NHANES body measures data are used to:
- estimate the prevalence of overweight and obesity in the U.S. population;
- study the association between body measures and NHANES body composition assessments;
- study health conditions and health-risk factors and conditions, including cardiovascular disease, diabetes, hypertension, physical inactivity, and dietary patterns; and
- monitor growth and development in infants and children.
The target age groups for the NHANES 2003–04 body measurement assessments are as follows:
Body weight: All ages
Recumbent length: 0–47 months
Standing height: 2+ years
Upper leg length: 8+ years
Upper arm length: 2+ months
Head circumference: 0–6 months
Mid-upper arm circumference: 2+ months
Waist circumference: 2+ years
Mid-thigh circumference: 8+ years
Maximal calf circumference: 8+ years
Triceps skin fold: 2+ months
Sub scapular skin fold: 2+ months
All survey participants are eligible for the body measurement component. There were no medical, safety, or other exclusions for the body measurements protocol. For persons who used a wheelchair, the health technicians used their discretion to obtain as many measures as practical. No changes have been made to the NHANES body measurement protocol since the continuous NHANES began in 1999.
Protocol and Procedure
The body measurement assessments performed on survey participants varied according to the age of the survey participant. The body measurement protocol used age at the time of the screening interview. In some instances, several weeks elapsed between the initial screening interview and the health examination visit. The Demographics Data File includes an age variable for age at examination (RIDAGEEX).
Arm and leg measurements were made on the right side of the body. If an examinee had an amputation, medical condition, or medical appliance such as a cast that prevented measurements from being taken on the right side of the body, the examiner took measurements on the left side. The body measurements file does not identify persons with amputations due to data disclosure concerns. Body weight data for individuals who had limb amputations were excluded from the release file. This data set includes body measurements for women who were pregnant at the time of the exam. Pregnancy status is denoted by the Demographic Data File variable, RIDEXPRG. RIDEXPRG denotes pregnancy status for 8–59 year old females at the time of the health examination. Please refer to the Demographics File documentation for important information about the pregnancy status recode variables that were included in the NHANES 1999-2000, NHANES 2001–2002, and NHANES 2003–2004 public data files.
For general guidelines on standard procedures that were followed for the anthropometric measures, please refer to the Anthropometric Standardization Reference Manual (Lohman, 1988). The body measurement techniques used in the survey are illustrated in the NHANES III Anthropometric Procedures Video. The video demonstrates the NHANES III anthropometric procedures in detail. The NHANES III video illustrates the proper technique for all of the measures taken in NHANES 2003–04 except the maximum calf circumference measurement. The video is available for purchase from the U.S. Government Printing Office.
For the procedures relevant to this component, please refer to NHANES website to obtain the Survey Operations Manual, Consent Documents, and brochures.
Quality Assurance & Quality Control
The NHANES health technicians completed a 2-day training program to learn the body measurement protocol. The training included an overview of the component using the NHANES III anthropometry videotape, demonstration by an expert examiner, and supervised practice exercises with an expert examiner and practice subjects of different ages. Health technician performance in the field was monitored using direct observation, data reviews, and expert examiner evaluations. The mobile examination centers (MECs) provided a standardized environment for the collection of high quality data. As part of their required duties, the health technicians verified body measurement instrument calibration in the field; a schedule for performing the equipment calibration checks is monitored. Data entry errors were reduced through the use of automated data entry methods. Please refer to the Body Measures Component Manual for a detailed description of the quality assurance and control measures used in NHANES.
Data Processing and Editing
Data were reviewed for unusual and erroneous values. The initial checks used range criteria that were based on the NHANES III or NHANES 1999–2002 body measures data. Range checks for infants under 2 months of age were based on NHANES 1999–2002 because the NHANES III dataset does not include infants less than 2 months of age. The range checks for persons 2 months of age and older were based on NHANES III. During the data review, values that were above the 99th percentile or below the 1st percentile for a particular age or age-gender group were flagged for review. When records were flagged, the entire body measurements record was reviewed for reasonableness subject characteristics such as height, weight, age and gender were taken into consideration. Values that were unrealistic were deleted from this file. Unusual, but plausible values were flagged. The variable list includes the flag variables. Analysts may wish to review the flagged records to determine if they should be included in their analysis. No body measures data were changed and there are no imputed values in this file.
Component status code: NCHS added a final body measures component status code (BMDSTAT) to the NHANES 2003–2004 file to provide analysts with a quick method of determining complete and partial body measures examination records.
Flag Variables: The flag variables in this file denote unusual values were noted during data review. Typically, unusual values occurred when a subject was extremely short, tall, overweight or underweight. Analysts may wish to review the flagged values prior to analyzing the data.
Comment Codes: Comment codes were noted by the health technicians during data collection. The comment codes were added to document problems or situations that prevented a measurement from being taken altogether or situations that could modify the results reported. For example, the variable BMIWT is a body weight comment code. If a subject did not change into the exam gown and wore their own clothing instead, a code of BMIWT of ‘3’ denoting ‘clothing worn’ was made. Analysts should review the comment code information prior to analyzing the data.
Sample Weights: The examination sample weights should be used to analyze the body measurement data. Please refer to the NHANES Analytic Guidelines document for guidance on the use of the sample weights.