Evaluating Race/Ethnicity in Moderating Baseline Cardiometabolic Risk and Body Composition Changes in North Carolina First-Year College Women

Abstract

The roles of race/ethnicity and geographical region in the context of first-year college weight gain remain largely under-examined. The present study evaluated whether race/ethnicity: (1) at baseline was associated with greater representation in cardiometabolic health risk categories for body mass index, percent body fat, waist circumference, and waist-to-hip ratio in the full sample of 54 Black/African American and 80 White/European American first-year female undergraduates attending a North Carolina state university; and (2) moderated body composition changes between the beginning of the fall and spring semesters among the 83 participants who completed baseline and follow-up visits (N = 39 Black/African Americans). More Black/African Americans than White/European Americans had percent body fat values ≥32% at baseline; a greater proportion of White/European Americans than Black/African Americans had a waist-to-hip ratio >0.80. Among those who completed baseline and follow-up visits, White/European Americans had higher waist-to-hip ratios (unadjusted: p <0.01, adjusted for family income: p < 0.01) and waist circumferences (adjusted for family income: p < 0.05) at both time points. No strong moderating effects of race/ethnicity were detected. Preliminary results suggested that greater consideration of racial/ethnic indicators and potential regional variation in these biometric attributes among first-year college students is warranted.

Keywords:

• ethnicity
• African American
• risk factors
• socioeconomic status
• weight
• body mass index
• obesity

Acknowledgments

This research was supported by a UNC Charlotte Faculty Research Grant and a UNC Charlotte ADVANCE Grant Bonnie E. Cone Fellowship Award. The author would like to thank Assistant Provost Steve Coppola and his colleagues at the UNC Charlotte Office of Institutional Research along with the UNC Charlotte Dean of Students Office for their critical assistance in stimulating participant recruitment for this study. The author would also like to thank Suzanne Schoenefeld, Mallory Forman, and Ying-Yi Chou for their diligence and conscientiousness in ensuring high data quality and integrity over the course of the project.

Notes

1.Although hip circumference was obtained, it was only used to calculate waist-to-hip ratio measures and was not included as a separate dependent variable in the primary analyses, due to the relatively small sample size of students who attended both sessions, and to give priority to including more traditional markers of cardiometabolic health. However, researchers acknowledge that the examination of hip circumference as an independent measure of health risk is indeed an emerging and currently debated area of scientific inquiry (e.g., Heitmann, Frederiksen, & Lissner, 2004).

2.As is discussed in Rauscher & Myers (2008), capturing socioeconomic status (SES) in a valid manner is a complex undertaking, perhaps especially so in younger populations. The traditional measure of years of maternal education was initially considered as the primary SES indicator for this study. However, as the two racial/ethnic groups did not differ significantly in mean levels of maternal education attained, (t [132] = –1.2, p = 0.25), researchers believed it was important to include family income as an alternative proxy because this indicator did distinguish between groups at baseline. Family income, though still an imperfect measure, may provide a crude window into the financial resources that students theoretically had access to (or perceived they did), which could have indirectly affected variability in the biometric parameters described here. This decision is further supported by other contemporary research which has used family income as a proxy for SES among ethnically-diverse college females (Gordon et al., 2010).

3.To attempt to explain this unanticipated finding, a post-hoc analysis of variance (ANOVA) was conducted to determine if a significant change in hip circumference occurred over time. Results indicated that mean hip circumference measurements increased during the observation period (F [1, 81] = 10.0, p < 0.01). No main effect of racial/ethnic group (F [1, 81] = 0.9, p = 0.92) or interaction effect (F [1, 81] = 1.0, p = 0.32) were detected for this body composition variable.