RESEARCH Open Access [626867]

RESEARCH Open Access
Body composition and physical activity in Italian
university students
Luciana Zaccagni*, Davide Barbieri and Emanuela Gualdi-Russo
Abstract
Background: Increased sedentary lifestyle and prevalence of overweight/obesity are common in western countries.
The purposes of this study were (i) to assess the main health-related anthropometric characteristics in a sample of
students in relation to sex, amount of physical activity and sport discipline, and (ii) to investigate the accuracy of theBody Mass Index (BMI) and Waist-to-Stature Ratio (WSR) as indicators of body fat percentage (%F) in young adults.
Methods: 734 university students, both sexes, participated in th e present research. A self-a dministered questionnaire
acquired socio-demographic information (sex, age) and sport participation (hours/week, sport discipline). Anthropometricmeasurements and grip strength values were acquired acco rding to standardized procedures. Body composition was
assessed by means of the skinfold method.
Results: Most students had normal BMI, WSR and %F. There were signi ficant statistical differences in all anthropometric
traits between the two sexes. One-way ANOVA s within sex showed statistically significant differences in biceps skinfold,
waist circumference (WC), WSR, body density (BD), %F and fa t mass (FM) among different lev els of physical activity in
males; and in weight, BMI, arm girths and fat free mass (FFM) in females. One-way ANOVAs within sex showed statistically
significant differences in arm girths, grip strength and FFM among different sport disciplines in males, and in height,
weight, BMI, WC, relaxed arm girth, grip strength, FM and FFM in females. Despite the significant and positive correlationof BMI and WSR with %F both indices had poor sensitivity.
Conclusions: Physical activity plays an important role in body composition parameters: the most active males had the
least amount of FM and the most active females had the greatest amount of FFM. BMI and WSR are not accurate indicesof adiposity in young adults.
Keywords: Body fat, BMI, WSR, Young adults, Phys ical activity, University students
Background
Body composition assessment is used to monitor perform-
ance and training in the athletic community, and to verifythe health status of the population in general. The Body
Mass Index (BMI) is often used to evaluate the weight
status, even if it does not discriminate between differentcomponents of the overall body mass by definition (BMI =
weight/height
2). Therefore, the adoption of BMI as a pre-
dictor of adiposity and of consequent health risk shouldbe used with caution [1], especially with physically active
individuals, who usually have a higher body density and
fat free mass (FFM) than the general population [2-4].Body fat percentage (%F) instead is directly correlated withincreased health risk, especially for metabolic and cardio-
vascular diseases [5-9]. Waist-to-Stature Ratio (WSR) and
Waist Circumference (WC) are supposed to have greaterdiscriminatory power compared to BMI [10,11] and are
more sensitive than BMI as an early predictor of health-
related risks [12]. In particular, WSR is probably the mostsensitive anthropometric index for the screening of the
metabolic syndrome in Mediterranean populations, com-
pared to both BMI and WC [13].
Low levels of physical activit y may place individuals at in-
creased risk of obesity and cardiovascular diseases [14]. On
the other side, physical activity has been suggested as ameans to reduce and control body fatness. More in general,
regular physical activity has proved to effectively reduce di-
verse health risk factors, especially those related to cardio-vascular diseases and the metabolic syndrome [15,16]. In
* Correspondence: luciana.zaccagni@unife.it
Department of Biomedical and Specialty Surgical Sciences, University ofFerrara, Corso Ercole I d ’Este 32, 44121 Ferrara, Italy
© 2014 Zaccagni et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative
Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, andreproduction in any medium, provided the original work is properly credited. The Creative Commons Public DomainDedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article,unless otherwise stated.Zaccagni et al. Journal of Translational Medicine 2014, 12:120
http://www.translational-medicine.com/content/12/1/120

particular, the American College of Sports Medicine
recommends that adults engage in at least 150 min ∙wk−1
of moderate intensity cardiovascular exercise and at
least 75 min ∙wk−1of vigorous intensity training, in order
to maintain a sufficient level of cardio-respiratory fit-
ness. Resistance training is also suggested 2 –3d∙wk−1
[17]. We can therefore assume that these recommendations
amount for a total of more than 4 h ∙wk−1of moderate-to-
intense physical activity.
The purpose of this study was to assess the main health-
related anthropometric characteristics of a group of univer-
sity students, in order to evalu ate their relationship with
quantity and type of physical activity according to sex. In
particular, FFM, %F, WC, WSR, BMI and grip strength
were taken into consideration. Furthermore, the accuracyof BMI and of WSR as predictors of %F was evaluated.
Methods
Participants and study design
This was a cross-sectional study carried out on a total of734 university students, 354 females aged 21.5 ± 2.9 yrs(mean ± standard deviation) and 380 males aged 22.1 ±
3.6 yrs, of the School of Sport Science (Faculty of Medi-
cine, University of Ferrara, Italy) who volunteered for thestudy. The sample was composed of North Italian students
(mainly coming from the regions of Emilia Romagna and
Veneto). Body image perception was previously assessedon the same sample [18]. The research protocol was ap-
proved by the Ethic Committee for Biomedical Research
of the University of Ferrara, and all participants providedwritten informed consent.
A questionnaire on training and physical activity pat-
terns was administered to participants. The mean weeklyamount of physical activity was 6.7 ± 4.2 hrs for males and
4.2 ± 3.8 hrs for females; 28 males (7.4% of the total male
sub-sample) and 83 females (23.4% of the total femalesub-sample) did not practice any sport activity.
Anthropometric survey
All measures were taken in th e Anthropometry Laboratory
at the University of Ferrara, during the tutorials for the stu-dents of the course of Anthropometry and Ergonomics in
the second year of the School of Sport Science.
Standing (H, cm) and sitting heights (SH, cm) were
measured to the nearest 0.1 cm using a wall-mounted sta-
diometer (Magnimeter, Raven Equipment Limited, UK).Weight (W, kg) was measured to the nearest 0.1 kg using
a calibrated electronic scale. BMI was calculated as W/H
2
(kg/m2). Skinfold thicknesses at biceps (B Sk) and triceps
(T Sk) were measured to the nearest 0.1 cm using a Lange
caliper (Beta Technology Inc.). All girths (Waist Circum-
ference WC, Contracted Arm Girth CAG, Relaxed ArmGirth RAG) were measured to the nearest 0.1 cm using anon-metallic and non-stretchable tape. WSR was calcu-
lated as WC/H.
All measurements were taken on the left side of the
body, according to standardized procedures [19]. During
the anthropometric measurements, all participants were
barefoot and clothed appropriately.
Left and right hand grip strength was measured to the
nearest 0.5 kg by means of a Takei dynamometer (T.K.K.
5001 grip-A Takei scientific instruments Co., LTD, Japan).The highest value of two trials was recorded, after an ad-
equate period of rest between attempts, for each hand.
Assessment of body composition
Body density (BD) was calculated using Durnin & Womers-
ley [20] equations with two skinfolds (biceps and triceps),according to sex and age of the student. %F was calculated
from BD using Siri equation [21]. Fat Mass (FM, kg) was
calculated as (%F*W)/100 and FFM (kg) as W-FM.
Indices and classifications
According to the World Health Organization [22], under-weight was defined as BMI < 18.5 kg/m
2,n o r m a lw e i g h ta s
18.5 kg/m2≤BMI < 25 kg/m2, overweight as 25 kg/m2≤
BMI < 30 kg/m2,a n do b e s i t ya saB M I ≥30 kg/m2.B e c a u s e
of the small number of students with a BMI ≥30 kg/m2
(only one female and 15 males), they were included in the
overweight group for further elaboration. Even if there iswidespread consensus on cut-p oints for weight status, this
is not the case for what concern fatness. According to Gal-
lagher et al. [23], %F ≥20% (males) and %F ≥33% (females)
are the cut-points adopted to define overfatness, corre-
sponding to overweight classification using BMI in a popu-
lation of young adults.
According to the National Institute for Health and
Clinical Excellence guidelines, WC ≥102 cm for men
and ≥88 cm for women are prerequisite risk factors for
the diagnosis of the metabolic syndrome, as WSR ≥0.5
for both males and females [12].
Statistical analysis
All variables were checked for normality and logarithmic-
ally (10-based) transformed where necessary (skinfold atbiceps and triceps). Results were expressed as mean ±
standard deviation. Comparisons between sexes were car-
ried out using a two-sample Student ’st – t e s tf o rc o n t i n u o u s
data and a chi-square ( χ
2) test for categorical data.
Subsequently, both females and males were divided into
3 tertiles, according to their level of weekly physical activity:low ( ≤3h∙wk
−1for females, ≤5h∙wk−1for males), medium
(3 < h ∙wk−1< 6 for females, 5 < h ∙wk−1< 8 for males) and
high ( ≥6h∙wk−1for females, ≥8h∙wk−1for males). One-way
ANOVAs were used to assess the differences in anthropo-
metric variables and grip strength among the 3 groups and
post hoc comparisons were performed using Tukey test.Zaccagni et al. Journal of Translational Medicine 2014, 12:120 Page 2 of 9
http://www.translational-medicine.com/content/12/1/120

Correlation analysis between total weekly hours of phys-
ical activity and anthropometric variables was carried out.
In order to assess the anthropometric differences among
subjects practicing different activities, one-way ANOVA
was performed on sports with at least 10 participants: soc-
cer, body building, basketball, swimming and volleyball inmales; gymnastics, other gym activities (O.G.A.), ballet,
volleyball, swimming and jogging in females. When a sig-
nificant F value was obtained, post-hoc comparisons wereperformed by means of Tukey test.
To determine the accuracy of BMI as a measure of over-
fatness – and therefore of poor health status – participantswere classified into one of four categories: 1) overweight
and overfat (True Positive, TP), 2) overweight and normal
fat (False Positive, FP), 3) normal weight and overfat (FalseNegative, FN), and 4) normal weight and normal fat (True
Negative, TN). The sensitivity, specificity, and predictive
values of BMI were calculated for each group. Sensitivitywas calculated as the proportion of overfat individuals
who were identified as overweight by BMI (i.e. TP/(TP +
FN)). Specificity was calculated as the proportion of nor-mal fat individuals who were identified as normal weight
by BMI (i.e. TN/(TN + FP)). Positive Predictive Value
(PPV) was calculated as the probability that a participantidentified as overweight by BMI was truly overfat: PPV =
TP/(TP + FP). Negative Predictive Value (NPV) was calcu-
lated as the probability that a participant who was identi-fied as normal weight by BMI was normal fat: NPV = TN/
(TN + FN) [24]. Test accuracy increases as the total num-
ber of FP and FN decreases.
To test the accuracy of WSR as a measure of overfat-
ness, the same procedure was adopted, substituting the
overweight category with excessive WSR.
The statistical significance was set at p < 0.05. All ana-
lyses were performed using “Statistica ”for Windows,
Version 11.0 (StatSoft Italia srl, Padua, Italy).
Results
There were significant differences among all anthropo-metric traits between sexes (Table 1). Males were on
average heavier, taller, leaner and stronger than females and
had wider girths. Females had t hicker skinfolds than males,
as expected [20,25], therefore they had lower BD and
higher %F. 72% of males and 89% of females were normal
fat, while 27.3% of males and 10% of females were overfat.
Only 4 females (1.2% of the sub-sample) had WC ≥
88 cm and 7 males (2.0% of the sub-sample) had WC ≥
102 cm; 5% of females and 13% of males had WSR ≥0.5.
BMI mean values were in the normal range according to
WHO weight status categories [22]. χ
2test proved there
was a significant difference (p < 0.001) between sexes inweight status distribution. No male student was under-
weight, compared to 5.6% of females who fell into this cat-
egory. Most males (71.7%) and females (80.9%) werenormal weight. Males were more overweight (24.2%) and
obese (4.2%), than females (13.2% overweight and only
0.3% obese).
ANOVAs within male sub-sample with different levels
of physical activity (Table 2) show significant statistical
differences in biceps skinfold, WC, WSR, BD, %F and
FM, supporting the positive effects of physical activityon health-related anthropometric traits. Tukey post-hoc
test shows significant differences only between the low
and high level groups.
ANOVAs within female sub-sample with different levels
of physical activity (Table 3) show significant statistical dif-
ferences in weight, BMI, contracted and relaxed arm girthsand FFM, supporting the positive effects of physical activ-
ity, particularly on FFM. Tukey post-hoc test shows sig-
nificant differences between the high level group and the
other two.
Statistical correlations between hours of physical activ-
ity and BMI, triceps and biceps skinfolds, WC, WSR,
BD, %F and FM were significant (p < 0.05) in males, and
biceps skinfold, left and right had grip strength, BD, %Fand FFM in females (Table 4).
ANOVAs between sport disciplines with more than 10
participants in males (Table 5) show significant statisticaldifferences in relaxed and contracted arm girths, left and
right hand grip strength and FFM. Tukey post-hoc test
shows significant differences between body building andother sports, especially soccer, for all the traits above.Table 1 Anthropometric traits by sex
Males Females
Trait mean ± SD mean ± SDH (cm) 177.6 ± 6.3 163.9 ± 6.0
W (kg) 75.6 ± 10.2 58.7 ± 8.2BMI (kg/m
2) 24.0 ± 2.8 21.8 ± 2.6
SH (cm) 92.9 ± 3.5 87.0 ± 3.6T Sk (mm) 10.8 ± 5.0 16.1 ± 5.5B Sk (mm) 5.5 ± 3.2 8.7 ± 4.5WC (cm) 81.7 ± 7.3 70.3 ± 6.5WSR 0.46 ± 0.04 0.43 ± 0.04CAG (cm) 32.5 ± 3.1 27.4 ± 2.7RAG (cm) 29.5 ± 3.0 25.7 ± 2.6
RHG (kg) 50.2 ± 8.0 30.8 ± 5.1
LHG (kg) 48.3 ± 8.1 29.2 ± 5.0BD (g/cc) 1.059 ± 0.011 1.039 ± 0.011%F 17.3 ± 4.9 26.6 ± 5.2FM (kg) 13.3 ± 5.1 16.0 ± 4.9FFM (kg) 62.4 ± 7.4 42.9 ± 4.9
H = height; W = weight; SH = sitting heigh t; T sk = triceps skinfold; B sk = biceps
skinfold; WC = waist circumference; WSR = w aist-to-stature ratio; CAG = contracted
arm girth; RAG = relaxed arm girth; RHG = right hand grip; LHG = left hand grip;BD = body density; %F = body fat percentage; FM = fat mass; FFM = fat free mass.Zaccagni et al. Journal of Translational Medicine 2014, 12:120 Page 3 of 9
http://www.translational-medicine.com/content/12/1/120

Table 2 Anthropometric traits in male sub-samples by level of physical activity
Males Low (1st tertile) Medium (2nd tertile) High (3rd tertile)
Trait mean ± SD mean ± SD mean ± SD P-value
H (cm) 177.2 ± 6.2 177.6 ± 6.4 178.1 ± 6.3 0.557
W (kg) 76.4 ± 11.3 75.7 ± 10.5 74.9 ± 8.8 0.507BMI (kg/m
2) 24.3 ± 3.3 24.0 ± 3.0 23.6 ± 2.2 0.151
SH (cm) 92.9 ± 3.4 92.8 ± 3.6 93.0 ± 3.6 0.872T Sk (mm) 11.6 ± 5.7 10.9 ± 4.2 10.0 ± 4.7 0.054B Sk (mm) 6.0 ± 3.1
a5.5 ± 3.3 5.2 ± 3.2 0.043
WC (cm) 83.1 ± 7.9a81.8 ± 8.2 80.3 ± 5.5 0.009
WSR 0.47 ± 0.05a0.46 ± 0.05 0.45 ± 0.03 0.002
CAG (cm) 32.8 ± 3.1 32.3 ± 3.1 32.4 ± 3.2 0.483RAG (cm) 29.7 ± 2.9 29.4 ± 3.0 29.3 ± 3.0 0.515
RHG (kg) 49.9 ± 7.6 49.9 ± 7.8 50.6 ± 8.5 0.691
LHG (kg) 48.7 ± 7.8 48.1 ± 7.9 48.1 ± 8.6 0.805BD (g/cc) 1.057 ± 0.012
a1.059 ± 0.010 1.061 ± 0.011 0.009
%F 18.2 ± 5.4a17.5 ± 4.3 16.3 ± 4.8 0.009
FM (kg) 14.2 ± 5.9a13.4 ± 4.7 12.4 ± 4.5 0.022
FFM (kg) 62.5 ± 8.0 62.0 ± 7.5 62.6 ± 6.8 0.835
Note: Tukey post-hoc test:ap<0 . 0 5c o m p a r e dw i t hh i g h .
H = height; W = weight; SH = sitting height; T sk = triceps skinfold; B sk = biceps skinfold; WC = waist circumference; WSR = waist-to-stature ratio; CAG = c ontracted arm
girth; RAG = relaxed arm girth; RHG = right hand grip; LHG = left hand grip; BD = body density; %F = body fat percentage; FM = fat mass; FFM = fat free mass.
Table 3 Anthropometric traits in female sub-samples by level of physical activity
Females Low (1st tertile) Medium (2nd tertile) High (3rd tertile)
Trait mean ± SD mean ± SD mean ± SD P-value
H (cm) 163.5 ± 5.9 163.8 ± 5.6 164.4 ± 6.6 0.521
W (kg) 57.9 ± 8.8 57.6 ± 7.5b60.3 ± 8.0 0.019
BMI (kg/m2) 21.6 ± 2.9 21.5 ± 2.4b22.3 ± 2.4 0.035
SH (cm) 86.6 ± 3.5 87.0 ± 3.2 87.4 ± 3.9 0.224T Sk (mm) 16.5 ± 5.4 15.8 ± 6.1 16.1 ± 5.1 0.497B Sk (mm) 9.2 ± 4.5 8.4 ± 5.1 8.4 ± 3.8 0.253WC (cm) 70.3 ± 7.9 69.5 ± 6.1 70.7 ± 5.4 0.403WSR 0.43 ± 0.05 0.43 ± 0.04 0.43 ± 0.03 0.488
CAG (cm) 26.9 ± 2.7
a27.1 ± 2.5b28.0 ± 2.7 0.004
RAG (cm) 25.4 ± 2.6a25.5 ± 2.4 26.1 ± 2.6 0.045
RHG (kg) 30.0 ± 5.2 31.0 ± 4.6 31.5 ± 5.3 0.076LHG (kg) 28.4 ± 5.0 29.4 ± 4.8 29.8 ± 5.1 0.068BD (g/cc) 1.038 ± 0.011 1.040 ± 0.012 1.039 ± 0.010 0.337%F 27.2 ± 5.4 26.1 ± 5.6 26.5 ± 4.8 0.337FM (kg) 16.1 ± 5.2 15.3 ± 4.9 16.3 ± 4.5 0.315FFM (kg) 41.9 ± 5.0
a42.4 ± 4.6b44.1 ± 4.8 0.001
Note : Tukey post-hoc test:ap < 0.05 compared with high;bp < 0.05 compared with high.
H = height; W = weight; SH = sitting height; T sk = triceps skinfold; B sk = biceps skinfold; WC = waist circumference; WSR = waist-to-stature ratio; CAG = c ontracted arm
girth; RAG = relaxed arm girth; RHG = right hand grip; LHG = left hand grip; BD = body density;%F = body fat percentage; FM = fat mass; FFM = fat free mass.Zaccagni et al. Journal of Translational Medicine 2014, 12:120 Page 4 of 9
http://www.translational-medicine.com/content/12/1/120

Body builders had the highest BMI –similar to that of
volleyball players -, arm girths, right and left hand grip
strength, BD and FFM, and the lowest H, skinfold thick-nesses, %F and FM. Soccer players had the lowest W, arm
girths and hand grip. Volleyball players had the highest W,
WC and FM, and the lowest SH. Basketball players hadthe highest H and SH, and the lowest BMI. Swimmers had
the thickest skinfolds and the highest %F, the lowest WC,
BD, and FFM.
ANOVAs between sport discipline with more than 10
participants in females (Table 6) show significant statistical
differences in height, weight, BMI, WC, RAG, left and righthand grip strength, FM and FFM. Tukey post-hoc test
shows significant differences between volleyball players,
gymnasts and dancers for the traits above. Volleyballplayers had the highest H, SH, W, BMI, triceps skinfold,
girths, hand grip strength, %F and FFM. Gymnasts were
the shortest and lightest and had the greatest BD, the low-est SH, skinfold thickness, WC, %F, FM and FFM. Dancers
had the smallest arm girths (RAG values being similar to
t h o s eo fg y m n a s t s )a n dg r i ps t r e n g t h .
A significant positive correlation between BMI and %F
was found in males (r = 0.476, p < 0.001), but it did not
reach significance in basketball players (p = 0.300) andbody builders (p = 0.906). In fact, one third of the subjects
who were classified as overweight according to BMI, but
who were actually normal fat, practiced body building.Twelve percent of total participants fell within the FP
quadrant and 10% in the FN one (Figure 1[a]). SensitivityTable 4 Correlation coefficients between anthropometric
traits and hours of physical activity in males and females
Trait Males Females
H (cm) 0.045 0.038
W (kg) −0.076 0.042
BMI (kg/m2) −0.114* 0.022
SH (cm) −0.015 0.048
T Sk (mm) −0.180** −0.072
B Sk (mm) −0.197*** −0.0173**
WC (cm) −0.163** −0.021
WSR −0.170** −0.048
CAG (cm) −0.074 0.091
RAG (cm) −0.073 0.034
RHG (kg) 0.028 0.123*
LHG (kg) 0.003 0.127*
BD (g/cc) 0.212*** 0.120*%F −0.212*** −0.121*
FM (kg) −0.200*** −0.063
FFM (kg) 0.033 0.131*
***p < 0.001 **p < 0.01 *p < 0.05.
H = height; W = weight; SH = sitting height; T sk = triceps skinfold; B sk = bicepsskinfold; WC = waist circumference; WSR = w aist-to-stature ratio; CAG = contracted
arm girth; RAG = relaxed arm girth; RHG = right hand grip; LHG = left hand grip;BD = body density; %F = body fat percent age; FM = fat mass; FFM = fat free mass.
Table 5 Anthropometric traits by sport in males
Males Soccer N = 132 Swimming N = 25 Basketball N = 26 Bodybuilding N = 41 Volleyball N = 13
Traits mean ± SD mean ± SD mean ± SD mean ± SD mean ± SD pH (cm) 177.0 ± 6.1 177.8 ± 6.0 180.0 ± 6.9 176.9 ± 6.9 178.2 ± 6.3 0.298
W (kg) 74.5 ± 9.3 74.5 ± 9.0 75.7 ± 9.9 77.7 ± 9.3 78.4 ± 13.5 0.295BMI (kg/m
2) 23.8 ± 2.4 23.6 ± 2.6 23.5 ± 3.2 24.8 ± 2.4 24.8 ± 4.7 0.141
SH (cm) 92.8 ± 3.6 93.4 ± 3.0 94.0 ± 3.6 92.5 ± 3.7 92.0 ± 3.3 0.391T Sk (mm) 10.8 ± 5.1 11.3 ± 5.5 10.9 ± 3.9 9.5 ± 4.2 11.0 ± 4.0 0.452B Sk (mm) 5.9 ± 3.8 6.0 ± 3.3 4.9 ± 2.1 4.6 ± 2.0 4.9 ± 1.9 0.145WC (cm) 81.3 ± 6.3 81.0 ± 5.8 82.9 ± 6.6 81.9 ± 6.8 83.3 ± 14.0 0.184WSR 0.46 ± 0.04 0.46 ± 0.03 0.46 ± 0.04 0.46 ± 0.04 0.47 ± 0.09 0.763
CAG (cm) 31.5 ± 2.7 33.0 ± 2.5 31.9 ± 2.5 36.0 ± 3.2
a32.1 ± 2.8 0.000
RAG (cm) 28.5 ± 2.6 29.7 ± 2.4 28.8 ± 2.5 32.4 ± 3.1a29.4 ± 2.4 0.000
RHG (kg) 47.9 ± 7.7 50.2 ± 5.6 50.2 ± 6.8 55.0 ± 8.5b48.9 ± 7.1 0.000
LHG (kg) 45.6 ± 7.4 48.4 ± 4.1 49.9 ± 6.6 52.5 ± 9.5 49.5 ± 8.4 0.000BD (g/cc) 1.059 ± 0.011 1.058 ± 0.011 1.060 ± 0.008 1.062 ± 0.011 1.059 ± 0.011 0.618%F 17.4 ± 4.9 18.0 ± 4.8 17.1 ± 3.7 16.2 ± 4.8 17.6 ± 4.8 0.620FM (kg) 13.1 ± 5.0 13.7 ± 4.9 13.0 ± 3.6 12.7 ± 4.4 14.2 ± 5.6 0.837FFM (kg) 61.2 ± 6.6 60.9 ± 6.1 62.4 ± 8.4 65.2 ± 8.2
b64.2 ± 10.2 0.035
Tukey post-hoc test:aBodybuilding vs all other sports p < 0.001bBodybuilding vs soccer p < 0.05.
H = height; W = weight; SH = sitting height; T sk = triceps skinfold; B sk = biceps skinfold; WC = waist circumference; WSR = waist-to-stature ratio; CAG = c ontracted arm
girth; RAG = relaxed arm girth; RHG = right hand grip; LHG = left hand grip; BD = body density; %F = body fat percentage; FM = fat mass; FFM = fat free mass.Zaccagni et al. Journal of Translational Medicine 2014, 12:120 Page 5 of 9
http://www.translational-medicine.com/content/12/1/120

was 0.62 and specificity was 0.83, while PPV was 0.58 and
NPV was 0.85. A significant positive correlation betweenBMI and %F was found in females (r = 0.622, p < 0.001)
but it did not reach significance in gymnasts (p = 0.752).
Seven percent were classified as FP and 4% as FN (Figure 1[b]). Sensitivity was 0.59 and specificity was 0.92, while
PPV was 0.45 and NPV was 0.95. Therefore, sensitivity
was poor for both sexes, reflecting the fact that the indi-viduals who were at the same time classified as overfat (ac-
cording to their %F) and overweight (according to their
BMI) were only a small proportion of those who were ac-tually overfat. Also PPV of BMI was poor, because really
fat individuals were about a half of those who were classi-
fied as overweight.
A significant positive correlation between WSR and %
F was found in males (r = 0.439, p < 0.001). Four percent
of total participants fell in the FP quadrant and 17% inthe FN one (Figure 1[c]). Sensitivity was 0.36 and speci-
ficity was 0.95, while PPV was 0.73 and NPV was 0.80.
A significant positive correlation between WSR and %Fwas found in females (r = 0.527, p < 0.001). Three per-
cent of total participants fell within the FP quadrant and
8% in the FN one (Figure 1[d]). Sensitivity was 0.24 andspecificity was 0.97, while PPV was 0.47 and NPV was
0.92. Therefore, sensitivity was poor for both sexes, and
PPV was poor especially in females.Discussion
In the present study, we found a different trend in thetwo sexes in relation to training volume: female students
performing more hours of weekly physical activity had a
significantly higher amount of FFM compared to the lessactive individuals, while male students showed a lower %
F and FM. A study by Westerterp et al. [26] found a
negative correlation between energy expenditure and %Fin males, but not in females. Also, a negative correlation
between physical activity and FM was found in males,
but not in females [27]. Even if FM can be reduced bymeans of increased physical activity, females seem to
compensate for excess energy expenditure with added
energy intake. Since women tend to preserve their en-ergy balance more than men [28], FM loss can be not
significant. Increased caloric intake could also justify
added FFM, as in the present research.
The different behaviours may be consistent with both
sex-related differences and sport preferences. The exam-
ined females are more often than males engaged in indi-vidual sports and in disciplines with a relevant aesthetic
component (gym activities, ballet, gymnastics). Males are
more often than females engaged in team sports (soccer,basketball, volleyball) or in strength-related activities, like
body building, both involving intense repeated efforts,
which have been positively correlated to fat loss [29,30].Table 6 Anthropometric traits by sport in females
Females Gymnastics N = 19 O.G.A. N = 50 Swimming N = 39 Jogging N = 16 Ballet N = 47 Volleyball N = 47 p
Traits mean ± SD mean ± SD mean ± SD mean ± SD mean ± SD mean ± SDH (cm) 161.2 ± 6.0 163.0 ± 5.8 164.7 ± 6.6 163.7 ± 4.7 163.4 ± 4.5 166.2 ± 7.0
a0.033
W (kg) 54.9 ± 5.9 58.3 ± 8.7 58.0 ± 8.1 58.3 ± 9.4 57.1 ± 6.1 62.8 ± 7.9a,b0.002
BMI (kg/m2) 21.0 ± 2.0 21.9 ± 2.9 21.3 ± 2.3 21.7 ± 2.7 21.4 ± 1.9 22.7 ± 2.3 0.049
SH (cm) 86.3 ± 3.0 86.4 ± 3.7 87.1 ± 4.0 87.3 ± 3.1 87.1 ± 3.3 88.4 ± 4.0 1.635T Sk (mm) 14.3 ± 4.1 14.9 ± 5.8 16.0 ± 6.5 15.2 ± 4.1 15.8 ± 4.7 17.5 ± 5.0 0.177B Sk (mm) 6.6 ± 2.4 8.9 ± 4.6 9.7 ± 5.0 9.6 ± 3.6 7.6 ± 3.5 8.7 ± 3.5 0.083WC (cm) 67.1 ± 4.8 69.4 ± 6.1 70.6 ± 7.6 70.6 ± 7.1 68.3 ± 4.1 71.9 ± 5.0
a0.023
WSR 0.42 ± 0.03 0.43 ± 0.04 0.43 ± 0.05 0.43 ± 0.04 0.42 ± 0.02 0.43 ± 0.03 0.359CAG (cm) 26.9 ± 1.2 27.4 ± 2.7 27.2 ± 2.6 27.1 ± 2.5 26.6 ± 2.4 28.3 ± 2.3 0.063RAG (cm) 24.9 ± 1.5 25.5 ± 2.6 25.7 ± 2.7 25.5 ± 2.8 24.9 ± 2.2 26.6 ± 2.3
b0.033
RHG (kg) 30.1 ± 4.6 31.8 ± 5.4 30.7 ± 4.8 31.3 ± 4.3 28.3 ± 3.6c31.8 ± 5.3b0.008
LHG (kg) 29.3 ± 4.8 29.8 ± 5.5 28.7 ± 4.6 29.9 ± 5.1 26.7 ± 3.9c30.1 ± 4.8b0.018
BD (g/cc) 1.043 ± 0.009 1.040 ± 0.010 1.038 ± 0.014 1.038 ± 0.009 1.040 ± 0.010 1.037 ± 0.010 0.333%F 24.4 ± 4.0 26.1 ± 5.4 26.8 ± 6.3 26.9 ± 4.2 26.0 ± 4.5 27.6 ± 4.5 0.329FM (kg) 13.2 ± 2.1 15.6 ± 5.1 15.8 ± 5.3 16.0 ± 4.4 14.9 ± 3.9 17.8 ± 4.4
a0.009
FFM (kg) 40.9 ± 4.6 43.1 ± 5.2 42.5 ± 5.2 42.2 ± 5.9 41.6 ± 3.1 45.4 ± 4.4a,b0.003
O.G.A. = other gym activities.
Tukey post-hoc test:avolleyball versus gymnastics p < 0.05bvolleyball vs ballet p < 0.01cballet vs other gym activities p < 0.05.
H = height; W = weight; SH = sitting height; T sk = triceps skinfold; B sk = biceps skinfold; WC = waist circumference; WSR = waist-to-stature ratio; CAG = c ontracted arm
girth; RAG = relaxed arm girth; RHG = right hand grip; LHG = left hand grip; BD = body density; %F = body fat percentage; FM = fat mass; FFM = fat free mass.Zaccagni et al. Journal of Translational Medicine 2014, 12:120 Page 6 of 9
http://www.translational-medicine.com/content/12/1/120

So the different adaptation could be sport-related. Body
building determines an evident increase on muscle hyper-
trophy, which is significant in comparison to soccer. This
fact may contribute to the limited accuracy of BMI as anindex of body fatness and general health status, since body
builders have a high BMI, close to overweight, even if they
have the lowest %F in our sample. Also, females are lessphysically active, on average, therefore it can be hypothe-
sized that physical adaptation in response to moderate
physical activity can be correlated to increased musclemass, and, vice-versa, that physical adaptation in response
to high volume of weekly physical activity can be corre-
lated to reduced %F.
The variance in weekly hours of physical activity within
the sample determined significant differences in body com-
position, and showed the limits of BMI and WSR as indicesof adiposity. Intersecting BMI values with %F, we have ob-
tained important indications on its limited applicability in a
sample of young adults with different levels of weekly train-ing hours.
The analysis of specificity and sensitivity showed that
neither BMI nor WSR can be considered accurateindices of the health status of the population of young
adults because they are not consistent measures of body
fatness. In fact, both BMI and WSR had good specificity
versus %F, but low sensitivity, suggesting that a signifi-cant percentage of overfat individuals were classified as
normal according to BMI or WSR.
A possible limiting factor of the present study is that
physical activity assessment ( weekly training hours and type
of sport) was based only on a self-reported questionnaire.
Also, the training volume does not account for training in-
tensity and quality (mainly aer obic, anaerobic etc.). A lower
volume of weekly training hours involving a strenuous
practice may have more significant outcomes than a highervolume with a less intense effort, in particular for what
concern body compositi on. Moreover, it must be
highlighted that the skinfold-thickness technique is an in-direct method for assessing body composition, based on
population-specific predictive e quations. Although relatively
inexpensive, non-invasive, and widely used in sportsmen,its accuracy cannot be granted especially in individuals with
adipose tissue that is not well separated from the underlying
muscle [31].
Figure 1 Scatterplot of anthropometric indices BMI ([a] and [b]) and WSR ([c] and [d]) and %F for each male ([a] and [c]) and female ([b]
and [d]) study participant – in each scatterplot the four quadrants are labelled FN (false negative), TP (true positive), TN (true negative),
and FP (false positive) to illustrate the correct classifications and misclassifications.Zaccagni et al. Journal of Translational Medicine 2014, 12:120 Page 7 of 9
http://www.translational-medicine.com/content/12/1/120

Conclusions
This study examined a large sample of Italian university
students from the same geographical area by means ofrigorous anthropometric procedures.
In conclusion, a different behaviour was highlighted
in the two sexes in relation to weekly amount of phys-ical activity: males mainly showed a decrease in %F,
whereas females showed an increase in FFM, which
could be explained by a stronger tendency to maintainenergetic balance or by diff erent sport preferences.
BMI and WSR have been suggested as indirect measures
of %F, because of the ease with which they can be col-lected. The present study confirms their low accuracy. In
fact, in females, misclassification (FP + FN) was 11% for
both BMI and WSR. In males, misclassification was 22%for BMI and 21% for WSR. Therefore, regardless of the
fact that WSR has been proposed as a better index of adi-
posity than BMI, both indices show similar low accuracyand they cannot be considered reliable predictors of body
fatness, especially in young males. Greater accuracy can be
found in females, possibly because of lower overall FFMcompared to males. In fact, high FFM contributes to in-
creased BMI, without any real detrimental effect (e.g. in
body builders).
The present study confirms that an active lifestyle, in-
cluding regular weekly physical activity, is significantly
correlated to body composition parameters.
Competing interests
The authors declare that they have no competing interests.
Authors ’contributions
EG-R and LZ have planned the study, ZL and BD participated to the data
collection and did all the statistical analysis. All authors were involved with
data interpretation, critical revisions of the paper. All authors read andapproved the final manuscript.
Acknowledgments
We would like to thank all the students who volunteered for the study.
Received: 6 March 2014 Accepted: 30 April 2014
Published: 9 May 2014
References
1. De Lorenzo A, Bianchi A, Maroni P, Iannarelli A, Di Daniele N, Iacopino L,
Di Renzo L: Adiposity rather than BMI determines metabolic risk. Int J
Cardiol 2013, 166(1):111 –117.
2. Zaccagni L, Onisto N, Gualdi-Russo E: Biological characteristics and ageing
in former elite volleyball players. J Sci Med Sport 2009, 12(6):667 –672.
3. Barbieri D, Zaccagni L, Cogo A, Gualdi Russo E: Body composition and
somatotype of experienced mountain climbers. High Alt Med Biol 2012,
13(1):46 –50.
4. Klungland Torstveit M, Sundgot-Borgen J: Are under- and overweight
female elite athletes thin and fat? A controlled study. Med Sci Sports Exerc
2012, 44(5):949 –957.
5. Tanaka S, Togashi K, Rankinen T, Pérusse L, Leon AS, Rao DC, Skinner JS,
Wilmore JH, Bouchard C: Is adiposity at normal body weight relevant
for cardiovascular disease risk? Int J Obes Relat Metab Disord 2002,
26(2):176 –183.
6. Cho YG, Song HJ, Kim JM, Park KH, Paek YJ, Cho JJ, Caterson I, Kang JG: The
estimation of cardiovascular risk factors by body mass index and bodyfat percentage in Korean male adults. Metabolism 2009, 58(6):765 –771.7. Onisto N, Teofoli F, Zaccagni L, Gualdi Russo E: Anthropometric traits and
aging: a cross-sectional survey in diabetic elderly women. Arch Gerontol
Geriatr 2009, 48(2):197 –200.
8. Gokulakrishnan K, Deepa M, Monickaraj F, Mohan V: Relationship of
body fat with insulin resistance and cardiometabolic risk factors
among normal glucose-tolerant subjects. JP o s t g r a dM e d 2011,
57(3):184 –188.
9. Chuang HH, Li WC, Sheu BF, Liao SC, Chen JY, Chang KC, Tsai YW:
Correlation between body composition and risk factors for
cardiovascular disease and metabolic syndrome. Biofactors 2012,
38:284 –291.
10. Gualdi-Russo E, Zironi A, Dallari GV, Toselli S: Migration and Health in Italy:
A Multiethnic Adult Sample. J Travel Med 2009, 16(2):88 –95.
1 1 . A s h w e l lM ,G u n nP ,G i b s o nS : Waist-to-height ratio is a better
screening tool than waist circumference and BMI for adultcardiometabolic risk factors: systematic review and meta-analysis.Obes Rev 2012, 13(3):275
–286.
12. Ashwell M, Hsieh SD: Six reasons why the waist-to-height ratio is a rapid
and effective global indicator for health risks of obesity and how its usecould simplify the international public health message on obesity. Int J
Food Sci Nutr 2005, 56(5):303 –307.
13. Mombelli G, Zanaboni AM, Gaito S, Sirtori CR: Waist-to-height ratio is a
highly sensitive index for the metabolic syndrome in a Mediterranean
population. Metab Syndr Relat Disord 2009, 7(5):477 –484.
14. Sacheck JM, Kuder JF, Economos CD: Physical fitness, adiposity, and
metabolic risk factors in young college students. Med Sci Sports Exerc
2010, 42(6):1039 –1044.
15. Reimers CD, Knapp G, Reimers AK: Does physical activity increase life
expectancy? A review of the literature. J Aging Res 2012, 2012: 243958.
doi:10.1155/2012/243958. Epub 2012 Jul 1.
16. Wagner A, Dallongeville J, Haas B, Ruidavets JB, Amouyel P, Ferrières J,
Simon C, Arveiler D: Sedentary behaviour physical activity and dietary
patterns are independently associated with the metabolic syndrome.Diabetes Metab 2012, 38(5):428 –435.
17. Garber CE, Blissmer B, Deschenes M R, Franklin BA, Lamonte MJ, Lee IM,
Nieman DC, Swain DP, American Colle ge of Sports Medicine, American
College of Sports Medicine position stand: Quantity and quality of
exercise for developing and maintaining cardiorespiratory,
musculoskeletal and neuromotor fitness in apparently healthy adults:
guidance for prescribing exercise. Med Sci Sports Exerc 2011,
43(7):1334 –1359.
18. Zaccagni L, Masotti S, Donati R, Mazzoni G, Gualdi-Russo E: Body image
and weight perceptions in relation to actual measurements by means of
a new index and level of physical activity in Italian university students.J Transl Med 2014, 12:42.
19. Weiner JS, Lourie JA: Practical Human Biology. London: Academic Press; 1981.
20. Durnin JV, Womersley J: Body fat assessed from total body density and its
estimation from skinfold thickness: measurements on 481 men and
women aged from 16 to 72 years. Br J Nutr 1974, 32(1):77 –97.
21. Siri WE: The gross composition of the body. Adv Biol Med Phys 1956,
4:239 –280.
22. James PT, Leach R, Kalamara E, Shayeghi M: The worldwide obesity
epidemic. Obes Res 2001, 9:S228 –S233.
23. Gallagher D, Heymsfield SB, Heo M, Jebb SA, Murgatroyd PR, Sakamoto Y:
Healthy percentage body fat ranges: an approach for developing
guidelines based on body mass index. Am J Clin Nutr 2000,
72(3):694 –701.
24. McNeil B, Keeler E, Adelstein J: Primer on certain elements of clinical
decision making. N Engl J Med 1975, 293:211 –215.
25. Gualdi Russo E, Gruppioni G, Gueresi P, Belcastro MG, Marchesini V:
Skinfolds and body composition of sports participants. J Sports Med Phys
Fitness 1992, 32:303 –313.
26. Westerterp KR, Goran MI: Relationship between physical activity related
energy expenditure and body composition: a gender difference. Int J
Obes Relat Metab Disord 1997, 21:184 –188.
27. Westerterp KR, Meijer GA, Kester AD, Wouters L, Ten Hoor F: Fat-free mass
as a function of fat mass and habitual activity level. Int J Sports Med 1992,
13(2):163 –166.
28. Westerterp KR, Meijer GA, Janssen EM, Saris WH, Ten Hoor F: Long-term
effect of physical activity on energy balance and body composition. Br J
Nutr 1992, 68:21–30.Zaccagni et al. Journal of Translational Medicine 2014, 12:120 Page 8 of 9
http://www.translational-medicine.com/content/12/1/120

29. Tremblay A, Simoneau JA, Bouchard C: Impact of exercise intensity on
body fatness and skeletal muscle metabolism. Metabolism 1994,
43(7):814 –818.
30. Sijie T, Hainai Y, Fengying Y, Jianxiong W: High intensity interval exercise
training in overweight young women. J Sports Med Phys Fitness 2012,
52(3):255 –262.
31. Selkow NM, Pietrosimone BG, Saliba SA: Subcutaneous thigh fat
assessment: a comparison of skinfold calipers and ultrasound imaging.J Athl Train 2011, 46:50–54.
doi:10.1186/1479-5876-12-120
Cite this article as: Zaccagni et al. :Body composition and physical
activity in Italian university students. Journal of Translational Medicine
2014 12:120.
Submit your next manuscript to BioMed Central
and take full advantage of:
• Convenient online submission
• Thorough peer review
• No space constraints or color figure charges
• Immediate publication on acceptance
• Inclusion in PubMed, CAS, Scopus and Google Scholar
• Research which is freely available for redistribution
Submit your manuscript at
www.biomedcentral.com/submitZaccagni et al. Journal of Translational Medicine 2014, 12:120 Page 9 of 9
http://www.translational-medicine.com/content/12/1/120