Files in this item

FilesDescriptionFormat

application/pdf

application/pdfQIN-DISSERTATION-2021.pdf (2MB)Restricted to U of Illinois
(no description provided)PDF

Description

Title:Physical activity participation and obesity in U.S. youth: a MIMS equating study
Author(s):Qin, Xiong
Director of Research:Zhu, Weimo
Doctoral Committee Chair(s):Zhu, Weimo
Doctoral Committee Member(s):Gao, Xiaotian; Rice, Ian; Sydnor, Synthia
Department / Program:Kinesiology & Community Health
Discipline:Kinesiology
Degree Granting Institution:University of Illinois at Urbana-Champaign
Degree:Ph.D.
Genre:Dissertation
Subject(s):Intensity
Linking
Validation
Sedentary-active
Abstract:Physical activity (PA) is associated with many health benefits. To better understand PA participation and health-related fitness of U.S. children and youth, the National Health and Nutrition Examination Survey (NHANES) National Youth Fitness Survey (NNYFS) tracked free-living PA of children in U.S. using ActiGraph in 2012. NNYFS also conducted a set of fitness tests, including cardiorespiratory endurance, muscular strength and endurance, flexibility and body composition. However, PA data was not released using ActiGraph Count, or simply the Count, an established unit with intensity cutoff scores, but rather in a new unit called “Monitor-Independent Movement Summary (MIMS).” While MIMS provides versatility as it can be used with any brand of accelerometer, it has not been validated with criterion measures, nor the cutoff scores for PA intensities were set up. This dissertations study consists of three studies and the purposes of these studies were to link MIMS with the ActiGraph Count, to validate the link and to determine the PA participation of U.S. children and youth by applying the link to the NNYFS data, respectively. Specifically, Study 1 used test equating to derive a transformation link between MIMS and the Count by using a set of ActiGraph data, in which both MIMS and the Count could be calculated, and to create an estimate for the Count called “MIMS-EQ-Count”, where “EQ” represents “equated”. Study 2 used the indirect calorimeter data from the same data of Study 1 to determine the equivalence of MIMS-EQ-Count. Finally, Study 3 used the link developed in Study 1 and verified by Study 2 to transform the MIMS data in NNYFS to MIMS-EQ-Count and to further evaluate the PA intensity, as well as to examine the relationship between body composition and PA participation. As a result of this study, MIMS was successfully linked to the Count by using the equipercentile test-equating method and creating a table of transformation. Moreover, Study 2 proved that MIMS-EQ-Count and the Count have the similar relationship with VO2, indicating that MIMS-EQ-Count could be a valid substitute for the Count. Finally, Study 3 found that the time in moderate-vigorous PA (MVPA) of U.S. children and youth decreased as age increased, but no clear differences were found between sex, as well as racial and BMI categories. This result contradicts the conclusions of previous studies that obese children and youth are less active. In addition, Study 3 found that U.S. children and youth were rather physically active and the rates to meet the PA recommendation, i.e., 60 minutes per day, were all close to or at 100%. This is clearly too high to be true and the supervising finding is likely caused by wearing accelerometers on wrist and the “sedentary active behavior”, such as when children move their wrists vigorously while playing video games, was mistakenly recognized as MVPA. In conclusion, this dissertation study successfully found a way to transform MIMS into the Count. However, due to the test site complications and the MIMS algorithm, the NNYFS data is not applicable yet as future studies need to focus on the methodology for evaluation of data from wrist-worn accelerometers.
Issue Date:2021-12-01
Type:Thesis
URI:http://hdl.handle.net/2142/113979
Rights Information:Copyright 2021 Xiong Qin
Date Available in IDEALS:2022-04-29
Date Deposited:2021-12


This item appears in the following Collection(s)

Item Statistics