j.jhsa.2008.01.008.pdf

8/10/2019 j.jhsa.2008.01.008.pdf

1/8

SCIENTIFIC ARTICLE

Grip Strength in Healthy Caucasian Adults:Reference Values

Christian M. Gnther, MD, Alexander Brger, MD, Markus Rickert, MD, Alexander Crispin, MPH,Christoph U. Schulz, MD

Purpose The aim of this study was to update reference data of handgrip strength for healthy adults of bothgenders spanning a wide age range and to analyze possible factors of inuence.Methods Intraindividual and interindividual variations of grip strength and their relation to severalanthropometric factors were analyzed in a standardized manner for 769 healthy adults (women, n 403; men, n

366) aged between 20 years and 95 years. Measurements were done in neutral position of arm, forearm, andwrist on setting II of a Baseline digital hydraulic dynamometer (NexGen Ergonomics Inc. Quebec, Canada).Results Mean strength was about 41% less in women (right 29 kg; left 27 kg) than in men (right 49 kg; left 47kg) resulting in a ratio of left to right hand slightly above .95 in both genders. During the course of life, hand strengthdevelops comparably in both genders peaking at 35 years of age and decreasing continuously further on.Anthropometric variables such as forearm circumference and length, hand size, or body mass showed a positive

correlation with grip strength. Body mass index, type of work, and hand dominance showed only a partial positivecorrelation or no correlation with grip strength. Gender and age, followed by parameters representing body lengthand obesity, were observed to have the highest predictive value for handgrip strength and were therefore enteredinto the generation of prediction equations.ConclusionsWe recommend side adjustment of measured values for intraindividual comparison andinclusion of information regarding anthropometric characteristics, as well as using gender- and age-adjustedreference values, whereas hand dominance can be neglected. The regression equations we generated mightprove to be useful for clinicians or for those who use normative values within software to provide moreaccurate predictions of strength scores for specic applications.( J Hand Surg 2008;33A:558565. Copyright 2008 by the American Society for Surgery of the Hand.)Key wordsHand strength, grip strength, normative values, anthropometric values, hand dominance.

GRIP STRENGTH IS A STANDARD parameter for theevaluation of hand function. Because measurementof grip strength with a hand dynamometer is

simple and inexpensive, this test is commonly used for outcome documentation after injuries of the upper extremities, 1,2 as a functional index of nutritional status, 3,4 toassess the efcacy of nutritional intervention in hospitalizedpatients, 5,6 as well as for impairment determinations.

In general the contralateral, uninjured hand serves ascontrol as suggested by Reikeras. 7 Even though thispragmatic management is helpful, it may be questioned froma more scientic point of view, as interpretation of data maybe difcult due to intraindividual and interindividualvariations of grip strength even in healthy individuals. Also,lately Massy-Westropp et al 8 found right- and left-handstrengths to be greatly different regardless of gender andhand-dominance. Crosby et al 9 reported right-handedindividuals to be stronger on their dominant side in contrastwith left-handed subjects, where mean grip was similar for

both hands. Of course, grip strength also may be inuencedby health status 10 and physical activity. 11 Furthermore, Jostyet al12 reported on variations of grip strength due todifferent types of work, and Chong et al 13 observed acorrelation between grip strength and anthropometric factorssuch as forearm circumference and body height and weightin healthy Chinese adults.

Last but not least, available data often are relatively old,such as the classic study of Mathiowetz et al 14 in 1985, andshould be proved again, as actual sociocultural changes andtechnical modernization might have inuence on gripstrength because of changes of nutritional status, increasing

From the Department of Orthopaedics, Klinikum Grosshadern,Ludwig-Maximilian-University Munich, Munich, Germany.Received for publication February 7, 2007; accepted inrevised form January 9, 2008.No benets in any form have been received or will bereceived from a commercial party related directly orindirectly to the subject of this article.Corresponding author: Christoph U. Schulz, MD,Department of Orthopaedics Klinikum Grosshadern,Ludwig-Maximilian-University Munich, Marchioninistr. 15,81377 Munich, Germany; e-mail: [email protected] .

0363-5023/08/33A04-0013$34.00/0doi:10.1016/j.jhsa.2008.01.008

558 ASSH Published by Elsevier, Inc.
mailto:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]

8/10/2019 j.jhsa.2008.01.008.pdf

2/8

8/10/2019 j.jhsa.2008.01.008.pdf

3/8

initially reduced anthropometric data (hand width, handlength, wrist circumference, forearm length, forearmcircumference, height, weight, body mass index) viaprincipal components analysis, extracting factors witheigenvalues 1. A rotated component matrix was producedusing the varimax rotation algorithm. For each rotatedcomponent, we identied the original variable with thehighest factor loading as lead variables to be included inthe linear regression models. Variable selection relied on ap-based forward selection algorithm. Separate models wereformulated for right-hand as well as left-hand grip strength.

Because of the exploratory character of the analyses, noadjustments for multiple testing were carried out (the testsshould not be mistaken as conrmatory). Raw p-values .05were considered to be signicant. SPSS version 14.0.1 (SPSSInc., Chicago, IL) served as the tool for statistical analysis.

RESULTSConsidering the writing hand as dominant, 93% of women(n 373) and 95% of men (n 346) were right-handed,and 7% of women (n 30) and 6% of men (n 20) wereleft-handed.

A distribution close to normal (Gaussian) was conrmedfor grip strength and all anthropometric measures (size of hand and forearm; weight).

Inuence of GenderGrip strength is signicantly less in women than in men (p

.001). Observed values ranged between 9 kg and 51 kg(right hand) and 7 kg and 45 kg (left hand) in women andbetween 15 kg and 85 kg (right hand) and 13 kg and 84 kg(left hand) in men.

Inuence of AgeIn both women and men, we observed an increase of gripstrength until a maximum about the age of 35 years. Further on, increasing age was inversely related with grip strength(Fig. 2 ).

Inuence of Side and HandednessGrip strength is signicantly higher in the right hand inwomen (p .001) and in men (p .001). Mean values inwomen are 29 7 kg for the right hand and 27 7 kg for the left hand, thus strength of the left hand averages 95% of the right hand. Mean values in men are 49 kg 11 for theright hand and 47 kg 10 for the left hand, thus strengthof the left hand averages 95% of the right hand.

Even though individuals showed slightly higher values on the dominant hand, univariate ANOVArevealed a signicant main effect difference with

FIGURE 2: A Relation between age and right-hand grip strength. B Relation between age and left-hand grip strength.

T ABLE 1: Grip Strength According to Gender, Side, and Handedness

Univariate Analysis of Variance

Left Side Right Side

Signicance (p) Partial Eta Squared Signicance (p) Partial Eta Squared

Gender .001 0.210 .001 0.208

Hand dominance .098 0.004 .585 0.000

Gender * hand dominance .402 0.001 .152 0.003

Corrected model .001 0.557 .001 0.571

p value: 2-tailed signicance.* , in combination with.

560 GRIP STRENGTH IN HEALTHY CAUCASIAN ADULTS

JHS Vol A , April

8/10/2019 j.jhsa.2008.01.008.pdf

4/8

gender (p .001) and no difference with handdominance (p .098) ( Table 1 ).

Inuence of Anthropometric MeasuresA positive correlation existed between grip strength andseveral anthropometric variables in women and men ( Table2). Indices for strength per body weight and strength per

hand size were found to differ signicantly between bothgenders ( Table 3 ).

Inuence of OccupationMarked differences regarding type of work were notdetectable for working men or for working women.Though we found retired individuals of both genders to

T ABLE 2: Correlations Between Grip Strength and Several Anthropometric Measures

Men (n 366) Women (n 403)

Right GripStrength

Left GripStrength

Right GripStrength

Left GripStrength

r p r p r p r pHeight 0.391 * .001 0.354 * .001 0.422 * .001 0.372 * .001

Weight 0.247 * .001 0.225 * .001 0.092 .064 0.079 .115

BMI 0.031 .557 0.032 .545 0.094 .061 0.083 .094

Right forearm length 0.294 * .001 0.450 * .001

Right forearm circumference 0.375 * .001 0.113 .023

Right hand width 0.306 * .001 0.258 * .001

Right hand length 0.350 * .001 0.420 * .001

Right wrist circumference 0.093 .075 0.060 .233

Left forearm length 0.245 * .001 0.428 * .001

Left forearm circumference 0.373 * .001 0.050 .319Left hand width 0.272 * .001 0.234 * .001

Left hand length 0.278 * .001 0.375 * .001

Left wrist circumference 0.116 .027 0.072 .148

r, Bravais-Pearson = s correlation coefcient; p-value, 2-tailed probability.* Correlation is signicant at the .01 level (2-tailed).

T ABLE 3: Grip Strength per Body Weight and per Hand Size

Gender Mean SD 2-Tailed Signicance (p)

R grip strength/weight M 0.606 0.136 .001F 0.446 0.129

L grip strength/weight M 0.575 0.134 .001

F 0.426 0.127

R grip strength/R hand length (kg/cm) M 2.545 0.518 .001

F 1.618 0.382

L grip strength/L hand length (kg/cm) M 2.408 0.519 .001

F 1.540 0.370

R grip strength/R hand width (kg/cm) M 5.191 1.060 .001

F 3.450 0.837

L grip strength/L hand width (kg/cm) M 4.956 1.056 .001

F 3.314 0.813

GRIP STRENGTH IN HEALTHY CAUCASIAN ADULTS 561

JHS Vol A , April

8/10/2019 j.jhsa.2008.01.008.pdf

5/8

have much lower grip strength values, this should beattributed to collinearity with age.

Regression EquationsFactor analysis of the anthropometric data showed 2components with eigenvalues above 1 explaining more than85% of the variance ( Tables 4 and 5). The rst onecorrelated best with height, followed by length of forearm

and hand, presenting obviously a substitute for body length.The second factor showed a close relation to body massindex (BMI), followed by weight and forearmcircumference, making it a parameter describing obesity.

Thus, we decided to enter BMI and body height intothe regression analysis, in addition to gender and age.Having observed best t for cubic regression of severalparameters earlier during bivariate analysis, we nallyentered into stepwise regression analysis the followingparameters: gender, age, age 2 , age 3 , height, height 2 , height 3 ,BMI, BMI 2 , and BMI 3 . For both right- and left-hand gripstrength, the stepwise variable selection algorithm of SPSSchose the following parameters to be entered into theregression equation: gender, age 3 , height 3 , and BMI. For theleft side, additionally age and age 2 for the right BMI 3 wereselected. For the complete regression equations, see Table 6 .

DISCUSSIONMeasurement of handgrip strength is a simple method toassess forearm and hand function or clinical aspects such asnutritional status. Even though many reports exist on gripstrength in normal and pathologic situations, 115,2024 mostlycohorts are small, age distribution is limited, or ethnicderivation or cultural differences are unclear. Furthermore,the large number of possible factors of inuence on gripstrength itself is problematic with respect to datainterpretation regarding the predictive values of each of these factors.

Therefore, the current study is based on a greatnumber of healthy adults of all ages and both genders.Because all were of similar ethnic (Caucasian)derivation, factors affecting hand function in other populations such as chopstick arthritis in Asian

T ABLE 4: Total Variance of Grip Strength Explained by Extracted Components

Total Variance Explained

Component

Right Hand Left Hand

Initial Eigenvalues Initial Eigenvalues

Total % of Variance Cumulative % Total % of Variance Cumulative %

1 5.331 66.642 66.642 5.360 67.000 67.000

2 1.493 18.664 85.306 1.510 18.871 85.871

3 .422 5.270 90.576 .433 5.416 91.286

4 .240 3.004 93.580 .221 2.763 94.049

5 .191 2.385 95.965 .175 2.191 96.240

6 .174 2.181 98.146 .166 2.075 98.315

7 .144 1.806 99.952 .131 1.637 99.952

8 .004 .048 100.000 .004 .048 100.000

Extraction method: principal component analysis.

T ABLE 5: Rotated Component Matrix Revealing Anthropometric Lead Variables

Component

Right Left

1 2 1 2

Hand width (cm) .733 .495 .715 .521

Hand length (cm) .891 .210 .899 .219

Wrist circumference(cm)

.615 .667 .593 .697

Forearm length (cm) .900 .187 .901 .191

Forearm circumference(cm)

.414 .810 .405 .825

Height (cm) .915 .139 .918 .142

Weight (kg) .442 .846 .437 .845

BMI .074 .968 .082 .963

Extraction method: principal component analysis. Rotation method:varimax with Kaiser normalization. Rotation converged in 3 iterations.


JHS Vol A , April

8/10/2019 j.jhsa.2008.01.008.pdf

6/8

populations 19 were excluded indirectly. On the other hand, a comparable study already w ith adults of Chinese derivation was available, 13 thus allowing adirect comparison especially of anthropometricmeasures. Even though and despite the fact thatGaussian distribution was veried for grip strength andeach of the measured anthropometric factors, theselection of participants might have inuenced our data. We therefore suggest using our results as areference rather than as normative for Caucasianpopulations ( Tables 7 and 8 ).

For data analysis, we used the mean of 3 consecutive

measurements; however, in contrast with Haidar , t hedynamometer generally was used with setting II. 18 Eventhough maximal grip strength may be underestimated under these cond itio ns, between 2% and 20% in about 10% of individuals, 17 it seemed best as a simple and reproducibleprocedure.

Grip strength increases in healthy individuals of bothgenders until the third decade of life and then starts todecrease continuously with further aging. This may beattributed to the decrease of mus cle diameter as well as toasymptomatic joint degeneration. 25 Women tend to peakearlier (rst half of fourth decade vs men in the second half),

T ABLE 6: Grip Strength Regression Equations

Side Adjusted R 2 Regression Equation

R 0.762 y 21.57 16.14 Female 3.13 10 5 Age3 3.15 10 6 Height 3 0.74BMI 2.32 10 4 BMI 3

L 0.752 y 13.90 16.36 Female 1.26 Age 2.32 10 2 Age2 1.01 10 4

Age3

2.35 106

Height3

0.16 BMI y, predicted grip strength; R 2 , amount of variance accounted for by the model; Age (years); Height (cm); Female 0 for male patient; Female 1for female patient.

T ABLE 7: Reference Values for Male Grip Strength

Age(y)

Right Grip Strength (kg) Left Grip Strength (kg)

Mean SD Minimum Maximum Mean SD Minimum Maximum

2029 53 8 36 70 51 8 29 65

3039 54 10 36 83 52 9 33 774049 54 7 34 70 52 8 28 70

5059 51 9 29 79 49 8 27 73

6069 45 7 32 63 43 7 29 65

7079 38 9 17 51 35 8 16 47

8095 31 8 16 44 28 7 18 42

T ABLE 8: Reference Values for Female Grip Strength

Age(y)

Right Grip Strength (kg) Left Grip Strength (kg)

Mean SD Minimum Maximum Mean SD Minimum Maximum

2029 32 5 19 44 30 5 16 42

3039 33 5 21 49 32 5 22 45

4049 32 6 19 46 30 5 19 44

5059 28 5 14 39 27 5 13 38

6069 26 5 10 40 25 5 11 36

7079 21 4 12 29 20 4 9 27

8095 16 4 10 27 15 4 9 25


JHS Vol A , April

8/10/2019 j.jhsa.2008.01.008.pdf

7/8

decline earlier, and seem to decrease at a slower rate thanmen as reported before by Hurley. 26 Because the relation of grip strength and age was similar in both genders, gender-specic changes of hormonal status seem unlikely toinuence grip strength in healthy individuals. Thus, for interpretation of grip strength, age is an indispensable factor.

Data conrm the general assumption that men are

stronger than women, who reach nearly 60% of absolutemale grip power. Additionally, we found men to havesignicantly higher grip strength when adjusted to weightand hand size ( Table 3 ). This strongly indicates gender-specic qualitative differences of muscle bers, although thismay depend, at least partially, on the extent of training 27,28

as well as differences between the proportions of fat, bone,and muscle as known for women and men.

In general, the right hand was stronger than the lefthand. Data further revealed higher levels of grip strength onthe dominant than on the nondominant side in both right-and left-handed individuals of both genders. However, as

differences were not signicant, hand dominance does notinuence grip strength. These results compare favorablywith a meta-analysis of Bohannon. 29

Current data conrm that, as similarly observed inother populations such as Indian or Chinese, 13,30 in aCaucasian population a positive correlation existsbetween several anthropometric aspects such as handwidth, hand length, forearm circumference and length,as well as body height and weight. However,correlations were shown to be relatively weakanalyzing the relationship between anthropometricmeasures and grip strength separately. This already

indicated grip strength to be inuenced by multiplefactors, which was conrmed during regressionanalysis.

In contrast with the general opinion, littledifference could be seen between blue-collar andwhite-collar workers, analyzing both gendersseparately. Signicant differences found betweenworking and retired populations are likely to be highlycorrelated with weakening during normal aging. Thedata therefore suggest that absolute grip strength doesnot depend on type of work; the actual lifestyle withincreasing focus on personal tness could be one

reason in this predominately urban population.Nevertheless, because absolute values reported byMathiowetz et al 14 more than 20 years ago for womenand men are nearly the same as the actual valuesobserved by us, this aspect requires further investigation.

Entering the patients values of the predictor variables(age, gender, height, and BMI) into the equations yieldsestimates of the individuals left and right grip strength.Therefore, the regression equations we generated mightprove to be useful for clinicians or for those who usenormative values within software to provide more accurate

predictions of normative strength scores for specicapplications.

The common method of intraindividual comparison isgood but requires side-adjustment and that interindividualcomparison requires information about gender andanthropometric variables such as body height or handdimensions.

REFERENCES1.Rosen B. Recovery of sensory and motor function after

nerve repair. A rationale for evaluation. J Hand Ther 1996;9:315327.

2.Chau N, Petry D, Bourgkard E, Huguenin P, Remy E,Andre JM. Comparison between estimates of hand volumeand hand strengths with sex and age with and withoutanthropometrics data in healthy working people. Eur JEpidemiol 1997;13:309 316.

3.Vaz M, Thangam S, Prabhu A, Shetty PS. Maximalvoluntary contraction as a functional indicator of adultchronic undernutrition. Br J Nutr 1996;76:9 15.

4.Jeejeebhoy KN. Nutritional assessment. Gastroenterol ClinNorth Am 1998;27:347369.

5.Wilson DO, Rogers RM, Sanders MH, Pennock BE, Reilly JJ. Nutritional intervention in malnourished patients withemphysema. Am Rev Respir Dis 1986;134:672677.

6.Efthimiou J, Fleming J, Gomes C, Spiro SG. The effect of supplementary oral nutrition in poorly nourished patientswith chronic obstructive pulmonary disease. Am Rev Respir Dis 1988;137:10751082.

7.Reikeras O. Bilateral differences of normal hand strength.Arch Orthop Trauma Surg 1983;101:223224.

8.Massy-Westropp N, Rankin W, Ahern M, Krishnan J,

Hearn TC. Measuring grip strength in normal adults:reference ranges and a comparison of electronic andhydraulic instruments. J Hand Surg 2004;29A:514519.

9.Crosby CA, Wehbe MA, Mawr B. Hand strength:normative values. J Hand Surg 1994;19A:665670.

10.Heimburger O, Qureshi AR, Blaner WS, Berglund L,Stenvinkel P. Hand-grip muscle strength, lean body mass,and plasma proteins as markers of nutritional status inpatients with chronic renal failure close to start of dialysistherapy. Am J Kidney Dis 2000;36:12131225.

11.Nevill AM, Holder RL. Modelling handgrip strength in thepresence of confounding variables: results from the AlliedDunbar National Fitness Survey. Ergonomics 2000;43:1547 1558.

12.Josty IC, Tyler MP, Shewell PC, Roberts AH. Grip andpinch strength variations in different types of workers.

J Hand Surg 1997;22B:266269.13.Chong CK, Tseng CH, Wong MK, Tai TY. Grip and

pinch strength in Chinese adults and their relationship withanthropometric factors. J Formosa Med Assoc 1994;93:616 621.

14.Mathiowetz MS, Kashman N, Volland G, Weber K, DoweM, Rogers S. Grip and pinch strength: normative data for adults. Arch Phys Med Rehabil 1985;66:6972.

15.Bohannon RW. Reference values for extremity muscle


JHS Vol A , April

8/10/2019 j.jhsa.2008.01.008.pdf

8/8

strength obtained by hand-held dynamometry from adultsaged 20 to 79 years. Arch Phys Med Rehabil1997;78:2632.

16.Trossman PB, Li PW. The effect of the duration of intertribal rest periods on isometric grip strength performancein young adults. Occupational Therapy Journal of Research1989;362378.

17.Firrell JC, Crain GM. Which setting of the dynamometer provides maximal grip strength? J Hand Surg 1996;21A:397 401.

18.Haidar SG, Kumar D, Bassi RS, Deshmukh SC. Averageversus maximum grip strength: which is more consistent?

J Hand Surg 2004;29B;1:8284.19.Hunter DJ, Zhang Y, Nevitt MC, Xu L, Niu J, Lui LY, et

al. Chopstick arthropathy: the Beijing Osteoarthritis Study.Arthritis Rheum 2004;50:14951500.

20.Burke WE, Tuttle WW, Tompson CW, Janney CD, Weber RJ. The relation of grip strength and grip-strengthendurance to age. J Appl Physiol 1953;5:628 630.

21.Hanten WP, Chen WY, Austin AA, Brooks RE, Carter

HC, Law CA, et al. Maximum grip strength in normalsubjects from 20 to 64 years of age. J Hand Ther 1999;12:193200.

22.Hinson M, Gench BE. The curvilinear relationship of gripstrength to age. Occup Ther J Res 1989;9:5360.

23.Merkies IS, Schmitz PI, Samijn JP, Meche FG, Toyka KV,

van Doorn PA. Assessing grip strength in healthy individualsand patients with immune-mediated polyneuropathies.Muscle Nerve 2000;23:13931401.

24.Montoye HJ, Lamphiear DE. Grip and arm strength in malesand females, age 10-69. Res Quart 1977;48:109120.

25.Chaisson CE, Zhang Y, Sharma L, Kannel W, Felson DT.Grip strength and the risk of developing radiographic hand

osteoarthritis: results from the Framingham Study. ArthritisRheum 1999;42:3338.

26.Hurley BF. Age, sex and muscular strength. J Gerontol ABiol Sci Med Sci 1995;50(Spec No):4144.

27.Wilmore JH. The application of science to sport:physiological proles of male and female atlete. Can J ApplSport Sci 1979;4:103115.

28.Simoneau JA, Lortie G, Boulay MR, Thibault MC,Thriault G, Bouchard C. Skeletal muscle histochemical andbiochemical characteristics in sedentary male and femalesubjects. Can J Physiol Pharmacol 1985;63:3035.

29.Bohannon RW. Grip strength: a summary of studies

comparing dominant and nondominant limb measurements.Percept Mot Skills 2003;96:728730.

30.Vaz M, Hunsberger S, Diffey B. Prediction equations for handgrip strength in healthy Indian male and female subjectsencompassing a wide age range. Ann Hum Biol 2002;29:131141.


JHS Vol A , April

Documents

j.jhsa.2008.01.008.pdf