Skip to main content
ARS Home » Plains Area » Houston, Texas » Children's Nutrition Research Center » Research » Research Project #426613

Research Project: Assessing Food Intake and Physical Activity of Children

Location: Children's Nutrition Research Center

2017 Annual Report


Objectives
Objective 1: Determine the effect of an improved NSLP and SBP on the dietary intake and BMI of youth ages 5-18. A major focus will be comparing results from low income and food insecure youth to those with higher incomes. The baseline survey years will be NHANES 2007-2008 and 2009-2010. NHANES 2013-2014 and 2015-2016 will provide data after the implementation. Objective 2: Determine the effect of new rules on competitive foods on the dietary intake and BMI of youth ages 5-18. The baseline survey years will be NHANES 2009-2010 and 2011-2012. NHANES 2015-2016 will provide data after the implementation. Objective 3: Objective no longer being addressed due to investigator departure. Objective 4: Objective no longer being addressed due to investigator departure. Objective 5: Assess and enhance the validity and reliability of Personal Activity Location Measurement System, developed at the University of California-San Diego, through multiple, iterative studies to accurately measure preschool children's physical activity, transportation, and location. Subobjective 5A: Assess the accuracy of PALMS using GPS signal to noise ratio to correctly identify if children are indoors or outdoors as compared to staff observation. Subobjective 5B: Assess the ability of PALMS to accurately identify pedestrian travel and motor vehicle travel by preschool aged children as compared to staff observation. Subobjective 5C: Assess the reliability of locations identified by PALMS, for appropriate location tracking of children over time. Subobjective 5D: Assess whether there is seasonal variation in GPS data that can affect the identification of indoor and outdoor time in children.


Approach
Childhood obesity is a serious problem in the U.S., and has far reaching consequences for children. This research addresses dietary behaviors and physical activity among children, the two factors in the energy balance equation. Researchers will use National Health and Nutrition Examination Survey (NHANES) data to assess changes in child dietary behaviors that occur with the new school food regulations being implemented in 2012-2014, as well as the proposed competitive food guidelines that will be implemented in 2014. These new regulations provide more fruit, vegetables, and whole grain foods, and limit added sugar/fat foods in schools. Whether an improved school food environment improves student dietary intake will provide important information for the USDA. Scientists will also establish the necessary dosage level of a sustainable school-based obesity prevention program for Mexican-American adolescents, and the minimal dose of prevention associated with improved outcomes. The program demonstrated long-term (two-year) efficacy to decrease the rate of weight gain with Mexican-American adolescents. The costs associated with the program will be identified, providing important information for schools for future implementation. Scientists will also look at the validity of using the Personal Activity Location Measurement System (PALMS), developed at the University of California-San Diego, to simultaneously process accelerometer-based physical activity data and GPS-based location data to identify preschool children’s physical activity, transportation, and location, will be evaluated. Further developing and validating the PALMS protocols, processing algorithms and technologies for young children in the Houston, TX, area will provide critical information to inform policy and interventions to promote greater physical activity for young children.


Progress Report
For Objective 1, we have downloaded the latest national dietary data (2013-2014) to obtain children's dietary intake after the new school lunch guidelines were implemented in the fall of 2012. This will enable us to identify the impact of new school lunch guidelines on the dietary intake (total day and lunch meals) and body mass index (BMI) of youth ages 5-18. Analyses are underway to compare school meal intake before (using national dietary data from 2007-2012) and after implementation with the new data from 2013-2014. For Objective 5, our research team continued to make progress on the task-based observation protocol to assess and enhance the validity and reliability of the Personal Activity Location Measurement System (PALMS) to accurately measure preschool children's physical activity, transportation, and location in Houston, Texas. The study consisted of collecting location data with GPS data loggers and physical activity data with accelerometers among preschool children while the children performed specific tasks in the neighborhood around their home. Children were instructed by research staff to perform specific tasks, such as going inside and outside of their house, taking a walking trip, and taking a car trip with their parent, while wearing the monitors. While performing the tasks, the children were observed by research staff who recorded exactly what they were doing. Since the inception of the study, all data have been collected for Objective 5, processed in the PALMS system and the analyses for sub-objective 5-A and 5-B are completed. A total of 73 preschool aged children participated, with an average age of 4.7 (SD 0.82) years old, and 41% were girls. To address sub-objective 5-A, analyses showed that the existing procedures used by the PALMS software were adequate for identifying whether a child was indoors or outdoors, with a sensitivity of 74-75% and a specificity of 84-85% depending on how often the child's location was measured (every 15 seconds or every 30 seconds). These values are adequate, with the specificity meeting pre-set criteria of 80% and sensitivity approaching the preset criteria of 80%. In order to assess whether we could improve the sensitivity and specificity, additional analysis were conducted to identify whether a different threshold for signal-to-noise ratio should be used when processing the data in order to test our hypothesis. Changing the signal to noise threshold from the default value of 250 to 260, which was optimal for the 15 second data collection interval, improved the sensitivity slightly to 78.3%, but decreased the specificity to 81.5%. The optimal cut-off for the 30 second data collection interval was 249, which did not change the sensitivity or specificity in a meaningful way. To address sub-objective 5-B, analyses demonstrated the speed cut-offs used by PALMS adequately identified whether the child was traveling by car, standing or walking when analyzing the data for when the child was outdoors. The overall average specificities were 91.9% and overall average sensitivity was 80.7% for when data were collected every 15 seconds. Given the delay in data collection in year 2 as previously reported, these analyses have just been completed. Abstracts of these findings will be submitted this summer/fall.


Accomplishments