Body Adiposity Index as a Predictor of Body Fat in an Oldest Old and Independent Cohort of Brazilian Older Adults

Evelyn Dearo Spinoza; Flavia Kurebayashi Fonte; Vanessa Amarante Carvalho; Regiane Aparecida Dos Santos; Gisele W. B. Colleoni; Maysa Seabra Cendoroglo

doi:10.4235/agmr.24.0008

Ann Geriatr Med Res > Volume 28(3); 2024 > Article

Spinoza, Fonte, Carvalho, Dos Santos, Colleoni, and Cendoroglo: Body Adiposity Index as a Predictor of Body Fat in an Oldest Old and Independent Cohort of Brazilian Older Adults

Original Article

Annals of Geriatric Medicine and Research 2024;28(3):284-290.

Published online: May 17, 2024

DOI: https://doi.org/10.4235/agmr.24.0008

Body Adiposity Index as a Predictor of Body Fat in an Oldest Old and Independent Cohort of Brazilian Older Adults

Evelyn Dearo Spinoza, Flavia Kurebayashi Fonte, Vanessa Amarante Carvalho, Regiane Aparecida Dos Santos, Gisele W. B. Colleoni^*, Maysa Seabra Cendoroglo^*

Discipline of Geriatrics and Gerontology, Department of Medicine, Paulista School of Medicine, Federal University of São Paulo, São Paulo, Brazil

Corresponding Author: Maysa Seabra Cendoroglo, MD, PhD Discipline of Geriatrics and Gerontology, Department of Medicine, Paulista School of Medicine, Federal University of São Paulo, Rua dos Otonis, 863 Vila Clementino, São Paulo, SP 04025-002, Brazil
E-mail: projeto.longevos@gmail.com

^* These authors contributed equally to this work.

Received January 10, 2024 Revised March 21, 2024 Accepted May 7, 2024

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (https://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Background

Anthropometry and body mass index (BMI) do not assess body composition or its distribution in older adults; thus, individuals may have different fat percentages but similar BMI values. The body adiposity index (BAI) was recently proposed as a feasible and inexpensive method for estimating the percentage of body fat based on measurements of hip circumference and height. The present study evaluated whether BAI and BMI are useful alternatives to dual-energy X-ray absorptiometry (DXA), which is rarely used in clinical practice, for predicting body fat in independent long-lived older adults.

Methods

In this cross-sectional study, we used DXA to calculate the percentage of body fat, which was compared with BAI and BMI values. We performed Pearson correlation analyses and used Cronbach’s alpha, described by Bland and Altman, to compare the reliability between the indexes.

Results

Among 157 evaluated individuals (73.2% women, mean age 87±3.57 years), men had a lower percentage of total fat, as assessed by DXA, and lower BAI indices than women. The correlation between BAI and DXA was moderate (r=0.59 for men and r=0.67 for women, p<0.001). We confirmed the reliability based on Cronbach’s alpha coefficients of 0.67 in men and 0.77 in women. We also observed that the BAI was strongly positively correlated with BMI in both men and women.

Conclusion

The BAI, used in combination with BMI, can be an alternative to DXA for the assessment of body fat in the oldest old in clinical practice, mainly women, and can be used to add information to BMI.

Key words: Oldest old, Adiposity, Dual-energy X-ray absorptiometry

INTRODUCTION

Obesity is characterized by an excessive accumulation of body fat resulting from a positive energy balance¹⁾ and is a risk factor for many diseases.²⁾

Venturini et al.³⁾ suggested that the lower prevalence of obesity in older adults aged >80 years may be attributed to increased mortality before older adults reach more advanced ages. However, even in the oldest old, obesity is also associated with great risk, as demonstrated by Leigh et al.,⁴⁾ who reported that obesity increased the risk of mortality and considerably increased the risk of poor health in the 18-year follow-up of older adult women born between 1921 and 1926.

Anthropometry is an important nutritional indicator in addition to weight, height, circumferences, and body mass index (BMI) in older adults. However, anthropometric measurements and BMI, which are widely used in clinical practice, do not assess body composition or distribution. Thus, older adults with similar BMIs may have different body compositions.^5-⁸⁾

The distribution of body fat changes with age, becoming more centralized in the trunk, as well as increased infiltration into muscle tissue. Accumulation of fat in the hips appears to have a benign cardiovascular effect, whereas accumulation in the waist and a high waist-to-hip ratio are associated with mortality.^9,¹⁰⁾

Mortality is correlated with BMI in a U-shaped curve pattern,¹¹⁾ including in the oldest old patients.¹²⁾ This paradox involves lipid metabolism, N-terminal pro-B-type, natriuretic peptides, prothrombotic factors, ghrelin, production of cytokines such as tumor necrosis factor and adipokines, and increased mobilization of endothelial progenitor cells.¹¹⁾ However, in a systematic review of older adults, 20 of 58 studies did not find evidence of an obesity paradox. Nevertheless, obesity is associated with comorbidities or acute diseases.¹³⁾

The UK National Institute for Health and Care Excellence guidelines indicate that the waist-to-height ratio is more accurate than BMI in estimating central adiposity, regardless of ethnic and sex differences.¹⁴⁾ Although bioimpedance analysis (BIA), ultrasonography (US), computed tomography (CT), magnetic resonance imaging (MRI), and dual-energy X-ray absorptiometry (DXA-gold standard) allow the determination of the percentages of fat and lean mass, they have the disadvantages of high cost and operational complexity,⁵⁾ which mainly affect their use in underdeveloped countries.

Considering the importance of body fat estimation, alternatives that may be useful and feasible have been investigated.⁶⁾ Bergman et al.¹⁵⁾ proposed the body adiposity index (BAI) as a method to estimate the percentage of body fat in both men and women by measuring hip circumference (HC) and height. The BAI was developed using data from the “BetaGene study” and validated in the “Triglycerides Cardiovascular Risk in African-Americans (TARA)” study; some of the advantages of this index, in addition to its use when an accurate weight measurement is not possible, are that it does not require normalization or numerical correction.

Chang et al.¹⁶⁾ were the first to apply the BAI in older adults, including 954 older adults aged 55–96 years from “The Baltimore Longitudinal Study of Aging” (BLSA). They showed that compared to DXA, the BAI was an accurate measure of adiposity in both white and black races and in women, but not in men (BMI worked better). Segheto et al.¹⁷⁾ applied the BAI to 706 Brazilians with a mean age of 36 years, residing in the state of Minas Gerais. In this study of a probabilistic population sample, the researchers observed an overestimation of fat percentage by BAI in men and an underestimation in women using DXA as the reference (concordance/Kaplan–Meier=41%).

Therefore, in the present study, we evaluated whether BAI and BMI may be useful for predicting body fat in independent long-lived older adults and be used as an alternative to (DXA), which is rarely used in clinical practice.

MATERIALS AND METHODS

Patients

This study is a cross-sectional study and included older adults aged ≥80 years who were independent, without cognitive impairment or serious illnesses, with compensated chronic illnesses, and living in the community. These individuals had been followed up since 2010 at the outpatient clinic of the Discipline of Geriatrics and Gerontology at the Federal University of São Paulo, UNIFESP, São Paulo, Brazil. All participants signed an informed consent form after being informed about the study, and the research project was approved by the ethics committee (Research project approved by the ethics committee of Federal University of São Paulo, UNIFESP under number 0225/2017). This study complied the ethical guidelines for authorship and publishing in the Annals of Geriatric Medicine and Research.¹⁸⁾ This study analyzed data from 2014 to 2016, which included 167 individuals. Of these, 10 individuals were excluded: eight for incomplete data and one individual each for body fat percentages <20% and >55%, respectively, measured by DXA.¹⁶⁾ None of the patients smoked.

Study Variables

Data on comorbidities and medications in use were reported by the patients and also obtained from their medical records. Serum creatinine levels were measured in the peripheral blood using an automated technique. Body weight was measured using a Plenna digital scale (with a maximum capacity of 150 kg) with the individuals barefoot and wearing at least their own clothes. While stepping on the scale, the participants were instructed to remain upright. The average of three consecutive weight measurements was used for analysis. Height of the participants was measured using a non-extensible measuring tape fixed to the wall without a skirting board. The participants remained in erect standing position without shoes. The measurement was taken during expiration with the participants’ gaze on the horizon (forming a 90° angle between the chin and neck), and the value was marked with a measuring scale at the top of the head. The BMI was calculated as the ratio of weight (kg) to the square of height (m). The BMI cutoffs for normality or eutrophy, malnutrition, overweight, and obesity were 23–28 kg/m², <23 kg/m², 28–30 kg/m², and >30 kg/m², respectively.¹⁹⁾ This BMI classification included patients living in São Paulo, Brazil, and was similar to other classifications proposed for older adults.⁸⁾ HC was measured using a non-extensible measuring tape that circled the hip in the region with the greatest perimeter between the waist and thigh, with the individual in an orthostatic position wearing thin clothes.²⁰⁾ The BAI was calculated according to the formula proposed by Bergman et al.,¹⁵⁾ as follows.

BAI = [(HC (cm) / (height (m) × (square root) height (m)] – 18.

No cutoff points for BAI normality are available for older adults. DXA was performed using Hologic equipment by a trained professional with experience in radiological measurements at the Research Laboratory of the Discipline of Endocrinology at the Federal University of São Paulo. This non-invasive technique is the gold standard for measuring body fat, is considered safe, and can measure three body components: fat mass, fat-free mass, and bone mass.^21-²³⁾

Statistical Analysis

We performed the statistical analyses using IBM SPSS Statistics for Windows, version 21.0. The Kolmogorov–Smirnov test was used to verify the normality of the distributions of continuous variables. The Z-score was used to standardize data that did not follow the normality curve. Continuous variables are represented as means and standard deviations, whereas categorical variables are represented as absolute frequencies and percentages. We applied the univariate general linear model (GLM) test to compare variables between sexes. The significance level was set at p<0.05. We analyzed categorical variables using the chi-squared test. We also used Pearson’s correlations, Bland–Altman scatterplots, and Cronbach’s α coefficients to compare the reliability between the methods for measuring body composition.

RESULTS

We analyzed data from 157 long-lived older adults, including 42 men (26.8%) and 115 women (73.2%), aged 80–100 years. We did not observe any differences between the sexes in terms of the number of comorbidities (Table 1). Men had a lower percentage of total fat, as assessed by DXA, and lower BAI indices than women. The BAI values in both men and women showed a moderate positive correlation with DXA and a strong positive correlation with BMI (Table 2).²⁴⁾

We observed a positive correlation between BAI and DXA in the total sample and according to sex (Fig. 1A, 1C, 1E). This dispersion of data suggests an agreement between methods. The Bland–Altman dispersion plots in Fig. 1B, 1D, and 1F confirm the trends observed in the Pearson correlation plots. For the total sample of 157 individuals, the difference between the methods was −15.06 (−15.80 to −14.81) with limits of agreement for 95% of the cases (mean±1.96 standard deviation [SD]) between −24.32 and −5.79. We observed an underestimation of the values using the BAI compared with those using DXA, with only 4.45% of the sample “not meeting” the 95% agreement limit range. We confirmed the reliability of the measurements using Cronbach’s α (α=0.78). In men (n=42), the difference between the methods was −11.85 (−13.12 to −10.53) with limits of agreement for 95% of the cases (mean±1.96 SD) between −19.97 to −3.67. We also observed an underestimation of the BAI values compared with those using DXA, with only 2.38% not adhering to the 95% agreement limit. We confirmed the reliability using Cronbach’s α (α=0.67). In women (n=115), we observed a difference of −16.24 (−17.05 to −15.43) between the methods, with limits of agreement for 95% of the cases (mean±1.96 SD) between −24.81 and −7.67. We also observed data underestimation for BAI values compared with those using DXA, with only 4.34% of these older adult women not adhering to the 95% concordance limit. We confirmed the reliability based on the Cronbach’s α value (α=0.77).

DISCUSSION

This cross-sectional study was conducted on a uniquely old independently living population in Sao Paulo, Brazil to compare anthropometric measurements of BAI and BMI with fat percentage measured using DXA. This study aimed to identify whether the BAI would add information as a surrogate measure of body fat for this population. In the present study, BAI showed a moderate correlation with the fat percentage assessed by DXA in both sexes, mainly in women, similar to other studies conducted in younger Brazilian populations^17,^25-³¹⁾ and in other countries that included a small proportion of older adults.^15,^32-³⁴⁾ However, no studies have included only long-lived older adults; thus, no comparison with the present study is possible.

Studies on older adults are heterogeneous as they consider both robust and frail patients, with multiple comorbidities and different smoking habits being factors that modify the impact of obesity on mortality.^11,³⁵⁾ Even in the BLSA study, the average age was 70.4±9.5 (in those aged 65 and over), i.e., the study did not include exclusively long-lived people.¹⁶⁾ Thus, the novelty of the present study and its main difference with the BLSA is the inclusion of exclusively long-lived older adults. In the BLSA study,¹⁶⁾ BAI was not a good predictor of fat percentage; BAI tended to be overestimated in older adults with <15% fat (as assessed by DXA), and underestimated for those with a high body fat percentage (40% as assessed by DXA). The linear regression model that included BAI explained 56% of the variance in fat percentage, whereas the model that included BMI explained 33% of the variance (interaction terms of sex×BMI were significant).

In the present study, we observed an underestimation of BAI values compared with those using DXA for both men and women, contrary to the findings in older adults reported by Chang et al.,¹⁶⁾ which showed that BAI was an accurate measure of adiposity in older women. In contrast, in a study of a younger Brazilian population, Segheto et al.¹⁷⁾ reported an overestimation of the BAI in men and an underestimation in women compared with DXA values as the reference (concordance/Kaplan–Meier=41%). Thus, determining body adiposity in older individuals can be challenging, both in large population studies and in clinical practice, owing to the gold-standard DXA method being expensive; moreover, the equipment is not portable and requires several procedures to perform the test.^16,³²⁾ Bergman et al.¹⁵⁾ suggested using the BAI to minimize the challenges of costs and technical training requirements in clinical practice, proposing an index with simple measurements and low cost to identify the amount of body fat. In older adults, this scenario must be evaluated with greater caution because of the body transformations that occur with age. Height tends to reduce with increasing age due to changes in vertebral morphology, which results in an overestimation of BMI. In contrast, the BAI, which considers HC, offers an advantage because it can be obtained with less difficulty than other anthropometric measures.

Why is HC and not waist circumference a key variable for estimating body fat? Bergman et al.¹⁵⁾ used data from the “BetaGene” population study to develop the BAI. The percentage of body fat, as measured using DXA, was used as the “gold standard” for validation, and the authors used two characteristics to select the parameters for the new index. First, they considered the values with the strongest correlation with adiposity percentage: HC and height. They expected that the new surrogate index would be related to the ratio of HC to height, owing to the correlation being positive for HC and negative for height. The second consideration was independence; the minimal absolute correlation between the chosen variables suggested that they may contribute independent information to the prediction of adiposity percentage. Therefore, the authors concluded that the BAI can reflect body fat percentage in adult men and women of different ethnicities without numerical correction. In contrast, waist circumference, another frequently used anthropometric measure, reflects abdominal adipose tissue and cannot be used to distinguish deposits of visceral (VAT) and subcutaneous fat (SAT).^36-³⁸⁾ However, no published studies have proposed a physiological explanation for the relevance of HC in estimating adiposity percentage. Therefore, further research on this topic is required.

BAI can be a good predictor of body fat^15,^16,^26-^28,^33,³⁴⁾; however, it should only be used in combination with other data.^17,^30-^32,^39,⁴⁰⁾ BAI estimates body fat, whereas BMI estimates nutritional status ^17,³³⁾; furthermore, BMI does not differentiate between fat and lean mass,⁴¹⁾ and is not as accurate.^40,⁴²⁾ Therefore, the question remains: what is the best anthropometric index for measuring obesity in the oldest old? We hope that our study findings help answer this question.

The main strength of this study is its casuistic nature: we included older adults aged ≥80 years without serious health problems. During follow-up, we observed that the participants compensated for their chronic diseases. We believe that these patients survived despite being obese because they had fewer factors that could aggravate obesity. However, as this was a cross-sectional study, we could not explore whether BAI and BMI have different predictive powers for adverse outcomes in this group of patients.

In conclusion, BAI in combination with BMI can be an alternative to DXA in clinical practice for the assessment of body fat in the oldest old patients, mainly women.

ACKNOWLEDGMENTS

CONFLICT OF INTEREST

The researchers claim no conflicts of interest.

The authors declare that no generative AI or AI-assisted technologies were used in the writing process.

FUNDING

None.

AUTHOR CONTRIBUTIONS

Conceptualization, EDS, MSC; Data curation, EDS, FKF, VAC, RADS; Formal analysis, EDS, MSC; Investigation, FKF, VAC, RADS, MSC; Methodology, FKF, VAC, RADS; Project administration, MSC; Supervision, MSC; Validation, EDS, GWBC, MSC; Visualization, MSC; Writing-original draft, EDS, MSC; Writing-review & editing, GWBC, MSC.

Fig. 1.

Graphic comparison of results. (A) Pearson correlation between BAI and DXA of the total sample (r=0.72, p<0.001). (B) Bland–Altman plot (BAI and DXA of the total sample), with an alpha of 0.78. (C) Moderately significant correlation between BAI and DXA in the male group (r=0.59, p<0.001; Pearson correlation). (D) Bland–Altman plot (BAI and DXA from the male group), with an alpha of 0.67. (E) Significant correlation between BAI and DXA in the female group (r=0.67, p<0.001; Pearson correlation); (F) Bland–Altman plot (BAI and DXA from the female group), with an alpha of 0.77. DXA (fat %), percentage of body fat by dual-energy X-ray absorptiometry; BAI, body adiposity index.

Table 1.

Descriptive data of 157 independent long-lived elderly

Variable	Total (n=157)	Male (n=42)	Female (n=115)	p-value
Sex	157 (100)	42 (26.8)	115 (73.2)
Age (y)	86.77±3.92	88.12±4.54	86.28±3.57	0.09
DXA (fat %)	39.45±6.75	33.46±5.15	41.64±5.89	<0.01
Weight	67.3±13.0	71.9±14.45	60.7±10.99	0.01
Height	1.55±0.10	1.65±0.08	1.52±0.06	0.01
HC (cm)	98.58±9.37	98.12±8.07	98.72±9.83	0.76
BAI (%)	24.39±4.17	21.64±2.91	25.39±4.11	<0.01
BMI (kg/m²)	26.30±4.18	26.22±4.23	26.33±4.17	0.88
Nutrition diagnosis (OPAS)
Malnutrition	32 (20.4)	8 (19.0)	24 (20.9)	0.91
Eutrophic	73 (46.5)	21 (50.0)	52 (45.2)	0.91
Overweight	23 (14.6)	5 (11.9)	18 (15.7)	0.91
Obesity	29 (18.5)	8 (19.0)	21 (18.3)	0.91
≥3 comorbidities	134 (85.35)	34 (80.95)	100 (86.95)	0.74
Medications, n≥3	146 (93.0)	39 (92.8)	108 (93.9)	0.43
Creatinine (mg/dL)	0.99±0.33	1.13±0.30	0.94±0.33	0.43

Values are presented as number (%) or mean±standard deviation.

DXA (fat %), percentage of body fat by dual-energy X-ray absorptiometry; HC, hip circumference; BAI, body adiposity index; BMI, body mass index; OPAS, Pan American Health Organization.

Table 2.

Correlation indexes (Pearson correlation r) between percentage of fat by DXA and anthropometric measurements of the oldest old in relation to sex

	DXA	BAI
Male (n=42)
BMI	0.67*	0.84*
BAI	0.59*	1
DXA	1	0.59*
Female (n=115)
BMI	0.74*	0.88*
BAI	0.67*	1
DXA	1	0.67*

DXA, dual-energy X-ray absorptiometry; BMI, body mass index; BAI, body adiposity index.

^* p<0.001.

REFERENCES

1. Buffa R, Floris GU, Putzu PF, Marini E. Body composition variations in ageing. Coll Antropol 2011;35:259–65.

2. Eskinazi FM, Marques AP, Leal MC, Duque AM. [Aging and epidemic obesity]. UNOPAR Cient Ciênc Biol Saúde 2011;13:295–8.

3. Venturini CD, Engroff P, Gomes I, De Carli GA. [Obesity prevalence associated with caloric intake, glycemia, and serum lipides in a sample population of elderly adults from Southern Brazil]. Rev Bras Geriatr Gerontol Rio de Janeiro 2013;16:591–601.

4. Leigh L, Byles JE, Jagger C. BMI and healthy life expectancy in old and very old women. Br J Nutr 2016;116:692–9.

5. Santos RR, Bicalho MA, Mota P, de Oliveira DR, de Moraes EN. Obesity in the elderly. Rev Med Minas Gerais 2013;23:64–73.

6. Previato HD, Dias AP, Nemer AS, Nimer M. [Association between body mass index and waist circumference in elderly women, Ouro Preto, Minas Gerais, Brasil]. Nutr Clin Diet Hosp 2014;34:25–30.

7. Batsis JA, Mackenzie TA, Bartels SJ, Sahakyan KR, Somers VK, Lopez-Jimenez F. Diagnostic accuracy of body mass index to identify obesity in older adults: NHANES 1999-2004. Int J Obes (Lond) 2016;40:761–7.

8. Kiskac M, Soysal P, Smith L, Capar E, Zorlu M. What is the optimal body mass index range for older adults? Ann Geriatr Med Res 2022;26:49–57.

9. de Koning L, Merchant AT, Pogue J, Anand SS. Waist circumference and waist-to-hip ratio as predictors of cardiovascular events: meta-regression analysis of prospective studies. Eur Heart J 2007;28:850–6.

10. Coutinho T, Goel K, Correa de Sa D, Kragelund C, Kanaya AM, Zeller M, et al. Central obesity and survival in subjects with coronary artery disease: a systematic review of the literature and collaborative analysis with individual subject data. J Am Coll Cardiol 2011;57:1877–86.

11. Donini LM, Pinto A, Giusti AM, Lenzi A, Poggiogalle E. Obesity or BMI paradox? Beneath the tip of the iceberg. Front Nutr 2020;7:53.

12. Lv Y, Mao C, Gao X, Ji JS, Kraus VB, Yin Z, et al. The obesity paradox is mostly driven by decreased noncardiovascular disease mortality in the oldest old in China: a 20-year prospective cohort study. Nat Aging 2022;2:389–96.

13. Drame M, Godaert L. The obesity paradox and mortality in older adults: a systematic review. Nutrients 2023;15:1780.

14. National Institute for Health and Care Excellence (NICE): Obesity: identification, assessment and management [Internet]. London, UK: NICE; c2014 [cited 2024 Mar 18]. Available from: https://www.nice.org.uk/guidance/cg189/resources/obesity-identificat ion-assessment-and-management-pdf-35109821097925.

15. Bergman RN, Stefanovski D, Buchanan TA, Sumner AE, Reynolds JC, Sebring NG, et al. A better index of body adiposity. Obesity (Silver Spring) 2011;19:1083–9.

16. Chang H, Simonsick EM, Ferrucci L, Cooper JA. Validation study of the body adiposity index as a predictor of percent body fat in older individuals: findings from the BLSA. J Gerontol A Biol Sci Med Sci 2014;69:1069–75.

17. Segheto W, Coelho FA, Cristina Guimaraes da Silva D, Hallal PC, Marins JC, Ribeiro AQ, et al. Validity of body adiposity index in predicting body fat in Brazilians adults. Am J Hum Biol 2017;29:e22901.

18. Noh JH, Jung HW, Ga H, Lim JY. Ethical guidelines for publishing in the Annals of Geriatric Medicine and Research. Ann Geriatr Med Res 2022;26:1–3.

19. Lebrao ML, Duarte YA. [SABE- Health, Well-being and Aging: the SABE project in the city of São Paulo: an initial approach] [Internet]. Brasilia, Brazil: Pan American Health Organization; 2003 [cited 2024 mar 18]. Available from: https://repositorio.usp.br/item/001410546.

20. International Society for the Advancement of Kinanthropometry (ISKA). International standards for anthropometric assessment [Internet]. Potchefstroom, South Africa: ISKA; 2001 [cited 2024 Mar 18]. Available from: https://trove.nla.gov.au/work/9868163?q&versionId=11459452.

21. Jackson AS, Pollock ML, Ward A. Generalized equations for predicting body density of women. Med Sci Sports Exerc 1980;12:175–81.

22. Lohman TG. Advances in body composition assessment. Current issues in exercise science series. Monograph Number 3. Champaign, IL: Human kinetics Human Kinetics; 1992.

23. Paiva CR, Gaya AC, Bottaro M, Neto JT. [Bioimpedance vs dual-energy X-ray absorptiometry to assess body composition in children]. Revista Unimontes Científica 2002;3:1–10.

24. Dancey C, Reidy J. Estatistica Sem Matematica para Psicologia: Usando SPSS para Windows [Statistics without math for psychology using SPSS for Windows]. 3rd ed. Porto Alegre, Brazil: Artmed; 2006.

25. Barreira TV, Harrington DM, Staiano AE, Heymsfield SB, Katzmarzyk PT. Body adiposity index, body mass index, and body fat in white and black adults. JAMA 2011;306:828–30.

26. Godoy-Matos AF, Moreira RO, Valerio CM, Mory PB, Moises RS. A new method for body fat evaluation, body adiposity index, is useful in women with familial partial lipodystrophy. Obesity (Silver Spring) 2012;20:440–3.

27. Silva MI, Vale BS, Lemos CC, Torres MR, Bregman R. Body adiposity index assess body fat with high accuracy in nondialyzed chronic kidney disease patients. Obesity (Silver Spring) 2013;21:546–52.

28. Dias J, de Avila M, de Oliveira Damasceno V, Goncalves R, Barbosa FP, Lamounier JA, et al. Applicability of body adiposity index in estimating body fat of young Brazilian women. Rev Bras Med Esporte 2014;20:17–20.

29. Goncalves R, Mascarenhas LP, Liebl EC, Lima VA, Souza WB, Grzelczak MT, et al. [Degree of agreement with BAI and BMI body fat percentage]. Revista Brasileira de Qualidade de Vida 2014;6:8–16.

30. Souza WC, Mascarenhas LP, Reiser FC, Souza WB, Lima VA, Muniz MA, et al. [Relationship between the body adiposity, neck circumference and body mass index in sedentary women]. Rev Bras Obes Nutr Emagrecimento 2014;8:159–64.

31. Guedes AC, Biscuola AP, Lima MC. [Comparison between body mass index and adiposity index body in adult male]. Rev Bras Obes Nutr Emagrecimento 2015;9:235–42.

32. Lopez AA, Cespedes ML, Vicente T, Tomas M, Bennasar-Veny M, Tauler P, et al. Body adiposity index utilization in a Spanish Mediterranean population: comparison with the body mass index. PLoS One 2012;7:e35281.

33. Johnson W, Chumlea WC, Czerwinski SA, Demerath EW. Concordance of the recently published body adiposity index with measured body fat percent in European-American adults. Obesity (Silver Spring) 2012;20:900–3.

34. Sun G, Cahill F, Gulliver W, Yi Y, Xie Y, Bridger T, et al. Concordance of BAI and BMI with DXA in the Newfoundland population. Obesity (Silver Spring) 2013;21:499–503.

35. Watanabe D, Yoshida T, Watanabe Y, Yamada Y, Miyachi M, Kimura M. Frailty modifies the association of body mass index with mortality among older adults: Kyoto-Kameoka study. Clin Nutr 2024;43:494–502.

36. Kullberg J, von Below C, Lonn L, Lind L, Ahlstrom H, Johansson L. Practical approach for estimation of subcutaneous and visceral adipose tissue. Clin Physiol Funct Imaging 2007;27:148–53.

37. Bosy-Westphal A, Booke CA, Blocker T, Kossel E, Goele K, Later W, et al. Measurement site for waist circumference affects its accuracy as an index of visceral and abdominal subcutaneous fat in a Caucasian population. J Nutr 2010;140:954–61.

38. Cameron AJ, Magliano DJ, Shaw JE, Zimmet PZ, Carstensen B, Alberti KG, et al. The influence of hip circumference on the relationship between abdominal obesity and mortality. Int J Epidemiol 2012;41:484–94.

39. Gibson CD, Atalayer D, Flancbaum L, Geliebter A. Body adiposity index (BAI) correlates with BMI and body fat pre- and post-bariatric surgery but is not an adequate substitute for BMI in severely obese women. Int J Body Compos Res 2012;10:9–14.

40. Freedman DS, Thornton JC, Pi-Sunyer FX, Heymsfield SB, Wang J, Pierson RN Jr, et al. The body adiposity index (hip circumference ÷ height(1.5)) is not a more accurate measure of adiposity than is BMI, waist circumference, or hip circumference. Obesity (Silver Spring) 2012;20:2438–44.

41. Sung YA, Oh JY, Lee H. Comparison of the body adiposity index to body mass index in Korean women. Yonsei Med J 2014;55:1028–35.

42. Appelhans BM, Kazlauskaite R, Karavolos K, Janssen I, Kravitz HM, Dugan S, et al. How well does the body adiposity index capture adiposity change in midlife women? The SWAN fat patterning study. Am J Hum Biol 2012;24:866–9.