The prevalence of childhood obesity is increasing globally(1). While BMI is commonly used to define obesity, it is unable to differentiate between fat and muscle mass, leading to calls to measure body composition specifically(2). While several tools are available to assess body composition in infancy, it is unclear if they are directly comparable. Among a subset of healthy infants born to mothers participating in a randomised controlled trial of a preconception and antenatal nutritional supplement(3), measurements were made at ages 6 weeks (n = 58) and 6 months (n = 70) using air displacement plethysmography (ADP), whole-body dual-energy X-ray absorptiometry (DXA), and bioelectrical impedance spectroscopy (BIS). Estimates of percentage fat mass (%FM) were compared using Cohen’s kappa statistic (κ) and Bland-Altman analysis (4,5). There was none to weak agreement when comparing tertiles of %FM (κ = 0.15–0.59). When comparing absolute values, the bias (i.e., mean difference) was smallest when comparing BIS to ADP at 6 weeks (+1.7%). A similar bias was observed at 6 months when comparing DXA to ADP (+1.8%). However, when comparing BIA to DXA at both ages, biases were much larger (+7.6% and +4.7% at 6 weeks and 6 months, respectively). Furthermore, there was wide interindividual variance (limits of agreement [LOA] i.e., ± 1.96 SD) for each comparison. At 6 weeks, LOA ranged from ± 4.8 to ± 6.5% for BIA vs. DXA and BIA vs. ADP, respectively. At 6 months, LOA were even wider, ranging from ± 7.3 to ± 8.1% (DXA vs. ADP and BIA vs. DXA, respectively). Proportional biases were apparent when comparing BIS to the other tools at both ages, with BIS generally overestimating %FM more among infants with low adiposity. In addition to differences according to tool type, within-tool factors impacted body composition estimation. For ADP measurements, the choice of FFM density reference (Fomon vs. Butte) had minimal impact; however, choice of DXA software version (GE Lunar enCORE basic vs. enhanced) and BIS analysis approach (empirical equation vs. mixture theory prediction) led to very different estimates of body composition. In conclusion, when comparing body composition assessment tools in infancy, there was limited agreement between three commonly used tools. Therefore, researchers and clinicians must be cautious when conducting longitudinal analyses or when comparing findings across studies, as estimates are not comparable across tools.

Various body indicators are used to predict health risks. However, controversies still exist regarding the best indicators to predict CVD. Using a large number of measurements, our aim was to assess their associations with blood pressure (BP) and to identify the most relevant parameters to be used in health surveillance studies. The population included 589 students (67·2 % women) aged 20–25 years from Constantine (Algeria). Sixteen parameters were considered, including crude body measurements, ratios and body fat indicators based on bioelectrical impedance analysis (BIA). We used multi-adjusted linear regression models to assess the associations between body measurements and BP. According to WHO definitions, underweight, overweight-without obesity, obesity and hypertension (HT) were identified in 6·1, 18·0, 2·4 and 5·1 % of the subjects, respectively. Prevalence of HT was higher in men than in women (11·9 % v. 1·8 %; P < 0·001). In the whole sample, almost all indicators were positively associated with systolic and diastolic BP. The suprailiac skinfold had the strongest associations with systolic (β = 3·498; P < 0·001) and diastolic (β = 2·436; P < 0·001) BP, and as a whole, arm circumferences and weight were also good candidates. The currently used BMI, waist-to-hip, waist-to-height ratio and BIA indictors also predicted BP, but they did not seem to be better determinants of BP than crude anthropometric measurements. This study showed that overweight and HT were already found in the present population of young Algerian adults. Most body indicators were highly associated with BP, but simple anthropometric measurements appeared to be particularly useful to predict BP.

Early life exposures and growth patterns may affect long-term risk of chronic non-communicable diseases (NCD). We followed up in adolescence two Zambian cohorts (n 322) recruited in infancy to investigate how two early exposures – maternal HIV exposure without HIV infection (HEU) and early growth profile – were associated with later anthropometry, body composition, blood lipids, Hb and HbA1c, blood pressure and grip strength. Although in analyses controlled for age and sex, HEU children were thinner, but not shorter, than HIV-unexposed, uninfected (HUU) children, with further control for socio-demographic factors, these differences were not significant. HEU children had higher HDL-cholesterol than HUU children and marginally lower HbA1c but no other biochemical or clinical differences. We identified three early growth profiles – adequate growth, declining and malnourished – which tracked into adolescence when differences in anthropometry and body fat were still seen. In adolescence, the early malnourished group, compared with the adequate group, had lower blood TAG and higher HDL, lower grip strength (difference: −1·87 kg, 95 % CI −3·47, −0·27; P = 0·02) and higher HbA1c (difference: 0·5 %, 95 % CI 0·2, 0·9; P = 0·005). Lower grip strength and higher HbA1c suggest the early malnourished children could be at increased risk of NCD in later life. Including early growth profile in analyses of HIV exposure reduced the associations between HIV and outcomes. The results suggest that perinatal HIV exposure may have no long-term effects unless accompanied by poor early growth. Reducing the risk of young child malnutrition may lessen children’s risk of later NCD.

There is limited information as to whether people who experience severe acute malnutrition (SAM) as young children are at increased risk of overweight, high body fat and associated chronic diseases in later life. We followed up, when aged 7–12 years, 100 Zambian children who were hospitalised for SAM before age 2 years and eighty-five neighbourhood controls who had never experienced SAM. We conducted detailed anthropometry, body composition assessment by bioelectrical impedance and deuterium dilution (D2O) and measured blood lipids, Hb and HbA1c. Groups were compared by linear regression following multiple imputation for missing variables. Children with prior SAM were slightly smaller than controls, but differences, controlling for age, sex, socio-economic status and HIV exposure or infection, were significant only for hip circumference, suprailiac skinfold and fat-free mass index by D2O. Blood lipids and HbA1c did not differ between groups, but Hb was lower by 7·8 (95 % CI 0·8, 14·7) g/l and systolic blood pressure was 3·4 (95 % CI 0·4, 6·4) mmHg higher among the prior SAM group. Both anaemia and high HbA1c were common among both groups, indicating a population at risk for the double burden of over- and undernutrition and associated infectious and chronic diseases. The prior SAM children may have been at slightly greater risk than the controls; this was of little clinical significance at this young age, but the children should be followed when older and chronic diseases manifest.

We have previously shown that children with a bioelectrical impedance spectroscopy phase angle at 50° (PA 50°) of <2.7 on postoperative day 2 had a four-fold increase in the risk of prolonged paediatric intensive care length of stay. In this study, we demonstrate a relationship between a baseline measure of phase angle 200/5° and postoperative length of stay.

Carcass data were collected from 24 kids (average live weight of 12.5±5.5 kg; range 4.5 to 22.4 kg) of Jarmelista Portuguese native breed, to evaluate bioelectrical impedance analysis (BIA) as a technique for prediction of light kid carcass and muscle chemical composition. Resistance (Rs, Ω) and reactance (Xc, Ω), were measured in the cold carcasses with a single frequency bioelectrical impedance analyzer and, together with impedance (Z, Ω), two electrical volume measurements (VolA and VolB, cm2/Ω), carcass cold weight (CCW), carcass compactness and several carcass linear measurements were fitted as independent variables to predict carcass composition by stepwise regression analysis. The amount of variation explained by VolA and VolB only reached a significant level (P<0.01 and P<0.05, respectively) for muscle weight, moisture, protein and fat-free soft tissue content, even so with low accuracy, with VolA providing the best results (0.326⩽R2⩽0.366). Quite differently, individual BIA parameters (Rs, Xc and Z) explained a very large amount of variation in dissectible carcass fat weight (0.814⩽R2⩽0.862; P<0.01). These individual BIA parameters also explained a large amount of variation in subcutaneous and intermuscular fat weights (respectively 0.749⩽R2⩽0.793 and 0.718⩽R2⩽0.760; P<0.01), and in muscle chemical fat weight (0.663⩽R2⩽0.684; P<0.01). Still significant but much lower was the variation in muscle, moisture, protein and fat-free soft tissue weights (0.344⩽R2⩽0.393; P<0.01) explained by BIA parameters. Still, the best models for estimation of muscle, moisture, protein and fat-free soft tissue weights included Rs in addition to CCW, and accounted for 97.1% to 99.8% (P<0.01) of the variation observed, with CCW by itself accounting for 97.0% to 99.6% (P<0.01) of that variation. Resistance was the only independent variable selected for the best model predicting subcutaneous fat weight. It was also selected for the best models predicting carcass fat weight (combined with carcass length, CL; R2=0.943; P<0.01) and intermuscular fat weight (combined with CCW; R2=0.945; P<0.01). The best model predicting muscle chemical fat weight combined CCW and Z, explaining 85.6% (P<0.01) of the variation observed. These results indicate BIA as a useful tool for prediction of light kids’ carcass composition.

The present paper will use fat mass percentage (FM%) obtained via BOD POD® air-displacement plethysmography (FMADP%) to examine the relative validity of (1) anthropometric measurements/indices and (2) of FM% assessed with equations (FMeq%) based on skinfold thickness and bioelectrical impedance (BIA). In 480 Belgian children (aged 5–11 years) weight, height, skinfold thickness (triceps and subscapular), body circumferences (mid-upper arm, waist and hip), foot-to-foot BIA (Tanita®) and FMADP% were measured. Anthropometric measurements and calculated indices were compared with FMADP%. Next, published equations were used to calculate FMeq% using impedance (equations of Tanita®, Tyrrell, Shaefer and Deurenberg) or skinfold thickness (equations of Slaughter, Goran, Dezenberg and Deurenberg). Both indices and equations performed better in girls than in boys. For both sexes, the sum of skinfold thicknesses resulted in the highest correlation with FMADP%, followed by triceps skinfold, arm fat area and subscapular skinfold. In general, comparing FMeq% with FMADP% indicated mostly an age and sex effect, and an increasing underestimation but less dispersion with increasing FM%. The Tanita® impedance equation and the Deurenberg skinfold equation performed the best, although none of the used equations were interchangeable with FMADP%. In conclusion, the sum of triceps and subscapular skinfold thickness is recommended as marker of FM% in the absence of specialised technologies. Nevertheless, the higher workload, cost and survey management of an immobile device like the BOD POD® remains justified.

The aim of the present study was to analyse the agreement of bioelectrical impedance analysis (BIA) compared with dual-energy X-ray absorptiometry (DXA) and MRI in estimating body fat, skeletal muscle and visceral fat during a 12-month weight loss intervention. A total of nineteen obese adults (twelve females, seven males) aged 20·2–48·6 years, mean BMI 34·6 (se 0·6) kg/m2, participated in the study. Body fat, skeletal muscle and visceral fat index were measured by BIA (Omron BF-500; Omron Medizintechnik) and compared with DXA (body fat and skeletal muscle) at baseline, 5 and 12 months, and with MRI (visceral fat) at baseline and 5 months. The subjects lost 8·9 (se 1·8) kg (9·0 (se 1·7) %) of body weight during the 12-month intervention. BIA, as compared to DXA, accurately assessed loss of fat (7·0 (se 1·5) v. 7·0 (se 1·4) kg, P= 0·94) and muscle (1·0 (se 0·2) v. 1·4 (se 0·3) kg, P= 0·18). While body fat was similar by the two methods, skeletal muscle was underestimated by 1–2 kg using BIA at each time point. Compared to MRI, BIA overestimated visceral fat, especially in males. BIA and DXA showed high correlations for kg fat, both cross-sectionally and longitudinally (r 0·91–0·99). BIA, compared with DXA and MRI, detected kg muscle and visceral fat more accurately cross-sectionally (r 0·77–0·87 and r 0·40–0·78, respectively) than their changes longitudinally (r 0·24–0·61 and r 0·46, respectively). BIA is at its best when assessing the amount or changes in fat mass. It is a useful method for measuring skeletal muscle, but limited in its ability to measure visceral fat.

Accurate assessment of neonatal body composition is essential to studies investigating neonatal nutrition or developmental origins of obesity. Bioelectrical impedance analysis or bioimpedance analysis is inexpensive, non-invasive and portable, and is widely used in adults for the assessment of body composition. There are currently no prediction algorithms using bioimpedance analysis in neonates that have been directly validated against measurements of fat-free mass (FFM). The aim of the study was to evaluate the use of bioimpedance analysis for the estimation of FFM and percentage of body fat over the first 4 months of life in healthy infants born at term, and to compare these with estimations based on anthropometric measurements (weight and length) and with skinfolds. The present study was an observational study in seventy-seven infants. Body fat content of infants was assessed at birth, 6 weeks, 3 and 4·5 months of age by air displacement plethysmography, using the PEA POD body composition system. Bioimpedance analysis was performed at the same time and the data were used to develop and test prediction equations for FFM. The combination of weight+sex+length predicted FFM, with a bias of < 100 g and limits of agreement of 6–13 %. Before 3 months of age, bioimpedance analysis did not improve the prediction of FFM or body fat. At 3 and 4·5 months, the inclusion of impedance in prediction algorithms resulted in small improvements in prediction of FFM, reducing the bias to < 50 g and limits of agreement to < 9 %. Skinfold measurements performed poorly at all ages.

Indices for lean and fat mass adjusted for height derived from bioelectrical impedance for children aged 7 years have been published previously and their usefulness in the clinical assessment of undernutrition has been demonstrated. However, there is a need for norms that cover a wider age range and to explore their functional significance. The aim of the present study is to derive lean and fat indices for children aged 7–11 years and investigate associations with objective measures of cardio-respiratory fitness and grip strength. Subjects were 9574 children participating in the Avon Longitudinal Study of Parents and Children (ALSPAC). Bioelectrical impedance analysis (BIA) data collected longitudinally between ages 7 and 11 were used to derive lean and fat indices using the method of standardised residuals. Cardio-respiratory fitness (CRF) (9 years) and grip strength (11 years) were also measured. Correlation coefficients and 95 % CI were calculated to assess the strength of association between lean index, fat index and CRF and grip strength. Equations for calculating lean and fat indices in children aged 7–11 years relative to the ALSPAC population are presented. Lean index was linearly associated with CRF (rboys 0·20 (95 % CI 0·15, 0·25), rgirls 0·26 (95 % CI 0·22, 0·30)) and grip strength (rboys 0·29 (95 % CI 0·26, 0·32), rgirls 0·26 (95 % CI 0·23, 0·29)). BMI showed slightly weaker associations, while fat index was unrelated to either CRF or grip strength. Lean indices relate to muscle function and fitness while fat index does not.

Measurements of body composition are crucial in identifying HIV-infected patients at risk of malnutrition. No information is available on the validity of indirect body composition methods in African HIV-infected outpatients. Our first aim was to test the validity of fifteen published equations, developed in whites, African-Americans and/or Africans who were or not HIV-infected, for predicting total body water (TBW) from bioelectrical impedance analysis (BIA) in HIV-infected patients. The second aim was to develop specific predictive equations. Thirty-four HIV-infected patients without antiretroviral treatment and oedema at the beginning of the study (age 39 (sd 7) years, BMI 18·7 (sd 3·7) kg/m2, TBW 30·4 (sd 7·2) kg) were measured at inclusion then 3 and 6 months later. In the resulting eighty-eight measurements, we compared TBW values predicted from BIA to those measured by 2H dilution. Range of bias values was 0·1–4·3 kg, and errors showed acceptable values (2·2–3·4 kg) for fourteen equations and a high value (10·4) for one equation. Two equations developed in non-HIV-infected subjects showed non-significant bias and could be used in African HIV-infected patients. In the other cases, poor agreement indicated a lack of validity. Specific equations developed from our sample showed a higher precision of TBW prediction when using resistance at 1000 kHz (1·7 kg) than at 50 kHz (2·3 kg), this latter precision being similar to that of the valid published equations (2·3 and 2·8 kg). The valid published or developed predictive equations should be cross-validated in large independent samples of African HIV-infected patients.

We tested the validity and reliability of the BioSpace InBody 320, Omron and Bod-eComm body composition devices in men and women (n 254; 21–80 years) and boys and girls (n 117; 10–17 years). We analysed percentage body fat (%BF) and compared the results with dual-energy X-ray absorptiometry (DEXA) in adults and compared the results of the InBody with underwater weighing (UW) in children. All body composition devices were correlated (r 0·54–0·97; P ≤ 0·010) to DEXA except the Bod-eComm in women aged 71–80 years (r 0·54; P = 0·106). In girls, the InBody %BF was correlated with UW (r 0·79; P ≤ 0·010); however, a more moderate correlation (r 0·69; P ≤ 0·010) existed in boys. Bland–Altman plots indicated that all body composition devices underestimated %BF in adults (1·0–4·8 %) and overestimated %BF in children (0·3–2·3 %). Lastly, independent t tests revealed that the mean %BF assessed by the Bod-eComm in women (aged 51–60 and 71–80 years) and in the Omron (age 18–35 years) were significantly different compared with DEXA (P ≤ 0·010). In men, the Omron (aged 18–35 years), and the InBody (aged 36–50 years) were significantly different compared with DEXA (P = 0·025; P = 0·040 respectively). In addition, independent t tests indicated that the InBody mean %BF in girls aged 10–17 years was significantly different from UW (P = 0·001). Pearson's correlation analyses demonstrated that the Bod-eComm (men and women) and Omron (women) had significant mean differences compared with the reference criterion; therefore, the %BF output from these two devices should be interpreted with caution. The repeatability of each body composition device was supported by small CV ( < 3·0 %).

Insulin use is common in type 2 diabetes and is frequently accompanied by weight gain, the composition of which is poorly understood. The present study evaluates insulin-induced body composition changes. Body weight and composition of thirty-two type 2 diabetic patients undergoing their first 12 months of insulin therapy were compared with those observed in thirty-two type 2 diabetic patients previously treated on insulin (minimum 1 year). Body composition was determined by simultaneous body water spaces (bioelectrical impedance analysis) and body density measurements. After 6 months, glycosylated Hb (HbA1c) significantly improved in the newly treated group (P<0·0001), but remained stable in those treated previously. HbA1c did not differ between 6 and 12 months in the two groups. Body weight significantly (P=0·04) changed over 12 months in those newly treated only (+2·8 kg), essentially comprising fat-free mass (P=0·044). Fat mass remained unchanged (P=0·85) as did total body water, while extracellular: total body water ratio tended to increase in those newly treated (P=0·059). Weight changes correlated with HbA1c changes (R2 0·134, P=0·002) in the initial 6 months only. Insulin therapy leads to weight gain (2·8 kg), predominantly fat-free mass, over 12 months. After 6 months, newly treated patients continued gaining weight despite an unchanged HbA1c, suggesting the potential anabolic role of insulin in subsequent gains. Therefore, in the initial 6 months, weight gain can be attributed to a ‘glucose control-related effect’ and further gain appears to be due to a ‘non-glucose control-related’ effect of insulin treatment.

The present study examined the consistency of bioelectrical impedance analysis (BIA) and anthropometric measurements in body composition analysis in a field study of prepubertal children using a representative group of 2286 5–7-year-old children from Kiel, north-west Germany. Body composition was assessed using anthropometric measures (A; four skinfolds) and BIA. Various published algorithms (according to and for A, , and for BIA and for a combined approach) were used to estimate body composition. Using A resulted in a sum of four skinfolds varying between age-dependent median values of 24·0 and 28·2 mm in boys and 30·5 and 33·3 mm in girls. When fat mass (FM) was calculated from A, age- and algorithm-dependent differences in median values were observed, with values varying between 8·5 and 14·6 % for boys and 11·1 and 14·9 % for girls. Using different algorithms ( v. ) only minor inconsistencies were observed. BIA-derived resistance index (height2/resistance) varied between 18·8 and 24·4 cm2/Ω for boys and 17·1 and 19·0 cm2/Ω for girls. Using four different algorithms to estimate FM from BIA data resulted in high intra-individual variances in percentage FM (from 13·8 to 33·4) as well as in the prevalence of overweight (from 14·7 to 98·4 % for boys and from 42·3 to 98·5 % for girls). Data obtained using the different BIA algorithms showed some, or even marked, inconsistencies as well as systematic deviations (an overestimation of FM at low percentage FM, v. ). When comparing BIA with A, BIA systematically overestimated FM. The differences between the results were influenced by BMI, gender and height. Considerable inconsistencies were observed at low BMI (<10th percentile) for girls and for small children. Although the within-observer as well as between-observer CV for both techniques are acceptable, we recommend caution in relation to the algorithms used for data analysis. The use of an interchange table of percentage FM derived from different algorithms for different percentile groups of skinfold thicknesses is recommended.

Multi-frequency bioimpedance analysis (MFBIA) was used to determine the impedance, reactance and resistance of 103 lamb carcasses (17·1–34·2 kg) immediately after slaughter and evisceration. Carcasses were halved, frozen and one half subsequently homogenized and analysed for water, crude protein and fat content. Three measures of carcass length were obtained. Diagonal length between the electrodes (right side biceps femoris to left side of neck) explained a greater proportion of the variance in water mass than did estimates of spinal length and was selected for use in the index L2/Z to predict the mass of chemical components in the carcass. Use of impedance (Z) measured at the characteristic frequency (Zc) instead of 50 kHz (Z50) did not improve the power of the model to predict the mass of water, protein or fat in the carcass. While L2/Z50 explained a significant proportion of variation in the masses of body water (r2 0·64), protein (r2 0.34) and fat (r20·35), its inclusion in multi-variate indices offered small or no increases in predictive capacity when hot carcass weight (HCW) and a measure of rib fat-depth (GR) were present in the model. Optimized equations were able to account for 65–90% of the variance observed in the weight of chemical components in the carcass. It is concluded that single frequency impedance data do not provide better prediction of carcass composition than can be obtained from measures of HCW and GR. Indices of intracellular water mass derived from impedance at zero frequency and the characteristic frequency explained a similar proportion of the variance in carcass protein mass as did the index L2/Z50.

Total body water (TBW) was measured by deuterium oxide (D2O) dilution and predicted from bioelectrical impedance (Z) in nineteen anorexic and twenty-seven control women. The equation of Kushner et al. (1992) based on the impedance index (ZI = height2/Z) gave biases of 0.9 (sd 2.5) and 0.8 (sd 2.5) litres in controls and patients respectively (NS, ANOVA). The ZI-based equation of Deurenberg et al. (1993) gave biases of 1.5 (sd 2.4) litres (NS) and 3.0 (sd 2.1) litres (P <0.001) in controls and patients respectively. Despite the fact that weight was the most powerful predictor of TBW on the study sample (n 46, r2 0.90, P < 0.0001, se of the estimate 1.6 litres, CV 5.7%), the formulas of Segal et al. (1991) and Kushner et al. (1992) based on the association of weight and ZI gave an inaccurate prediction of TBW in both control and anorexic subjects, with a bias ranging from -3.2 (sd 2.4) to 2.9 (sd 2.1) litres (P ≤0.001). Population-specific formulas based on ZI (n 46) gave a more accurate prediction of TBW by bioelectrical impedance analysis on the study subjects, with biases of -0.1 (sd 1.8) and 0.5 (sd 1.7) litres in controls and patients respectively (NS). However, the individual bias was sometimes high. It is concluded that bioelectrical impedance analysis can be used to predict TBW in anorexic women at a population level, but the predictions are less good than those based on body weight alone.

Abstract Body composition was measured by densitometry in a group of eighty-one healthy Chinese males and 124 healthy Chinese females, aged 18–67 years. Biceps, triceps, subscapular and suprailiac skinfolds were measured as well as total body bioelectrical impedance at 50 kHz. Mean heights were 1·71 (SD 0·05) and 1·61 (SD 0·06) m, body weights 66·1 (SD 9·1) and 57·2 (SD 9·4) kg and BMI 22·7 (SD 3·0) and 22·0 (SD 3·3) kg/m2 in males and females respectively. Body fat values from body density were 20·1 (SD 6·2) and 29·7 (SD 7·5)% in males and females respectively. Mean predicted body-fat values from skinfolds, bioelectrical impedance and BMI, using prediction formulas developed in Western populations, did not differ or differed only slightly from body-fat values estimated from body density. The SD of the difference for each method was 5% body fat or 3·3 kg fat-free mass (CV 6%), which is approximately equal to the accuracy level of each predictive method. Compared with densitometry the predictive methods overestimated body fat at the lower levels of body fatness, whereas at the higher levels of body fat the predictive methods tended to underestimate body fatness. It is concluded that prediction formulas for estimating body fat from skinfolds, impedance or BMI developed in Western populations are applicable and valid in the adult Chinese population. However, in very lean subjects the predictive methods overestimate body fat compared with values obtained from body density

In twenty-eight healthy subjects, ten men and eighteen women, with a range in body mass index (BMI) of 17.9–31.6 kg/m2 and an age range 20–60 years, body composition was estimated by dual-energy X-ray absorptiometry (DEXA), skinfold anthropometry (SFA) and bioelectrical impedance analysis (BIA) of the ‘whole body’and body segments. In thirteen subjects muscle mass was also estimated by 24 h urinary creatinine excretion. The relationship between fat-free mass (FFM) determined by DEXA and the impedance index of each body segment (calculated as Iength2/impedance (Z)) was analysed. The strongest correlation was between FFM (DEXA) and height2/‘whole-body’Z (Zw) (r 0.97 for the combined sexes, standard error of estimate (SEE) 2.72 kg). Separate prediction equations were found to be necessary for males and females when estimating FFM from BIA measurement of the arm (for men, r 0.93, SEE 1.98 kg; for women, r 0.75, SEE 2.87 kg). Muscle mass derived from 24 h creatinine excretion showed weak correlation with FFM (DEXA) (r 0.57, P = 0.03) and no correlation with FFM (SFA). FFM (SFA) correlated well with both FFM (DEXA) (r 0.96, SEE = 3.12 kg) and with height2/Zw (r 0.92, SEE 4.52 kg). Measurement of the impedance of the arm offers a simple method of assessing the composition of the whole body in normal individuals, and it appears comparable with other methods of assessment.

Fat-free mass (FFM) was estimated in forty-seven premenopausal Chinese women, aged 18–43 years, from anthropometric data (skinfolds, body mass index (weight/height2; BMI)) or bioelectrical impedance, using several prediction formulas for body composition from the literature, and values compared with the mean of these three individual methods used as a frame of reference. In thirty-six women these values could be compared with FFM calculated from total body water (TBW) determined by D2O dilution. The prediction formulas used were developed from studies on Caucasian adults and their validity will have to be shown in populations with different ethnic backgrounds. The mean difference between FFM predicted from BMI and the frame of reference was 0.1 kg (95 % confidence interval (CI) -0.1, 0.4), from bioelectrical impedance it was 0.5 kg (95% CI 0.3, 0.7), and from skinfolds it was -0.6 kg (95% CI -0.9, -0.4). The mean difference between FFM calculated from TBW and the frame of reference was higher (2.2 kg, 95 % CI 1.2,3.3). The results of the present study indicate that the three methods may be valid for predicting body composition in adult Chinese females, but further research is needed on development and cross-validation of prediction equations for body composition for Chinese.