Body Mass Index Correlates with Body Fat Percentage in Children and Adolescents in the Democratic Republic of the Congo: A cross sectional study

Page created by Maria Lawrence
 
CONTINUE READING
Body Mass Index Correlates with Body Fat Percentage
    in Children and Adolescents in the Democratic
    Republic of the Congo: A cross sectional study
Guy Ikambo Wanghi1*, Leslie Lytle2, Augustin Rudahaba Buhendwa3 and Ernest Kiswaya Sumaili1, 4

   1. Department of Basic Sciences, Unit of Physiology, Faculty of Medicine, University of
      Kinshasa, Kinshasa, RD Congo.
   2. Department of Health Behavior, Gillings School of Global Public Health University of
      North Carolina at Chapel Hill, Chapel Hill, NC, USA.
   3. Department of Physical Medicine and Rehabilitation, Faculty of Medicine, University of
      Kinshasa, Kinshasa, RD Congo
   4. Renal Unit, Department of Internal Medicine, University of Kinshasa Faculty of Medicine,
      Kinshasa, RD Congo

 *Corresponding author: Guy Ikambo Wanghi, Department of Basic Sciences, Unit of Physiology,
Faculty of Medicine, University of Kinshasa, Kinshasa, RD Congo. Email: guy.wanghi@unikin.ac.cd.

                                         Summary
       INTRODUCTION
                Paediatric obesity is a worldwide health challenge and parameters to
       evaluate population obesity risk are needed. The purpose of this study is to report
       on anthropometric measures for assessing obesity risk for children in sub-
       Saharan Africa and to confirm that measuring height and weight to calculate
       BMI can be used as a good indicator of obesity risk.
       MATERIALS AND METHODS
                This was a cross-sectional study of a sample of 1,442 students aged 6 to 18
       years enrolled in schools in the city of Kinshasa, the capital of the Democratic
       Republic of Congo. These school children were selected using a multistage
       sampling method. BMI for-age-sex was calculated using WHO AnthroPlus and
       expressed in Z‑scores. Skinfolds were used to calculate percent body fat (%BF).
       Data were analyzed using SPSS version 21.
       RESULTS
                This study showed that the percentage body fat (BF) was higher in girls
       than in boys (20.5 ± 9.6 vs 12.4 ± 6.3 p< 0.001). The BMI for girls was higher
       compared to boys (18.5 ± 3.7 vs 17.3 ± 3.0 p
Keywords: Body Composition, percentage of body fat, Overweight, Obesity, Adolescents, Children

                                          [Afr. J. Health Sci. 2021 34(1):2-11]

                                                        associated with level of adiposity but getting
Introduction                                            an accurate measure of fat is very challenging
         Childhood and adolescent overweight
                                                        [12]. Direct and highly accurate measurement
and obesity is a major public health problem in
                                                        of body fat or adiposity requires sophisticated
the developed world [1]. Those conditions
                                                        equipment such as a Dexa scan using dual-
have also significantly increased in low and
                                                        energy X-ray absorptiometry. Other tools such
middle income countries (LMIc) in the past
                                                        as bio-electrical impedance, the use of skin
decades. Overweight and obesity in children
                                                        calipers, and body circumference measures are
and adolescents in LMIc now co-exist with
                                                        also used to assess adiposity. But these tools
underweight, as a result of nutrition transition
                                                        are difficult to use in epidemiological surveys
[2-5].
                                                        due to the cost of equipment, the time required,
         In sub-Saharan Africa, the prevalence
                                                        or the challenging measurement protocols [5,
of overweight and obese children and
                                                        13]. To fill this gap, WHO recommends Body
adolescents varies from 5 to 17 % and 1 to 5%,
                                                        Mass Index (BMI) as the most useful
respectively [6-7]. Several reports have shown
                                                        population level measure of underweight,
that overweight and obesity are now increasing
                                                        overweight and obesity. BMI is also the
in low and middle-income populations
                                                        surrogate measure for prediction of percentage
(including in Africa), especially in urban areas.
                                                        of body fat (% BF) [14].
From 1990 to 2010, the growth rate of
                                                                 Several studies have shown a
overweight and obesity doubled, indicating
                                                        relationship between BMI and percentage body
that Africa has the fastest growth rates of
                                                        fat (BF) [15-17] but such studies have rarely
overweight and obesity [8-9] and is expected
                                                        been performed in sub-Saharan Africa.
to reach 11% in 2025 [8].
                                                                 The objective of our study was to
         Childhood and adolescence are
                                                        report on assessment of body fat and height
important periods of life, as many
                                                        and weight for children in the Democratic
physiological         and          psychological
                                                        Republic of the Congo (DRC) and to confirm
transformations take place at these ages.
                                                        that measuring height and weight to calculate
Moreover, healthy and unhealthy lifestyles and
                                                        BMI can serve as a good indicator of risk for
behaviors are established during these years,
                                                        obesity. Comparisons were made between BMI
which can influence the behavior and health
                                                        and skinfold thickness, a criterion measure of
status of adults [10-11]. Therefore, it is vitally
                                                        adiposity. In addition, percentage body fat was
important to understand the trends in the
                                                        compared among boys and girls and by age
development of body weight disorders in
                                                        groups. The results of this present work can
young people and their negative effects on
                                                        provide a baseline for adiposity in youth in
overall health [5].
                                                        Sub-Saharan Africa. In addition, it can help
         To evaluate population trends, tools to
                                                        encourage health practitioners in clinics to use
assess weight-related risk need to be simple,
                                                        the simple measure of BMI to assess the risk of
cost effective and feasible to use across a wide
                                                        obesity in their patients and to provide public
variety of settings.
                                                        health officials with a good surveillance tool
         Overweight and obesity are defined as
                                                        for obesity risk in youth.
abnormal or excessive fat accumulation that
may impair health [5]. Chronic disease risk is

African Journal of Health Sciences Volume 34, Issue No. 1, January - February, 2021                        3
Materials and Methods                                Anthropometric
Study Sample and Design                              Measurement
         This was a cross-sectional study                     To measure height, weight and
enrolling 6–18-year-old students living in           percentage BF, we used the World Health
Kinshasa, the capital of DRC. The sample was         Organization and the International Society for
selected using multistage sampling method            the Advancement of Kinanthropometry
from November 15th, 2018 to June 20th, 2019.         guidelines [18]. Height was measured in the
A total of 568 eligible primary and secondary        sagittal plane with a portable SECA 214
schools were grouped according to their              anthropometer (Seca®, Hamburg, Germany) to
location into three strata which correspond to       the nearest 0.1 cm. The subjects’ height was
the existing three educational provinces in          measured with each subject standing upright,
Kinshasa. The schools in each stratum were           the head in Frankfort plane, without shoes,
further stratified into primary and secondary.       with both feet flat on the platform and apposed
         The name of each school was written         at the medial malleoli; the heels, buttocks, and
on a piece of paper, wrapped, and put into the       occiput placed against the scale with arms
corresponding ballot bags representing its           hanging freely by the sides. The instrument
stratum. A total of 18 schools, one school from      was checked before each measurement
each of the 3 strata, were selected by a simple      ensuring that both headboard and footboard
random method. Within each selected school,          were at 90° to the vertical rule.
classrooms were selected to participate and all               Weight was measured in kilograms
students in selected classes were offered the        (kg) with an OMRON body composition BF
opportunity to participate.                          511 balance (OMRON Healthcare Europe BV,
         To be eligible, participant were to         Hoofddord, the Netherlands) to the nearest 0.1
meet the following selection criteria:               kg. The students stood on the scale without
Congolese citizen, consent of parents or tutors      shoes wearing light cotton material. The scale
and age between 6 and 18 years old. School           automatically adjusted to zero before each
record of birth certificate of each enrolled child   measurement. It was also checked daily using
was used to establish age. Students with             an object of known weight as a quality control
disabilities and pregnant girls were excluded.       measure to ensure validity and reliability.
                                                              Body mass index (BMI) was
Ethical Consideration                                calculated as weight in kilograms divided by
        To conduct this study, ethical approval
                                                     height in meters squared (kg/m2) [19].BMI‑for
was obtained from the Ethical Review
Committee of the Ministry of Health (Ref             ‑age indices were calculated with WHO
N°.143/CNES/BN/PMMF/2018).              Written      AnthroPlus®        (v1.0.4,     World     Health
Permission was obtained from Minister of             Organization, Geneva 2009) [20] and
Education (the Ministry of primary and               expressed in Z‑scores.
secondary schools), as well from each                         Skinfolds in the biceps, triceps,
participating school and from all classroom          suprailiac, subscapularis and abdominal
teachers. Parents or tutors were informed of         regions were taken on the right side of the
the study prior to the survey date. The parents      body with Harpenden skinfold caliper (Holtain
were consented for this intervention and the         Limoted, U.K) with a constant pressure of
youth gave their assent.                             10g/mm2 to the nearest 0.2 mm [21].

African Journal of Health Sciences Volume 34, Issue No. 1, January - February, 2021                     4
Index-for-age-sex was calculated using WHO
Operational Definitions                             Anthro plus software V.1.0.4.             All data
         Using BMI, the children were
                                                    analyses were performed with Statistical
classified into the following categories:
                                                    Package for the Social Sciences (SPSS),
underweight: thin: +2), and “normal weight” (BMI‑for‑age Z‑           body composition variables were computed.
score − 2 to + 1). The values were set from the     After the computation of the descriptive
median values of the WHO international              analyses, independent t-test was calculated to
growth reference 2007 for 5–19‑year‑old             test differences of means with respect to
children [19-20].                                   statistical significance. The statistical level was
         Percentage body fat (% BF) was based       set at p-value ≤ 0.05.
on the sum of triceps and calf skinfolds derived
from skinfolds using the equation developed
                                                    Results
                                                             A total of 1,442 pupils (668 females
by Slaughter et al. (1988) [5] which is
                                                    and 774 males) participated in the study. Table
internationally accepted for use in children and
                                                    1 shows the descriptive statistics of the sample.
adolescents from different ethnic groups. All
                                                    There were slightly more boys represented in
anthropometric measurements were taken
                                                    the total sample than girls and the sample of
twice     by      Level     2      ISAK-certified
                                                    boys was significantly older than the sample of
anthropometrics’ and the average scores of the
                                                    girls. Girls were shorter than boys and girls
two measurements. The values were set from
                                                    were significantly heavier, had significantly
the median values of the WHO international
                                                    higher mean BMI and had higher body fat as
growth reference 2007 for 5–19‑year‑old
                                                    measured by skinfold thickness as compared to
children [19-20].                                   boys. In the group of younger children, the
         Percentage body fat was compared           differences between girls and boys were only
with BMI according to Lohman et al. (2000)          statistically significant for the percentage of
[5] across six categories, as the following: very   body fat and skin folds with girls having
low 7 - 11, low 12 - 14, optimal range 15 - 18,     significantly higher values than boys (p <
moderate high 19 - 25, high 26 - 32, and very       0.001).
high 32 and above.                                           When comparing older children by
Statistical Analysis                                gender, weight, BMI, percentage of body fat
        Underweight, overweight and obesity         and skin folds were higher in girls than in boys
were defined based on the Body Mass Index           (p < 0.001). Boys were, on average, taller than
(BMI) for age-sex of the children. Body Mass        girls (p < 0.05).

African Journal of Health Sciences Volume 34, Issue No. 1, January - February, 2021                       5
Table 1: Characteristics of Study Population by Age Groups
                          All                    Age 6-9 years        Age 10-18 years
Variables Girl       Boy        P-       Girl      Boy       P-       Girl      Boy         p-
             (n=668 (n=774) value (n=21 (n=196) value (n=451) (n=578) Value
             )                           7)
Age          11.8±3. 12.2±3. 0.012 7.6             7.7 ±1.0 0.481 13.7          13.7        0.978
(years)      5       4                   ±1.1                         ±2.5      ±2.4
Height       1.47±0. 1.51±0.
Table 2: Categories of Percentage BF by Sex and Age Group
                          All                         Age 6-9 years             Age 10-18 years
Category    Girl        Boy        P       Girl        Boy         P      Girl       Boy        p-
of body     (n=668)     (n=774)            (n=217)     (n=196)            (n=451)    (n=578)    Value
fat
Very low    83 (12.4)   455
60

                                  50

                   Body Fat (%)   40

                                  30

                                  20

                                  10

                                  0
                                       10   15         20         25          30         35
                                                        BMI (Kg/m2)
                                                 boy                   girl
Figure 1: Scatter Plot of the Relationship between Body Mass Index (BMI) and Percentage of
Body Fat (%BF) of School-Aged Girls and Boys in Kinshasa.

         Figure 1 shows the positive                           communicable diseases is roughly equivalent
relationships between percentage BF and BMI                    to measuring percentage body fat through skin
by sex. In addition, this relationship appears to              folds. These results are consistent with those
be linear in nature. This relationship between                 reported in other studies outside of Africa [15-
BMI and percentage BF was independent at                       17]. However, while the technique for
age and increased with BMI. In this sample of                  measuring skin folds is inexpensive, easy, and
youth from DRC, BMI appears to be a good                       reproducible for an experienced technician, it
proxy for adiposity.                                           is not very accurate and is subject to inter
                                                               observer error [13].
Discussion                                                              The results of this study also showed
         Our study sample composed of school-
                                                               that in the younger age group (6 - 9 years),
aged children and adolescents (6-18 years) in
                                                               BMI was non-significantly different between
the city province of Kinshasa tested the
                                                               the two sexes, while percentage BF was
hypothesis that body mass index correlates
                                                               significantly different between girls and boys.
with body fat percentage in children and
                                                               However, for the oldest age group (10-18
adolescents in Sub-Saharan Africa. The results
                                                               years), the differences between BMI and the
of this work, shown in figure 1 proved this
                                                               percentage BF were statistically significant
hypothesis. Indeed, in the figure 1, BMI places
                                                               between the two sexes.
youth in the same category of being at risk for
                                                                        These large differences in body fat
fatness as does the data that more directly
                                                               were also observed in table 2 specifically in the
measure fatness.
                                                               younger and older age groups where optimal
         The age and sex risk models for
                                                               and moderately high body fat was higher in
obesity-related diseases are similar when we
                                                               girls than in boys. Some authors explain that
use a body fat assessment which is a more
                                                               for the same body mass index (BMI), women
difficult assessment task and requires large
                                                               have ~10% more body fat than men [21, 23].
screening and intervention efforts compared to
                                                               Striking gender differences appear during
a BMI assessment which is much easier to
                                                               puberty: the increase in body weight in boys is
measure height and weight in large screening
                                                               mainly due to increases in lean body mass
and intervention efforts. It is clear that these
                                                               while in girls due to increased fat mass; typical
results have shown that measuring BMI to
                                                               android and gynoid fat distributions also
change obesity and the risk of chronic non-

African Journal of Health Sciences Volume 34, Issue No. 1, January - February, 2021                                8
appear for the first time during this period [23   to be a simple and easy measure that clinicians
– 26]. Two studies in black Africa, mainly in      can use to analyze the risk of obesity and
Zimbabwe and Kenya, have shown that girls          chronic non-communicable diseases in
have high percentage BF values compared to         children and adolescents in sub-Saharan Africa
boys. Michelin et al in their study found that     and particularly in DR Congo.
there are typical gender differences in body fat
composition for both younger and older
                                                   Conflict of Interest
                                                            All authors declared no competing
adolescents [27-28].
                                                   interests.
        We also found large differences in
very low body fat in boys, especially in the       Author Contributions
older age group (58 % boys vs. 6 % girls). This             GIW and ESK were involved in the
could be due to a combination of adverse           conception, design, data analysis and
environmental constraints to good growth           interpretation of the study. GIW collected in
amplified in boys such as under-nutrition, high    addition data. LL contributed to interpretation.
burden of infectious or parasitic diseases, poor   ESK and ARB contributed to data analysis and
living conditions including housing, economic      interpretation. All authors were involved in the
insecurity, food crisis and lack of sanitary or    writing of the manuscript and provided final
educational facilities [29]. However, the          approval.
difference between girls and boys in the very
                                                   References
low-fat category for the total sample (12.4 %
                                                   1. Ji CY, Cheng TO: Prevalence and
of girls versus 58.8 % of boys) is more
                                                      geographic distribution of childhood
explained by differences observed in the older        obesity in China in 2005. Int J Cardiol
age group (6.0 % of girls versus 58.0 % of            2008, 131 (1):1-8.
boys). The gender difference is slightly smaller   2. Reddy SP, Resnicow K, James S,
in the younger age group (25.8 % versus 61.2          Kambaran N, Omardien R, Mbewu AD:
%).                                                   Underweight, overweight and obesity
Strength and Limitations of                           among South African adolescents: results
                                                      of the 2002 National Youth Risk
Study                                                 Behaviour Survey. Public Health Nutr
         The strengths of our study are the           2009, 12 (2):203-207.
relatively large sample size, the quality of the   3. Hands B, Larkin D, Parker H, Straker
anthropometric measurements, and the fact that        L, Perry M: The relationship among
all social categories were present in the             physical activity, motor competence and
sample. The cross-sectional nature of the study       health-related fitness in 14-year-old
could be seen as a limitation and skinfold            adolescents. Scand J Med Sci Sports 2009,
measures used as the criterion measure of             19 (5):655-663.
adiposity while DEXA or bioelectrical              4. Garcia-Hermoso A, Correa-Bautista JE,
impedance may be the more rigorous measures           Olloquequi J, Ramirez-Velez R: Health-
of adiposity.                                         related physical fitness and weight status
                                                      in 13- to 15-year-old Latino adolescents. A
Conclusion                                            pooled analysis. J Pediatr (Rio J) 2019, 95
        Our study showed a correlation                (4):435-442.
between BMI and percentage BF in children          5. Monyeki MA, Neetens R, Moss SJ,
and adolescents in sub-Saharan Africa. Given          Twisk J: The relationship between body
the difficulties of measuring skin folds and          composition and physical fitness in 14 year
training experienced technicians, BMI appears         old adolescents residing within the Tlokwe

African Journal of Health Sciences Volume 34, Issue No. 1, January - February, 2021                   9
local municipality, South Africa: the               Sectional Cohort Study. Diabetes Metab
      PAHL study. BMC Public Health 2012,                 Syndr Obes, 13:1587-1597.
      12:374.                                       13.   Mechanick JI, Farkouh ME, Newman
6.    Martorell R, Kettel Khan L, Hughes                  JD, Garvey WT: Cardiometabolic-Based
      ML, Grummer-Strawn LM: Overweight                   Chronic Disease, Adiposity and
      and obesity in preschool children from              Dysglycemia Drivers. J Am Coll
      developing countries. Int J Obes Relat              Cardiol,2020, 75 (5):525-538.
      Metab Disord 2000, 24 (8):959-967.            14.   West J, Manchester B, Wright J, Lawlor
7.    Kouanda S, Doulougou B, De Coninck                  DA, Waiblinger D: Reliability of routine
      V, Habimana L, Sondo B, Tonglet R,                  clinical measurements of neonatal
      Ketelslegers JM, Robert A: Insulin                  circumferences and research measurements
      Growth Factor-I in Protein-Energy                   of neonatal skinfold thicknesses: findings
      Malnutrition during Rehabilitation in Two           from the Born in Bradford study. Paediatr
      Nutritional Rehabilitation Centres in               Perinat Epidemiol, 2011,25 (2):164-171.
      Burkina Faso. J Trop Med 2009, 2009:          15.   Kyle UG, Genton L, Slosman DO,
      832589                                              Pichard C: Fat-free and fat mass
8.    de Onis M, Blossner M, Borghi E:                    percentiles in 5225 healthy subjects aged
      Global prevalence and trends of                     15 to 98 years. Nutrition 2001, 17 (7-
      overweight and obesity among preschool              8):534-541.
      children. Am J Clin Nutr 2010, 92             16.   Deurenberg P, Hautvast JG: Prevalence
      (5):1257-1264.                                      of overweight and obesity in The
9.    Negash S, Agyemang C, Matsha TE,                    Netherlands in relation to
      Peer N, Erasmus RT, Kengne AP:                      sociodemographic variables, lifestyle and
      Differential prevalence and associations of         eating behavior: starting points for the
      overweight and obesity by gender and                prevention and treatment of obesity. Bibl
      population group among school learners in           Nutr Dieta 1989; (44):8-21.
      South Africa: a cross-sectional study. BMC    17.   Rush EC, Chhichhia P, Hinckson E,
      Obes 2017, 4:29.                                    Nabiryo C: Dietary patterns and vitamin
10.   Galan-Lopez P, Ries F, Gisladottir T,               B(12) status of migrant Indian
      Dom-nguez R, Sánchez-Oliver AJ:                    preadolescent girls. Eur J Clin Nutr 2009,
      Healthy Lifestyle: Relationship between             63 (4):585-587.
      Mediterranean Diet, Body Composition          18.   Physical status: the use and interpretation
      and Physical Fitness in 13 to 16-Years Old          of anthropometry. Report of a WHO
      Icelandic Students. Int J Environ Res               Expert Committee. World Health Organ
      Public Health 2018, 15 (12):2632                    Tech Rep Ser 1995, 854:1-452.
11.   Twig G, Yaniv G, Levine H, Leiba A,           19.   de Onis M, Onyango AW, Borghi E,
      Goldberger N, Derazne E, Ben-Ami                    Siyam A, Nishida C, Siekmann J:
      Shor D, Tzur D, Afek A, Shamiss A et                Development of a WHO growth reference
      al: Body-Mass Index in 2.3 Million                  for school-aged children and adolescents.
      Adolescents and Cardiovascular Death in             Bull World Health Organ 2007, 85(9):660-
      Adulthood. N Engl J Med 2016, 374                   667.
      (25):2430-2440.                               20.   WHO: AnthroPlus for Personal
12.   Macek P, Biskup M, Terek-Derszniak                  Computers Manual: Software for
      M, Stachura M, Krol H, Gozdz S, Zak                 Assessing Growth of the World’s Children
      M: Optimal Body Fat Percentage Cut-Off              and Adolescents. Bull World Health Organ
      Values in Predicting the Obesity-Related            2007, Geneva: WHO; 2009. Available
      Cardiovascular Risk Factors: A Cross-               from:

African Journal of Health Sciences Volume 34, Issue No. 1, January - February, 2021                     10
http://www.who.int/growthref/tools/en/       25. Hattori K, Tahara Y, Moji K, Aoyagi K,
      [Last accessed on 2019, January 15th].           Furusawa T: Chart analysis of body
21.   Brown N, Scurr J: The need for a                 composition change among pre- and
      standardised anthropometric protocol for         postadolescent Japanese subjects assessed
      objective assessment of pre- and                 by underwater weighing method. Int J
      postoperative breast surgery. Gland Surg         Obes Relat Metab Disord 2004, 28:520–
      2012, 1(3):142-145.                              524.
22.   Eriksson Hogling D, Bäckdahl J, Thorell      26. Wells JC: Sexual dimorphism of body
      A, Rydn M, Andersson DP: Metabolic               composition. Best Pract Res
      Impact of Body Fat Percentage                    ClinEndocrinol Metab 2007, 21:415–430.
      Independent of Body Mass Index in            27. Olivieri F, Semproli S, Pettener D,
      Women with Obesity Remission After               Toselli S Growth and malnutrition of rural
      Gastric Bypass. 2020; Obes Surg, 30              Zimbabean children ( 6 – 17 feaors of age).
      (3):1086-1092.                                   Am J Phys Anthrol 2008, 136 : 214 – 222
23.   Grundy SM, Williams C, Vega GL:              28. Semproli S, Gualdi-Russo E. Childhood
      Upper body fat predicts metabolic                malnutrition and growth in a rural area of
      syndrome similarly in men and women.             western Kenya. Am J Phys Anthropol
      Eur J Clin Invest,2018 ; 48(7).                  2007; 132 : 463 - 469.
24.   Maynard LM, Wisemandle W, Roche              29. Buhendwa RA, Roelants M, Thomis M,
      AF, Chumlea WC, Guo SS, Siervogel                Nkiama CE: Nutritional status and height,
      RM: Childhood body composition in                weight and BMI centiles of school-aged
      relation to body mass index. Pediatrics          children and adolescents of 6-18-years
      2001, 107:344–330.                               from Kinshasa (DRC). Ann Hum Biol 2017
                                                       44 (6):554-561.

African Journal of Health Sciences Volume 34, Issue No. 1, January - February, 2021                  11
You can also read