Skip to main content

Table 1 Comparison of ISCOLE and 2003–2006 NHANES on accelerometer data collection and data management factors

From: Improving wear time compliance with a 24-hour waist-worn accelerometer protocol in the International Study of Childhood Obesity, Lifestyle and the Environment (ISCOLE)

  ISCOLE (2012–2013) NHANES 2003-2006
Data collection
 Instrument Actigraph GT3X+ Actigraph 7164
 Instrument initialization Started data collection at midnight of the day the device was received, data collected at 80 Hz. Started data collection when device was received, data collected at 60 sec epochs.
 Instrument wear regimen Worn at mid-axillary line, lying on the iliac crest for 24 hours, removing while bathing and swimming (even though GT3X+ is a waterproof device). Worn at mid-axillary line, lying on the iliac crest during wake time, removing while bathing and swimming.
 Instrument wear instructions Instructions were read and a hard-copy paper was handed out. Instructions were read and a hard-copy paper was handed out.
 Compliance enhancing strategies In U.S. ISCOLE: Phone calls, daily visits to school, small daily incentives (e.g., erasers, stickers). Monetary incentive, available by phone to answer questions or concerns, reminder postcard for return mailing of instrument.
 Instrument return Research staff collected the devices from the children. Instruments were mailed back to NHANES warehouse.
 Data download Raw data downloaded and processed simultaneously. Processed data were in 1 sec epoch with the low frequency filter and included the 3 axis of orientation, steps, lux, and inclinometer. Processed data in 60 sec epochs contained activity counts in the x-axis and steps.
 Immediate determination of valid data Using a simple algorithm, data checked to include ≥ 4 days, including 1 weekend day with > 10 h/day of wear time Only checked whether downloaded or not downloaded
 Participant checklist Used to track instruments’ serial numbers, distribution and return dates, as well as record compliance checks. Computerized management system
 Data transfer Data were uploaded to a secure website to be retrieved by the coordinating center Not necessary
Data management
 Visual quality control check File names and file sizes were checked Screened for possible outliers and code data as reliable or questionable
 SAS dataset creation Files were re-integrated to 60 sec epochs and combined into a country specific dataset Details not available
 Automated quality control checks Using an automated process, instrument serial number, distribution and return dates were checked against the participant checklist Details not available
 Final dataset Once data checks were completed 3 datasets were created, 60 sec epoch with low frequency extension, 60 sec epoch with regular filter, and 15 sec epoch with low frequency extension. Details not available
 Data organization Date, time, sequence of data and day of the week variables were created. Date, time, sequence of data and day of the week variables were created.
Additional variables (e.g., vector magnitude) were created and processes (e.g., separating date and time) were undertaken
 Initialization errors Verified devices were initialized midnight of the first day Details not available
 Additional cleaning The last day of data was deleted as well as any day after 7 days of data collection Details not available
 Finalization Once all queries were resolved sequential 24-hour days were identified and labeled for midnight-to-midnight and noon-to-noon analysis Details not available
  1. N.B. Details for ISCOLE data collection and data management presented in Additional file 1. Details for NHANES data collections located at http://www.cdc.gov/nchs/data/nhanes/nhanes_05_06/BM.pdf.
  2. Details for NHANES data management located at http://www.cdc.gov/nchs/nhanes/nhanes2003-2004/PAXRAW_C.htm.