difference in change in Junk Food scores over time between the intervention and control arms (DID = -.02, SE = .03, , 95% CI [-.08, .04]) . Conclusions: Although the Junk Food scale demonstrates mixed evidence for convergent validity, this 5-item measure may provide a useful assessment of unhealthy food consumption among English- and Spanish-speakers with T2D. To our knowledge, this is the first study to investigate the factor structure of the STC, which may inform its utility for research and practice. Disclosure L.D.Pappalardo: None. C.Schechter: None. R.Fang: None. C.J.Hoogendoorn: None. J.S.Gonzalez: Consultant; Virta Health Corp. E.A.Walker: None. Funding National Institutes of Health (R18 DK098742) Einstein–Mount Sinai Diabetes Research Center (P30 DK020541) New York Regional Center for Diabetes Translation Research (P30 DK111022)