|Year : 2021 | Volume
| Issue : 3 | Page : 269-274
Discriminatory ability of mid-upper arm circumference in identifying overweight and obese adolescents: Findings from the comprehensive national nutrition survey, India
Department of Community Medicine, Independent Public Health Consultant, 9, Nirmal Vihar Ambala Cantt, Haryana, India
|Date of Submission||20-Jan-2021|
|Date of Decision||09-Jul-2021|
|Date of Acceptance||16-Jul-2021|
|Date of Web Publication||22-Sep-2021|
9, Nirmal Vihar Ambala Cantt, Haryana, Pin - 13301
Source of Support: None, Conflict of Interest: None
| Abstract|| |
Background: Overweight and obesity during adolescence is an important public health problem. However, little is known about the age-and sex-specific mid-upper arm circumference (MUAC) cut-offs for identifying overweight and obese adolescents. Objectives: The present study was planned to assess diagnostic performance of MUAC in identifying overweight and obese adolescents and estimating age specific MUAC cut-offs, separately for males and females, taking body mass index for age Z-score (BAZ) as the gold standard. Methods: The present study is secondary data analysis using Comprehensive National Nutrition Survey, India, on 31,471 adolescents. The, area under curve receiver operating characteristic curve (AUC), and Youden Index were used to estimate MUAC cut-offs for overweight (BAZ > +1) and obesity (BAZ > +2). Results: The MUAC cut-offs to identify overweight were: For 10–14 years– 22.9/23.4 cm, for 15–19 years – 27.0/25.6 cm for males and females, respectively; and for obesity were: For 10–14 years – 24.5/25.1, for 15–19 years – 28.5/28.0 cm for males and females, respectively. For overweight, among males, the age-specific cut-off ranged between 21.2 cm (10 years) and 29.8 (19 years), and for females ranged between 21.2 cm (10 years) and 26.7 cm (19 years). For obesity, it ranged between 22.4 cm (10 years) and 31.1 cm (18 years) for males, and 23.9 cm (10 years) to 26.9 cm (19 years) for females. For obesity, AUC ranged between 0.81 and 0.92, indicating good to excellent diagnostic accuracy. Conclusion: Age- and sex-specific MUAC cut-offs could be considered as a screening tool for identifying overweight and obese adolescents.
Keywords: 10–19 years' adolescents, anthropometry, body mass index, overnutrition
|How to cite this article:|
Nitika N. Discriminatory ability of mid-upper arm circumference in identifying overweight and obese adolescents: Findings from the comprehensive national nutrition survey, India. Indian J Public Health 2021;65:269-74
|How to cite this URL:|
Nitika N. Discriminatory ability of mid-upper arm circumference in identifying overweight and obese adolescents: Findings from the comprehensive national nutrition survey, India. Indian J Public Health [serial online] 2021 [cited 2021 Dec 7];65:269-74. Available from: https://www.ijph.in/text.asp?2021/65/3/269/326396
| Introduction|| |
Globally, overweight and obesity during the adolescence are increasingly recognized as an important public health problem, imposing a substantial burden on individual health. These adolescents are at an increased risk of developing metabolic syndromes, and adverse cardiovascular outcomes including hypertension, and diabetes in the adult life. Furthermore, rising prevalence of obesity is affecting adolescents of all socioeconomic status, largely attributed to changes in dietary pattern and sedentary lifestyle. Identification of overweight and obese adolescents is a foremost important step in preventing obesity-related health problems.
Body mass index (BMI) and BMI for age Z-scores (BAZ) are generally used to identify overweight and obesity. However, the use of BMI poses certain challenges in resource constrained settings. First, it requires calibrated weighing scales. Second, it requires measuring of weight and height, and then computation of BMI followed by Z-scores using growth charts. Thus, the availability of alternative measure, which is more practical, handy, and easy to use, can be considered in resource-limited settings.
Mid-upper arm circumference (MUAC) holds promise in this regard. MUAC is being used for identifying severe acute malnutrition in under-five population, also, its use had been suggested in adolescents for screening thinness. Furthermore, its use had been shown in pregnant women for classifying the nutritional status in some of the countries. Recent studies had shown that MUAC correlated well with body fat percentage and insulin resistance.,In addition, the possibility of use of MUAC and its performance in identifying obesity had been shown in 5–14-years of age group children and adolescents.,,, However, there is a paucity of research on the potential role of MUAC among adolescents, 10–19-year age group. With this background, the present study was conducted to assess the diagnostic performance of MUAC in identifying overweight and obese adolescents and estimating the cut-off values for MUAC among adolescents (age-specific, early adolescents: 10–14 years, and late adolescents: 15–19 years), separately for males and females compared with BAZ as the gold standard. This is a step toward opening a discussion on the need of simplified MUAC charts that could be used in field settings for screening overweight and obesity and further incorporation in adolescent health programs.
| Materials and Methods|| |
Study setting, design, and participants
The present study is secondary data analysis of Comprehensive National Nutrition Survey (CNNS). CNNS was conducted during 2016–2018. A multi-stage, stratified, probability proportional to size sampling approach was used for selection of rural and urban sampling units, across all 29 states of India and national capital Delhi. Then, from rural and urban sampling units, only the households with children or adolescents aged 0–19 years were randomly selected. The complete details of the CNNS survey had been published earlier. CNNS individual level data have been used for the current analysis.
Of 35,830 adolescent's information available in CNNS data, the current analysis was done on 31,471 adolescents' information (16,158 males, and 15,313 females), excluding observations with MUAC Z-score <−5 or more than +5, flagged observations for BAZ or HAZ, BMI/BAZ not available, and BMI more than 100 kg/m2.
All the anthropometric measurements were taken by trained anthropometrist working as a team (of two) in field, and standardization exercises were conducted before the start of data collection. All anthropometric equipment were calibrated daily before the start of data collection. All the measurements were taken in the household of each participant.
Standard fiberglass tapes were used for measuring MUAC on the right arm after identifying the midpoint of the upper arm and applying standard pressure while measuring the circumference. MUAC was measured to the nearest 0.1 cm. MUAC was measured twice and the mean was used for the analysis. Standing height was measured using three-piece wooden height boards with legs extended and head set in the Frankfurt position, to the nearest 0.1 cm. Body weight was assessed using digital SECA weighing scale in light clothes, to the nearest 0.1 kg. During data collection, height was measured twice, and weight was measured once. Furthermore, the average of two readings of height was considered for the analysis. Weight and mean height measures were used to compute the BMI, using the formula: Weight (kg)/height (m)2 and Z-scores were calculated using the WHO international growth reference data for children and adolescents 5–19 years of age. BMI Z-score more than +1 standard deviation (SD), and more than +2 SD was considered as overweight and obesity, respectively.
State-wise inter-observer and intra-observer technical error of the measurement had been computed for height, and MUAC, and published earlier.
The analysis was done using Stata 15.0 (StataCorp, College Station, TX, USA) and statistical software R version 3.3.3 (The R Foundation for Statistical Computing, Vienna, Austria). Descriptive statistics, including mean and SD for continuous variables, and number (percentages) for categorical variables were estimated for summarizing characteristics of participants. For estimating the prevalence of overweight and obesity, weighted analysis using national weights as provided in the data files were used. Pearson correlation coefficient, a measure of strength of linear relationship, was estimated for MUAC/BMI and MUAC/BAZ separately for different age-group and gender combinations. Receiver operating characteristic (ROC) curve was used to assess the discriminatory ability and determining the cut-off of MUAC for correctly identifying overweight and obese adolescents. Overweight and obesity defined based on BMI-z score was used as the reference or gold-standard, as it had been shown to correlate well with body fat,,, and hence used as gold standard or reference in the current analysis, against which performance of MUAC to assess the body fatness has been evaluated.
Area under ROC (AUC), an index of the test's ability to discriminate between true positives (overweight/obese) and true negatives (normal weight/non-obese) was calculated for MUAC for each combination of young/late adolescent and separately for males and females. A test with AUC value more than 0.8 is considered as good.
In addition, to determine the cut-off point for MUAC to correctly differentiate overweight/normal weight and obese/non-obese, Youden Index, J (sensitivity + specificity − 1), was calculated. It was calculated using “cutpt” command in Stata. J equals 1 for a perfect diagnostic test and 0 for a poor diagnostic test. Gender- and age-specific MUAC cut-offs, and cut-offs separately for young (10–14 years) and older adolescents (15–19 years) were estimated for overweight and obesity.
The CNNS proposal was assessed and approved by Ethics Committee of the Postgraduate Institute for Medical Education and Research in Chandigarh, India, and the Institutional Review Board of the Population Council, New York. For the current analysis, de-identified version of dataset had been used, hence, ethical approval was not required for this analysis.
| Results|| |
The mean age was 14.0 years (SD 2.6). The mean (SD) for MUAC, BMI, and BAZ was 21.9 (3.5), 18.0 (3.4), and 0.84 (1.3) respectively. The prevalence (weighted) of overweight and obesity among adolescents was 4.8% (95% confidence interval [CI]: 4.5, 5.0%) and 1.0% (95% CI: 0.9, 1.2%), respectively.
Correlation between mid-upper arm circumference versus body mass index, and body mass index for age Z-score
The relationship between MUAC versus BMI/BAZ was linear and there was significant correlation (P < 0.001) between measurements of MUAC and BMI; MUAC and BAZ, with Pearson correlation coefficients ranging between 0.69–0.78 for males, and 0.72–0.81 for females.
Diagnostic accuracy of mid-upper arm circumference for overweight and obesity
[Figure 1] depicts the ROC and AUC for MUAC to correctly identify overweight males and females among early and late adolescents against the gold standard/reference based on BAZ. The AUC ranged between 0.82–0.87, signifying good diagnostic accuracy. Similarly, [Figure 2] depicts for obesity, with AUC ranging between 0.85–0.92, indicating good to excellent diagnostic accuracy.
|Figure 1: Receiver operating characteristics curve depicting performance of mid-upper arm circumference for identifying overweight among (a) 10–14 years male; (b) 15–19 years males; (c) 10–14 years female; and (d) 15–19 years female.|
Click here to view
|Figure 2: Receiver operating characteristics curve depicting performance of mid-upper arm circumference for identifying obesity among (a) 10–14 years male; (b) 15–19 years males; (c) 10–14 years female; and (d) 15–19 years female.|
Click here to view
Age specific mid-upper arm circumference cut-offs for overweight and obesity
Overall, for 10–19 years, based on Youden index, MUAC cut-off for overweight was 24.3 cm and 24.5 cm for males and females, respectively. For males, the MUAC cut-off for overweight was 22.9 cm and 27.0 cm for 10–14 years, and 15–19 years, respectively. For females, the MUAC cut-off for overweight was 23.4 cm and 25.6 cm for 10–14 years, and 15–19 years, respectively. [Table 1] summarizes the age specific MUAC cut-offs for overweight along with the diagnostic accuracy measures. For males, the cut-off ranged between 21.2 cm (10 years) and 29.8 (19 years). For females, the range was 21.2 cm (10 years) and 26.7 cm (19 years). The sensitivity and specificity ranged between 74.5%–97.6%. MUAC age-specific cut-offs correctly classified more than 88% adolescents.
|Table 1: Receiver operating characteristics curve analysis of mid-upper arm circumference cut-off values among adolescents for overweight|
Click here to view
Overall, for 10–19 years, the MUAC cut-off for obesity was 26.2 cm and 25.6 cm for males and females, respectively. For males, the MUAC cut-off for obesity was 24.5 cm and 28.5 cm for 10–14 years, and 15–19 years, respectively. For females, the MUAC cut-off for obesity was 25.1 cm and 28.0 cm for 10–14 years, and 15–19 years, respectively. [Table 2] summarizes the age specific MUAC cut-offs for obesity along with the diagnostic accuracy measures. For males, the cut-off ranged between 22.4 cm (10 years) and 31.1 cm (18 years). For females, it ranged from 23.9 cm (10 years) to 29.9 cm (18 years). The sensitivity and specificity ranged between 68.8% and 97.9%. MUAC age-specific cut-offs correctly classified more than 90% adolescents.
|Table 2: Receiver operating characteristics curve analysis of mid-upper arm circumference cut-off values among adolescents for obesity|
Click here to view
| Discussion|| |
The present study has important inferences pertaining to the age- and sex-specific MUAC cut-offs for the identification of overweight and obese adolescents (defined based on BAZ). The MUAC was found to be correlating well with BAZ, pointing toward its applicability for screening overweight/obese adolescents. Furthermore, MUAC was found to have good to excellent AUC values for age-specific cut-offs, suggesting the equivalent diagnostic accuracy in detecting overweight/obesity compared to BMI z-score.
The present study found a significant Pearson correlation coefficient of 0.69–0.75 between MUAC and BAZ, and this corroborated with the findings from the earlier studies., The present study found AUC in the range of 0.82–0.96, signifying good-excellent diagnostic accuracy of age-specific cut-offs of MUAC in correctly classifying BAZ defined overweight/obese. Similarly, a study conducted in India reported AUC value ranging between 0.92–0.98 for MUAC in identifying overweight among 5–14-year age group. Another study from Ethiopia reported MUAC had an AUC of 0.96 for correct identification of overweight among adolescents aged 15–19 years.Other studies from South Africa, China, and Thailand also reported similar findings,, and thus points toward the potential role of MUAC as a proxy for BAZ in identification of overweight and obese adolescents in resource-limited settings.
The MUAC cut-off to identify overweight was: For 10–14 years– 22.9/23.4, for 15-19 years– 27.0/25.6 cm for males and females respectively; and for obesity were: For 10–14 years– 24.5/25.1, for 15-19 years – 28.5/28.0 cm for males and females respectively. Similarly, a study from Ethiopia reported a MUAC cut-off point for overweight of 27.75 and 27.9 cm for males and females, respectively. In another study from India, MUAC cut-off point proposed was 23 cm (males) and 23.3 cm (females) for obesity among 10–14-year age group. Overall, MUAC cut-offs depending on age and sex, for overweight/obesity ranged between 22.2 cm– 25.5 for early adolescents., In summary, the proposed cut-offs corroborated with the findings from the previous studies. Further, the proportion of adolescents correctly classified based on MUAC were in general more for late adolescents (15–19 years) in comparison to early adolescents (10–14 years), possibly due to less variation in cut-offs for late-adolescents as their growth is more or less stable.
In the present study, both the age-specific cut-offs and cut-offs for early and late adolescents have been estimated. However, due to considerable variation in the two cut-offs, age-specific cut-offs should be preferred. For example, for overweight, the MUAC cut-off for young adolescent (10–14 years) males was 22.9 cm and for a 14-year-old male, it was 25.3 cm. Thus, using a cut-off of 22.9 cm for a 14-year-old male would classify a normal weight adolescent as overweight. However, in situation where age is not known, age-group (10–14 years or 15–19 years) MUAC cut-offs could be considered. Keeping this in mind, age-specific and sex-specific cut-offs are proposed as a substitute (of BMI and BAZ) for the correct identification of overweight and obese in adolescents.
Furthermore, MUAC had been shown to correlate well with dyslipidemia indicators, and insulin resistance, a well know biomarkers of cardiovascular diseases.,Additionally, considering the ease of measurement of MUAC in field settings, it may facilitate early detection of overweight and obesity and corrective measures accordingly. Such adolescents will need further management depending on other factors, though it is outside the purview of current manuscript. Undoubtedly, MUAC is a relatively simple measure requiring only a measuring tape compared to assessing BMI. However, accurate MUAC measurements will need training of health staffs to ensure the accurate assessment of MUAC, keeping measuring tape at a right point with a specific tension, also, keeping it not too loose and not too tight. Although it should not be a problem, as health staff are generally trained in assessing MUAC in under-5 age group population. For under 5, color coded tapes are used for MUAC assessment, on the similar lines, color coded tapes could be considered for adolescents – red for obesity, amber for overweight, and green for normal weight. From a programmatic perspective, and the integration in adolescent health programs, MUAC holds promise in correct identification of overweight and obese. MUAC is a relatively inexpensive measure, as it only requires a measuring tape, and can be easily used in field settings, and in schools. Further, the results can be easily understood by caregivers and adolescents.
The study has its own strengths and limitations. The strengths of the study are the MUAC cut-off estimates are based on sample representative of entire country. Second, standardized field staffs and equipment's were used for assessing MUAC and BMI with strict quality control and monitoring in the CNNS. Third and most importantly, the sex-specific, age-specific, and age-group (10–14, and 15–19 years) specific cut-offs have been proposed, for both overweight and obesity. This is important as the body composition of adolescents tends to change with age, moreover, MUAC was found to be weakly correlating with age. However, the study has few limitations: Gold standard measure of assessing body fat like total body water hydrodensitometry, multicomponent methods or bioelectrical impedance analysis were not used, rather BMI z-score was used. However, BMI Z-score is most commonly used method to assess overweight and obesity and had been shown to reflect the body fatness assessed through advanced techniques like bioelectrical impedance analysis,, though it does not differentiate between fat mass and fat-free (lean) mass.
| Conclusion|| |
MUAC, a relatively simple and easy to use measure has equivalent diagnostic accuracy in identifying overweight and obesity compared to BAZ. Thus, age- and sex-specific MUAC cut-offs could be considered as a screening tool for the identification of overweight and obese adolescents in India and similar settings. However, the strategies to effectively integrate MUAC in adolescent health programs will need careful planning, though it could be done through availability of color-coded tapes or age- and gender-specific MUAC charts. In addition, future studies should focus on assessing the performance of MUAC and BMI/BAZ compared to standard measures of body-fat measurement like multi-component methods.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Kosti RI, Panagiotakos DB. The epidemic of obesity in children and adolescents in the world. Cent Eur J Public Health 2006;14:151-9.
Li Y, Zhai F, Yang X, Schouten EG, Hu X, He Y, et al.
Determinants of childhood overweight and obesity in China. Br J Nutr 2007;97:210-5.
Emerging Risk Factors Collaboration; Wormser D, Kaptoge S, Di Angelantonio E, Wood AM, Pennells L, et al.
Separate and combined associations of body-mass index and abdominal adiposity with cardiovascular disease: Collaborative analysis of 58 prospective studies. Lancet 2011;377:1085-95.
Sethi V, Gupta N, Pedgaonkar S, Saraswat A, Dinachandra Singh K, Rahman HU, et al.
Mid-upper arm circumference cut-offs for screening thinness and severe thinness in Indian adolescent girls aged 10-19 years in field settings. Public Health Nutr 2019;22:2189-99.
Furtado JM, Almeida SM, Mascarenhas P, Ferraz ME, Ferreira JC, Vilanova M, et al.
Anthropometric features as predictors of atherogenic dyslipidemia and cardiovascular risk in a large population of school-aged children. PLoS One 2018;13:e0197922.
Zhu Y, Lin Q, Zhang Y, Deng H, Hu X, Yang X, et al.
Mid-upper arm circumference as a simple tool for identifying central obesity and insulin resistance in type 2 diabetes. PLoS One 2020;15:e0231308.
de Almeida CA, Del Ciampo LA, Ricco RG, Silva SM Jr., Naves RB, Pina JF. Assessment of mid-upper arm circumference as a method for obesity screening in preschool children. J Pediatr (Rio J) 2003;79:455-60.
Jaiswal M, Bansal R, Agarwal A. Role of mid-upper arm circumference for determining overweight and obesity in children and adolescents. J Clin Diagn Res 2017;11:C05-8.
Mazıcıoğlu MM, Hatipoğlu N, Oztürk A, Ciçek B, Ustünbaş HB, Kurtoğlu S. Waist circumference and mid-upper arm circumference in evaluation of obesity in children aged between 6 and 17 years. J Clin Res Pediatr Endocrinol 2010;2:144-50.
Sisay BG, Haile D, Hassen HY, Gebreyesus SH. Performance of mid-upper arm circumference as a screening tool for identifying adolescents with overweight and obesity. PLoS One 2020;15:e0235063.
Ministry of Health and Family Welfare GoI, UNICEF, Population Council. Comprehensive National Nutrition Survey 2016–2018; 2019. Available from: https://nhm.gov.in/showfile.php?lid=712
. [Last accessed on 2020 Dec 22].
Zou KH, O'Malley AJ, Mauri L. Receiver-operating characteristic analysis for evaluating diagnostic tests and predictive models. Circulation 2007;115:654-7.
Akindele MO, Phillips JS, Igumbor EU. The relationship between body fat percentage and body mass index in overweight and obese individuals in an urban African setting. J Public Health Afr 2016;7:515.
Craig E, Bland R, Ndirangu J, Reilly JJ. Use of mid-upper arm circumference for determining overweight and overfatness in children and adolescents. Arch Dis Child 2014;99:763-6.
Ranasinghe C, Gamage P, Katulanda P, Andraweera N, Thilakarathne S, Tharanga P. Relationship between Body Mass Index (BMI) and body fat percentage, estimated by bioelectrical impedance, in a group of Sri Lankan adults: A cross sectional study. BMC Public Health 2013;13:797.
Kumar R, Indrayan A. Receiver operating characteristic (ROC) curve for medical researchers. Indian Pediatr 2011;48:277-87.
Fluss R, Faraggi D, Reiser B. Estimation of the Youden Index and its associated cutoff point. Biom J 2005;47:458-72.
Dasgupta A, Butt A, Saha TK, Basu G, Chattopadhyay A, Mukherjee A. Assessment of malnutrition among adolescents: Can BMI be replaced by MUAC. Indian J Community Med 2010;35:276-9.
] [Full text]
Lu Q, Wang R, Lou DH, Ma CM, Liu XL, Yin FZ. Mid-upper-arm circumference and arm-to-height ratio in evaluation of overweight and obesity in Han children. Pediatr Neonatol 2014;55:14-9.
Rerksuppaphol S, Rerksuppaphol L. Mid-upper-arm circumference and arm-to-height ratio to identify obesity in school-age children. Clin Med Res 2017;15:53-8.
[Figure 1], [Figure 2]
[Table 1], [Table 2]