Date of Award

5-2011

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Program

Nursing

Research Advisor

Patricia Cowan, Ph.D.

Committee

Michael Carter, DNSc, DNP Diane Pace, Ph.D. Elizabeth Tolley, Ph.D. Frances Tylavsky, DrPH.

Keywords

Adolescents, BMI, Bone mass, CDC, DXA, NHANES

Abstract

The effect of body mass index (BMI) on bone mineral content (BMC) among adolescents has been researched yielding mixed results. This study explored the relationship of left lower extremity (LLE) BMC on BMI across the spectrum of weight in a large nationally representative group of adolescents. This descriptive study used secondary data from the publically accessible, cross-sectional survey files of the 1999-2004 Continuous National Health Assessment and Nutritional Examination Survey (NHANES) that contained whole body dual energy X‑ray absorptiometry (DXA) data as well as BMI calculations looking at the adolescent population 12- to19-year-olds at the time of the exam. The sample contained 5416 adolescents with males 59.7%; with Caucasian 27.3%, African American 32.1%, and Mexican American 34.0%; and with underweight 3.5%, normal weight 61.8%, overweight 15.8% and obese 18.9%. The analysis of the data used SAS survey regression examining 90 domains from the sample based on the three demographic groups of gender, race/ethnicity, and age category and their possible permutations. Each domain survey regression was performed five times, once for each of the five imputations of DXA data from each survey participant's DXA scan. The five regression results were averaged in accordance with NHANES guidelines to yield a composite regression estimates with associated standard errors as results for the different domains. Results revealed a positive, linear relationship between BMI to LLE BMC with p < 0.0001 for most of the domains. Results also revealed that the relationship between BMI to LLE BMC depended greatly on the demographic factors of gender, race/ethnicity and age category. Conclusions: The greater the slope of the regression line for a particular domain meant that the domain's LLE BMC was more influenced by change in BMI. The rate that BMI affects BMC varies according to gender, race/ethnicity, and age and must be examined accordingly.

DOI

10.21007/etd.cghs.2011.0250

Download

Share

COinS