PTJ
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 QUICK SEARCH:   [advanced]


     


PHYS THER
Vol. 84, No. 7, July 2004, pp. 608-621

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when Rapid Responses are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ropponen, A.
Right arrow Articles by Battié, M. C
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ropponen, A.
Right arrow Articles by Battié, M. C
Related Collections
Right arrow Musculoskeletal System/Orthopedic: Other
Right arrow Tests and Measurements
Social Bookmarking
 Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Research Reports

The Role of Genetics and Environment in Lifting Force and Isometric Trunk Extensor Endurance

Annina Ropponen, Esko Levälahti, Tapio Videman, Jaakko Kaprio and Michele C Battié

A Ropponen, PT, MSc, is Senior Assistant of Ergonomics, Department of Physiology, University of Kuopio, PO Box 1627, 70211 Kuopio, Finland (annina.ropponen{at}uku.fi), and is a doctoral student, Department of Health Sciences, University of Jyväskylä, Jyväskylä, Finland.
E Levälahti, MSc, is Statistician, Department of Public Health, University of Helsinki, Helsinki, Finland
T Videman, PhD, is Professor and Alberta Heritage Foundation for Medical Research Scientist, Faculty of Rehabilitation Medicine, University of Alberta, Edmonton, Alberta, Canada
J Kaprio, PhD, is Professor, Department of Public Health, University of Helsinki
MC Battié, PT, PhD, is Professor and Canada Research Chair, Faculty of Rehabilitation Medicine, University of Alberta

Address all correspondence to Ms Ropponen at the first address


Submitted December 31, 2002; Accepted January 7, 2004


    Abstract
 
Background and Purpose. Our understanding of what different back performance tests are measuring is limited. The purpose of this study was to investigate the relative contributions of genetics and unique and common environmental factors for 3 tests of back muscle performance in a classic twin analysis. Subjects. The subjects were a population-based sample of 122 monozygotic and 131 dizygotic male twin pairs aged 35 to 69 years (X=49.9, SD=7.7). Methods. Variance component analysis was applied to estimate genetic and environmental influences on isokinetic and psychophysical lifting and isometric trunk extensor endurance test performance. The Cholesky decomposition genetic factor model was used to estimate genetic and environmental correlations of these variables. Path analysis was applied to study determinants of isokinetic and psychophysical lifting and isometric trunk extensor endurance test performance. Results. Genetic effects accounted for 60%, 33%, and 5% of the total variance of isokinetic and psychophysical lifting forces and isometric trunk extensor endurance, respectively, and unique environmental factors accounted for 35%, 49%, and 61% of the variance. Discussion and Conclusion. Genetics had a dominant role in isokinetic lifting and unique environmental factors in isometric trunk extensor endurance. The relatively high role of genetics in lifting force suggests the potential to increase and sustain changes in back muscle force in the general population may be particularly challenging.

Key Words: Back muscle performance • Endurance • Environment • Genetics • Muscle force


    Introduction
 Top
 Abstract
 Introduction
 Method
 Results
 Discussion
 Conclusions
 References
 
Tests of back muscle performance have been used to reflect impairments in people with low back problems as well as risk for future problems. Although a variety of tests of trunk muscle force, isometric trunk extensor endurance, and lifting are commonly used for similar purposes, the measurements obtained with these tests are poorly correlated with one another, indicating to us that the tests are measuring different phenomena with different underlying determinants.17 A better understanding of factors influencing performance on such tests and their relative importance, in our view, could help to guide the selection of back tests and the interpretation of data obtained with them.

A number of constitutional, behavioral, and environmental factors have been associated with tests of performance of low back muscles. Lifting force is lower in adults over 45 years of age than in younger adults.3,5,8,9 Similar findings have been reported for isometric trunk extensor endurance time5,10,11 and for muscle force and trunk extension torque measured during isometric contractions while standing, sitting, and lifting using isokinetic devices and testing protocols with a range of motion of 0 to 80 degrees of flexion and speeds of 30°/s to 120°/s.6,10,1214 However, the influence of age, spanning from young through middle adulthood, on trunk muscle force is less clear.7,15

Anthropometric factors also appear to be of importance in low back muscle performance. In a study of male monozygotic (MZ) twins,3 subjects who were taller, heavier, and with larger paraspinal muscle cross-sectional area than their twin brothers also had higher isokinetic lifting peak force as measured with 0 to 60 degrees of flexion and a speed of 0.5 m/s. A multivariate analysis showed that lean body mass alone explained 41% of the variance, and weight had a small additional effect.3 However, after including familial aggregation (representing effects of shared genetic and family influences) in the model, the variance explained by lean body mass dropped to 8%.3 The low effect of lean body mass after accounting for familial aggregation could be expected on the basis of earlier findings that relative weight, assessed as body mass index (in kilograms per square meter), had a genetic component.16 Subjects of both sexes without low back pain (LBP) with a high percentage of body fat have demonstrated shorter isometric trunk extensor endurance times as measured lying prone without support for the upper body and with hands behind the neck.3,11 However, weight had no effect on isokinetic trunk extension and lifting mean peak force in patients with LBP.9,13

Effects of physical activity on tests of low back muscle performance are controversial, which may be explained, in part, by variations in definitions of physical activity and difficulties with its accurate measurement. In some studies,2,3 no difference in peak force (measured with 0°–80° of flexion and speeds of 60°–120°/s and 0.5 m/s) was found between active and inactive subjects with or without back problems. In other studies, however, researchers have shown an association between greater leisure-time physical activity and longer trunk extensor endurance time (measured with subjects lying prone with hands across the chest and sitting)10,17 and between greater leisure-time physical activity and better isometric and isoinertial trunk extension torque.10,15

Different aspects (intensity and duration) of LBP history have been shown to have an influence on measurements of low back muscle performance in the majority of studies of nonpatient samples,7,10,15,1720 but not in all studies.3,9,12,21 We contend that the apparently conflicting results were due, at least in part, to definitions of current back pain (pain present either during testing or during the test day) or past back pain (an episode of LBP or frequency of LBP over the prior year)3,9,12,21 and to severity of the episodes of LBP (as determined by visual analog scale rating for pain or difficulty in performing different tasks).3,21 Severity of pain and other aspects of LBP in studies of patients with chronic LBP also have been associated with low back muscle performance.2,8,13,17,2225 A model of fear-avoidance of pain suggests the avoidance of physical activity (eg, lifting effort) due to prior experiences with back pain can affect back muscle performance.2628 Pain and disability have been associated with isokinetic lifting and trunk extension peak force and average torque while standing29 and with isometric trunk extensor endurance and repetitive lifting of the upper body in extension from a prone position30 in women, but associations among men are less clear.3,29,31 One group of researchers25 found an association between occurrence of back pain and isometric trunk extensor endurance in both sexes.

Back muscle performance also appears to be influenced by attitudes and beliefs about self-efficacy (a person's beliefs about his or her own capacities) and self-assessed health.3,8,15,32 Self-efficacy beliefs have been found to be positively correlated to isokinetic lifting peak force,8,30,33 isometric lifting peak force,15,33 and isokinetic trunk extension total work.13 Self-assessed health has been found to correlate to isokinetic lifting force3 and isometric trunk extensor endurance,3,17 but not to psychophysical lifting.3

While rarely examined, familial influences appear to play a major role in isokinetic lifting peak force.3 Gibbons and colleagues3 found that familial aggregation accounted for 56% of the overall variance of isokinetic lifting peak force, 32% of the variance in psychophysical lifting force, and 15% of the variance in isometric trunk extensor endurance. The results were based on analyses of 65 pairs of MZ twins and did not allow for the separation of familial aggregation into its genetic and nongenetic components, the latter being those environmental experiences and effects common to both siblings.3 By adding dizygotic (DZ) twins to a similar study setting, exploring genetic versus shared environmental influences, the components of familial aggregation would be possible. This study design would provide overall estimates on the relative roles of genes, childhood and adulthood, for back muscle performance. A better understanding of the underlying determinants for each of the tests could have both practical and theoretical relevance.

Three tests of back muscle performance were used in our study. One test measured isometric performance, another test measured isokinetic performance, and the third test measured psychophysical aspects of acceptable loads. The isometric back endurance performance test remains one of the few back muscle performance tests to be associated with future occurrence of back problems and is, therefore, of particular interest. The isokinetic lifting test appears to measure a basic capacity that is determined, in large part, by genetic influences. The specific goal of this study was to investigate the relative contributions of genetic, common (shared family), and unique environmental and behavioral factors as determinants of isokinetic and psychophysical lifting force and isometric trunk extensor endurance test performances using a classic twin study method. Our intent was to extend the previous work by Gibbons et al3 through expanding the sample of MZ twins and adding a sample of DZ twins, making it possible to separate familial aggregation into its genetic and shared familial environment (nongenetic) components.


    Method
 Top
 Abstract
 Introduction
 Method
 Results
 Discussion
 Conclusions
 References
 
Subjects

A sample of 147 MZ and 153 DZ male twin pairs, 35 to 70 years of age (X=49.9, SD=7.7), was drawn from the Finnish Twin Cohort. The Finnish Twin Cohort was established by identifying all same-sexed twin pairs born before 1958 and alive in 1975 from the Central Population Register of Finland, which has data on all Finnish citizens and residents.34 Thus, the sample was population-based. Selection criteria were based solely on within-pair differences in occupational materials handling, sedentary work, exercise, driving history, or cigarette smoking, as has been described elsewhere.35 Additional pairs were selected to increase the sample size, and they were selected at random with the goal of enhancing the representativeness of the sample. The 65 pairs of MZ twins studied by Gibbons et al3 were included in the study sample as a part of the MZ twin pairs. Back muscle performance tests were conducted at a laboratory in central Finland as a part of a larger study on the effects of common exposures on back and other musculoskeletal problems during 1991–1993 and 1997–1999.35 Fifty-eight subjects were not tested because of acute or severe back pain, heart problems, hand impairment, or cancer metastases causing pain during performance (n=50) or malfunction of the testing device (n=8). For inclusion in analyses, complete data were required for the 3 back muscle performance tests, isokinetic and psychophysical lifting force, and isometric trunk extensor endurance, for both twins of a pair, leading to inclusion of 122 MZ and 131 DZ pairs. The mean age (53 years, 95% confidence interval [CI]=51–56) of those excluded from testing was greater than that of those included (49 years, 95% CI=48–50), but there were no differences in exclusion by zygosity.

Interview

Information on lifetime work and exercise histories and other regularly performed activities involving physical loading and a detailed history of neck and back pain and other health problems were gathered in a thorough, structured interview. All interviews were done as part of the current study. Five trained interviewers carried out the interviews. The same person interviewed both the siblings of a pair, and each interview took approximately 2.5 hours to complete.

Work history.
Each subject discussed in detail every job held for at least 3 months since age 12 years to estimate the lifetime physical loading exposures experienced at work. The subject's profession, time spent driving and sitting on the job (hours per day), average weight lifted (in kilograms) and frequency of lifting (number of times per hour, per day, and per week), different working positions (minutes per hour), standing and walking during work (hours per day), weekly working hours, recalled trauma at work, and commuting to and from work were recorded. A limited evaluation of reliability of work history data was conducted by telephone interview 12 months later.35 The subjects' responses to questions asked during the telephone interview were compared with their responses to questions asked during the initial interview for those who noted that their job had not changed since that time. The Spearman correlation coefficients were .75 for sitting, .77 for driving, and .60 for total lifting per day.35

Leisure-time physical activity.
The subjects' lifetime histories of regularly performed exercise and other leisure-time physical activities during at least 3 months per year were reviewed by interviewers as a part of the interview. Mode, length of time of participation (years during lifetime, months per year), duration (minutes per session), frequency (times per week), intensity (light, moderate, or strenuous), and associated injuries were recorded. The overall reliability of data obtained in the structured interview of lifetime exercise using a 5-year test-retest interval has been previously reported.36 Weekly exercise hours and years of participation were found to be the most reliably measured variables, with intraclass correlation coefficients (ICCs) of .63 to .90.36

Procedure

Three different aspects of back muscle performance were evaluated. The isokinetic lifting test is a test of fast, maximal lifting effort at a constant speed, whereas no maximal effort is called for in the psychophysical lifting test, and the isometric trunk extensor endurance test measures a person's ability to maintain a position through an isometric contraction. The performance measurements from the tests have been found to poorly correlate to one another, with Pearson correlation coefficients ranging from .03 to .24.5 Either a trained physician or a physical therapist administered the tests to both twins within a pair.

The isokinetic lifting test was performed from a forward-bending position with knees straight. The forward-bending position was measured by use of an MIE goniometer* on the L3 vertebra at 60 degrees of spinal flexion. The MIE goniometer is commercially available and widely used among physical therapists, but we are not aware of studies of the reliability or validity of measurements obtained with the device. The L3 vertebra was selected because we believed it best approximates the spinal motion axis when the thoracic spine and pelvis are in a stationary position.37 The subjects received standardized instructions and a demonstration and were asked to lift as rapidly and forcefully as possible. No other encouragement was given. The isokinetic lifting tester includes a platform with footprints 21 cm apart to stand on and a bar to lift 3 cm in front of footprints. To minimize the contribution of the arm and leg muscles in force production, the subjects kept their knees, wrists, and elbows straight. Prior to 4 recorded lifts, subjects performed 5 training lifts at constant speeds of 0.5 and 0.7 m/s, the speeds available with the tester. Between recorded lifts were 1-minute rest intervals. Lifts were performed first at the 0.5-m/s speed and then at the 0.7-m/s speed. Lifting force was recorded (in newtons), and speed and height of the lift were entered to calculate work done (in joules). The isokinetic tester was built at the University of Jyväskylä and had a precise gauge with a measurement error of 1% of maximum. The tester was calibrated before and after every measurement session, 4 times per a day. The reliability of isokinetic lifting measurements was determined in a study of working-age adults without known pathology or impairments, and ICCs of .97 and .87 were obtained for intratest and intertest reliability of peak force and total work measurements.5

The psychophysical lifting test was similar to the "acceptable isometric lifting force test" described by Foreman et al,38 except the lifting position was at 60 degrees of spinal flexion instead of at knee level. According to Foreman et al, the "acceptable" lifting force is the maximum force a subject exhibits when asked to demonstrate the maximum force that he or she can comfortably maintain for 5 seconds. Sixty degrees of spinal flexion was selected to standardize the test position for isokinetic and psychophysical lifting tests. Subjects were asked to use the maximum force (in newtons) they thought they could comfortably maintain for 5 seconds while performing the isometric lift. No encouragement was given. The subjects kept their knees and elbows straight for 3 lifts, with short rest intervals (approximately 5 seconds) between lifts. Pearson correlation coefficients of .87 to .96 were reported for intertester reliability of measurements among working-age subjects without known pathology or impairments.38,39

Isometric trunk extensor endurance was evaluated by timing (in seconds) how long a subject was able to hold the upper part of his body horizontal while lying prone with no support beyond the upper border of the iliac crest. The subject's hands were kept behind the neck, and the thighs and ankles were fixed to the Table by 2 wide straps. Subjects were instructed to hold the position as long as they could and then to tell why they stopped. During testing, subjects received encouragement once if their position fell below the horizontal level as indicated by a plumb bob hanging from the ceiling that was adjusted to contact the back when the horizontal position was maintained. A deviation of 1 cm was allowed. If the position was not immediately corrected, or if the subject claimed he could no longer hold the position due to fatigue or discomfort, the test was ended and holding time was recorded. Pearson correlation coefficients for intertest reliability have ranged from .66 to .89 in earlier studies.7,21

Body fat measurement.
Bioelectric impedance (BIA) was used to obtain percentage of body fat.40 With this study protocol, the coefficient of variation between 2 consecutive BIA measurements was in the order of 2% to 3%. The validity of BIA measurements of adults has been supported by studies of construct41 assuming pretesting and testing procedures are well standardized42 and the BIA prediction equation is population-specific.41,43 Lean body mass was computed based on the percentage of body fat and was calculated by subtracting fat weight from total body weight. Body fat and lean body mass were obtained based on earlier results of their relative influence on back muscle performance test results.3

Data Analysis

For statistical analysis, interview variables were selected based on frequency of answers (to meet the criteria of extensive) and results of the study by Gibbons et al.3 Interview data were coded into categorical variables based on the following criteria. Physical occupational loading was graded on a 5-point scale from "retired/unemployed presenting no occupational physical loading" to "heavy physical loading at work for current job status" at the time of the interview. The occupational physical loading categories were created based on different physical job demands that focused primarily on materials handling and included the subjects' description and estimate of sitting, standing and walking, lifting, bending, and driving involved in the job. Regular aerobic exercise was categorized for the purposes of analyses using a 4-point scale ranging from "≤1 year of regular aerobic exercise" to "30 years of regular aerobic exercise." Participation in power sports (eg, weight lifting) and other sports was coded to binary variables as participating for a minimum of 1 year at least 2 times per week or not. The exercise variables were selected based on their potential training effects for force or endurance as might be relevant to performance on the back muscle tests. Health as compared with that of others of the same age was graded on a 5-point scale ("much better" to "much worse"). In addition, fatigue as a reason to end isometric trunk extensor endurance testing was dichotomized (yes/no).

Normality of back muscle performance test variables was assessed by skewness and kurtosis tests (using the Stata sktest statistical program{dagger}).44 Back muscle performance test variables that were not normally distributed were first transformed (see Tab. 1) using appropriate functions (log and square root) to approximate normal distributions. Observations of suspected determinants (except weight, which was transformed) were summarized as either categorical or binary variables because no transformation was found that adequately normalized the distribution. The equality of means of continuous variables by zygosity was tested using an adjusted Wald test (using the Stata svymean statistical program{dagger}) to take into account that the twin individuals had been sampled as twin pairs and thus did not represent fully independent observations.44 A Wald test also was used to test the equality of proportions of binary variables (using the Stata svytab statistical program{dagger}). The equality of variances of continuous variables was tested using the variance ratio test. The equality of distributions of categorical variables by zygosity was tested using a design-based independence test, equivalent to a chi-square test of independence but taking the sampling design into account (using the Stata svytab program45).


View this table:
[in this window]
[in a new window]

 
Table 1. Means, Proportions, and Frequencies (95% Confidence Interval [95% CI] in Parentheses) and Test of Equality of Means, Proportions, and Frequencies by Zygosity for Continuous, Binary, and Categorical Variablesa

 
The effect of age on back muscle performance test results among twin individuals was modeled using a survey regression model (using the Stata svytab program) that takes into account the covariance within twin pairs and calculates the proportion explained by age, P value, 95% CI of regression coefficient, and 95% CI of regression constant adjusted to that within-pair covariance. Because age was categorized (for the skewness of distribution), the regression constant is the predicted mean in lowest age category and the regression coefficient is predicted mean change per 10 years of age. Stata svytab procedures were used to carry out the regression analysis and the equality tests.44

Users of the classical twin study design assume that genetic factors and environmental factors are not correlated, there is no genotype x environment interaction, and there is random mating with respect to the traits under study in the population (ie, no tendency to mate with someone with similar traits).46 If these assumptions hold true, the twin model can be used to determine the relative contribution of the additive effect of genes (A), dominance effect of genes (D), common environment (C), and unique environment (E) (Tab. 2). MZ pairs share all genes, whereas DZ pairs share half of their genes. Both types of twin pairs share fully a common environment (ie, exposures and experiences shared by siblings in a family), and this is assumed to be equal in magnitude in MZ and DZ pairs. Thus, the increased similarity of both MZ and DZ twins can be used to estimate common environmental effects, while the greater similarity of MZ pairs compared with DZ pairs provides evidence for genetic effects. The extent to which MZ pairs are more than twice as similar as DZ pairs permits estimation of additive and dominance effects.47


View this table:
[in this window]
[in a new window]

 
Table 2. Examples of Factors Constituting the Variance Components in Back Muscle Performance Tests

 
In both univariate and multivariate analyses of continuous variables, data were summarized with covariance matrices and means, and models were fit using the maximum likelihood estimation method. In data analysis with binary or categorical variables, data were summarized to polychoric correlations, asymptotic variances, and asymptotic covariance, while models were fit using asymptotic weighted least-squares estimation. The Mx statistical program for twin and family data was used for all estimations.46

Univariate path models were estimated to identify the potential sources of variance (Fig. 1) in each trait. Effects due to dominance and common environment cannot be simultaneously estimated solely with data from MZ and DZ pairs reared together.48 The squares of the standardized path coefficients (a, c, d, e) from the latent variables A, C, D, and E, respectively, are estimates of the variance components.


Figure 1
View larger version (13K):
[in this window]
[in a new window]

 
Figure 1. Univariate path model with additive genetic (a), genetic dominance (d), common environmental (c), and unique environmental (e) effects. {alpha} is additive genetic correlation (monozygotic [MZ] twins: {alpha}=1, digyzotic [DZ] twins: {alpha}=.5) and ß is dominance genetic correlation (MZ twins: ß=1, DZ twins: ß=.25). P1 (P2)=phenotypic value of twin 1 (twin 2).

 
We used the Cholesky decomposition genetic factor model to analyze the relationships among variables for the multivariate modeling investigating the genetic and environmental relationships among several traits (ie, asking whether the genetic factors affecting one trait are fully or partly correlated to genetic factors affecting another trait), and likewise for environmental effects. Based on structures of most parsimonious univariate models and fit statistics, {chi}2 statistic and Akaike information criterion (AIC), the most parsimonious multivariate model was found (shown graphically in Fig. 2). Proportions of variance and covariance explained by genetic and environmental factors and the genetic and environmental correlations of the most parsimonious multivariate model are reported here. A higher proportion of variance indicates that the same genes or environmental factors influence more than one trait at a time, whereas a lower proportion of variance indicates an influence of different genes or environmental effects. The contribution of the same genes or environmental factors to the observed phenotypic correlation between 2 variables is measured by the genetic or environmental correlations.


Figure 2
View larger version (17K):
[in this window]
[in a new window]

 
Figure 2. Graphical representation of Cholesky decomposition genetic factor model (additive genetic factors [A], genetic dominance factors [D1–2], common environmental factors [C1–2], and unique environmental factors [E1–4]). Means are not estimated to be equal.

 
Path modeling was used to model the relationships between genetic and environmental factors of the results of back muscle tests and suspected determinants (except age). To estimate the contribution of age, a simpler model, the survey regression model, was used. In the full model, the same genetic and environmental factors as in most parsimonious univariate models for back muscle test and suspected determinant were included, and all possible paths between these factors were included (Fig. 3). Paths between genetic factors, between unique environmental factors, and between common environmental factors were dropped one by one from the model, and the model with the lower AIC or the most simple model (if AICs were appropriately equal) was chosen. The information of difference in mean or proportion of variance between MZ and DZ twins is valuable. In addition, the information of possible difference between MZ and DZ twins will be accounted for in the models, if it is possible to model the mean or proportion of variance.


Figure 3
View larger version (17K):
[in this window]
[in a new window]

 
Figure 3. Path model for relationships between factors of back muscle performance test and suspected determinant. A=corresponding genetic factors, C=corresponding common environmental factors, and E=corresponding unique environmental factors. {zeta}C, {zeta}E, and {zeta}A are corresponding residuals for factors of back muscle test. Numbers (1–9) present possible estimated paths.

 

    Results
 Top
 Abstract
 Introduction
 Method
 Results
 Discussion
 Conclusions
 References
 
Analyses of Individuals

One test of the assumptions of the classic twin model is that means and variances of the traits do not differ by zygosity. Only 3 of the 17 general characteristics tested differed by zygosity (Tab. 1). Height (175 cm for MZ twins versus 177 cm for DZ twins, P=.014), occupational lifting during the previous year (59% for MZ twins versus 48% for DZ twins, P=.022), and fatigue as the reason to end isometric trunk extensor endurance testing (62% for MZ twins versus 72% for DZ twins, P=.033) differed by zygosity. Means of back muscle performance tests also showed some differences by zygosity. Compared with MZ twins, DZ twins had higher mean psychophysical lifting force (by 14%) and isometric trunk extensor endurance (by 37%) and 9% lower isokinetic lifting work. No difference was observed for isokinetic lifting force (Tab. 1). Corrections for multiple testing were not done. Survey regression models were used to obtain the proportion of variance explained by 10 years of additional age (6% in isokinetic lifting force, 12% in work, and 4% in isometric trunk extensor endurance). Ten years of additional age had no influence on psychophysical lifting.

Pair-wise Descriptive Results

The ICCs for isokinetic lifting force and work were greater for the MZ twins than for the DZ twins. However, the ICCs for psychophysical lifting and isometric trunk extensor endurance did not differ by zygosity (Tab. 3).


View this table:
[in this window]
[in a new window]

 
Table 3. Pearson Correlation Coefficients Estimated to Quantify Resemblance of Twins on Back Muscle Performance (95% Confidence Interval [95% CI] in Parentheses) and Test of Equality of Correlations by Zygosity for Back Muscle Test Variablesa

 
Variance Component Estimation by Multivariate Twin Models for Back Tests

In the final model (Fig. 2), the results of the 3 back tests had a common additive genetic effect (A). Isokinetic lifting force and work had a shared genetic dominance factor (D1), and isokinetic lifting work also had a test-specific genetic dominance factor (D2). Psychophysical lifting force and isometric trunk extensor endurance were affected by test-specific common environmental factors (C1–C2). The variation of each test was explained by environmental factors that are unique to each twin but shared by the test (E1). Isometric trunk extensor endurance and isokinetic lifting work were affected by test-specific unique environmental factors (E2 and E4). Isokinetic lifting force and work were affected by the same unique environmental factor (E3).

For isokinetic lifting force and work, genetic factors (both additive genetic and genetic dominance effects) had the greatest influential role, explaining a total of 60% to 65% of the variance. Unique environmental factors accounted for 35% to 41% of the variance of force and work, but common environmental factors had no contribution. Additive genetic factors accounted for 33% of overall variance in psychophysical lifting force, unique environmental factors accounted for 49% of the variance, and common environmental factors accounted for 18% of the variance. Genetic factors appeared to have only a minor influence on isometric trunk extensor endurance variance, with unique environmental factors accounting for 61% of the variance and common environmental factors accounting for 34% of the variance (Tab. 4).


View this table:
[in this window]
[in a new window]

 
Table 4. Proportions of Variance (95% Confidence Interval in Parentheses) Explained by Additive Genetic Factors (A), Genetic Dominance Factors (D1–2), and Common Environmental Factors (C1–2) and by 4 Different Unique Environmental Factors (E1–4)a

 
The same additive genetic factors explained 57% to 92% of the covariance among the 3 back tests and 31% of the covariance of isokinetic lifting force and work (Tab. 5). In addition, nearly the same genetic dominance factors explained the covariance between dominance effects of isokinetic lifting force and work; the genetic dominance correlation was .67. The covariation explained by unique environmental factors among the 3 back tests showed much variation, from a minimal correlation of .03 between the environmental components of isometric trunk extensor endurance and isokinetic lifting to a moderate correlation of .43 between the environmental effects on isokinetic lifting work and psychophysical lifting force (Tab. 5).


View this table:
[in this window]
[in a new window]

 
Table 5. Proportions of Covariance (95% Confidence Interval in Parentheses) Explained by Additive Genetic Factors, Unique Environmental Factors, and Genetic Dominance Factors and Correlationsa

 
Accounting for Specific Back Tests in Relation to Specific Determinants

The variance in isokinetic lifting force accounted for by unique environmental factors was estimated to be 11% for weight, 1% for self-assessed health (compared with others of the same age), 2% for occupational physical loading, and 4% for participation in power sports. In isokinetic lifting work, unique explanatory environmental factors followed the pattern of force, being weight (8% of the variance), occupational loading (2% of the variance), and power sports (41% of the variance). In psychophysical lifting force, the variance accounted for by unique environmental factors was estimated to be 2% for self-assessed health, 4% for occupational loading, and 1% for power sports. A common explanatory environmental factor was body weight, explaining 37% of the variance in psychophysical lifting force. The variance in isometric trunk extensor endurance accounted for by unique environmental factors was estimated to be 10% for weight, 6% for self-assessed health, 2% for aerobic sports, 7% for power sports, and 5% for the reason of ending the isometric test (Tab. 6). The variance in isometric trunk extensor endurance accounted for by common explanatory environmental factors, such as weight, other sports, and education, could not be explained by variables considered in this study.


View this table:
[in this window]
[in a new window]

 
Table 6. Path Models for the 3 Back Muscle Performance Tests: Path Coefficients and Proportions of Variance (95% Confidence Interval in Parentheses) Explained (R2e and R2c) for Unique and Common Environmental Factors, Fit Statistics, and Difference in Akaike Information Criterion (AIC) Values Between Full Model and Selected Modela

 

    Discussion
 Top
 Abstract
 Introduction
 Method
 Results
 Discussion
 Conclusions
 References
 
The 3 back muscle performance tests had substantially different relative contributions: the proportion of variance accounted for by additive genetic effects varied from 39% for isokinetic lifting force to 5% for isometric trunk extensor endurance. The genetic dominance factor contributed only to the variance of isokinetic lifting. Unique and common environmental factors had a substantially different influence on the back tests. Overall, these factors appeared to have the most influential role on psychophysical lifting force and isometric trunk extensor endurance time, collectively explaining 67% to 95% of the variance in test performance. Age, with explanatory percentages from 4% to 12%, had in comparison little or no influence on the variance of the back tests, similar to other suspected determinants and earlier findings by Gibbons et al3 for MZ twins, which were a subsample of the present data set.

Methodological Considerations

The structured interview allowed us to obtain detailed information on subjects' lifetime exposures to occupational and leisure-time physical loading from adolescence through adulthood; however, the degree to which the measurements accurately represent the potential determinants of interest is largely unknown and may have affected estimates of effect sizes. We did have knowledge of the reliability of data obtained for work and physical activity items from the structured interview36 as well as the repeatability of the 3 back tests.5 Another limitation is the sample size. A larger number of subjects would have provided more precise estimates and may have allowed a greater exploration of factors constituting environmental variance in tests.

The back muscle performance test results among the subjects of this study are similar to those reported by other researchers.21,38,39 The 3 back tests we used were commonly used in the early 1990s in the initial phase of our study data collection.35 Other back tests (such as submaximal tests and functional performance measures) are more commonly used today due to ongoing development and introduction of new tests and training devices and their wider availability. Another explanation for diminished use of the muscle performance tests may be the lack of usefulness in predicting other aspects of function or performance. It is likely that testing, training, and exercise protocols will continue to change over time. However, the tests used in our study were designed to examine a basic movement pattern (ie, lifting) and different modes of muscle testing (isometric and isokinetic). In addition, isometric trunk extensor endurance tests remain one of the few physical measures for adults without known pathology or limitations associated with future risk of LBP. The tests used in our study still have relevance to increasing our knowledge base of the determinants of these distinctly different types of back muscle performance tests.

There was an unexpected difference in the means of back test results by zygosity. Some of the difference was due to "zygosity effect": MZ twins are more similar than DZ twins. Although steps were taken to ensure comparable results of MZ twin testing in 1991–1993 and DZ twin testing in 1997–1999, such as the standardization of the test protocol and regular calibration of the testing device, some unidentified variations between the testing periods may have occurred. There were 2 testers during the 6-year period of testing. One tester tested those twins measured in 1991–1993, and the other tester tested those twins measured during 1997–1999. There was a possibility of a "tester-effect," where subjects might perform differently around one tester than around another tester. However, the 2 testers were the same sex and followed standardized testing procedures, which should have minimized possible differences due to testers.

Variance Component Estimation by Multivariate Twin Models for Back Tests

Genetic factors, we believe, had an influential role on isokinetic lifting force and work. The heritability or percentage of variance accounted for by genetic factors was 60% to 65%, as compared with 6% to 12% due to age. This was not unexpected by us because of the results of the study by Gibbons et al,3 who found familial aggregation to have a large influence on isokinetic lifting in the MZ twin sample. In our study, unlike that of Gibbons et al,3 we were able to investigate both the genetic and shared environmental components of familial aggregation and determine that isokinetic lifting performance is likely to be primarily determined by genetics, with common (childhood) environment having almost no contribution. The isokinetic lifting test requires fast and forceful performance, and it is the only test of the 3 back muscle performance tests that calls for maximal muscle contraction.

The proportion of variance of isokinetic lifting performance in the study subjects explained by unique environment (35% of force and 41% of work) might be related to training effects. It could be expected that those with work or leisure-time activities requiring maximal effort would have heightened isokinetic lifting performance as compared with those with lesser physical demands. In addition, lifting performance with maximal efforts may be less affected by environmental factors (but more effected by factors such as age) than other tests where effort is determined by submaximal or isometric contractions. Submaximal and isometric force levels have been associated with factors such as self-efficacy beliefs,13,15,33 sensation of pain,8,17,21,26,27 and experience with lifting performance.2628

One common genetic factor underlying the 3 back tests explained most of the covariance. A common genetic factor explaining the variance shared by the tests suggests the same additive genes account for the variation in the tests. A future step in investigating genetic determinants of isokinetic lifting force will be to search for some of the specific genes responsible. If the genes associated with the test results could be identified, this finding would have several implications. First, and foremost, it would help us better understand the biological basis (physiology) of variability in muscle function. Although much is already known, discovering new genes may reveal new physiologic mechanisms and metabolic pathways relevant in muscle force development. Eventually, knowledge of such genes may assist in planning rehabilitation and muscle force training programs. Such programs could be further tailored and targeted when we have information about gene-environment interactions (eg, knowledge of how those who have genes for "a strong back" benefit from different types of muscle force training as compared with those without the relevant gene forms). However, it is unlikely that only a single gene or a small number of genes will account for the observed genetic variation, and more likely there will be gradients of genetic susceptibility to specific training programs.

Accounting for Specific Back Tests in Relation to Specific Determinants

Environmental factors, such as physical activity and lifestyle, have been shown to influence isometric trunk force and trunk extensor endurance test results.10,15,21,32 Psychology and personality also have been shown to have an association with lifting force test results.15,30 These findings are in line with our results. We found that unique environmental factors accounted for most of the variance of psychophysical lifting and isometric trunk extensor endurance (Fig. 2). However, the proportions of variance of environmental factors explained by specific factors, such as occupational loading category, participating in aerobic activities and sports, and health, remained low (Tab. 6). There appear to be several factors affecting the environmental variance component, and it is difficult to derive the separate effect of each variable, especially when these variables seem to be associated. In addition, some of the environmental factors suspected of playing a role in environmental variance are known to have a genetic component. Physical activity32,49 and health,45 for example, have a genetic component, and, therefore, the impact of these factors can be attributed, in part, to a genetic effect. Weight, which had a substantial influence (8%–37%), also is known to have a genetic component.16

Unique environmental factors explained between 8% and 43% of the covariance among the 3 back tests. The highest environmental correlation (estimated correlation between environmental components based on the model), as expected, was between isokinetic lifting force and isokinetic lifting work (.70), which are part of the same performance because the force, speed, and height of the lift were used to calculate work done. The lower correlations between other back tests also were predictable. The nature of these tests is clearly different. The isokinetic lifting test is a maximal lifting effort and likely depends on individual biological factors, such as muscle fiber type, which has a demonstrated genetic component,46 and neuromotor function. In the psychophysical lifting test, the subject determines the effort, and the isometric trunk extensor endurance test may measure motivation, pain tolerance, and endurance more than back force. Similar environmental correlations for the 3 back tests were found by Gibbons et al3 (who studied a subsample of the present sample) with several different determinants such as physical loading, percentage of body fat, and back pain. Other researchers46 also have found low between-test correlations.


    Conclusions
 Top
 Abstract
 Introduction
 Method
 Results
 Discussion
 Conclusions
 References
 
The role of genetics was dominant in isokinetic lifting test performance and unique environmental factors had some effect. This high impact of genetics suggests that the isokinetic lifting test results reflect basic physical capacity for lifting. The relatively high impact of genetics also suggests challenges for potential interventions to alter isokinetic lifting back function. Psychophysical lifting was affected mostly by adulthood exposures, to some extent by genetics, and to a minor extent by childhood (common environment) influences. The role of genetics was minor in isometric trunk extensor endurance, which probably reflects a multitude of factors, such as general current health, physical activity as adults, and current body weight. In summary, these 3 tests measure very different attributes with different determinants, underlining the importance of careful test selection. For instance, the isometric trunk extensor endurance test could be used for screening effects of rehabilitation or back strengthening because the isometric trunk extensor endurance test is mostly affected by environmental factors that are possibly influenced by interventions.


    Footnotes
 
Ms Ropponen, Dr Videman, and Dr Battié provided concept/idea/research design. All authors provided writing. Ms Ropponen provided data collection, and Ms Ropponen, Mr Levälahti, and Dr Kaprio provided data analysis. Ms Ropponen, Dr Videman, and Dr Battié provided fund procurement. Dr Videman, Dr Kaprio, and Dr Battié provided project management and subjects. Dr Videman provided facilities/equipment and institutional liaisons

This study was approved by the ethical committees of the Department of Public Health, University of Helsinki, and the Faculty of Rehabilitation Medicine, University of Alberta.

* MIE Medical Research Limited, Leeds, United Kingdom. Back

{dagger} StataCorp LP, 4905 Lakeway Dr, College Station, TX 77845. Back


    References
 Top
 Abstract
 Introduction
 Method
 Results
 Discussion
 Conclusions
 References
 

  1. Mayer TG, Gatchel R, Betancur J, Bovasso E. Trunk muscle endurance measurement: isometric contrasted to isokinetic testing in normal subjects. Spine.1995; 20:920–927.[Web of Science][Medline]
  2. 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; 18:2491–2501.[Web of Science][Medline]
  3. Gibbons LE, Videman T, Battié MC. Determinants of isokinetic and psychophysical lifting strength and static back muscle endurance: a study of male monozygotic twins. Spine.1997; 22:2983–2990.[Web of Science][Medline]
  4. Hupli M, Sainio P, Hurri H, Alaranta H. Comparison of trunk strength measurements between two different isokinetic devices used at clinical settings. J Spinal Disord.1997; 10:391–397.[Web of Science][Medline]
  5. Latikka P, Battié MC, Videman T, Gibbons LE. Correlations of isokinetic and psychophysical back lift and static back endurance tests in men. Clin Biomech.1995; 10:325–330.
  6. Smith SS, Mayer TG, Gatchel RJ, Becker TJ. Quantification of lumbar function, part 1: isometric and multispeed isokinetic trunk strength measures in sagittal and axial planes in normal subjects. Spine.1985; 10:757–764.[Web of Science][Medline]
  7. Troup JD, Foreman TK, Baxter CE, Brown D. The perception of back pain and the role of psychophysical tests of lifting capacity. Spine.1987; 12:645–657.[Web of Science][Medline]
  8. Keller A, Johansen JG, Hellesnes J, Brox JI. Predictors of isokinetic back muscle strength in patients with low back pain. Spine.1999; 24:275–280.[Web of Science][Medline]
  9. Mostardi RA, Noe DA, Kovacik MW, Porterfield JA. Isokinetic lifting strength and occupational injury: a prospective study. Spine.1992; 17:189–193.[Web of Science][Medline]
  10. Holmström E, Moritz U, Andersson M. Trunk muscle strength and back muscle endurance in construction workers with and without low back disorders. Scand J Rehabil Med.1992; 24:3–10.[Web of Science][Medline]
  11. Kankaanpää M, Laaksonen D, Taimela S, et al. Age, sex, and body mass index as determinants of back and hip extensor fatigue in the isometric Sörensen back endurance test. Arch Phys Med Rehabil.1998; 79:1069–1075.[Web of Science][Medline]
  12. Battié MC, Bigos SJ, Fisher LD. Isometric lifting strength as a predictor of industrial back pain reports. Spine.1989; 14:851–856.[Web of Science][Medline]
  13. Estlander AM, Vanharanta H, Moneta GB, Kaivanto K. Anthropometric variables, self-efficacy beliefs and pain and disability ratings on the isokinetic performance on low back pain patients. Spine.1994; 19:941–947.[Web of Science][Medline]
  14. Jerome JA, Hunter K, Gordon P, McKay N. A new robust index for measuring isokinetic trunk flexion and extension: outcome from a regional study. Spine.1991; 16:804–808.[Web of Science][Medline]
  15. Malchaire JB, Masset DF. Static and dynamic performances of the trunk and associated factors. Spine.1995; 20:1649–1656.[Web of Science][Medline]
  16. Bouchard C. Heredity and the path to overweight and obesity. Med Sci Sports Exerc.1991; 23:285–291.[Web of Science][Medline]
  17. Taimela S, Kankaanpää M, Airaksinen O. A submaximal back extension endurance test utilizing subjective perception of low back fatigue. Scand J Rehabil Med.1998; 30:107–112.[Web of Science][Medline]
  18. Latimer J, Maher CG, Refshauge K, Colaco I. The reliability and validity of the Biering-Sorensen test in asymptomatic subjects and subjects reporting current or previous nonspecific low back pain. Spine.1999; 24:2085–2090.[Web of Science][Medline]
  19. Leino P, Aro S, Hasan J. Trunk muscle function and low back disorders: a ten-year follow-up study. J Chronic Dis.1987; 40:289–296.[Web of Science][Medline]
  20. Luoto S, Heliövaara M, Hurri H, Alaranta H. Static back endurance and the risk of low-back pain. Clin Biomech.1995; 10:323–324.
  21. Jörgensen K, Nicolaisen T. Trunk extensor endurance: determination and relation to low-back trouble. Ergonomics.1987; 30:259–267.[Medline]
  22. Suni J, Oja P, Miilunpalo S, et al. Health-related fitness test battery for adults: associations with perceived health, mobility, and back function and symptoms. Arch Phys Med Rehabil.1998; 79:559–569.[Web of Science][Medline]
  23. Hupli M, Hurri H, Luoto S, et al. Isokinetic performance capacity of trunk muscles, part I: the effect of repetition on measurement of isokinetic performance capacity of trunk muscles among healthy controls and two different groups of low back pain patients. Scand J Rehabil Med.1996; 28:201–206.[Web of Science][Medline]
  24. Rantanen P. Physical measurements and questionnaires as diagnostic tools in chronic low back pain. J Rehabil Med.2001; 33:31–35.[Web of Science][Medline]
  25. Lindström I, Öhlund C, Nachemson A. Physical performance, pain, pain behavior and subjective disability in patients with subacute low back pain. Scand J Rehabil Med.1995; 27:153–160.[Web of Science][Medline]
  26. Lethem J, Slade PD, Troup JDG, Bentley G. Outline of a fear-avoidance model of exaggerated pain perception: I. Behav Res Ther.1983; 21:401–408.[Web of Science][Medline]
  27. Slade PD, Troup JDG, Lethem J, Bentley G. The fear-avoidance model of exaggerated pain perception, II: preliminary studies of coping strategies for pain. Behav Res Ther.1983; 21:409–416.[Web of Science][Medline]
  28. Crombez G, Vervaet L, Lysens R, et al. Avoidance and confrontation of painful, back-straining movements in chronic back pain patients. Behav Modif.1998; 22:62–77.[Abstract/Free Full Text]
  29. Rissanen A, Alaranta H, Sainio P, Härkönen H. Isokinetic and non-dynamometric tests in low back pain patients related to pain and disability index. Spine.1994; 19:1963–1967.[Web of Science][Medline]
  30. Lackner JM, Carosella AM. The relative influence of perceived pain control, anxiety, and functional self-efficacy on spinal function among patients with chronic low back pain. Spine.1999; 24:2254–2261.[Web of Science][Medline]
  31. Grönblad M, Hurri H, Kouri JP. Relationship between spinal mobility, physical performance tests, pain intensity and disability assessments in chronic low back pain patients. Scand J Rehabil Med.1997; 29:17–24.[Web of Science][Medline]
  32. Beunen GP, Thomis MAI. Genetic determinants of sports participation and daily physical activity. Int J Obes.1999; 23:S55–S63.[Web of Science]
  33. Hazard RG, Reeves VM, Fenwick JW, et al. Test-retest variation in lifting capacity and indices of subject effort. Clin Biomech.1993; 8:20–24.
  34. Kaprio J, Sarna S, Koskenvuo M, Rantasalo I. The Finnish Twin Registry: formation and compilation, questionnaire study, zygosity determination procedures, and research program. Prog Clin Biol Res.1978; 24(pt B):179–184.
  35. Battié MC, Videman T, Gibbons LE, et al. Determinants of lumbar disc degeneration: a study relating lifetime exposures and MRI findings in identical twins. Spine.1995; 21:2601–2612.
  36. Ropponen A, Levälahti E, Simonen R, et al. Repeatability of lifetime exercise reporting. Scand J Med Sci Sports.2001; 11:185–192.[Web of Science][Medline]
  37. Dvir Z. Clinical applicability of isokinetics: a review. Clin Biomech.1991; 6:133–144.
  38. Foreman TK, Baxter CE, Troup JDG. Ratings of acceptable load and maximal isometric lifting strengths: the effects of repetition. Ergonomics.1984; 27:1283–1288.
  39. Alaranta H, Soukka A, Harju R, Heliövaara M. Developing the Techniques Used for Diagnosing Musculoskeletal Diseases [in Finnish]. Helsinki, Finland: Finnish National Social Insurance Institution;1990 :45–47.
  40. Van Loan MD. Bioelectrical impedance analysis to determine fat-free mass, total body water and body fat. Sports Med.1990; 10:205–217.[Web of Science][Medline]
  41. Janssen I, Heymsfield SB, Baumgartner RN, Ross R. Estimation of skeletal muscle mass by bioelectric impedance analysis. J Appl Physiol.2000; 89:465–471.[Abstract/Free Full Text]
  42. Broeder CE, Burrhus KA, Svanevik LS, et al. Assessing body composition before and after resistance or endurance training. Med Sci Sport Exerc.1997; 29:705–712.[Web of Science][Medline]
  43. Ravaglia G, Forti P, Maioli F, Boschi F, et al. Measurement of body fat in healthy elderly men: a comparison of methods. J Gerontol, A Biol Sci Med Sci.1999; 54:M70–M76.[Abstract]
  44. Stata Statistical Software: Release 6.0. College Station, Tex: Stata Corp;1999 .
  45. Rao JNK, Scott AJ. The analysis of categorical data from complex sample surveys: chi-squared tests for goodness of fit and independence in two-way tables. Journal of the American Statistical Association.1981; 76:221–230.
  46. Neale MC. Mx: Statistical Modeling. 2nd ed. Richmond, Va: Department of Human Genetics, Medical College of Virginia;1994 .
  47. Simoneau J, Bouchard C. Genetic determinism of fiber type proportion in human skeletal muscle. FASEB J.1995; 9:1091–1095.[Abstract]
  48. Neale MC, Cardon LR. Methodology for Genetic Studies of Twins and Families. Dordrecht, the Netherlands: Kluwer Academic;1992 .
  49. Lauderdale DS, Fabsitz R, Meyer JM, et al. Familial determinants of moderate and intense physical activity: a twin study. Med Sci Sports Exerc.1997; 29:1062–1068.[Web of Science][Medline]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
J. Appl. Physiol.Home page
M. C. Battie, E. Levalahti, T. Videman, K. Burton, and J. Kaprio
Heritability of lumbar flexibility and the role of disc degeneration and body weight
J Appl Physiol, February 1, 2008; 104(2): 379 - 385.
[Abstract] [Full Text] [PDF]


This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when Rapid Responses are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ropponen, A.
Right arrow Articles by Battié, M. C
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ropponen, A.
Right arrow Articles by Battié, M. C
Related Collections
Right arrow Musculoskeletal System/Orthopedic: Other
Right arrow Tests and Measurements
Social Bookmarking
 Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Copyright © 2004 by the American Physical Therapy Association.