Full text of DOI:10.1007/BF03344033

J. Endocrinol. Invest. 25: 431-435, 2002
The impact of gender, body dimension and body
composition on hand-grip strength in healthy children
A. Sartorio*,**, C.L. Lafortuna***, S. Pogliaghi***, and L. Trecate*
*Laboratory for Experimental Endocrinological Research (LSRE), Istituto Auxologico Italiano, IRCCS,
Milano, **Division of Metabolic Diseases III, Istituto Auxologico Italiano, IRCCS, Piancavallo (VB),
***Institute of Bioimaging and Molecular Physiology, CNR, Milan, Italy
ABSTRACT. Maximum hand-grip (HG) strength,
body composition and main anthropometric vari-
ables were evaluated in 278 children with normal
weight and growth, aged 5-15 yr divided into 3
age groups: group 1, age SD: 7.6 0.9 yr 7.6 0.9
SD (Tanner stage 1); group 2, age: 10.8 0.7 yr
(Tanner stage: 2-3); group 3, age: 13.2 0.9 yr
(Tanner stage: 4-5). Weight, height, body surface
area (BSA), BMI, percent body fat (BF) and fat free
mass (FFM) increased progressively and signifi-
cantly from the younger to the older age group. A
significant difference between genders was de-
tected only for BF and FFM, females having a
higher fat mass and a lower FFM compared to
males. Most children were right-handed (91%). In
either genders, a curvilinear relation was detected
between HG strength and age, with best fit for
the dominant (d) hand given by the equations:
dHG=5.891 *100.051 age, r2=0.986, p<0.001 in
males and dHG=6.163 *100.045 age, r2=0.973,
p<0.001 in females. The increase in HG strength
after 11 yr appears to be steeper in males as com-
pared with that found in females. In both d and
non-dominant (nd) hand, a significant difference
in HG strength was detected between males and
females, the average difference being about 10%
at all ages. For both genders, nd hand was signif-
icantly weaker than d hand in the older age
groups (2 and 3), but not in the younger group 1.
Age and gender-dependent differences in HG
strength (but not differences between d and nd
hand) disappear if HG strength is normalized for
FFM. Thus, in general, dHG strength normalized
for FFM resulted on average to be 0.67 0.11
kg/kg. A multiple linear regression analysis indica-
ted that HG was positively correlated with BMI,
BSA, stature, stature2 and FFM (p<0.001 for all
correlations) without differences between gen-
ders, while a negative correlation was found be-
tween HG strength and %BF. The most significant
correlation was found between HG strength and
FFM, without any significant difference between
genders, so that the overall equation describing
the line for the d hand was: dHG strength=
2.32+0.63 FFM, r2=0.72, p<0.001. In conclusion,
the present study indicates that the age-depen-
dent increase of HG strength as well as the be-
tween-gender differences are strongly related to
changes of FFM values occurring during child-
hood. Moreover, the study provides a standard
normative value of maximal HG strength for the
healthy children population in Northern Italy.
(J. Endocrinol. Invest. 25: 431-435, 2002)
©2002, Editrice Kurtis
INTRODUCTION
is considered the single clinical measure most rep-
resentative of total body strength in humans (3-5).
Normative data for healthy children have been re-
ported previously by several Authors (6-8), although
using different instruments (spherical vigorimeter
and Jamar dynamometer) and in restricted age-
ranges (6, 8). In these studies, anthropometric da-
ta were not systematically examined, number of tri-
als for each hand and resting periods between tri-
als varied considerably, thus hampering the gener-
al applicability of the normal values obtained.
Furthermore, to our knowledge no data are avail-
able on the relationships between maximal HG
Maximal hand-grip (HG) strength is commonly used
as an index of general health and as a screening test
for the integrity of both the upper motor neurons
and function of the motor unit (1, 2). Furthermore, it
Key-words: Body composition, children, fat free mass, hand-grip.
Correspondence: Dr. Alessandro Sartorio, Laboratorio Sperimentale di
Ricerche Endocrinologiche (LSRE), Istituto Auxologico Italiano, IRCCS,
Via Ariosto 13, 20145 Milano, Italy.
E-mail: sartorio@auxologico.it
Accepted December 14, 2001.
431

A. Sartorio, C.L. Lafortuna, S. Pogliaghi, et al.
make their best effort (strongly encouraged by the medical staff)
in a spirit of competition, the measurements were performed in
presence of all the classmates (of the same sex).
All assessments took place in a quiet and temperature-con-
trolled room (approximately 20 C), between 08:30-11:30 h.
Hand dominance was determined by asking the children to
throw a ball or use a writing implement. All children were seat-
ed in appropriately sized chairs that allowed their feet to be flat
on the floor.
Maximum HG strength was measured with a pediatric hand dy-
namometer (Lafayette Instrument, U.S.A.), with standardized po-
sitioning and instructions, as previously described (7). The subjects
were seated with their shoulder adducted and neutrally rotated,
elbow flexed at 90°, forearm in neutral position, and wrist between
0° and 30° dorsiflexion and between 0° and 15° of ulnar deviation
(7, 10). The hand to be tested first was chosen randomly, each
hand being then tested alternatively (with a 60-second rest be-
tween trials). Three trials were performed on each hand and the
highest attained value (kg) for each hand was used for analysis.
strength and body composition modifications over
age so far.
The purpose of the present study was to develop
normative maximum HG strength in normally grow-
ing children aged 5-15 yr, using a single dynamo-
meter and a rigorously standardized procedure.
Furthermore, we evaluated the influence of gen-
der, anthropometric variables [body mass (BM),
BMI, body surface area (BSA), body fat (BF), fat-free
mass (FFM)] and hand dominance on age-related
HG strength modifications.
MATERIALS AND METHODS
A total of 332 (197 males, 135 females) healthy, urban school
children from Milan, Italy (age range: 5-15 yr, mean age SD:
11.4 2.4 yr) participated in the study. The study protocol was
approved by the Ethical Committee of the Istituto Auxologico
Italiano, Milan, Italy and written informed consent was obtained
from the children’s parents.
Statistics
In a first step of analysis, subjects were stratified according to
pubertal development, so that 3 groups for each gender were
formed, corresponding to Tanner stages 1 (group 1), 2-3 (group
2) and 4-5 (group 3), respectively.
Average and standard deviation (SD) were calculated for all the
parameters. Groups were compared for significant differences by
means of two-way ANOVA (Tanner stage and gender) (Sigma-
stat, Jandel Scientific), followed by Tukey test.
In a second step of analysis, subjects were grouped in one-year
age classes and average and SD of HG values were calculated.
Finally, linear correlation was tested between individual HG val-
ues and anthropometric variables (BMI, BSA, stature, stature2
and FFM) of both males and females.
A multiple linear regression, based on all anthropometric vari-
ables, gender and age was also tested. The least significant vari-
ables were discarded stepwise, while strongly inter-correlated
variables were excluded.
Anthropometric measurements
Standing height was measured with a stadiometer (Harpenden
Stadiometer, Holtain, U.K.) to the nearest 0.1 cm. BM was mea-
sured with an electronic scale (Tanita Inc., Japan) to the nearest
0.1 kg; BMI was calculated as the weight (kg) divided by the
squared height (m2).
Total body water was measured by a tetrapolar foot-to-foot
impedance apparatus (model TBF 305, Tanita Inc., Japan), using
4 pressure contact stainless steel foot pads; this apparatus
records bw, and when this is stable, predicts total body water
via the measurement of impedance and hence percent body fat
(%BF) and fat-free mass (%FFM), as previously described (9).
A BMI value higher or lower than 2 SD from the mean value for
age and sex was taken as an exclusion criteria.
HG strength measurement
The differences between regression lines were tested using con-
ventional regression equation comparison methods (11).
Significance was set at p<0.05 for all comparisons.
Each school class was given written instructions and a demon-
stration before being tested and further verbal instructions were
provided at the time of the test. To encourage the children to
Table 1 - Descriptive data of the study group.
Group 1 (Tanner stage 1)
Group 2 (Tanner stage 2, 3)
Group 3 (Tanner stage 4, 5)
Gender
No.
M
F
M
F
M
F
25
29
46
20
94
64
Age (yr)
BM (kg)
Height (cm)
BMI (kg/m2)
BSA (m2)
BF (%)
7.7 0.9
25.9 3.6
127.7 6.8
15.8 1.1
0.96 0.09
13.7 3.6
22.5 2.4
7.5 0.8
10.8 0.7
35.7 6.1
144.4 6.8
17.0 1.7
1.21 0.12
17.4 5.3
29.3 3.8
10.7 0.7
34.8 4.7
143.6 8.0
16.8 1.1
1.19 0.11
24.1 4.5**
26.3 3.5*
13.4 0.9
48.3 8.5
158.4 8.5
19.1 2.2
1.47 0.16
16.8 7.3
39.6 8.7
13.1 0.9
47.9 8.1
157.4 6.3
19.3 2.5
1.50 0.14
25.4 6.1**
35.4 5.2**
24.4 3.6
124.8 7.1
15.6 1.3
0.93 0.09
20.9 4.2**
19.2 2.2**
FFM (kg)
Values are given as average SD. M: males; F: females; BF: body fat; BM: body mass; BSA: body surface area; FFM: fat-free mass. Asterisks indicate sig-
nificant differences between M and F. (*p<0.01; **p<0.001).
432

Hand-grip strength in children
RESULTS
group 3: 26.1 5.5 kg; females: group 1: 12.9 3.1 kg,
group 2: 17.3 2.2 kg, group 3: 22.9 4.8 kg]. A sig-
nificant difference in HG strength, in both d and nd
hand, was detected between males and females at
all ages (average difference -10%). For both genders,
nd hand was significantly weaker than d hand in the
older age groups, while no difference was detected
in younger children (group 1, 5-8 yr). Age- and gen-
der-dependent differences, but not those depend-
ing on dominance, disappear if HG strength is nor-
malized for FFM, as shown in Figure 2. Therefore, the
value of maximal HG strength normalized for FFM
was averaged across all ages and irrespective of gen-
der and resulted to be 0.67 0.11 kg/kg.
Figure 3 shows the average value of dHG strength
values observed in males (top panel) and females (bot-
tom panel) as a function of age (in yr). The ranges of
variability ( 1 SD and 2 SD) are also indicated in the
figure. For both genders, HG strength increased in a
non-linear fashion as a function of the age, and the
best fitting of the function based on d hand HG
strength was given by the equations: dHG=5.891
*100.051 age, r2=0.986, p<0.001 in males, and dHG=
About 15% of the recruited population was discarded
(30 males and 20 females), due to overweight (BMI
values higher than 2 SD for age and sex), while an-
other 1% (2 males and 3 females) was discarded for
excessive leanness (BMI lower than -2 SD for age and
sex). Therefore, analysis was performed on the remain-
ing 165 males and 113 females, for a total of 278 sub-
jects. Group characteristics are summarized in Table 1.
Weight, height, BSA, BMI, %BF and FFM increased
progressively and significantly from the younger to
the older age group (Table 1). A significant differ-
ence between genders was detected only for %BF
and FFM, females of all age groups having a high-
er fat mass and a lower FFM, compared to males.
Most children were right-handed (91%). Maximal HG
strength increased progressively and significantly
(p<0.001) in the 3 groups (Fig. 1), both in males and
in females [dominant (d) hand, males: group 1:
14.3 2.4 kg, group 2: 19.7 2.9 kg, group 3: 27.3 6.0
kg; females, group 1: 12.5 2.7 kg, group 2: 18.3 2.2
kg, group 3: 24.3 4.5 kg; non-dominant (nd) hand,
males: group 1: 14.7 2.4 kg, group 2: 18.8 3.1 kg,
Males
Females
40.0
35.0
30.0
25.0
20.0
15.0
10.0
5.0
Males
Females
1
0.8
0.6
0.4
0.2
0
**
*
*
0.0
Group 1
Group 2
Group 3
Group 1
Group 2
Group 3
40.0
35.0
30.0
25.0
20.0
15.0
10.0
5.0
1
°
*
*
** °
0.8
0.6
0.4
0.2
0
*
*
°
* °
*
0.0
Group 1
Group 2
Group 3
Group 1
Group 2
Group 3
Fig. 1 - Average values of maximal hand-grip (HG) strength de-
veloped with dominant (d, top panel) and non-dominant (nd, bot-
tom panel) hand by males (light bars) and females (dark bars) of
the 3 age groups. Asterisks indicate significant differences be-
tween genders (*p<0.05; **p<0.001); the open symbols indicate
significant differences between d and nd hand (°p<0.05).
Fig. 2 - Average values of maximal hand-grip (HG) strength
normalized by fat-free mass (FFM) unit developed with dom-
inant (d, top panel) and non-dominant (nd, bottom panel)
hand by males (light bars) and females (dark bars) of the 3
age groups.
433

A. Sartorio, C.L. Lafortuna, S. Pogliaghi, et al.
Fig. 4 - Relationship between maximal dominant hand-grip
(dHG) strength and fat-free mass (FFM) determined in 165 males
(closed circles) and 113 females (open circles). Equations de-
scribing the linear function are indicated for the 2 genders.
rent instruments (spherical vigorimeter, Jamar dyna-
moter, etc.), scarcely standardized measurement pro-
cedures (single trial, variable resting periods between
trials, etc.), lack and/or scarce anthropometric data.
An age-dependent increase of HG strength was re-
ported by Robertson et al. (6), Link et al. (8) in
young children (between 3-6 yr) and by Mathiowetz
et al. (7) in older children and adolescents (up to
19 yr). Although the majority of Authors reported
that males are usually stronger than females in all
age groups (1, 7), Link et al. (8) did not show sig-
nificant differences between the 2 genders.
Fig. 3 - Average values of maximal dominant hand-grip (dHG)
strength (heavy line) as a function of age, observed in males
(top panel) and females (bottom panel). Ranges of variability:
1SD (broken line) and 2SD (dotted line). Due to the relative
small number of females aged 15 yr, the corresponding data
for HG strength are not shown in the figure.
The results of the present study confirm the find-
ing of a progressive age-dependent increase of HG
strength in both sexes, a significant 10% difference
being evident between males and females at all
ages (for both d and nd hand). Since it has been
shown that both muscle fibre composition (12) and
ATP and phosphocreatine muscle concentration
(13) at the age of 6-8 are practically identical with
that of adults, it is likely that the observed HG
strength increase with age is largely dependent on
the parallel increase in muscle mass, although a rel-
ative immaturity in the contractile properties of the
muscles in the younger group might also play a
contributory role (14). Nonetheless, present data
indicate that the positive relationship between HG
strength and age is highly correlated with the age-
dependent increase of FFM, which occurs both in
males and in females. It is of interest to notice that
the differences of FFM values (about 10%) between
males and females (evident at all ages) were similar
to the gender differences of HG strength. This find-
ing appears to be in line with the results of the
study of Kanehisa et al. (15) in young subjects (aged
6.163 *100.045 age, r2=0.973, p<0.001 in females.
Around the age of 11, a marked increase in HG
strength occurs and it appears to be steeper in males
than in females.
A multiple regression analysis allowed to assess that
HG strength was positively correlated with BMI, BSA,
stature, stature2, and negatively correlated with %BF
(p<0.001 for all correlations) without differences be-
tween genders, but the strongest correlation was ob-
served with FFM. Figure 4 illustrates the positive cor-
relation obtained between dHG strength and FFM.
When the function obtained in males was compared
with that observed in females, no significant differ-
ence was found and the overall function for both gen-
ders was described by the equation: dHG strength=
2.32+0.63 FFM, r2=0.72, p<0.001.
DISCUSSION
Although several studies evaluated HG strength in
normal children, the general applicability of the ob-
tained values has been hampered by the use of diffe-
434

Hand-grip strength in children
REFERENCES
7-18), who report a similar increase in upper and
lower limb muscles maximal torque during isoki-
netic exercise as well as in the relevant muscle
cross-sectional area.
For both males and females, nd hand was signifi-
cantly weaker than d hand in the older age groups,
while no difference was detected in the 5-8 yr
group. This finding disagrees with the results by
Mathiowetz et al. (7), who reported similar HG
strength values of nd and d hand.
It is also of interest that gender differences, but not
d vs nd hand differences, disappeared if HG
strength is normalized for FFM. Thus it appears that
within the entire range of age investigated with the
present study, males and females can develop the
same muscle strength per unit FFM. Therefore the
differences in HG strength observed between the
genders can be considered as the result of the sig-
nificantly lower amount of FFM displayed by fe-
males, as also evidenced by the results presented in
Table 1.
1. Newman D.G., Pearn J., Barnes A., Young C.M., Kehoe
M., Newman J. Norms for hand grip strength. Arch. Dis.
Child. 1984, 59: 453-459.
2. Desrosiers J., Bravo G., Hebert R., Dutil E. Normative da-
ta for grip strength of elderly men and women. Am. J.
Occup. Ther. 1995, 49: 637-644.
3. Wessel J.A., Nelson R.C. Relationship between grip
strength and achievement in physical education among
college women. Res. Q. 1961, 32: 244-248.
4. Tinkle W.F., Montove H.J. Relationship between grip
strength and achievement in physical education among
college men. Res. Q. 1961, 32: 238-243.
5. Heyward V., McCleary L. Analysis of the static strength and
relative endurance of women athletes. Res. Q. 1975, 48:
703-710.
6. Robertson A., Deitz J. A description of grip strength in
preschool children. Am. J. Occup. Ther. 1988, 42: 647-
652.
7. Mathiowetz V., Wiemer D.M., Federman S.M. Grip and
pinch strength: norms for 6- to 19-years-olds. Am. J.
Occup. Ther. 1986, 40: 705-711.
As can be appreciated in Figure 3 (reporting the av-
erage curve with the variability ranges), a steeper in-
crease in HG strength occurs after 11 yr, both in males
(quantitatively greater) and females. It can be re-
marked that this pattern parallels that of urinary (u)
GH in normal children, as recently described by our
group (16). The strict temporal relationships between
u-GH levels and HG strength values, observed in both
sexes, confirms the gradual importance of GH (and/or
of the increased age-dependent physical training) in
determining adult muscle mass.
As far as normative data for HG strength is con-
cerned, it is possible that the results of the present
study might be slightly higher than those a clinician
would find when testing a child individually, be-
cause they were tested in small groups (which itself
could facilitate performances in simple motor tasks).
However, the small group setting can be consid-
ered the standard condition for these large scale
epidemiological investigations.
8. Link L., Lukens S., Bush M.A. Spherical grip strength in chil-
dren 3 to 6 years of age. Am. J. Occup. Ther. 1995, 49:
318-326.
9. Bell N.A., McClure P.D., Hill R.J., Davies P.S.W. Asses-
sment of foot-to-foot bioelectrical impedance analysis for
the prediction of total body water. Eur. J. Clin. Nutr. 1998,
52: 856-859.
10. Mathiowetz V., Weber K., Volland G., Kashman N.
Reliability and validity of grip and pinch strength evalua-
tion. J. Hand Surg. 1984, 9: 222-226.
11. Zar J.H. Biostatistical analysis, ed. 2. Prentice-Hall, Eagle-
wood Cliffs, 1984, pp. 292-305.
12. Bell R.D., Mac Dougall J.D., Billeter R., Howald H. Muscle fi-
bre types and morphometric analysis of skeletal muscle in
six-year-old children. Med. Sci. Sports Exerc. 1980, 12: 28-31.
13. Gariod L., Binzoni T., Ferretti G., Le Bas J.F., Reutenauer
H., Cerretelli P. Standardisation of 31 phosphorus-nuclear
magnetic resonance spectroscopy determination of high
energy phosphates in humans. Eur. J. Appl. Physiol. 1994,
68: 107-110.
In conclusion, the present study indicates that the
age-dependent increase of HG strength as well as
the between-gender differences are strongly relat-
ed to the age and gender-dependent variations in
FFM values occurring during childhood. Moreover,
the study provides a standard normative value of
maximal HG strength for the healthy children pop-
ulation in Northern Italy.
14. Ramsey J.A., Blimkie C.J.R., Smith K., Garner S., Mac
Dougall J.D., Sale D.G. Strength training effects in pre-
pubescent boys. Med. Sci. Sports Exerc. 1990, 22: 605-
614.
15. Kanehisa H., Ikegawa S., Tsunoda N., Fukunaga T. Strength
and cross-sectional areas of reciprocal muscle groups in the
upper arm and thigh during adolescence. Int. J. Sports Med.
1995, 16: 54-60.
16. Bona G., Petri A., Rapa A., Conti A., Sartorio A. on behalf
of the Italian Multicentre Study Group for “Urinary growth
hormone”. The impact of gender, puberty and body mass
on reference values for urinary growth hormone (GH) ex-
cretion in normally growing non-obese and obese chil-
dren. Clin. Endocrinol. (Oxf.) 1999, 50: 775-781.
ACKNOWLEDGEMENTS
The Authors wish to thank all the children and their families,
teachers and school headmasters, volunteers for their kind and
active cooperation.
435