Date of Award

Fall 2025

Project Type

Thesis

Program or Major

Nutritional Sciences

Degree Name

Master of Science

First Advisor

Maria Carlota Dao

Second Advisor

Noereem Z Mena

Third Advisor

Sherman Bigornia

Abstract

AbstractBackground: Hispanic/Latino adults in the United States experience disproportionate rates of obesity and diet-related chronic diseases yet remain underrepresented in obesity research. Traditional “one-size-fits-all” nutrition interventions have demonstrated limited effectiveness in this population, largely due to substantial heterogeneity in cultural, socioeconomic, and behavioral factors that influence dietary patterns. Objectives: The overarching goal of this project was to further characterize previously identified subgroups within the Hispanic/Latino population to provide a foundation for future tailored interventions that address the specific needs and preferences of these subpopulations. Specific aims were to: 1. Describe cluster typologies using a virtual nominal group technique (vNGT) with key informants, and 2. Analyze how dietary fiber acquisition patterns differ across the five clusters. Methods: Using data from the Hispanic Community Health Study/Study of Latinos (HCHS/SOL), this study analyzed 9,974 participants previously grouped into five clusters through k-means cluster analysis based on 15 non-biological variables across five domains: health behaviors, psychological symptoms, social factors, economic characteristics, and cultural factors. A vNGT session with four Hispanic/Latino community-based organization staff was conducted to generate consensus descriptions of each cluster. Dietary fiber sources were analyzed using two 24-hour dietary recalls, with Kruskal-Wallis tests examining differences across clusters. Results: The vNGT process generated distinct typologies reflecting both the experiences of the participants, and the salient cluster characteristics. Cluster 1: “Well-adjusted to new environment and focus on education, but low diet quality and low physical activity.” Cluster 2: “Physically active but poor diets.” Cluster 3: “Older adults that don’t want to change their habits due to aging and isolation.” Cluster 4: “Hardworking healthy eaters/Blue collars are very active and eat healthy but work so many hours they barely have time to sleep.” Cluster 5: “Anxious due to no job and education, bad diet and activity results in depressed and anxious.” Fiber acquisition patterns differed significantly across clusters (p<0,001 for 13 of 15 food groups), with legumes/nuts showing the largest effect size (ε²=0.022). Clusters 3 and 4 maintained less refined fiber sources (legumes/nuts: 22.8% and 25.6% of daily fiber, median values), while Cluster 1 relied more on processed sources to obtain fiber (desserts, snack foods, grains/cereals). Conclusions: This research showed the feasibility of empirically identifying distinct subgroups within the Hispanic/Latino population and characterizing their unique dietary fiber acquisition strategies. The identified fiber patterns, along with the unique combination of acculturation levels, socioeconomic contexts, and health behaviors provide actionable targets for tailored interventions. These findings reinforce the notion that nutrition interventions must consider the complex interplay of factors that shape dietary behaviors within distinct Hispanic/Latino subgroups rather than treating this population as monolithic.

Download

Share

COinS