REVIEW ARTICLE

Sports Med 2002; 32 (10): 615-631 0112-1642/02/0010-0615/$25.00/0 © Adis International Limited. All rights reserved.

Muscle Strength Testing Use of Normalisation for Body Size Slobodan Jaric1,2 1 Centre for Musculo-Skeletal Research, National Institute for Working Life, Umea, Sweden 2 Institute for Medical Research, Belgrade, Yugoslavia

Contents Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Muscle Strength Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Muscle Strength Normalisation . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 The Body-Size–Independent Index of Strength . . . . . . . . . . . . . . 2.2 Presenting Muscle Strength Data: The Normalisation Methods Applied 2.3 Distinction Between Muscle Force and Muscle Torque . . . . . . . . . 2.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Relationship to Functional Performance . . . . . . . . . . . . . . . . . . . . 3.1 The Strength Normalisation Applied . . . . . . . . . . . . . . . . . . . . 3.2 A Possible Role of the Functional Movement Performance . . . . . . . 3.3 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 Human Participants Are Not ‘Geometrically Similar’ . . . . . . . . . . . 4.2 Specific Populations: Adolescents and the Elderly . . . . . . . . . . . . 4.3 Other Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Abstract

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

615 616 617 617 619 621 623 623 623 624 626 626 626 627 627 628 628

Assessment of muscle strength tests has been a popular form of testing muscle function in sports and exercises, as well as in other movement-related sciences for several decades. Although the relationship between muscle strength and body size has attracted considerable attention from researchers, this relationship has been often either neglected or incorrectly taken into account when presenting the results from muscle strength tests. Two specific problems have been identified. First, most of the studies have presented strength data either non-normalised for body size, or normalised using inappropriate methods, or even several different normalisations have been applied on the same sets of data. Second, the role of body size in various movement performances has been neglected when functional movement performance was assessed by muscle strength. As a consequence, muscle function, athletic profiles, or functional movement performance assessed by tested muscle strength have been often confounded by the effect of body size. Differences in the normalisation methods applied also do not allow for comparison of the data obtained in different studies. Using the following allometric formula for obtaining index of muscle strength, S, independent of body size

616

Jaric

(assessed by body mass, m) should be recommended in routine strength testing procedures: Sn =

S mb

The allometric parameter should be either b = 0.67 for muscle force (recorded by a dynamometer), or b = 1 for muscle torque (recorded by an isokinetic apparatus). We also recommend using body-size–independent indices of both muscle strength and movement performance when assessing functional performance from recorded muscle strength or vice versa.

Tests of muscle strength have been extensively employed in sports, physical education, ergonomic and clinical practice. Their popularity has been predominantly based on their obvious validity for muscle-function assessment and relative simplicity, as well as on abundance of the relevant literature. However, the methods, as well as the reliability and external validity of strength testing have been often questioned.[1-3] A number of factors that affect the outcome of the applied strength tests have been studied. Some of these factors are related to the tested participants (e.g. differences in gender, age, physical activity or body composition), while others are predominantly related to various methodological aspects, such as type of contraction, starting position, stabilisation, gravity correction or test sequence.[2] Body size represents a well-known factor associated with the tested muscle strength. A number of different methodological procedures have been both studied and applied to account for the bodysize effect. The present review will deal with the problem of normalisation of muscle strength for body size applied within the contemporary literature. The main aims are: (i) to review the literature related to normalisation of muscle strength for body size; (ii) to identify potential flaws within the applied procedures; and (iii) to propose solutions to some of the identified problems of strength normalisation. The first section will deal with the purpose of muscle strength testing in general, as well as with the basic testing procedures applied. The following section will provide a review of the strength normalisation methods both suggested © Adis International Limited. All rights reserved.

and applied within the literature. The importance of the usually neglected distinction between the tested muscle force and muscle torque will be particularly stressed. The third section will focus on the external validity of strength tests when assessing functional movement performance. The importance of taking into account the relationship between the tested movement performance and body size to improve the external validity of muscle strength tests will be particularly stressed. The fourth section will deal with various confounding factors that may affect the relationship between muscle strength and body size. 1. Muscle Strength Testing Muscle strength has been defined as the maximum force (in N) or torque (in Nm) developed during maximal voluntary contraction under a given set of conditions.[4] To test muscle strength, a variety of approaches have been applied. Maximum muscle force has been usually measured by various kinds of dynamometers,[5-9] while maximum muscle torque has been recorded either directly by a standard ‘isokinetic apparatus’ (see Abernethy et al.[1] and Keating and Matyas[2] for review) or infrequently by calculating the torque from the recorded force and lever arm.[10,11] Muscle strength can be also recorded in different contraction regimens, such as the most often applied isometric, but also the concentric and eccentric contraction regimens. The later ones have been usually provided by standard isokinetic equipment that allows well-controlled mechanical conditions for contraction of the tested muscles. In addition to maximal Sports Med 2002; 32 (10)

Normalisation of Muscle Strength for Body Size

force and torque, some strength tests have also included ‘rate of force development’ that refers to the ability of the tested muscles to exert the force or torque in the shortest possible time.[1,3,6,8,11-14] Muscle strength tests have usually been performed under conditions that the recorded force or torque predominantly represents the results of action of a single muscle group. However, some tests based on contraction of several muscle groups of a particular kinetic chain have also been considered as muscle strength tests. Well known examples are the maximal force of leg extension,[13,15,16] or maximum weight lifted in some standard lifting tasks, such as bench-press, or dead lift,[14,17-19] or a force exerted at certain timepoints during athletic performance.[20-23] However, lifting maximal weights or manipulating heavy objects have also been considered as functional movement tasks important for sport- or ergonomic-related activities.[13,14,24,25] Therefore, the notion of strength tests will be generally restricted only to those based on contraction of a single muscle group. Strength testing has been extensively employed in a number of human-movement–related disciplines. The aim of athletic strength testing has been to provide normative values for particular sport disciplines (also called ‘athletic profiling’),[13,14,26-30] to select young athletes,[29,31] to distinguish among different performance levels,[29,31-33] or to evaluate the effects of physical exercise or athletic training procedures.[34-40] Ergonomic studies have often tested muscle strength to perform the ‘pre-employment selection’ for particular jobs.[41,42] In medicalrelated fields, strength has been tested to assess muscle function[1,43,44] and provide normative values for healthy populations,[45,46] to evaluate results from surgical or therapeutic procedures,[7,47] or to estimate the risk of injuries or health problems.[48-50] Finally, an important purpose of muscle strength testing common for athletic, ergonomic and medical-related studies has also been the assessment of functional movement performance.[24,25,50-53] The relationship between the strength of active muscle groups and selected movement performance has © Adis International Limited. All rights reserved.

617

often been interpreted as external validity of muscle strength tests.[1,3,44,54] The present review will focus on the normalisation of muscle strength for body size that could serve two important purposes of strength testing. The first one is to provide the body-size–independent indices of muscle strength that can serve as a valid assessment of muscle function. Therefore, these indices could be used as normative values for particular populations or discriminate among different groups of individuals (e.g. athletes of different specialisations or different levels of proficiency, individuals of different age and gender). The second purpose is to assess indices of strength that could provide the strongest possible relationship with performance of movements important for sport success, as well as for everyday life. 2. Muscle Strength Normalisation 2.1 The Body-Size–Independent Index of Strength

The effect of body size on various movement performances, as well as on various physiological functions in general has been studied for decades.[55-58] Therefore, in addition to a number of other factors that could affect the outcome of muscle strength tests (for review see Keating and Matyas[2] and Wilson and Murphy[3]), the effect of body size has attracted considerable interest from both researchers and professionals from various movement-related areas. Even the basic athletic, clinical, or everyday life experience suggests that taller or heavier individuals are usually stronger than the shorter and lighter ones. This effect becomes particularly prominent when animals of similar body stature, but very different sizes are compared (e.g. small and big cats).[56] It has been suggested that as the range of size for individuals increases, the strength of the relationship between strength and body size is likely to increase.[59] Therefore, it seems conceivable that the moderate relationship between the tested muscle strength and body size observed in a number of studSports Med 2002; 32 (10)

618

ies[28,45,60,61] is partly caused by a relatively narrow range of human body sizes. From the theoretical prospective, to assess the role of body size in outcome of muscle strength tests, the ‘effects of scale’ on muscle force have been analysed.[55,56,62] According to the most often used presumption usually called ‘geometric similarity’, but also ‘biological similarity’ or ‘isometric scaling’ (i.e. all human bodies have the same shape, so they only differ in size),[56,63,64] muscle force should be proportional either to body height squared (H2) or to body mass to power two-thirds (m2/3). Since the leverage of human limbs should not depend of body dimensions, muscle force increases exclusively because of an increase in muscle cross-sectional area. However, any area of geometrically similar objects is a priori proportional to body linear dimensions squared, or proportional to body volume (or mass) to power two-thirds. This finding, obtained on a hypothetical body muscle was, thereafter, extended to muscle strength tests in general. Therefore, to obtain an index of muscle strength independent of body size, the recorded strength should be divided by any body length (e.g. body height) squared, or any specific area (e.g. cross-section or physiological section area of muscle), or any body-volume–related index (e.g. body mass, bodyweight, lean body mass, volume of any body segment) to power two-thirds. It is suggested (section 2.3) that the discussed approach could be valid for a tested muscle force, but not for a tested muscle torque. In addition to the aforementioned results from theoretical analysis, most of the experimental studies have also suggested that muscle strength increases at a lower rate than body mass (see table I). The experimental approach aimed to reveal that the muscle strength index independent of body size has been usually based on allometric modelling.[14,19,56,58,65] In short, the assumed relation between muscle strength S (measured as force or torque) and body mass m (as the most often employed index of body size) was (equation 1): S = a • mb © Adis International Limited. All rights reserved.

Jaric

where b is the allometric parameter, while further considerations will show that the other equation parameter a corresponds to the normalised strength. Therefore, the log-transformation of equation 1 provides a regression model where log a and b are the intercept and slope, respectively (equation 2): log S = log a + b log m The final result gives the normalised strength Sn, which represents the body-size–independent index of strength for a particular participant population (equation 3): Sn =

S mb

Some studies have applied different methods, such as multivariate allometric modelling that includes other variables in addition to the index of body size.[61,67] However, most of them have had the same main goal of assessing the value of the allometric parameter b that enables calculation of the body-size–independent index of muscle strength according to equation 3. Table I depicts the results from a number of studies aimed to assess the optimal normalisation method that provides a body-size–independent index of muscle strength. One should note that for the sake of further consideration within section 3, some assessments of force exerted by several muscle groups are also included. Furthermore, most of the available data have been obtained on the force and torque of lower body muscles. A close inspection of the table suggests that the allometric approach has been the most commonly employed, while the obtained values of the allometric parameter b were mainly below one. Most of the authors agree, at least implicitly, that the discussed method based on the allometric approach (i.e. equation 3) could be used for normalisation of muscle strength for body size. Issues related to intercepts (see equation 2), dimensionless power functions,[65] using residuals between predicted and recorded results,[68] as well as possible advantages of some more complex normalisation approaches (see McMahon[56] for review) are beyond the scope of Sports Med 2002; 32 (10)

Normalisation of Muscle Strength for Body Size

619

Table I. Summary of research investigating the allometric parameter that provides a body-size–independent index of strength Study

Participants

Strength test

Method applied

Allometric coefficient

Comments

Batterham and George[65]

Men and women weightlifters (World Championship)

Weight lifted, biathlon results

Allometry

b = 0.45-0.48 (b = 0.68 when excluding the heaviest participants)

Obtained residuals suggest using more complex method

Challis[63]

Olympic records and power-lifting records

Weight lifted

Allometry

b = 0.64-0.65

Davies and Dalsky[64]

Men and women, age 65-78y (n = 148)

Torque of knee ext Allometry

b = 0.74

Bone-free lean tissue mass recommended for normalisation

Dooman and Power-lifting record holders Weight lifted Vanderburgh[17]

Allometry

b = 0.57 (bench-press), b = 0.60 (squat), NS relationship for dead-lift

Jaric et al.[28]

13 groups of young and adult athletes (n = 401)

Force of knee ext, hip flx and hip ext

Allometry

b = 0.67 provides better fit than b = 0 or b = 1

Adolescent athletes demonstrate higher b

Jaric et al.[5]

Men (n = 16)

Force and torque Allometry of 6 muscle groups

b = 0.67 (force), b = 1.02 (torque)

b higher for torques than for forces

Neder et al.[66]

Men and women (n = 61)

Torque of knee ext Allometry

b = 0.91-1.10

Nevill et al.[67]

Boys and girls, age groups Torque of knee ext Multilevel 11-16y (n = 453) and elbow flx regression analysis

b = 0.36-0.38 for m, b = 1.06-1.23 for height

Corrected for age and gender

Vanderburgh Female power-lifting and Dooman[19] records

Weight lifted

Allometry

b = 0.63-0.87

Vanderburgh et al.[58]

College men and women (n = 205)

Force of handgrip

Allometry

b = 0.54 (men), b = 0.48 (women)

Weir et al.[61]

Young club and high school wrestlers (n = 258)

Torque of knee ext Multivariate and knee flx allometry (age and FFM)

b = 0.94-1.31

Corrected for age

Wisloff et al.[14]

Elite and subelite male soccer players (n = 29)

Weight lifted (squat and bench press)

b = 0.55-0.74

b significantly lower than 1

Allometry

b = allometric coefficient; ext = extensors; FFM = fat-free mass; flx = flexors; m = body mass; NS = not significant.

this study. The importance of the distinction between the measured force and torque will be stressed within the following sections. However, the possible causes for wide scatter of the obtained allometric parameter b will be discussed within section 4. 2.2 Presenting Muscle Strength Data: The Normalisation Methods Applied

The primary goal of strength testing has often been to assess the objective value of muscle function independent of possible confounding factors. This value should discriminate among different groups of individuals and, therefore, it could contribute in profiling of certain groups of athletes, serve as a basis for talent identification, or provide normative strength values for various participant © Adis International Limited. All rights reserved.

populations. Thus, among the other participants and test-related factors that can affect the recorded strength (for review see Keating and Matyas,[2] and Wilson and Murphy[3]), the results from muscle strength tests should be normalised for body size. Otherwise, a part of the differences obtained between the strength recordings of individuals that usually demonstrate prominent differences in body size (e.g. younger and older adolescent athletes, or male and female participants, or particular groups of athletes) could be considerably confounded by the body-size effect. Although the effect of body size has been known for a long time (see section 1), a review of the literature reveals that the normalisation of the tested muscle strength has been inconsistently applied (see table II). A minority of studies normalised Sports Med 2002; 32 (10)

620

Jaric

Table II. Summary of research using muscle strength to provide normative values and/or to discriminate between different groups of individuals Study

Participants

Strength test

Normalisation applied

Abe et al.[71]

Female skiers, top and college level (n = 19)

Torque of knee flx Non-N, per CSA S (for both non-N and ext and per CSA)

Agre et al.[26]

Hockey players: goalies, defenders, forwards (n = 27)

Force and torque of various leg muscles

Non-N, per m

Akima et al.[43]

Men and women, age groups 20-84y (n = 164)

Torque knee flx and ext

Non-N, per CSA S (for non-N, between men also per CSA), NS (between women per CSA)

Andersson et al.[27]

Male and female athletes Torque of various trunk and hip (soccer, wrestling, muscles gymnastics, tennis) and nonathletes (n = 147)

Per m

S (between athletes and non-athletes, males and females)

Gymnasts strongest among the athletes

Bale[73]

Female basketball players (n = 18)

Force of handgrip

Non-N

S (between different positions in team)

m = 52-78kg

Clarke et al.[74]

Physically active and Force of elbow flx inactive men, age groups 20-50y (n = 62)

Non-N

S (both between activity levels and age groups)

m = 75-85kg, %F = 10-26%

Cometti et al.[32]

Male soccer players: elite, subelite, amateur (n = 95)

Torque of knee ext and knee flx

Non-N

S

De Ste Croix et al.[76]

Boys and girls, age 8-9 and 13-14y (n = 141)

Torque of knee flx Non-N, per m, and ext ANCOVA

S

Faria and Faria[31]

Male junior gymnasts: I and II level (n = 65)

Force of handgrip

NS

Froese and Houston[75]

Male and female students (n = 30)

Torque of knee ext Non-N

S

m = 60-73kg

Frontera et al.[77]

Men and women, age groups 45-78y (n = 200)

Torque of knee ext Non-N, per muscle mass, per FFM

S

Differences smaller after normalisation

Fry and Morton[33]

Male kayakists, I and II level (n = 38)

Torque of shoulder-arm complex

Non-N

S

m = 71-81kg

Hakkinen et al.[72]

Male power-lifters, bodybuilders and wrestlers (n = 14)

Force knee ext

Per m

NS

Hulens et al.[78]

Obese and lean women (n = 253)

Force and torque of different muscle groups

Non-N, per FFM S

Izquierdo et al.[79]

Men, age groups 42 and 65y (n = 47)

Torque of knee ext Non-N, per m, per CSA

Jaric et al.[69]

Elite male basketball, Force of hip flx, handball, soccer and hip ext and knee volleyball players (n = 68) ext

Paasuke et al.[80]

Women, age groups 30-80y (n = 63)

Paasuke et al.[12]

Pfeifer and Banzer[7]

Non-N

Per m2/3

Discrimination

NS (for both non-N and per m)

Comments

m = 77-89kg, %F = 7.7-12.2%

Differences smaller after normalisation

Normalisation affected difference

S (for non-N, per m), NS (per CSA) S

Soccer players strongest

Force of plantar flx Non-N

S

m = 60-68kg

Male Nordic combined athletes and untrained controls (n = 21)

Force and torque of knee ext

Non-N, per m

S

NS correlation between force and torque

Patients after ACL and healthy controls (n = 59)

Force of knee ext

Non-N

S

m not reported Continued over page

© Adis International Limited. All rights reserved.

Sports Med 2002; 32 (10)

Normalisation of Muscle Strength for Body Size

621

Table II. Contd Study

Participants

Strength test

Normalisation applied

Discrimination

Comments

Sipila and Suominen[10] Older women, nonathletes and athletes (n = 94)

Force and torque of knee ext

Non-N, per m, per CSA

S

Sleivert et al.[13]

Male athletes (volleyball, middle-distance runners) and controls (n = 30)

Torque of knee ext and plantar flx

Non-N

S

m = 66.5-81kg (volleyball players strongest)

Sunnegardh et al.[70]

School boys and girls, age groups 8 and 13y (n = 131)

Torque of several groups, force of handgrip

Non-N, per height2, per m, per FFM, per CSA

S

Strength per m and per FFM can be higher for younger group

Takala and Viikari-Juntura[50]

Patients and controls (n = 430)

Force of trunk flx and ext

Non-N

S

m not reported

Taylor et al.[30]

Male elite power and endurance athletes (n = 13)

Torque of leg ext muscles

Per m

S

Viljanen et al.[11]

Men and women, age groups 25-55y (n = 778)

Torque of trunk flx Per m and ext

NS (in trunk flx)

ACL = anterior cruciate ligament surgery; ANCOVA = analysis of covariance; CSA = muscle cross-sectional area; ext = extensors; FFM = fat-free mass; flx = flexors; m = body mass; non-N = non-normalised; NS = not significant; S = significant; %F = percentage fat.

the recorded strength per m2/3, but also per unit of muscle cross-sectional area, or per H2.[46,69-71] More often researchers applied so-called ratio standards by normalising recorded strength per kilogram of body mass, or fat-free mass.[11,27,30,72] Surprisingly, most of the studies did not apply normalisation for body size at all,[7,31,32,50] even when the compared samples of participants demonstrated a prominent difference in body mass.[12,33,73-75] Finally, several different normalisation methods have often been applied on the same sets of muscle strength recording.[10,12,26,43,70,71,76-79] The reviewed inconsistencies in muscle strength normalisation cause various problems within the professional literature related to muscle strength testing. An important problem is that most of the authors do not present the body-size–independent indices of muscle strength, either because of a lack of normalisation or because an inappropriate normalisation was applied. Therefore, although nonnormalised strength could be of importance for some athletic performance (e.g. shot put), these results do not allow for comparisons of the findings obtained in different studies. While the lack of strength normalisation in studies aimed to distin© Adis International Limited. All rights reserved.

guish among the individuals of similar body size could be accepted,[32] the same approach applied in studies performed on individuals of quite different body size could lead to wrong conclusions. For example, the data presented within table II suggest that the significant differences in muscle strength obtained between different groups of basketball players,[73] or between different groups of athletes[13] could become insignificant if a proper normalisation for body mass was applied. Some studies do not report any index of body size for the tested groups of individuals.[7,50] Finally, several normalisation procedures applied on the same set of strength recordings inevitably lead to a redundant set of results, which is difficult either to present, or to interpret, or to compare with results from other studies. 2.3 Distinction Between Muscle Force and Muscle Torque

An important source of problems in presenting muscle strength data has been the recently stressed distinction between the methods for normalisation of the recorded muscle force and muscle torque.[5] The data depicted in table II clearly demonstrate Sports Med 2002; 32 (10)

622

Jaric

that the applied methods of normalisation for body size remained similar when either muscle force or muscle torque were tested. Moreover, the same methods of normalisation were applied even when the muscle force and torque measures were obtained within the same studies.[12,26,78] The distinction between the measured muscle force and muscle torque when related to body size can be illustrated by a simple mechanical model of a hypothetical muscle strength test. Figure 1 illustrates the action of a muscle (shaded area) of a particular participant while exerting either force F (measured by a force transducer FT) or torque T (measured by an isokinetic apparatus; not depicted). The lever arm of the muscle force Fm with respect to the centre of the joint O is p, while the lever arm of the measured force F with respect to the same joint is q. The force recorded by the force transducer is (equation 4): F=

p •F q m

where p/q represents leverage of the muscle force. If all tested individuals are geometrically similar, the leverage does not depend on body size since both lever arms change proportionately. Since muscle force depends on the muscle cross-sectional (or, alternatively, physiological-sectional) area, it should be proportional to m2/3. According to equation 4, the same should be true for measured force, suggesting that measured muscle force should be proportional to m2/3. As a consequence, results from the muscle strength tests based on the recorded force should use the allometric parameter b = 2/3 = 0.67 when applying equation 3 to normalise strength for body size (see also Wisloff et al.[14]). However, if an isokinetic apparatus is used to test muscle strength, the recorded torque T is equal to the muscle torque (equation 5): T = Fm • p

Under the presumption of geometric similarity, muscle force is proportional to m2/3 (see the previous paragraph), while the lever arm q, as any other length, should be proportional to m1/3. Their product gives (equation 6): © Adis International Limited. All rights reserved.

Fm

F

O p q

FT

Fig. 1. A schematic illustration of testing muscle strength. F = measured force (by a force transducer); Fm = muscle force; FT = force transducer; O = elbow joint; p = muscle force moment arm; q = measured force moment arm.

T µ m 2 / 3 • m1 / 3 = m

As a consequence, the results from muscle strength tests based on the recorded torque should apply the allometric parameter b = 1 to normalise strength for body size. Despite the wide scatter of the available data, the literature review also suggests that the value of the allometric parameter b could be, in general, higher for tested torques than for tested forces. When studying optimal normalisation of muscle strength for body size (see equation 3), Vanderburgh et al.[58] suggested b = 0.48 to 0.54 for handgrip force, while Jaric et al.[5,28] found the value of the allometric parameter close to the suggested b = 0.67 for the tested force of several leg and arm muscles. Studies based on various weightlifting techniques (see table I for details) have also suggested the value of the allometric parameter for muscle forces exerted against heavy weights within the rage 0.45 < b < 0.87. However, when muscle torque was tested, Davies and Dalsky[64] suggested b = 0.74, Jaric et al.[5] found b = 1.02, Sports Med 2002; 32 (10)

Normalisation of Muscle Strength for Body Size

Neder et al.[66] found b = 0.91 to 1.10, while Weir et al.[61] suggested b = 0.94 to 1.31. Finally, when recording various muscle forces and torques, Jaric et al.[5] found higher values of the allometric parameter for torques than for forces in each of six tested muscles. The only exception was the findings obtained for adolescents by Nevill et al.[67] that suggested an exceptionally low value of the allometric parameter for tested muscle torques (b = 0.36 to 0.38). However, a confounding effect of maturation that could affect the obtained result will be discussed in section 4.2. Therefore, based on both the theoretical analysis and the limited experimental data, the allometric parameters b = 0.67 or b = 1 are prone to provide a size-independent index for muscle force and torque, respectively. 2.4 Summary

Body size has been recognised as an important factor that affects muscle strength. Most of the studies aimed to investigate the relationship between muscle strength and body size have suggested using the normalisation method based on the allometric approach shown in equation 3 (S is muscle strength recorded as either muscle force or muscle torque, m is body mass, b is the allometric parameter) to assess index of strength independent of body size. Surprisingly, a review of the professional literature revealed that most of the studies have presented either non-normalised strength, or strength normalised in an inappropriate way, while numerous studies have even applied several different normalisations on the same sets of strength data. As a consequence, a considerable part of the presented data provided ‘strength profiles’ of various participant populations or assessments of muscle function confounded by the effect of body size. Using b = 0.67 and b = 1 for normalisation of the recorded force and torque, respectively, is recommended in routine muscle strength tests. © Adis International Limited. All rights reserved.

623

3. Relationship to Functional Performance

3.1 The Strength Normalisation Applied

An important purpose of muscle strength testing within the sport, ergonomic- and medical-related literature has been to assess functional movement performance. The relationship between muscle strength and functional performance has often been interpreted as external validity of muscle strength tests,[2,44] while the nature of the relationship has been frequently discussed.[1,3,6,15,81] Table III depicts a summary of studies that tested muscle strength to assess different functional performances. Excluding very few examples that provided an exceptionally strong relationship,[12,82-84] most of the studies suggested either moderate or even no relation between muscle strength and tested functional performance.[9,15,28,33,49,85-89] Similar findings have been observed in ergonomic- and physiotherapeutic-related literature.[50] Therefore, one could conclude that muscle strength measured as force or torque of active muscle groups is a relatively weak predictor of functional movement performance. It has been already suggested that assessment of performance in homogeneous groups of elite athletes, in particular, could require a more sport-specific approach.[3,20,40,88,89] Inspection of the methods applied to normalise muscle strength for body size suggests the existence of the same methodological problems that have already been identified within section 2.2 (see also table II). In short, different normalisation methods have been arbitrarily applied on either muscle force or muscle torque tested. The nonnormalised data have been more often presented than the normalised, while several studies have also applied several normalisation methods on the same sets of strength recordings.[13,28,82,83] Similar problems have been pointed out within the ergonomic and medical-related strength testing.[40,90] Sports Med 2002; 32 (10)

624

Jaric

Table III. Summary of research using muscle strength to relate to functional movement performance Reference

Participants

Strength test

Normalisation applied

Augustsson and Thomee[85]

Men (n = 16)

Torque of knee ext Non-N

Birch et al.[86]

Male students (n = 31)

Force of hip ext

Blackburn and Morrissey[15]

Women (n = 20)

Cometti et al.[32]

Male soccer players (elite, subelite, amateur; n = 95)

Torque of knee ext and knee flx

Fry and Morton[33]

Male kayakists, I and II level Torque of (n = 38) shoulder-arm complex

Hakkinen[82]

Male and female basketball players (n = 20)

Jaric et al.[28]

13 groups of adolescent and Force of knee ext, Non-N, adult athletes (n = 401) hip flx and hip ext per m2/3, per m

Jaric et al.[69]

Elite male basketball, handball, soccer and volleyball players (n = 68)

Force of hip flx, hip ext and knee ext

Jaric et al.[87]

Male students (n = 39)

Kukolj et al.[88]

Functional Correlation (r) performance assessed Lifting weights, VJ

0.51-0.57

Non-N

Lifting weight, SLJ, VJ

0.43-0.62

Torque of knee ext Non-N

Lifting weight, VJ, SLJ

0.07-0.10

Non-N

Sprint time (10, 30 metres), VJ, ball kick

NS

Non-N

Kayaking time

From NS up to 0.69

VJ

0.80-0.81

VJ, SLJ

–0.03-0.48 (per m2/3 and per m provide stronger relationship than non-N)

Per m2/3

VJ

–0.29-0.40

Force of hip ext, knee ext and plantar ext

Non-N

VJ

0.22-0.42

Male university students (n = 24)

Force of knee ext and hip ext

Per m

Sprint time (0-15, 15-30 metres)

0.12-0.22

Magnusson et al.[49]

Men and women, former swimmers (n = 24)

Torque of various arm and leg muscles

Per m

Swimming performance

0.45-0.52

Ostenberg et al.[89]

Female soccer players (n = 101)

Torque of knee ext Corrected for m, Various jumps H and age

–0.35-0.46

Paasuke et al.[12]

Male Nordic combined athletes and untrained controls (n = 21)

Torque of knee ext Per m

VJ

0.70-0.82

Sleivert et al.[13]

Male athletes (volleyball, middle-distance runners) and controls (n = 30)

Torque of knee ext and plantar flx

Leg press, leg power

Up to 0.83 for non-N, NS for normalised

Suei et al.[83]

Boys, age 9.7-17y (n = 122)

Torque of knee ext Non-N, per m, per FFM

Sargent test, Wingate test

0.82-0.92 (non-N), 0.00-0.70 (per m, per FFM)

Ugarkovic et al.[9]

Elite junior male basketball players (n = 33)

Force of knee ext and hip ext

Per m

VJ

0.38-0.52

Wiklander and Lysholm[84]

Male and female runners (n = 184)

Torque of knee ext and knee flx

Non-N

Various jumps

0.61-0.84

Force of knee ext, Non-N, per m hip flx and hip ext

Non-N, per m

ext = extensors; FFM = fat-free mass; flx = flexors; H = body height; m = body mass; non-N = non-normalised; NS = not significant; SLJ = standing long jump; VJ = vertical jump.

3.2 A Possible Role of the Functional Movement Performance

Since the relationship between muscle strength and body size represents a well documented phe© Adis International Limited. All rights reserved.

nomenon, one could conclude that just the application of the proper normalisation methods recommended within the previous sections should provide not only a reliable index of muscle strength, but also the strongest possible strength-performance Sports Med 2002; 32 (10)

Normalisation of Muscle Strength for Body Size

relationship independent of the confounding effect of body size. However, the possibility remains that functional movement performance could also be related to body size. Moreover, different types of movement performance could be differently related to body size. These phenomena could generally lead to poor correlations between muscle strength and functional movement performance. It appears that some authors have been aware of the problem and, as a consequence, have made an effort to normalise various functional performance tests for body size (see table III). For example, some authors employed several normalisations of the results obtained in both muscle strength and functional performance tests,[13,31,83] others applied several different normalisations of the same results from functional performance tests,[14,91] while some authors presented both normalised and non-normalised indices of movement performance.[13,83] However, most of the authors have presented only nonnormalised results from functional performance tests. Although this is not among the main aims of the present study, the following paragraphs will briefly describe several groups of functional movement tasks that could require different normalisations for body size. The performance of a number of typical functional tests is based on exerting a force against external objects. The examples could be either different kinds of weightlifting often applied in athletic or physical education testing,[21,79,92] or two-hand lift or manual material handling applied in ergonomic studies.[93] Virtually all studies have demonstrated that the recorded performances are positively related to body size. Therefore, the body-size–independent index of performance based on exerting force against external objects should be obtained using a similar approach as indices of muscle strength (see equation 3 and section 2.3). The obtained allometric parameters have usually been close to the theoretically predicted values for muscle force b = 2/3 = 0.67 (see table I), suggesting that either the weight lifted or force exerted should be divided by body mass or any other proportional index to the power of two-thirds. © Adis International Limited. All rights reserved.

625

Another group of movement performance tasks consists of those based on maximum speed of either body centre of mass (e.g. sprinting and jumping),[28,32,88,89] or particular body segments (e.g. throwing, kicking and tennis serve).[33,52,94] Although some complex scaling methods have suggested a weak relationship between the performance of the aforementioned movements and body size, allometric modelling based on geometric similarity suggests no relationship.[56,62] Athletic experience also suggests no relationship between maximum movement velocity and body size. For example, the fastest running, or the longest jump, or the fastest tennis or volleyball serve, are expected neither from the smallest nor from the biggest athletes. Therefore, the performance of rapid body movements is not likely to require normalisation for body size. Finally, numerous functional tests are based either on keeping some strength demanding postures (e.g. enduring positions in official gymnastic competitions, or in ergonomic or physical-fitness tests),[24,39,92] or on overcoming the weight of the individual’s own body under difficult conditions (e.g. squats, pull-ups, push-ups, sit-ups and oneleg lift).[39,95] Since muscle forces are likely to increase at a slower rate than bodyweight (see section 2.3), it is reasonable to assume that the performance of such tasks decreases with an increase in body size. Interestingly, the discussed phenomenon had been noticed centuries ago by Galileo Galilei (see McMahon,[56] page 234) on a much wider scale of animal body sizes, but was also supported by the relatively small stature of elite gymnasts and acrobats.[27,31,96] Therefore, the performance based on enduring some strength demanding postures or performing movement tasks under difficult conditions against one’s own bodyweight may require a negative value of the allometric parameter (see equation 3) to assess a bodysize–independent functional performance. A number of findings in previous studies support the results from the presented qualitative analysis. The review of Hogan[97] suggested that the tests based on force exertion (maximal push, pull, Sports Med 2002; 32 (10)

626

dynamometric forces) and the tests based on body movement tasks (leg lifts, push-ups, squats) belong to ‘independent abilities’ (i.e. the correlations between them are low, particularly when compared with the correlations obtained among the tests belonging to the ‘same abilities’). The classical study of Fleishman[98] also suggested that ‘static strength’ (exerting forces), ‘explosive strength’ (running) and ‘dynamic strength’ (push-ups) belong to independent movement abilities. Several non-normalised functional tests of exerting force or power demonstrated differences among the tested groups of individuals, but the differences became insignificant after being normalised for body mass (suggesting a close association between body size and outcome of the applied tests).[13,99] Finally, lighter individuals could demonstrate worse results in functional tests based on exerting external force (bench-press, hipsled, medicine ball throw, two-hand lift), but better results when overcoming their own weight (pullups, sit-ups, keeping horizontally extended back) than the heavier individuals.[92,100] Therefore, it could be concluded that not only the scaling models, but also the experimental results support the concept of using the body-size–independent indices of both muscle strength and movement performance to assess the strength-performance relationship. It seems that the role of body size in movement performance has been particularly neglected and, therefore, could require special attention in future research. Specifically, an extensive and reliable classification of functional movement tasks from the perspective of the role of body size should be developed to relate these tasks with the tested muscle strength. Assessment of the appropriate values of the allometric parameter b for various groups of movement performance tasks could deserve the attention of future studies. 3.3 Summary

One of the most important purposes of strength testing is to assess functional movement performance. However, as with presenting strength data, researchers have often neglected the body-size effect on the tested muscle strength. In addition, the © Adis International Limited. All rights reserved.

Jaric

possible effect of body size on the assessed functional performance has also been neglected. As a result, a number of assessments of functional movement performance based on muscle strength tests have been confounded by the effect of body size. Applying body-size–independent indices of both muscle strength and movement performance to assess the relationship between muscle strength and functional movement performance should be recommended. 4. Limitations 4.1 Human Participants Are Not ‘Geometrically Similar’

Since the normalisation of muscle strength for body size has been arbitrarily applied through the professional literature, potentially the most important outcome of the present review could be the normalisation methods proposed for use in future studies. However, most of the arguments for and against the particular normalisation methods in both the present and previous studies have been based on the allometric modelling involving the principle of geometric similarity of the tested individuals (see sections 2.1, 2.3 and 3.2). However, it is well known that human bodies are neither similar in shape nor in body composition. For example, even biomechanical models of human bodies have to be presented separately for males and females, while muscle mass is gradually redistributed during maturation or aging.[101] These differences and/or changes are inevitably associated with changes in relative values (with respect to body size) of mass and cross-sectional area of particular muscles, inevitably affecting the results from muscle strength tests. Another problem reflects the differences in body composition. Specifically, the most frequently applied normalisation methods with respect to body mass or weight do not account for differences in percentage of fat or muscle tissue among the tested individuals. This problem has been already recognised when studying the relationship between muscle strength and body size,[63-65,78] but Sports Med 2002; 32 (10)

Normalisation of Muscle Strength for Body Size

also when profiling athletes.[10,69,91] With respect to the particularly large body mass of some individuals, it has been suggested that there are certain upper limits for lean body mass in human populations and, therefore, further increases in body mass are predominantly based on gain in fat tissue.[102] Some studies even demonstrated a negative relationship between muscle strength and percentage body fat.[10] Indices like cross-sectional area of muscle,[43,71,79] lean limb tissue[103] and bone-free lean mass[64] have been recommended to be used for muscle strength normalisation instead of body mass. This problem requires further study to provide a deeper insight into muscle strength, as well as in movement performance testing. However, it remains unlikely that routine strength testing in the future will often use, in addition to body mass, any other index of body size that requires a complex anthropometric assessment. 4.2 Specific Populations: Adolescents and the Elderly

The normalisation methods both reviewed and discussed within the present review are expected to provide the body-size–independent index of muscle strength. However, the literature suggests that the methods recommended for adult populations could not be effective when applied to adolescents and/or the elderly. Since aging is associated with a prominent decrease in both muscle strength and body size,[12,25,54,74] some studies have demonstrated that strength normalised for body size remains unchanged with age.[79,103] However, other studies suggest a moderate decrease in normalised strength associated with age.[11,43,61,77] The later effect has been interpreted by changes in the physiological function of muscles,[104] redistribution of muscle mass,[43] or differences in rate of strength decline in different muscles.[11,105] A similar phenomenon has been observed in young individuals. Both muscle strength and body size demonstrate an increase over the age of young participants, therefore normalisation of strength for body size becomes particularly important when © Adis International Limited. All rights reserved.

627

groups of different age are compared.[70,76] This increase is particularly accelerated during maturation, while the participants demonstrating both early and delayed maturation may be tested within the same age group.[28,101] As a consequence, the correlation between muscle strength and body size may be considerably higher than in adults,[28,61] while the optimal value of the allometric parameter for muscle strength normalisation could also be affected.[28,61,67] Interestingly, it seems that the strength per cross-sectional area of muscle or per fat-free body mass has not been compared between young and adult participants. These findings generally suggest that the relationship between muscle strength and body size is confounded by various processes associated with early maturation and aging. As a consequence, the standard normalisation method may not provide the body-size–independent indices of muscle strength when applied in either children or the elderly. 4.3 Other Factors

A number of other factors can also affect body size and/or muscle strength thus affecting the relationship between them. When comparing different groups of athletes, prominent differences in muscle histology could be a consequence of both the selection and training procedures (for review see Abernethy et al.,[35] and Ruby and Robergs[106]). Since body size and body shape could also be an important factor for early selection of athletes, the strength to body-size relationship can also be affected. The specificity of muscle contraction regimens has often been studied. One of the general findings is that the maximum forces exerted under isometric and ‘dynamic’ (i.e. concentric and eccentric) conditions are relatively weakly related.[18,85,106,107] Moreover, this phenomenon can also be affected by both the applied training and early selection.[27,40,94,108] As a consequence, numerous problems and phenomena discussed within the present review can be affected, such as differences in particular indices of strength between various participant populations, or different strength assessments Sports Med 2002; 32 (10)

628

obtained in the same population when different strength tests are applied (e.g. isometric versus isokinetic, or isoinertial).[1] Finally, it should be mentioned that in addition to muscle strength, the present study also proposes normalisation of the tested movement performance. It seems conceivable that most of the aforementioned factors that could affect the relationship between muscle strength and body size (e.g. differences in body shape and composition, effects of maturation or aging), could also affect the relationship between functional movement performance and body size. Theoretically, the problem could be solved by any normalisation of muscle strength that provides the highest possible relationship with the tested functional movement performance (or vice versa) in each particular set of experimental data. However, it is not likely that most routine testing procedures in the future will provide room for assessment of accurate values of the allometric coefficient for each particular participant population to apply the most effective normalisation method. 4.4 Summary

Among the most important factors that can affect the relationship between muscle strength and body size are systematic differences among particular populations in body shape or body composition, the effects of maturation, age, sport selection and training, specificity of the contraction regimen, and so on. These factors may not only be the most important cause of a high variability obtained in the allometric exponent b, but also in the strength of the relationship between muscle strength and functional performance. Therefore, it should be taken into account that different populations could require partly different normalisation methods (i.e. different values of the allometric parameter b) to obtain body-size–independent indices of muscle strength. However, it is reasonable to assume that most of the professional studies based on routine strength examination will provide no conditions for studying strength-size relationships in each particular participant population. In this case, the © Adis International Limited. All rights reserved.

Jaric

strength normalisation methods recommended within the previous paragraphs should be applied. 5. Conclusion Although body size represents a well-recognised factor that affects the outcome of muscle strength tests, the normalisation of muscle strength for body size has been applied arbitrarily through the literature. Therefore, the results from strength tests assumed to select young athletes, distinguish between different performance levels, or provide normative data for various athletic or patient populations have often been confounded by the body-size effect. The role of body size has also been neglected when assessing various functional movement performances. Although different participant populations could require partly different normalisation of muscle strength for body size, routine strength testing through the sport, ergonomic and medicalrelated literature requires standardised methods. The present literature review recommends applying the allometric method based on equation 3 (S is muscle strength recorded as either muscle force or muscle torque, while m is body mass). The allometric parameter b should be either b = 0.67 for muscle force (recorded by a dynamometer) or b = 1 for muscle torque (recorded by an isokinetic apparatus) to assess the body-size–independent indices of muscle strength. In addition, the effects of body size on various functional movement performances should also be taken into account. Specifically, to assess the relationship between muscle strength and movement performance, both the index of muscle strength and the index of functional movement performance should be body-size independent. The recommended normalisation methods should provide more reliable outcomes of muscle strength tests that could be compared among different studies, as well as a more reliable assessment of functional movement performance based on muscle strength tests. Acknowledgements The study was supported in part by grants from the Swedish Sport Research Council (Centrum for Idrottsforskning), the

Sports Med 2002; 32 (10)

Normalisation of Muscle Strength for Body Size

Swedish Council for Work Life Research and from the Serbian Research Council.

References 1. Abernethy P, Wilson G, Logan P. Strength and power assessment: issues, controversies and challenges. Sports Med 1995; 19: 401-17 2. Keating JL, Matyas TA. The influence of subject and test design on dynamometric measurements of extremity muscles. Phys Ther 1996; 76: 866-89 3. Wilson GJ, Murphy AJ. The use of isometric tests of muscular function in athletic assessment. Sports Med 1996; 22: 19-37 4. Sale DG. Testing strength and power. In: MacDougal JD, Wenger HA, Green HJ, editors. Physiological testing of the highperformance athlete. Champaign (IL): Human Kinetics, 1991: 21-75 5. Jaric S, Radosavljevic-Jaric S, Johansson H. Muscle force and muscle torque may require different methods when adjusting for body size. Eur J Appl Physiol Occup Physiol. In press 6. Murphy AJ, Wilson GJ, Pryor JF. The use of isoinertial force mass relationship in the prediction of dynamic muscle performance. Eur J Appl Physiol Occup Physiol 1994; 69: 250-7 7. Pfeifer K, Banzer W. Motor performance in different dynamic tests in knee rehabilitation. Scand J Med Sci Sports 1999; 9: 19-27 8. Pryor JF, Wilson GJ, Murphy AJ. The effectiveness of eccentric, concentric and isometric rate of force development tests. J Hum Mov Stud 1994; 27: 153-72 9. Ugarkovic D, Matavulj D, Kukolj M, et al. Body composition and strength predictors of jumping performance in junior basketball players. J Strength Cond Res. In press 10. Sipila S, Suominen H. Knee extension strength and walking speed in relation to quadriceps muscle composition and training in elderly women. Clin Physiol 1994; 14: 433-42 11. Viljanen T, Viitasalo JT, Kujala UM. Strength characteristics of a healthy urban adult population. Eur J Appl Physiol Occup Physiol 1991; 63: 43-7 12. Paasuke M, Ereline J, Gapeyeva H. Knee extension strength and vertical jumping performance in Nordic combined athletes. J Sports Med Phys Fitness 2001; 41: 354-61 13. Sleivert GG, Backhus RD, Wenger HA. Neuromuscular differences between volleyball players, middle distance runners and untrained controls. Int J Sports Med 1995; 16: 390-8 14. Wisloff U, Helgrund J, Hoff J. Strength and endurance of elite soccer players. Med Sci Sports Exerc 1998; 30: 462-7 15. Blackburn JR, Morrissey MC. The relationship between open and closed kinetic chain strength of the lower limb and jumping performance. J Orthop Sport Phys 1998; 27: 430-5 16. Driss T, Vandewalle H, Monod H. Maximal power and forcevelocity relationships during cycling and cranking exercises in volleyball players. J Sports Med Phys Fitness 1998; 38: 286-93 17. Dooman CS, Vanderburgh PM. Allometric modeling of the bench press and squat: who is the strongest regardless of body mass? J Strength Cond Res 2000; 14: 32-6 18. Murphy AJ, Wilson GJ. Poor correlations between isometric tests and dynamic performance: relationship to muscle activation. Eur J Appl Physiol Occup Physiol 1996; 73: 353-7

© Adis International Limited. All rights reserved.

629

19. Vanderburgh PM, Dooman C. Considering body mass differences, who are the world’s strongest women? Med Sci Sports Exerc 2000; 32: 197-201 20. Van Ingen Schenau GJ, De Koning JJ, Bakker FC, et al. Performance-influencing factors in homogeneous groups of top athletes: a cross-sectional study. Med Sci Sports Exerc 1996; 28: 1305-10 21. Jensen RL, Freedson PS, Hamill J. The prediction of power and efficiency during near maximal rowing. Eur J Appl Physiol Occup Physiol 1996; 73: 98-104 22. Young W, McLean B, Ardagna J. Relationship between strength qualities and sprinting performance. J Sports Med Phys Fitness 1995; 35: 13-9 23. Young W, Wilson G, Byrne C. Relationship between strength qualities and performance in standing and run-up vertical jumps. J Sports Med Phys Fitness 1999; 39: 285-93 24. Akesson I, Hansson G-A, Balogh I, et al. Quantifying work load in neck, shoulders and wrists in female dentists. Int Arch Occup Environ Health 1997; 69: 461-74 25. Kwon IS, Oldaker S, Schrager M, et al. Relationship between muscle strength and the time taken to complete a standardized walk-turn-walk test. J Gerontol A Biol Sci Med Sci 2001; 56 (9): B398-404 26. Agre JC, Casal DC, Leon AS, et al. Professional ice hockey players: physiologic, anthropometric, and musculoskeletal characteristics. Arch Phys Med Rehabil 1988; 69: 188-92 27. Andersson E, Sward L, Thorstensson A. Trunk muscle strength in athletes. Med Sci Sports Exerc 1988; 20: 587-93 28. Jaric S, Ugarkovic D, Kukolj M. Evaluation of methods for normalizing strength in elite and young athletes. J Sports Med Phys Fitness 2002; 42: 141-51 29. Reilly T, Bangsbo J, Franks A. Anthropometric and physiological predispositions for elite soccer. J Sports Sci 2000; 18: 669-83 30. Taylor NA, Cotter JD, Stanley SN, et al. Functional torquevelocity and power-velocity characteristics of elite athletes. Eur J Appl Physiol Occup Physiol 1991; 62: 116-21 31. Faria IE, Faria EW. Relationship of the anthropometric and physical characteristics of male junior gymnast to performance. J Sports Med 1989; 29: 369-78 32. Cometti G, Maffiuletti NA, Pousson M, et al. Isokinetic strength and anaerobic power of elite, subelite and amateur French soccer players. Int J Sports Med 2001; 22: 45-51 33. Fry RW, Morton AR. Physiological and kinanthropometric attributes of elite flatwater kayakists. Med Sci Sports Exerc 1991; 23: 1297-301 34. Matavulj D, Kukolj M, Ugarkovic D, et al. Effects of plyometric training on jumping performance in junior basketball players. J Sports Med Phys Fitness 2001; 41: 159-64 35. Abernethy PJ, Jurimae J, Logan PA, et al. Acute and chronic response of skeletal muscle to resistance exercise. Sports Med 1994; 17: 22-38 36. Bell GJ, Wenger HA. Physiological adaptations to velocitycontrolled resistance training. Sports Med 1992; 13: 234-44 37. Bemben MG, Murphy RE. Age related neural adaptation following short-term resistance training in women. J Sports Med Phys Fitness 2001; 41: 291-9 38. Kanehisa H, Mitsumasa M. Specificity of velocity in strength training. Eur J Appl Physiol Occup Physiol 1983; 52: 104-6

Sports Med 2002; 32 (10)

630

39. Kraemer WJ, Mazzetti SA, Nindl BC, et al. Effect of resistance training on women’s strength/power and occupational performances. Med Sci Sport Exerc 2001; 33: 1011-25 40. Murphy AJ, Spinks WL. The importance of movement specificity in isokinetic assessment. J Hum Move Stud 2000; 38: 167-83 41. Mathiassen SE, Ahsberg E. Prediction of shoulder flexion endurance from personal factors. Int J Ind Ergon 1999; 24: 315-29 42. Stevenson JM, Greenhorn DR, Bryant JT, et al. Gender differences in performance of a selection test using the incremental lifting machine. Appl Ergon 1996; 27: 45-52 43. Akima H, Kano Y, Enomoto Y, et al. Muscle function in 164 men and women aged 20-84 yr. Med Sci Sports Exerc 2001; 33: 220-6 44. Marcora S, Miller MK. The effect of knee angle on the external validity of isometric measures of lower body neuromuscular function. J Sports Sci 2000; 18: 313-9 45. Andrews AW, Thomas MW, Bohannon RW. Normative values for isometric force measurements obtained with hand-held dynamometers. Phys Therapy 1996; 76: 248-61 46. Beenakker EAC, van der Hoeven JH, Fock JM, et al. Reference values of maximum isometric muscle force obtained in 270 children aged 4-16 years by hand-held dynamometry. Neuromuscul Disord 2001; 11: 441-6 47. Reuter I, Engelhardt M, Stecker K, et al. Therapeutic value of exercise training in Parkinson’s disease. Med Sci Sports Exerc 1999; 31: 1544-9 48. Fleck SJ, Falkel JE. Value of resistance training for the reduction of sports injuries. Sports Med 1986; 3: 61-8 49. Magnusson SP, Constantini NW, McHugh MP, et al. Strength profiles and performance in masters’ level swimmers. Am J Sports Med 1995; 23: 626-31 50. Takala EP, Viikari-Juntura E. Do functional tests predict low back pain? Spine 2000; 25: 2126-32 51. Imrhan SN. Muscular strength in the elderly: implications for ergonomic design. Int J Ind Ergon 1994; 13: 125-38 52. Kraemer WJ, Piorkowski PA, Bush JA, et al. The effects of NCAA division 1 intercollegiate competitive tennis match play on recovery of physical performance in women. J Strength Cond Res 2000; 14: 265-72 53. Mandell PJ, Weitz E, Bernstein JI, et al. Isokinetic trunk strength and lifting strength measures: differences and similarities between low-back-injured and noninjured workers. Spine 1993; 18: 2491-501 54. Desrosiers J, Bravo G, Hebert R, et al. Normative data for grip strength of elderly men and women. Am J Occup Ther 1995; 7: 637-44 55. Astrand P-O, Rodahl K. Textbook of work physiology. 3rd ed. New York (NY): McGraw-Hill, 1986 56. McMahon TA. Muscles, reflexes and locomotion. Princeton (NJ): Princeton Press, 1984 57. Nevill AM, Ramsbottom R, Williams C. Scaling physiological measurements for individuals of different body size. Eur J Appl Physiol Occup Physiol 1992; 65: 110-7 58. Vanderburgh PM, Mahar MT, Chou CH. Allometric scaling of grip strength by body mass in college-age men and women. Res Q Exerc Sport 1995; 66: 80-4 59. Hortobagyi T, Katch FI, Katch VL, et al. Relationships of body size, segmental dimensions, and ponderal equivalents to mus-

© Adis International Limited. All rights reserved.

Jaric

60.

61.

62. 63.

64. 65.

66.

67.

68. 69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

cular strength in high-strength and low-strength subjects. Int J Sports Med 1990; 11: 349-56 Delitto A, Crandell CE, Rose SJ. Peak torque-to-body weight ratios in the trunk: a critical analysis. Phys Ther 1989; 69: 138-43 Weir JP, Housh TJ, Johnson GO, et al. Allometric scaling of isokinetic peak torque: the Nebraska Wrestling Study. Eur J Appl Physiol Occup Physiol 1999; 80: 240-8 Hill AV. The dimensions of animals and their muscular dynamics. Sci Prog 1950; 38: 209-30 Challis JH. Methodological report: the appropriate scaling of weightlifting performance. J Strength Cond Res 1999; 13: 367-71 Davies MJ, Dalsky GP. Normalizing strength for body size differences in older adults. Med Sci Sports Exer 1997; 29: 713-7 Batterham AM, George KP. Allometric modeling does not determine a dimensionless power function ratio for maximal muscular function. J Appl Physiol 1997; 83: 2158-66 Neder JA, Nery LE, Silva AC, et al. Maximal aerobic power and leg muscle mass and strength related to age in non-athletic males and females. Eur J Appl Physiol Occup Physiol 1999; 79: 522-30 Nevill AM, Holder RL, Baxter-Jones A, et al. Modeling developmental changes in strength and aerobic power in children. J Appl Physiol 1998; 84: 963-70 Jackson A. Strength measurement: controlling for individual differences. J Phys Edu Recreation Dance 1986; 57: 82-4 Jaric S, Ugarkovic D, Kukolj M. Anthropometric, strength, power and flexibility variables in elite male athletes: basketball, handball, soccer and volleyball players. J Hum Mov Stud 2001; 40: 453-64 Sunnegardh J, Bratteby LE, Nordesjo LO, et al. Isometric and isokinetic muscle strength, anthropometry and physical activity in 8-year-old and 13-year-old Swedish Children. Eur J Appl Physiol Occup Physiol 1988; 58: 291-7 Abe T, Kawakami Y, Ikegava S, et al. Isometric and isokinetic knee-joint performance in Japanese alpine ski racers. J Sports Med Phys Fitness 1992; 32: 353-7 Hakkinen K, Alen M, Komi PV. Neuromuscular, anaerobic, and aerobic performance characteristics of elite power athletes. Eur J Appl Physiol Occup Physiol 1984; 53: 97-105 Bale P. Anthropometric, body composition and performance variables of young elite female basketball players. J Sports Med Phys Fitness 1991; 31: 173-7 Clarke DH, Hunt MQ, Dotson CO. Muscular strength and endurance as a function of age and activity level. Res Q Exerc Sport 1992; 63: 302-10 Froese EA, Houston ME. Torque-velocity characteristics and muscle fiber type in human vastus lateralis. J Appl Physiol 1985; 59: 309-14 De Ste Croix MBA, Armstrong N, Weisman JR. Concentric isokinetic leg strength in pre-teen, teenage and adult males and females. Biol Sport 1999; 16: 75-86 Frontera WR, Hughes VA, Lutz KJ, et al. A cross-sectional study of muscle strength and mass in 45- to 78-yr-old men and women. J Appl Physiol 1991; 71: 644-50 Hulens M, Vansant G, Lysens R, et al. Study of differences in peripheral muscle strength of lean versus obese women: an allometric approach. Int J Obes 2001; 25: 676-81

Sports Med 2002; 32 (10)

Normalisation of Muscle Strength for Body Size

79. Izquierdo M, Hakkinen K, Anton A, et al. Maximal strength and power, endurance performance, and serum hormones in middle-aged and elderly men. Med Sci Sports Exerc 2001; 33: 1577-87 80. Paasuke M, Ereline J, Gapeyeva H, et al. Age-related differences in twitch contractile properties of plantarflexor muscles in women. Acta Physiol Scand 2000; 170: 51-7 81. Vandewalle H, Peres G, Monod H. Standard aerobic exercise tests. Sports Med 1987; 4: 268-89 82. Hakkinen K. Force production characteristics of leg extensor, trunk flexor and extensor muscles in male and female basketball players. J Sports Med Phys Fitness 1991; 31: 325-31 83. Suei K, McGillis L, Calvert R, et al. Relationships among muscle endurance, explosiveness, and strength in circum-pubertal boys. Pediatr Exerc Sci 1998; 10: 48-56 84. Wiklander J, Lysholm J. Simple tests for surveying muscle strength and muscle stiffness in sportsmen. Int J Sports Med 1987; 8: 50-4 85. Augustsson J, Thomee R. Ability of closed and open kinetic chain tests of muscular strength to assess functional performance. Scand J Med Sci Sports 2000; 10: 164-8 86. Birch K, Sinnerton S, Reilly T, et al. The relation between isometric lifting strength and muscular fitness measures. Ergonomics 1994; 37: 87-93 87. Jaric S, Ristanovic D, Corcos DM. Relations between kinetic parameters of active muscle groups and kinematic variables of a complex movement. Eur J Appl Physiol Occup Physiol 1989; 59: 370-6 88. Kukolj M, Ropret R, Ugarkovic D, et al. Anthropometric, strength and power predictors of sprinting performance. J Sports Med Phys Fitness 1999; 39: 120-2 89. Ostenberg A, Roos E, Ekdahl C, et al. Isokinetic knee extensor strength and functional performance in healthy female soccer players. Scand J Med Sci Sports 1998; 8: 257-64 90. Pincivero DM, Lephart SM, Karunakara RG. Relation between open and closed kinematic chain assessment of knee strength and functional performance. Clin J Sport Med 1997; 7: 11-6 91. Nindl BC, Mahar MT, Herman EA, et al. Lower and upperbody anaerobic performance in male and female adolescent athletes. Med Sci Sports Exerc 1995; 27: 235-41 92. Barnekow-Bergkvist M, Hedberg G, Janlert U, et al. Development of muscular endurance and strength from adolescence to adulthood and level of physical capacity in men and women at age of 34 years. Scand J Med Sci Sports 1996; 6: 145-55 93. Hattori Y, Ono Y, Shimaoka M, et al. Effects of box weight, vertical location and symmetry on lifting capacities and ratings on category scale in Japanese female workers. Ergonomics 2000; 43: 2031-42

© Adis International Limited. All rights reserved.

631

94. Cronin J, McNair PJ, Marshall RN. Velocity specificity, combination training and sport specific tasks. J Sci Med Sport 2001; 4: 168-78 95. Williford HN, Duey WJ, Olson MS, et al. Relationship between fire fighting suppression tasks and physical fitness. Ergonomics 1999; 42: 1179-86 96. Richards JE, Ackland TR, Elliott BC. The effect of training volume and growth on gymnastic performance in young women. Pediatr Exerc Sci 1999; 11: 349-63 97. Hogan J. Structure of physical performance in occupational tasks. J Appl Physiol 1991; 76: 495-507 98. Fleishman EA. The structure and measurement of physical fitness. Englewood Cliffs (NJ): Prentice-Hall, 1964 99. Hollings SC, Robson GJ. Body build and performance characteristics of male adolescent track and field athletes. J Sports Med Phys Fitness 1991; 31: 178-82 100. Schmidt WD. Strength and physiological characteristics of NCAA division III American football players. J Strength Cond Res 1999; 13: 210-3 101. Malina RM, Bouchard C. Growth, maturation and physical activity. Champaign (IL): Human Kinetics Books, 1991 102. Forbes GB. Human body composition: growing, aging, nutrition and activity. New York (NY): Springer-Verlag, 1987 103. Landers KA, Hunter GR, Wetzstein CJ, et al. The interrelationship among muscle mass, strength, and the ability to perform physical tasks of daily living in younger and older women. J Gerontol A Biol Sci Med Sci 2001; 56 (10): B443-8 104. Gray A, Feldman HA, Mckinlay JB, et al. Age, disease, and changing sex hormone levels in middle-ages men: results of the Massachusetts male aging study. J Clin Endocrinol Metab 1991; 73: 1016-25 105. Hasue M, Fujiwara M, Kikuchi S. A new method of quantitative measurement of abdominal and back muscle strength. Spine 1980; 5: 143-8 106. Ruby BC, Robergs RA. Gender differences in substrate utilisation during exercise. Sports Med 1994; 17: 393-410 107. Baker D, Wilson G, Carlyon B. Generality versus specificity: a comparison of dynamic and isometric measures of strength and speed-strength. Eur J Appl Physiol Occup Physiol 1994; 68: 350-5 108. Morrissey MC, Harman EA, Johnson MJ. Resistance training modes: specificity and effectiveness. Med Sci Sports Exerc 1995; 27: 648-60

Correspondence and offprints: Slobodan Jaric, Centre for Musculo-Skeletal Research, National Institute for Working Life, Box 7654, S-907 13, Umea, Sweden. E-mail: [email protected]

Sports Med 2002; 32 (10)