CHANCES : Anthropometry - weight, height, waist and hip

Prepared by: Mark O'Doherty, UKCRC Centre of Excellence for Public Health, Queens University Belfast, United Kingdom

1. Introduction

The cohorts were asked to report on the collection of anthropometric information so that an overview of the types and quality of the measurements employed by each cohort could be obtained. A summary of the responses to the questionnaire, and also information collected from published manuscripts, personal communication and cohort websites can be found in attached Excel file .

The assessment of height, weight, waist and hip circumference will be done separately for two groups of the measurements:

2. Weight and height measurements

Several cohorts have had technician-measured weight and height performed at baseline: all of the EPIC Elderly cohorts, HAPIEE, MORGAM, RES, SENECA, Tromsø, and Zutphen. EPIC Elderly Greece and HAPIEE also have self-reported weight and height at baseline, with a small few of the individual MORGAM cohorts using self-reported weight and height for non-ambulatory subjects or occasionally when some subjects are examined during home visits, though such cases within MORGAM were rare.

Of those cohorts with measured weight at baseline, the following have used a beam balance (considered a reliable instrument for population measurement): EPIC Elderly Bilthoven, HAPIEE, and the majority of the individual MORGAM cohorts (a small few have used digital scales). The other EPIC Elderly cohorts, SENECA, Tromsø, and Zutphen have all used calibrated scales, usually digital. RES has not commented yet on the type of scale used for recording weight measurements.

ESTHER, NIH_AARP, NHS, and SHARE have reported self-reported weight and height at baseline. However, it should be noted that NHS was a cohort of health care professionals, and a validation study focussed on self-reported weight in 1980 and 1986 found that the mean self-reported weights were about 1.5kg lighter than the measured values, which is compatible with the difference between a random casual weight in clothing and a nude, post-void morning weight. The correlations between technician-measured and self-reported weight were r=0.96 in 1980 and r=0.97 in 1986, and did not differ by level of BMI. ESTHER also had values reported by physicians using a medical conditions report, but it was found that more missing information was found on these reports than on self-reports. Therefore, information from the physicians’ medical conditions report was used to complete missing values on self-reports. However, a study that assessed the weight reported from both sources found that >65% of the weights reported by both sources were identical, with >96% being different by +/- 5kg. In addition >73% of the heights reported by both sources were identical, with >95% being different by +/- 5cm.

Weight was recorded by the European cohorts using kilograms, while the US cohorts (RES and NHS) used pounds. So far, the measurement resolution is known for the EPIC Elderly cohorts (all 100 g), SENECA, Tromsø and Zutphen (all 500 g).

For height measurements, the majority of cohorts used a stadiometer or a height rule taped vertically to a hard flat surface. Again, the type of instrument used to measure height within RES has yet to be clarified. Height was recorded by the European cohorts using centimetres, whilst the US cohorts (RES and NHS) used feet and inches. So far, the resolution is known for the EPIC Elderly cohorts (Greece - 0.1 cm; Denmark and Umeå - 1 cm; Bilthoven - 0.5 cm), SENECA (0.1 cm), Tromsø (1 cm) and Zutphen (0.1 cm).

Additionally, the majority of cohorts requested that participants removed any heavy outer garments (light clothing or underwear only allowed) and shoes for the measurement of weight and height. However, EPIC Elderly Umeå requested that participants removed shoes only, and normal clothing was allowed, though it is not clear if normal clothing included heavy outer garments.

The examination of technician-measured weight was repeated at the recontacts for RES, SENECA, Tromsø and Zutphen. The EPIC Elderly cohorts (currently unknown for EPIC Elderly Umeå), ESTHER and NHS recorded self-reported weight at each of their recontacts, whereas SHARE recorded self-reported weight at recontact 1 (wave 2) and recontact 3 (wave 4) only, and NIH_AARP recorded self-reported weight at recontact 2 only. HAPIEE did not record weight after baseline. Individual cohorts within MORGAM may have repeated measures for weight and/or height based on their own individual cohorts; MORGAM has baseline measurements standardised for weight and height.

Height was measured by the majority of cohorts at baseline only, as height is usually stable throughout adult life, though Tromsø and Zutphen measured height again in at least one of their recontacts.

3. Waist and hip measurements

Unlike weight and height, the cohorts which reported having baseline waist and hip measurements: all EPIC Elderly cohorts (except EPIC Elderly Umeå), HAPIEE, MORGAM, RES and SENECA, have used technician-measured procedures only. EPIC Elderly Umeå, ESTHER, NIH_AARP, NHS, SHARE, Tromsø and Zutphen do not have information on waist or hip measurements at baseline.

Of those cohorts with measurements recorded, the narrowest torso circumference (natural waist) or midway between the lowest rib and iliac crest was used for the waist measurement, while the widest circumference or maximum circumference over the buttocks was used for the hip measurement. The majority of cohorts reported that participants were asked to remove any heavy outer garments (light clothing or underwear only allowed) for the measurement of waist and hip. However, EPIC Elderly Bilthoven measured waist and hip over normal clothing, though it is not clear if normal clothing included heavy outer garments as they had stipulated light clothing, without shoes and no restrictive underwear for their weight and height measurement.

Waist and hip circumferences were recorded using centimetres for these cohorts at baseline. So far the measurement resolution used for waist and hip is known for EPIC Elderly Greece (0.1 cm).

RES and SENECA have recorded waist and hip measurements at each of their recontacts using similar procedures to their baseline examinations. EPIC Elderly Greece (which may be similar for the other EPIC Elderly cohorts) and HAPIEE did not record waist or hip measurements at their recontacts. As with weight and height, some individual cohorts within MORGAM may have repeated measures for waist and/or hip in their individual cohorts.

Some of the cohorts which did not record waist and hip measurements at baseline did so at some of their recontacts. Tromsø has collected technician-measured waist and hip measurements in centimetres at their T4 and T5 recontacts, with participants wearing light clothing only. Waist circumference was measured at the umbilical line (navel on stomach - usually midway between the lowest rib and iliac crest) and hip girth at the widest circumference. ESTHER began recording technician-measured waist and hip for a subsample of 3000 study participants who had a home visit at recontact 3 (2008-2010), with these results available 2011. The NIH_AARP also collected self-reported waist and hip measurements to the nearest 0.25 inches at recontact 1. They provided detailed instructions on how to take these measurements and asked participants not to respond if no tape measure was available. The NHS also followed these procedures for self-reported waist and hip measurements in their recontacts in 1986, 1996, and 2000. A validation study in HNS between self-reported and technician-measured circumferences resulted in correlation coefficients of 0.89 for the waist, 0.84 for the hip, and 0.70 for the WHR in the NHS.

4. Proposed harmonized variables

From the review of this information, a set of proposed harmonized variables has been suggested, and these have been defined in page Body composition variables.

Variables for height (cm), weight (kg), waist and hip circumference (cm) have been proposed. In addition, a variable for each of the above anthropometric measures has also been proposed which can be used to identify whether the measures are technician-measured or self-reported. It will also be possible to use two important derived anthropometric variables for the measurement of overall obesity and abdominal obesity, and these have also been proposed: body mass index (BMI) and waist-to-hip ratio (WHR).

WHR-BMI.xls (application/vnd.ms-excel)

1. Introduction

2. Weight and height measurements

3. Waist and hip measurements

4. Proposed harmonized variables

Attachments: