A simplified approach to analysing bio-electrical impedance data in epidemiological surveys

Int J Obes (Lond). 2007 Mar;31(3):507-14. doi: 10.1038/sj.ijo.0803441. Epub 2006 Aug 8.


Background: Bio-electrical impedance analysis (BIA) is widely used to estimate body composition. It is simple, quick and cheap, but less accurate than other methods. It has potential epidemiological value, but has conventionally required validation before application.

Aims: To develop a simple method of expressing weight, height and impedance data that avoids the need for population-specific validation equations in order to facilitate epidemiological application.

Methods: Body composition was measured using the four-component model in young adults (43 males, 90 females). Impedance (R) was measured hand-foot and foot-foot. Lean mass and fat mass were adjusted for height to give lean mass index (LMI) and fat mass index (FMI). Based on theoretical principles, we generated the index 1/R, which provides an index of body water adjusted for height. Sex-specific regression models were used to investigate the relationships between (a) 1/R and LMI, and (b) body mass index (BMI) adjusted for 1/R and FMI. The success of this approach was evaluated in relation to the conventional BIA approach, using correlation analysis.

Results: 1/R was a highly significant predictor of LMI. BMI adjusted for 1/R was a significant predictor of FMI. Our approach performed as well as the conventional approach for LMI, but not for FMI.

Discussion: Direct use of BIA data, rather than their combination with population-specific equations for the prediction of total body water, proved successful at ranking individuals of both sexes in terms of LMI and FMI. The index 1/R may prove particularly valuable in epidemiological studies where ranking of LMI is required.

MeSH terms

  • Adipose Tissue
  • Adult
  • Body Composition
  • Body Height
  • Body Weight
  • Electric Impedance*
  • Epidemiologic Methods*
  • Female
  • Follow-Up Studies
  • Humans
  • Male
  • Mathematics