Most commonlyassociated with the term is a speci®c test developed to provide a relatively quick assessment of a subject's maximal lifting capacity using a modi®ed weight-lifting device [
Trang 16 Extent of strength training done by
partici-pants, and their experience with isometric
test-ing
7 Health status of participants (medical exam
and/or health questionnaire recommended
3.2.14 Strength Data Reporting
The minimum data which should be reported for
strength-testing projects are:
1 Mean, median, and mode of data set
2 Standard deviation of data set
3 Skewness of data set (or histogram describing
data set)
4 Minimum and maximum values
3.2.15 Evaluation According to Physical
Assessment Criteria
A set of ®ve criteria have been purposed to evaluate the
utility of all forms of strength testing Isometric
strength testing is evaluated with respect to these
criteria in the following sections
3.2.15.1 Is It Safe to Administer?
Any form of physical exertion carries with it some risk
The directions for the person undergoing an isometric
test speci®cally state that the person is to slowly
increase the force until they reach what they feel is a
maximum, and to stop if at any time during the
exer-tion they feel discomfort or pain The direcexer-tions also
expressly forbid jerking on the equipment When
iso-metric testing is performed in this manner it is quite
safe to administer because the tested person is deciding
how much force to apply, over what time interval, and
how long to apply it The only known complaints
relat-ing to participation in isometric testrelat-ing are some
resi-dual soreness in the muscles which were active in the
test(s), and this is rarely reported
3.2.15.2 Does the Method Provide Reliable
Quantitative Values?
The test-retest variability for isometric testing is 5±
10% In the absence of a speci®c strength training
program, individual isometric strength remains tively stable over time When the number of trials isbased on the 10% criterion discussed earlier, therecorded strength is near or at the tested person'smaximum voluntary strength Assuming the abovefactors, and that test postures are properly con-trolled, isometric strength testing is highly reliableand quantitative
rela-3.2.15.3 Is Method Practical?
Isometric strength testing has already been usedsuccessfully in industry for employee placement, inlaboratories for the collection of design data, and inrehabilitation facilities for patient progress assessment.3.2.15.4 Is the Method Related to Speci®c Job
Requirements (Content Validity)?
Isometric strength testing can be performed in anyposture When it is conducted for employee placementpurposes, the test postures should be as similar as pos-sible to the postures that will be used on the job Theforce vector applied by the tested person should also besimilar to the force vector that will be applied on thejob When these two criteria are met, isometricstrength testing is closely related to job requirements.However, it should be noted that results obtained usingisometric strength testing loses both content and criter-ion-related validity as job demands become moredynamic
3.2.15.5 Does the Method Predict the Risk of
Future Injury or Illness?
A number of researchers have demonstrated that metric strength testing does predict risk of future injury
iso-or illness fiso-or people on physically stressful job [16,17].The accuracy of this prediction is dependent on thequality of the job evaluation on which the strengthtests are based, and the care with which the tests areadministered
3.3 PART II: MAXIMAL ISOINERTIALSTRENGTH TESTING
3.3.1 De®nition of Isoinertial StrengthKroemer [18±20] and Kroemer et al [4] de®ne the iso-inertial technique of strength assessment as one inwhich mass properties of an object are held constant,
as in lifting a given weight over a predetermined tance Several strength assessment procedures possess
Trang 2dis-the attribute in this de®nition Most commonly
associated with the term is a speci®c test developed
to provide a relatively quick assessment of a subject's
maximal lifting capacity using a modi®ed weight-lifting
device [18,21] The classic psychophysical methodology
of assessing maximum acceptable weights of lift is also
as an isoinertial technique under this de®nition [12]
While the de®nition provided by Kroemer [18] and
Kroemer et al [4] has been most widely accepted in the
literature, some have applied the term ``isoinertial'' to
techniques that dier somewhat from the de®nition
given above, such as in a description of the
Isotechnologies B-200 strength-testing device [22]
Rather than lifting a constant mass, the B-200 applies
a constant force against which the subject performs an
exertion The isoinertial tests described in this chapter
apply to situations in which the mass to be moved by a
musculoskeletal eort is set to a constant
3.3.2 Is Isoinertial Testing Psychophysical or Is
Psychophysical Testing Isoinertial?
As various types of strength tests have evolved over the
pasts few decades, there have been some unfortunate
developments in the terminology that have arisen to
describe and/or classify dierent strength assessment
procedures This is particularly evident when one
tries to sort out the various tests that have been
labelled ``isoinertial.'' One example was cited above
Another problem that has evolved is that the term
``isoinertial strength'' has developed two dierent
connotations The ®rst connotation is the conceptual
de®nitionÐisoinertial strength tests describe any
strength test where a constant mass is handled
However, in practice, the term is often used to denote
a speci®c strength test where subjects' maximal lifting
capacity is determined using a machine where a
con-stant mass is lifted [18,21] Partially as a result of this
dual connotation, the literature contains both
refer-ences to ``isoinertial strength test'' as a psychophysical
variant [23], and to the psychophysical method as an
``isoinertial strength test'' [4,24] In order to lay the
framework for the next two parts, the authors would
like to brie¯y discuss some operational de®nitions of
tests of isoinertial and psychophysical strength
When Ayoub and Mital [23] state that the isoinertial
strength test is a variant of the psychophysical method,
they refer to the speci®c strength test developed by
Kroemer [18] and McDaniel et al [21] Clearly, this
isoinertial protocol has many similarities to the
psy-chophysical method: both are dynamic; weight is
adjusted in both; both measure the load a subject is
willing to endure under speci®ed circumstances, etc.However, while both deal with lifting and adjustingloads, there are signi®cant dierences between the psy-chophysical (isoinertial) technique and the Kroemer±McDaniel (isoinertial) protocol, both procedurally and
in use of the data collected in these tests For purposes
of this chapter we will designate the Kroemer±McDaniel protocol maximal isoinertial strength tests(MIST) This part deals with the latter isoinertial tech-nique, which diers from the psychophysical technique
on the following counts:
1 In maximal isoinertial strength tests, theamount of weight lifted by the subject is system-atically adjusted by the experimenter, primarilythrough increasing the load to the subject's max-imum In contrast, in psychophysical tests,weight adjustment is freely controlled by the sub-ject, and may be upwards or downwards
2 The maximal isoinertial strength tests discussed
in this part are designed to quickly establish anindividual's maximal strength using a limitednumber of lifting repetitions, whereas psycho-physical strength assessments are typically per-formed over a longer duration of time (usually atleast 20 min), and instructions are that the sub-ject select an acceptable (submaximal) weight oflift, not a maximal one Due to the typicallylonger duration of psychophysical assessments,greater aerobic and cardiovascular componentsare usually involved in the acceptable workloadchosen
3 Isoinertial strength tests have traditionally beenused as a worker selection tool (a method ofmatching physically capable individuals todemanding tasks) A primary focus of psycho-physical methods has been to establish data thatcan be used for the purpose of ergonomic jobdesign [12]
3.3.3 Published DataThere are two primary maximal isoinertial strength testprocedures that will be described in this section Oneinvolves the use of a modi®ed weight-lifting machinewhere the subject lifts a rack of hidden weights to pre-scribed heights, as depicted inFig 4[21] Kroemer [18]refers to his technique as LIFTEST, while the AirForce protocol has been named the strength aptitudetest (SAT) The other test uses a lifting box, into whichweights are placed incrementally at speci®ed timesuntil the lifting limit is reached [25] The greatest
Trang 3phases: (1) a powerful upward pulling phase, where
maximal acceleration, velocity, and power values are
observed; (2) a wrist changeover manoeuvre (at
approximately elbow height), where momentum is
required to compensate for low force and acceleration;
and (3) a pushing phase (at or above chest height),
characterized by a secondary (lower) maximal force
and acceleration pro®le
The analysis by Stevenson et al [28] suggested that
successful performance of the criterion shoulder height
lift requires a technique quite dierent from the
con-cept of slow, smooth lifting usually recommended for
submaximal lifting tasks On the contrary, lifting of a
maximal load requires a rapid and powerful lifting
motion This is due in large part to the need to develop
sucient momentum to allow successful completion of
the wrist changeover portion of the lift Most lift
fail-ures occur during the wrist changeover procedure,
probably the result of poor mechanical advantage of
the upper limb to apply force to the load at this point
in the lift [28] Stevenson et al [28] found that certain
anatomical landmarks were associated with maximal
force, velocity, and power readings (see Fig 5)
Maximal force readings were found to occur at
mid-thigh, maximal velocity at chest height, minimum force
was recorded at head height, and the second maximal
acceleration (pushing phase) was observed at 113% of
the subject's stature
3.3.5 The Strength Aptitude Test
The strength aptitude test (SAT) [21] is a classi®cation
tool for matching the physical strength abilities of
individuals with the physical strength requirements
of jobs in the Air Force (McDaniel, personal
commu-nication, 1994) The SAT is given to all Air Force
recruits as part of their preinduction examinations
Results of the SAT are used to determine whether
the individual tested possesses the minimum strength
criterion which is a prerequisite for admission to
var-ious Air Force specialties (AFSs) The physical
demands of each AFS are objectively computed
from an average physical demand weighted by the
frequency of performance and the percent of the
AFS members performing the task Objects weighing
less than 10 lb are not considered physically
demand-ing and are not considered in the job analysis Prior
to averaging the physical demands of the AFS, the
actual weights of objects handled are converted into
equivalent performance on the incremental weight lift
test using regression equations developed over years
of testing These relationships consider the type of
task (lifting, carrying, pushing, etc.), the size andweight of the object handled, as well as the typeand height of the lift Thus, the physical job demandsare related to, but are not identical to, the ability tolift an object to a certain height Job demands forvarious AFSs are reanalyzed periodically for purposes
of updating the SAT
The ®rst major report describing this classi®cationtool was a study of 1671 basic trainees (1066 males and
605 females) [21] The incremental weight lift testsstarted with a 18.1 kg weight which was to be raised
to 1.83 m or more above the ¯oor This initial weightwas increased in 4.5 kg increments until subjects wereunable to raise the weight to 1.83 m Maximal weight
Figure 5 Analysis of the shoulder height strength test cates three distinct lift phases: (1) a powerful upward pullingphase (where maximal forces are developed), (2) a wrist chan-geover maneuver (where most failures occur), and (3) a push-ing phase (where a secondary, lower, maximal force isobserved)
Trang 4indi-lift to elbow height was then tested as a continuation of
the incremental weight lift test In the test of lifting the
weight to 1.83 m, males averaged 51.8 kg (10.5), while
females averaged 25.8 kg (5.3) The respective
weights lifted to elbow height were 58.6 kg (11.2)
and 30.7 kg ( 6.3) The distributions of weight lifting
capabilities for both male and female basic trainees in
lifts to 6 ft are provided in Fig 6 Results of the elbow
height lift are presented in Table 1 McDaniel et al [21]
also performed a test of isoinertial endurance This
involved holding a 31.8 kg weight at elbow height for
the duration the subject could perform the task Male
basic trainees were able to hold the weight for an
aver-age of 53.3 sec (22.11), while female basic trainees
managed to hold the weight an average of 10.3 sec
(10.5 SD)
When developing the SAT, the Air Force examined
more than 60 candidate tests in an extensive, four-year
research program and found the incremental weight lift
to 1.83 m to be the single test of overall dynamic
strength capability, which was both safe and reliable
(McDaniel, personal communication 1994) This
®nd-ing was con®rmed by an independent study funded by
the U.S Army [29] This study compared the SAT to a
battery of tests developed by the Army (including
iso-metric and dynamic tests) and compared these with
representative heavy demand tasks performed within
the Army Results showed the SAT to be superior to
all others in predicting performance on the criterion
tasks
Figure 6 Distribution of weight-lifting capabilities for male and female basic trainees for lifts to 6 ft (From Ref 21.)
Table 1 Weight-Lifting Capabilities of Basic Trainees forLifts to Elbow Height
Trang 53.3.6 Virginia Tech Data
Kroemer [18,20] described results of a study using a
similar apparatus as the one used by the U.S Air
Force The sample consisted of 39 subjects (25 male)
recruited from a university student population The
procedures were similar to McDaniel et al [21] with
the exception that the minimum starting weight was
11.4 kg, and that maximal lifting limits were
estab-lished to prevent overexertion These were 77.1 kg for
¯oor to knuckle height tests, and 45.4 for ¯oor to
over-head reach tests The following procedure was used for
establishing the maximal load: if the initial 11.4 kg
weight was successfully lifted, the weight was doubled
to 22.7 kg Additional 11.4 kg increments were added
until an attempt failed or the maximal lifting limit was
reached If an attempt failed, the load was reduced by
6.8 kg If this test weight was lifted, 4.5 kg was added; if
not, 2.3 kg were subtracted This scheme allowed quick
determination of the maximal load the subject could
lift
In Kroemer's study, six of 25 male subjects exceeded
the cuto load of 100 lb in overhead reach lifts [18,20]
All 14 females stayed below this limit The 19
remain-ing male subjects lifted an average of 27 kg The female
subjects lifted an average of 16 kg In lifts to knuckle
height, 17 of the 25 male (but none of the female)
subjects exceeded the 77.1 kg cuto limit The
remain-ing subjects lifted an average of about 54 kg, with
males averaging 62 kg and females 49 kg The
coe-cients of variation for all tests were less than 8%
Summary data for this study is given in Table 2
3.3.7 The Progressive Isoinertial LiftingEvaluation
Another variety of MIST has been described by Mayer
et al [25,30] Instead of using a weight 3 rack as shown
in Fig 3, the progressive isoinertial, lifting valuation(PILE) is performed using a lifting box with handlesand increasing weight in the box as it is lifted andlowered Subjects perform two isoinertial lifting/low-ering tests: one from ¯oor to 30 in (Lumbar) andone from 30 to 54 in (Cervical) Unlike the isoinertialprocedures described above, there are three possiblecriteria for termination of the test: (1) voluntary termi-nation due to fatigue, excessive discomfort, or inability
to complete the speci®ed lifting task; (2) achievement
of a target heart rate (usually 85% of age predictedmaximal heart rate); or (3) when the subject lifts a
``safe limit'' of 55±60% of his or her body weight.Thus, contrary to the tests described above, the PILEtest may be terminated due to cardiovascular factors,rather than when an acceptable load limit is reached.Since the PILE was developed as a means of evalu-ating the degree of restoration of functional capacity ofindividuals complaining of chronic low-back pain(LBP), the initial weight lifted by subjects using thisprocedure is somewhat lower than the tests describedabove The initial starting weight is 3.6 kg for womenand 5.9 kg for men Weight is incremented upwards at
a rate of 2.3 kg every 20 sec for women, and 4.6 kgevery 20 sec for men During each 20 sec period, fourlifting movements (box lift or box lower) are per-formed The lifting sequence is repeated until one of
Table 2 Results of Lifts to Shoulder and Knuckle Height for 25 Male and 14 Female Subjects
Trang 6the three endpoints is reached The vast majority of
subjects are stopped by the ``psychophysical''
end-point, indicating the subject has a perception of fatigue
or overexertion The target heart rate endpoint is
typi-cally reached in older or large individuals The ``safe
limit'' endpoint is typically encountered only by very
thin or small individuals
Mayer et al [25] developed a normative database
for the PILE, consisting of 61 males and 31 females
Both total work (TW) and force in pounds (F) were
normalized according to age, gender, and a body
weight variable The body weight variable, the adjusted
weight (AW), was taken as actual body weight in slim
individuals, but was taken as the ideal weight in
over-weight individuals This was done to prevent skewing
the normalization in overweight individuals Table 3
presents the normative database for the PILE
3.3.8 Evaluation According to Criteria for
Physical Assessment
3.3.8.1 Is It Safe to Administer?
The MIST procedures described above appear to
have been remarkably free of injury Isoinertial
pro-cedures have now been performed many thousands
of times without report of veri®able injury However,
reports of transitory muscle soreness have been
noted [25] The temporary muscle soreness associated
with isoinertial testing has been similar to that
experienced in isokinetic tests, but has been reported
less frequently than that experienced with isometric
strength tests
McDaniel et al [21] present some useful dations for design of safe isoinertial weight-lift testingprocedures The following list summarizes the recom-mendations made by these authors
recommen-1 Weight-lifting equipment should be designed
so that the weights and handle move only in
a vertical direction
2 Sturdy shoes should be worn; or the subjectmay be tested barefoot Encumbering clothingshould not be worn during the test
3 The initial weight lifted should be low: 20±
40 lb Weights in this range are within the ability of almost everyone Weight incrementsshould be small
cap-4 The upper limit should not exceed the largestjob related requirement or 160 lb, whichever isless
5 The starting handle position should be 1±2 ftabove the standing surface If the handle islower, the knees may cause obstruction Ifthe handle is too high, the subject will squat
to get their shoulders under it prior to lifting
A gap between the handles will allow them topass outside the subject's knees when lifting,allowing a more erect back and encouragingthe use of leg strength
6 The recommended body orientation prior tolifting should be (a) arms straight at theelbow, (b) knees bent to keep the trunk aserect as possible, and (c) head aligned withthe trunk The lift should be performedsmoothly, without jerk
Table 3 Normative Data
Trang 77 A medical history of the subject should be
obtained If suspicious physical conditions
are identi®ed, a full physical examination
should be performed prior to testing
Subjects over 50 years of age or pregnant
should always have a physical prior to
testing
8 All sources of overmotivation should be
mini-mized Testing should be done in private and
results kept con®dential Even the test subject
should not be informed until the testing is
completed
9 If the subject pauses during a lift, the strength
limit has been reached, and the test should be
terminated Multiple attempts at any single
weight level should not be allowed
10 The testing should always be voluntary The
subject should be allowed to stop the test at
any time The subject should not be
informed of the criteria prior to or during
the test
It is noteworthy that, as of 1994, over two million
subjects have been tested on the SAT without any
back injury or overexertion injury (McDaniel, personal
communication, 1994)
3.3.8.2 Does It Give Reliable, Quantitative
Values?
Kroemer et al [20] reported LIFTEST coecients of
variation (measures of intraindividual variability in
repeated exertions) of 3.5 for all subjects in overhead
lifts, and 6.9 in lifts to knuckle height The same study
showed somewhat higher variability in tests of
iso-metric strength (coecient of variations ranging from
11.6 to 15.4) Test±retest reliability was not reported by
McDaniel et al [21] Mayer et al [25] reported
correla-tion coecients of a reproducibility study of the PILE
which demonstrated good test±retest reliability for
both ¯oor to 30 in lifts (r 0:87, p < 0:001) and 30±
54 in lifts (r 0:93, p < 0:001) Thus, the reliability of
isoinertial procedures appears to compare favorably
with that demonstrated by other strength assessment
techniques
3.3.8.3 Is It Practical?
Isoinertial techniques generally appear practical in
terms of providing a test procedure that requires
minimal administration time and minimal time for
instruction and learning Even in a worst case nario, the isoinertial procedures used by Kroemerz[2] would take only a few minutes to determine themaximal weight lifting capability of the subject for aparticular condition The McDaniel et al [21](McDaniel, personal communication, 1994) procedurecan be performed in approximately 3±5 min ThePILE test administration time is reported to last onthe order of 5 min [25]
sce-Practicality is determined in part by cost of theequipment required, and on this account, the cost ofisoinertial techniques is quite modest In fact, thePILE test requires no more hardware than a liftingbox and some sturdy shelves, and some weight Theequipment needed to develop the LIFTEST devicesused by McDaniel et al [21] and Kroemer [18±20]would be slightly more expensive, but would not beprohibitive for most applications In fact, Kroemer[19] states that the device is easily dismantled andcould easily be transported to dierent sites in asmall truck or station wagon, or perhaps in a mobilelaboratory vehicle
3.3.8.4 Is It Related to Speci®c Job
Requirements?
Since industrial lifting tasks are performed cally, isoinertial strength tests do appear to providesome useful information related to an individual's abil-ity to cope with the dynamic demands of industriallifting McDaniel (personal communication, 1994)has reported that these tests are predictive of perfor-mance on a wide range of dynamic tasks, includingasymmetrical tasks, carrying, and pushing tasks.Furthermore, Jiang et al [26] demonstrated that theisoinertial lifting test to 6 ft was more highly correlatedwith psychophysical tests of lifting capacity thanisometric techniques The PILE test possesses goodcontent validity for industrial lifting tasks, as subjectsare able to use a more ``natural'' lifting technique whenhandling the lifting box
dynami-3.3.8.5 Does It Predict Risk of Future Injury or
Illness?
The ability of a strength test to predict risk of futureinjury or illness is dependent upon performance ofprospective epidemiological studies As of this writing,
no such studies have been conducted on the isoinertialtechniques described above
Trang 83.4 PART III: PSYCHOPHYSICAL
STRENGTH
3.4.1 Theory and Description of the
Psychophysical Methodology for
Determining Maximum Acceptable
Weights and Forces
According to contemporary psychophysical theory, the
relationship between the strength of a perceived
sensa-tion (S) and the intensity of a physical stimulus (I) is
best expressed by a power relationship [31]:
This psychophysical principle has been applied to
many practical problems, including the development
of scales or guidelines for eective temperature,
loud-ness, brightloud-ness, and ratings of perceived exertion
Based on the results of a number of experiments
using a variety of scaling methods and a number of
dierent muscle groups, the pooled estimate the
expo-nent for muscular eort and force is 1.7 [32]
When applying this principle to work situations, it is
assumed that individuals are capable and willing to
consistently identify a speci®ed level of perceived
sen-sation (S) For manual materials handling tasks, this
speci®ed level is usually the maximum acceptable
weight or maximum acceptable force The meaning of
these phrases are de®ned by the instructions given to
the test subject [33] ``You are to work on an incentive
basis, working as hard as you can without straining
yourself, or becoming unusually tired, weakened,
over-heated, or out of breath.''
If the task involves lifting, the experiment measures
the maximum acceptable weight of lift Similarly,
there are maximum acceptable weights for lowering
and carrying Such tests are isoinertial in nature;
how-ever, in contrast to the tests described in Part 2, they
are typically used to test submaximal, repetitive
hand-ling capabilities Data are also available for pushing
and pulling These are reported as maximum
accepta-ble forces and include data for initial as well as
sus-tained pulling or pushing
3.4.2 Why Use Psychophysical Methods?
Snook identi®ed several advantages and disadvantages
to using psychophysical methods for determining
maximum acceptable weights [34] The advantages
a fair day's pay.''
4 The results are reproducible
5 The results appear to be related to low-backpain (content validity)
Disadvantages include:
1 The tests are performed in a laboratory
2 It is a subjective method that relies on reporting by the subject
self-3 The results for very high-frequency tasks mayexceed recommendations for energy expendi-ture
4 The results are insensitive to bending and ing
twist-In terms of the application of the data derived fromthese studies, Liberty Mutual preferred to use it todesign a job to ®t the worker, since this applicationrepresented a more permanent, engineering solution
to the problem of low-back pain in industry [12].This approach not only reduces the worker's exposure
to potential low-back pain risk factors, but alsoreduces liability associated with worker selection [12]
3.4.3 Published Data3.4.3.1 Liberty MutualSnook and Ciriello at the Liberty Mutual InsuranceCompany have published the most comprehensivetables for this type of strength assessment [35] Themost recent data is summarized in nine tables, orga-nized as follows [35]:
1 Maximum acceptable weight of lift for males
2 Maximum acceptable weight of lift for females
3 Maximum acceptable weight of lower for males
4 Maximum acceptable weight of lower forfemales
5 Maximum acceptable forces of push for males(initial and sustained)
6 Maximum acceptable forces of push for females(initial and sustained)
7 Maximum acceptable forces of pull for males(initial and sustained)
8 Maximum acceptable forces of pull for females(initial and sustained)
9 Maximum acceptable weight of carry (malesand females)
Trang 93.4.3.2 Other Sources
Ayoub et al [36] and Mital [37] have also published
tables for maximum acceptable weights of lift Even
though their tables are similar in format and generally
in agreement with those from Liberty Mutual, there
are some dierences Possible sources for these
dier-ences may be dierdier-ences in test protocol, dierdier-ences in
task variables, and dierences in subject populations
and their characteristics
3.4.4 Experimental Procedures and Methods
For the sake of simplicity and convenience, the Liberty
Mutual protocol for lifting or lowering and an excerpt
from the lifting table will be used as examples for this
section The protocols used by Ayoub et al [36] and
Mital [37] were similar, but not exactly the same The
reader should refer to the original publications for
details
The Liberty Mutual experimental procedures and
methods were succinctly reviewed in their most recent
revision of the table [35] The data reported in these
revised tables re¯ect results from 119 second shift
workers from local industry (68 males, 51 females)
All were prescreened to ensure good health prior to
participation These subjects were employed by
Liberty Mutual for the duration of the project
(usually 10 weeks) All received 4±5 days of
condi-tioning and training prior to participation in actual
test sessions
Test subjects wore standardized clothing and shoes
The experiments were performed in an environmental
chamber maintained at 218C (dry bulb) and 45%
rela-tive humidity Forty-one anthropometric variables
were recorded for each subject, including several
iso-metric strengths and aerobic capacity
A single test session lasted approximately 4 h and
consisted of ®ve dierent tasks Each task session
lasted 40 min, followed by 10 min rest Most subjects
participated in at least two test sessions per week for 10
weeks In general, a subject's heart rate and oxygen
consumption were monitored during the sessions
3.4.4.1 Lifting or Lowering Tasks
In a lifting or lowering task session, the subject was
given control of one variable, usually the weight of
the box The other task variables would be speci®ed
by the experimental protocol These variables
include:
1 Lifting zone, which refers to whether the liftoccurs between ¯oor level to knuckle height(low), knuckle height to shoulder height(center), or shoulder height to arm reach(high)
2 Vertical distance of lift, which refers to the tical height of the lift within one of these liftingzones The speci®ed values for distance of lift inthe tables are 25 cm (10 in.), 51 cm (20 in.), and
ver-76 cm (30 in.) It is possible to use linear polation for lift distances not exactly equal toone of these values
extra-3 Box width, which refers to the dimension of thebox away from the body The three values ofbox width are 34 cm (13.4 in.), 49 cm (19.3 in.),and 75 cm (29.5 in.) It is possible to use linearextrapolation between these values
4 Frequency of lift, expressed as one lift per timeinterval, and include intervals of 5 sec, 9 sec,
14 sec, 1 min, 2 min, 5 min and 8 hr
These same de®nitions apply to a lowering task, exceptthe word ``lower'' is substituted for ``lift.'' The testprotocol for lowering was essentially identical to thatfor lifting, and the results are reported in a similarformat It should be noted, however, that the test pro-tocols for lifting and lowering involved using a specialapparatus that returned the box to its original speci®edlocation, so that the subject only lifted or lowered, notboth
Per the instructions, the subject was to adjust theweight of the box, according to his or her own percep-tions of eort or fatigue, by adding or removing steelshot or welding rods from a box The box had handlesand a false bottom to eliminate visual cues Each taskexperiment was broken into two segments so that theinitial weight of the box could be randomly variedbetween high versus low so that the subjectapproached his or her maximum acceptable weightfrom above as well as below If the results met a15% test±retest criterion, the reported result was theaverage of these two values If the results did not meetthis criterion, they were discarded and the test repeated
at a later time
In reporting the results, it was assumed thatthe gender-speci®c maximum acceptable weights for aparticular task were normally distributed As a con-sequence, the results were reported as percentages ofpopulation, strati®ed by gender The Liberty Mutualtables are organized around the following percentages:90%, 75%, 50%, 25%, and 10% [35] The 90th per-centile refers to a value of weight that 90% of indivi-
Trang 10duals of that gender would consider a maximum
accep-table weight (90% ``accepaccep-table''), while the 10th
per-centile refers to a value of weight that only 10% of
individuals of that gender would ®nd acceptable
(10% ``acceptable'')
3.4.5 Important Caveats
Snook and Ciriello have identi®ed several important
caveats that should be remembered when using the
Liberty Mutual tables [35]
1 The data for each experimental situation were
assumed to be normally distributed when the
maximum acceptable weights and forces
accep-table to 10%, 25%, 50%, 75%, and 90% of the
industrial population were determined
2 Not all values in the tables are based on
experi-mental data Some values were derived by
assuming that the variation noted for a
particu-lar variable for one type of task would be
simi-lar to that observed for another task, e.g., the
eects on lowering would be similar to that on
lifting
3 The tables for lifting, lowering, and carrying are
based on boxes with handles that were handled
close to the body They recommend that the
values in the tables be reduced by
approxi-mately 15% when handling boxes without
han-dles When handling smaller boxes with
extended reaches between knee and shoulder
heights, they recommend reducing the values
by approximately 50%
4 Some of the reported weights and forces exceed
recommended levels of energy expenditure if
performed for 8 hr or more per day These
data are italicized in the tables
5 The data in the tables give results for
indivi-dual manual materials handling tasks When a
job involves a combination of these tasks, each
component should be analyzed separately, and
the component with the lowest percent of
cap-able population represents the maximum
acceptable weight or force for the combined
task It should be recognized, however, that
the energy expenditure for the combined task
will be greater than that for the individual
components
Some recent data suggest that persons performing
lifting tasks are relatively insensitive to the perception
of high disk compression forces on the spine [38] As aresult, there may be some tasks in the tables thatexceed recommended levels of disk compression
3.4.6 Related Research3.4.6.1 Task and Subject Variables
A variety of researchers have examined the eects ofother task and subject variables using the psychophy-sical protocol Most of these studies involve a smallnumber (<10) of college students as test subjects.Some experiments used the Liberty Mutual protocol;others used the protocol described by Ayoub et al [36]and Mital [37] These ``re®nements'' are summarized inTable 4
Table 4 Miscellaneous Task Variables Evaluated Using thePsychophysical Methodology
Combinations of lift, carry, and lower 40, 41
Center of gravity relative to preferred
Trang 113.4.7 Recommended Applications
3.4.7.1 Job Evaluation
The Liberty Mutual tables were developed for the
pur-pose of evaluating work, not workers [39] In
particu-lar, the tables are intended to help industry in the
evaluation and design of manual materials handling
tasks that are consistent with worker limitations and
abilities [35] The explicit goal is the control of
low-back pain through reductions in initial episodes, length
of disability, and recurrences [39]
To apply the tables in the context of job evaluation,
it is ®rst necessary to specify the task variables of the
job For a lifting task, this would include the lift zone,
distance of lift, box width, frequency of lift, and the
presence or absence of box handles In addition, it
would be necessary to measure the weight of the object
to be handled, perhaps using a scale or dynamometer
Once these variables are speci®ed, the measured
weight can be compared to the data in the table to
determine the percent of capable population for
males and females The procedure is similar for pulling
or pushing The required force can be measured with a
dynamometer
Consider the following example The task is to lift a
box 49 cm wide that weighs 20 kg once every minute
between ¯oor level to knuckle height for a distance of
51 cm
From Table 5, an excerpt from the Liberty Mutual
tables, it is seen that the weight of the box, 20 kg, is
exactly equal to the maximum acceptable weight of lift
for 75% of males, i.e., 75% of males would consider
this task ``acceptable.'' By contrast, the highest
max-imum acceptable weight of lift reported for females is
18 kg As a result, this task is ``not acceptable'' to over90% of females
3.4.7.2 Job Design
To apply the tables in the context of job design, theprocess is essentially identical All task-speci®c para-meters must be identi®ed, except the required weight orforce (that is what you are determining) You select
a desired percent of capable of population, notinggender eects, then identify the maximum acceptableweight or force that corresponds to that desired per-cent This is the value recommended for job design
As an example, suppose you wish to design a liftingtask that requires a box 49 cm wide that must be lifted
51 cm once per minute within the ¯oor-to-knucklezone You desire to design this job to accommodate75% of females According to the data in Table 5, youwould recommend that the box weigh no more than 11
kg This weight would be acceptable to 75% of femalesand over 90% of males
Multiple task analysis consisting of a lift, carry, andlower, has also been investigated for the LibertyMutual data [40] In this circumstance, it was observedthat the maximum acceptable weight for the multipletask was lower than that for only the carrying taskwhen performed separately, but not signi®cantly dier-ent from the lifting or lowering maximum acceptableweights when performed separately For this type of amultiple task, the maximum acceptable weight for thetask should be the lowest maximum acceptable weight
of the lift or lower as if it were performed separately.One should be careful, however, because the energyexpenditure for the multiple task is probably under-
Table 5 Excerpt from the Liberty Mutual Tables for Maximum Acceptable Weight of Lift (kg) for Males and Females
Floor level to knuckle height, one lift everyBox Distance
Trang 12estimated when compared to performing the tasks
separately Similar results were reported by Jiang et
al [41]
3.4.8 Validation
3.4.8.1 Content Validity
The concept of content validity, also called face
valid-ity, addresses whether the content of the test is
identi-cal or highly similar to the content of the job This is
one of the major advantages of the psychophysical
methodology, but it is important for the user to realize
the limitations of the data, especially the caveats noted
earlier
It is noted that a 40 min test protocol is used to
predict an 8 hr maximum acceptable weight or force
The researchers at Liberty Mutual examined this
assumption by having subjects select their maximum
acceptable weight according to the usual protocol, then
having them continue to work, adjusting the weight or
force as desired, for a total of 4 hr [39] There was no
statistically signi®cant dierence between the values
selected after 40 min compared to the values selected
after 4 hr Karwowski and Yates reported similar
results [42]
Mital also examined this issue relative to the Ayoub
et al data [43] Mital found that the test subjects'
esti-mates of their 8 hr maximum acceptable weights of lift
were signi®cantly greater than that selected at the end
of an actual 8 hr period of work (an average 35%
reduction) He ``corrected'' for this eect in his tables
for 8 hr maximum acceptable weights of lift [37]
3.4.8.2 Criterion Related Validity
This type of validity, also called predictive validity,
deals with the question of whether the results of the
this type of job analysis predicts risk of future injury or
illness This is generally demonstrated by the presence
of a statistically signi®cant correlation between a test
``score'' and a particular outcome in an appropriately
conducted epidemiological study
There are two such studies relevant to the
criterion-related validity of the psychophysical methodology
Liberty Mutual Data In 1978, Snook et al published
an investigation of three preventive approaches to
low-back injuries in industry [44] They distributed 200
questionnaires to Liberty Mutual Loss Prevention
representatives throughout the United States These
representatives were asked to complete the
question-naire for the most recent compensable back injury
If the speci®c act or movement associated with the injury
were some form of manual handling task, a task tion was completed to estimate the percent of capableworking population that could perform the task withoutoverexertion, i.e., what percent of the population couldperform the task without exceeding their maximumacceptable weight or force The investigators received
evalua-192 questionnaires, one with incomplete data
They observed that 70% of these 191 low-backinjuries were associated with manual materials hand-ling tasks They also compared the observed number ofinjuries to an expected number of injuries according towhether the percent capable population was greaterthan or less than 75% This analysis is summarized
as follows:
Capablepopulation Observed Expected
The expected values were derived from control datathat revealed that 23.6% of jobs involve handlingtasks that less than 75% of the population could per-form without overexertion
X2 66:6 p < 0:01Based on these results, the authors concluded:
1 A worker is three times more susceptible to back injury if he or she performs a job that lessthan 75% of the working population can per-form without overexertion
low-2 At best, the ergonomic approach could reducelow-back injuries associated with manualmaterial handling tasks by 67% by designingthe jobs so that percent capable populationwas 75% The remaining 33% of back injurieswill occur regardless of the job demands
3 Since only 50% of the industrial back injuriesare related to manual materials handling taskswhere the percent capable population is lessthan 75%, the overall reduction in low-backinjuries would be 33% This reduction would
be higher if the percent capable populationwere raised to 90%
Ayoub et al Data Ayoub and coworkers proposedthe use of a severity index, called the job severity index(JSI), for purposes of validation [45] The JSI is a ratio
of job demands to worker capability Since a job mayconsist of multiple tasks, they de®ned the JSI as a time-
Trang 13and frequency-weighted average of the maximum
weight required by each task divided by the
task-speci®c worker capacity Their validation studies
included 101 jobs, performed by 385 males and 68
females, and involved four steps:
1 Selection of candidate jobs
2 Analysis of candidate jobs in terms of lifting
requirements and morbidity data
3 Determination of the JSI for jobs and operators
4 Determination of the relationship between JSI
and observed morbidity
Individual JSIs were calculated for each worker that
were subsequently grouped in to four categories:
0:00 JSI < 0:75; 0:75 JSI < 1:5; 1:5 JSI < 2:25;
and JSI 2:25
The morbidity data was classi®ed into ®ve groups:
musculoskeletal injuries to the back; musculoskeletal
injuries to other parts of the body; surface-tissue
injuries due to impact; other surface-tissue injuries;
and miscellaneous injuries, and reported as incidence
rates per 100 workers per year Data for severity (days
lost) and cost were also collected
Their results revealed that the incidence of back
injuries and the incidence of disabling back injuries
increased substantially if the JSI was greater than or
equal to 1.5 The relationships were nonlinear The
severity for disabling back injuries was increased if
the JSI was greater than 2.25 The authors did not
report any statistical analyses
Another aspect of their validation involved
classify-ing jobs accordclassify-ing to the percent of capable
popula-tion Each job was categorized according to the
percentage of the and population ``overstressed,'' i.e.,
JSI greater than 1.5 The ranges were: >75%; < 5%
and 75%; and 5% They observed that the
inci-dence of back injuries, inciinci-dence of disabling injuries,
days lost per injury, and total cost increased as the
percent of population ``overstressed'' increased The
authors did not report any statistical analyses
Both Sets of Data Another study that examined the
predictive validity of the psychophysical methodology
was published by Herrin et al [46] These investigators
performed detailed biomechanical and psychophysical
evaluations on 55 industrial jobs from ®ve major
industries The psychophysical analyses involved
deter-mining the minimum percent of capable population
from the Liberty Mutual tables for each individual
task (PSY.MIN) as well as an average percent of
cap-able population when the job involved multiple tasks
(PSY.AVG) Additional comparison variables
included the JSI and lifting strength ratio (LSR).These investigators modi®ed the de®nition of JSI torepresent a frequency- and time-weighted ratio ofweights lifted compared to the average task-speci®clifting strength of males and females, averaged acrossall tasks By contrast, the lifting strength ratio repre-sented the worst case scenario in that the LSR was thelargest single ratio identi®ed among all the tasks.After the jobs were characterized as describedabove, injury and illness data for 6912 incumbentworkers were monitored for two years retrospectivelyand one year prospectively (>12:6 million man-hours).Morbidity was categorized as contact incidents,musculoskeletal disorders (excluding the back), andback incidents, and expressed as incidence rates (num-ber of incidents per 100 workers per year) Severitydata was also examined (lost time versus no lost time).The results revealed a signi®cant negative correla-tion between the minimum percent capable population(PSY.MIN) and all three incidence rates, i.e., the inci-dence rates increased as the percent capable populationdecreased A similar correlation was noted betweenPSY.MIN and severity There was no correlationbetween the average percent capable population(PSY.AVG) with any incidence rate or severity Theincidence rates for musculoskeletal disorders andback disorders were positively and signi®cantly corre-lated with the LSR The LSR was also correlated withseverity The only correlated with severity, notincidence
The authors oered the following conclusions:
1 Overexertion injuries can be related to physicaljob stresses
2 Indices representing the extremes of the jobrequirements (PSY.MIN and LSR) are gener-ally more predictive of risk than indices repre-senting averages (PSY.AVG and JSI)
3 The percent capable population for the moststressful aspect of the job, either isometric orpsychophysical, is the most simple index ofthis type
3.4.9 Evaluation According to Criteria forPhysical Assessment
3.4.9.1 Is It Safe to Administer?
According to Snook, there has been one compensableinjury among the 119 industrial worker test subjects[47] This single episode involved a chest wall strainassociated with a high lift It was also associated
Trang 14with four days restricted activity, but no permanent
disability
3.4.9.2 Does the Protocol Give Reliable
Quantitative Values?
The Liberty Mutual protocol incorporates a criterion
for test±retest reliability (maximum dierence of 15%)
Legg and Myles reported that 34% of their data did
not meet this criterion [48] In contrast, Gallagher
reported that only 3% of tests in their study had to
be repeated because of violating the 15% test±retest
criterion [49] Clearly, the maximum acceptable
weights and forces are quantitative
3.4.9.3 Is It Practical?
There are two major sources of impracticality
asso-ciated with this type of strength assessment: (1) it is
conducted in a laboratory, and (2) the duration of
testing is somewhat prolonged compared to other
strength assessment methods It is possible, however,
to have the subjects use objects that are actually
handled in the workplace Equipment is not very
costly
3.4.9.4 Is It Related to Speci®c Job
Requirements (Content Validity)?
The content validity of this method of strength
assess-ment is one of its greatest assets One potential
weak-ness, however, is its insensitivity to bending and
twisting
3.4.9.5 Does It Predict Risk of Future Injury or
Illness (Predictive Validity)?
The results of two epidemiological studies suggest
that selected indices derived from the psychophysical
data are predictive of risk for contact injury,
muscu-loskeletal disorders (excluding the back), and back
disorders [44,45] These indices are correlated to the
severity of these injuries A third study demonstrated
predictive value [46] It should be noted, however,
that at high frequencies, test subjects selected weights
and forces that often exceeded consensus criteria
for acceptable levels of energy expenditure In
addi-tion, test subjects may also select weights and forces
that exceed consensus levels of acceptable disk
compression
3.5 PART IV: ISOKINETIC STRENGTH3.5.1 Theory and Description of IsokineticStrength Measurement
The concept of isokinetic measurement of strength wasoriginally related by Hislop and Perrine [71].Characteristics of an isokinetic exertionare constantvelocity throughout a predetermined range of motion.Strictly speaking, a means of speed control, and not aload in the usual sense, is applied in isokinetic exertion[71] However, load and resistance are de®nitely pre-sent in this technique In this case, the load is a result
of the energy absorption process performed by thedevice to keep the exertion speed constant Energycannot be dissipated through acceleration in isokineticexercise, because this is prevented by the device.Because the energy is not dissipated in the process, it
is converted into a resistive force, which varies in tion to the eciency of the skeletal muscle
rela-Since the speed of motion is held constant in netic exercise, the resistance experienced during a con-traction is equivalent to the force applied throughoutthe range of motion For this reason, the technique ofisokinetic exercise has sometimes been referred to asaccommodating resistance exercise This type of exer-cise allows the muscle to contract at its maximum cap-ability at all points throughout the range of motion Atthe extremes of the range of motion of a joint, themuscle has the least mechanical advantage, and theresistance oered by the machine is correspondinglylower Similarly, as the muscle reaches its optimalmechanical advantage, the resistance of the machineincreases proportionally It must be understood, how-ever, that while isokinetic devices control the speed ofthe exertion, this does not assure a constant speed ofmuscle contraction
isoki-It should be noted that while the speed of isokineticcontractions is constant during individual exertions, it
is also possible to compare muscular performance over
a wide range of isokinetic velocities Increasing theisokinetic speed of contraction will place increasingdemands on Type II muscle ®bers (fast twitch andfast oxidative glycolytic)
3.5.2 Workplace Assessment
It is clear that isometric strength testing cannot stitute for dynamic strength assessment when examin-ing highly dynamic occupational job demands Asmost industrial work tasks contain a signi®cantdynamic component, analysis of isokinetic strength
Trang 15sub-capabilities would appear to oer some advantage to
isometric testing in this regard However, it must be
recognized that isokinetic devices are not entirely
rea-listic in comparison with free dynamic lifting, where
subjects may use rapid acceleration to gain a weight
lifting advantage
The majority of isokinetic devices available on.the
market focus on quantifying strength about isolated
joints or body segments, for example, trunk extension
and ¯exion (see Fig 7) This may be useful for
re-habilitation or clinical use, but isolated joint testing
is generally not appropriate for evaluating an
indivi-dual's ability to perform occupational lifting tasks
One should not make the mistake of assuming, for
instance, that isolated trunk extension strength is
representative of an individual's ability to perform a
lift In fact, lifting strength may be almost entirely
unrelated to trunk muscle strength Strength of the
arms or legs (and not the trunk) may be the limiting
factor in an individual's lifting strength For this
rea-son, machines that measure isokinetic strengths of
iso-lated joints or body segments should not be used as a
method of evaluating worker capabilities related to job
demands in most instances
Many investigators have used dynamic isokinetic
lifting devices speci®cally designed to measure
whole-body lifting strength [72,73] (seeFig 8) These devices
typically have a handle connected by a rope to a winch,
which rotates at a speci®ed isokinetic velocity when the
handle is pulled Studies using this type of device havedemonstrated good correlations between isokineticdynamic lift strength (i.e., a lift from ¯oor to chestheight) and the maximum weights individuals werewilling to lift for infrequent tasks [72] Thus, undercertain circumstances, this device appears to possesssome validity for assessment of job related dynamiclifting strength capabilities of individuals However,many of these isokinetic lifting devices are limited toanalysis of relatively simple lifting tasks (i.e., a simplesagittal plane lift) Unfortunately, such rudimentarylifting tasks are rare in industry Some investigatorshave attempted to modify this type of instrument byproviding means to mount it so that isokinetic strengthcan be measured in vertical, horizontal, and transverseplanes [74] In spite of eorts to improve the versatility
of these devices, however, it remains clear that complexlifting tasks are not well simulated by current isokineticapparatus
3.5.3 Evaluation According to Criteria forPhysical Assessment
3.5.3.1 Is It Safe to Administer?
Given proper procedures and supervision, isokineticmusculoskeletal testing appears to be a reasonablysafe method of evaluating muscular strength andendurance Certain risks associated with use of freeweights, weight machines, and other isotonic methods
Figure 7 Many isokinetic devices are designed to evaluate isolated joint muscle strengths Such devices can be of great bene®t in
a clinical setting, but may not be as conducive to workplace assessment procedures
Trang 16muscle tension development Proper signal damping
procedures may suppress this ``overshoot''; however,
damping should not be used when absolute torque
values are required
Many other isokinetic devices have been developed
since the introduction of the CYBEX in 1980 Most of
these devices have demonstrated reliability similar to
the CYBEX Klopfer and Greij [75] analyzed the
lia-bility of torque production on the Biodex B-200 at high
isokinetic velocities (3008 4508 sec) and found that
coecients of ta correlation ranged from 0.95 to
0.97, re¯ecting a high degree of reliability of the test
equipment Other authors reported reliability of
between 0.94 and 0.99 with the same equipment [83]
A study analyzing the reliability of the Kinetic
Communicator (KINCOM) device reported intraclass
correlation coecients of 0.94±0.99 [84] Reliability of
the Lido isokinetic system appears somewhat lower
than the others reported above, ranging from 0.83±
0.94 [85] The reliability of the Mini-Gym (the
isoki-netic device best suited to analysis of occupational
tasks) does not appear to have been reported in the
literature
The foregoing data suggests that isokinetic
strength testing equipment generally exhibits a high
degree of reliability However, it should be noted
that results obtained using one system may not be
comparable to results collected on other systems
Several studies have attempted to compare results
between systems, and all have found signi®cant
dier-ences Torque values may vary as much as 10±15%
when comparing dierent systems [86,87] These
dis-crepancies indicate that data collected on dierent
devices cannot be compared, and that normative
data generated on one system cannot be used on
other systems
3.5.3.3 Is It Practical?
Several issues may impact the practicality of using
iso-kinetic devices to examine an individual's muscular
capabilities Not the least of these is the signi®cant
cost of purchasing an isokinetic measurement system
Many of the systems discussed in this section cost tens
of thousands of dollars, which may render such
sys-tems impractical for many applications Another
important issue related to practicality in terms of job
speci®c strength assessment is the ability of these
devices to easily simulate a variety of occupational
tasks While certain isokinetic devices have been
speci-®cally designed to mimic lifting tasks [72], many are
designed simply for quanti®cation of strength of
iso-lated muscle groups in a clinical setting without regard
to accurate simulation of work tasks
3.5.3.4 Is it related to speci®c job requirements?The answer to this question depends upon the type ofisokinetic device and how it is used As discussed pre-viously, isokinetic machines that test isolated musclegroups do not meet this criterion if the job requiresuse of many muscle groups or body segments On theother hand, the Mini-Gym can be used to evaluate thedynamic strength necessary to perform many types ofoccupational tasks, and results of strength tests usingthis device appear to be related to lifting capacity, atleast under certain conditions [72] However, manyindustrial tasks are clearly too complicated to be eval-uated using current isokinetic technologies Great caremust be taken to ensure that isokinetic strength mea-surements are appropriate for analysis of strengthrequirements associated with speci®c occupationaltasks
3.5.3.5 Does It Predict Risk of Future Injury or
3.6 SUMMARY
In spite of advances in measurement techniques and anexplosive increase in the volume of research, ourunderstanding of human strength remains in its intro-ductory stages It is clear that muscle strength is ahighly complex and variable function dependent on alarge number of factors It is not surprising, therefore,that there are not only large dierences in strength
Trang 17between individuals, but even within the same
indivi-dual tested repeatedly on a given piece of equipment
The issue is compounded by the fact that correlations
of strength among dierent muscle groups in the same
individual are generally low, and that tests of isometric
strength do not necessarily re¯ect the strength an
individual might exhibit in a dynamic test As a result
of these and other in¯uences, it is evident that great
care needs to be exercised in the design, evaluation,
reporting, and interpretation of muscular strength
assessments
Traditionally, tests of muscular strength were in the
domain of the orthopedist, physical therapist, and
exercise physiologist However, such tests are also an
important tool for the ergonomist due to the high
strength demands required of workers in manual
mate-rials handling tasks In some cases, it has been shown
that task demands may approach or even exceed the
strength that an individual is voluntarily willing to
exert in a test of strength In such cases, there is
evi-dence to suggest that the likelihood of injury is
signi®-cantly greater than when the task demands lie well
within an individual's strength capacity Because the
relationship between strength capabilities, job
demands, and musculoskeletal injury has been
estab-lished, it becomes apparent that tests of muscular
strength may be of bene®t to the ergonomist both in
the design of jobs, and in ensuring that individuals
have sucient strength to safely perform physically
demanding jobs
Several dierent strength assessment techniques
have been employed for these purposes, each
posses-sing unique characteristics and applicability to job
design and/or worker selection procedures The main
purpose of this chapter has been to elucidate these
strengths and weaknesses of the various procedures,
so that tests of strength may be properly applied in
the design of jobs and the selection of workers
One of the crucial points emphasized in this chapter
is that any test of strength used in job design or worker
selection must be directly related to the demands of the
job [89] For example, if an occupational lifting task
has a high dynamic component, a test of isometric
strength is not likely to provide the data necessary
for proper design of the job Of course, use of dynamic
strength tests to assess a job requiring isometric
exer-tions would also be a misapplication Another
poten-tial pitfall is the use of tests of strength on isolated
muscle groups, and assuming that these tests are
indi-cative of whole-body strength For example, one might
mistakenly assume that dynamic trunk extension
strength is representative of a person's capability to
perform a lifting task However, an individual's liftingcapacity may be entirely unrelated to trunk extensionstrength Instead, lifting capacity may be limited by anindividual's arm or leg strength, depending upon thetask being performed
It should be clear from the parts discussed in thischapter that tests of muscular strength are a tool thatcan be used in the prevention of occupational muscu-loskeletal disease However, it is imperative that theuse of these techniques be applied with a clear under-standing of the advantages and limitations associatedwith each technique The paragraphs that follow sum-marize the tests of muscular strength covered in thischapter Isometric strength is de®ned as the capacity toproduce force or torque with a voluntary isometric(muscles maintain a constant length) contraction Acharacteristic of this type of strength measurement isthe absence of body movement during the measure-ment period Isometric strength testing has a long his-tory, and it may be the easiest to measure andunderstand The basic procedures for testing isometricstrength are well established Risk of injury appears to
be small, and of relatively minor nature Residual ness of muscle groups tested is occasionally reported.Tests of isometric strength appear reliable, with test±retest variability on the order of 5±10% The approachappears quite practical and has been applied in manyindustrial situations The major limitation of isometricstrength testing is in its inability to accurately modelmaterials handling tasks that have a signi®cantdynamic component It is therefore recommendedthat tests of isometric strength be applied when there
sore-is little or no dynamic movement involved In spite ofthis limitation, it should be duly noted that of all theprocedures reviewed in this chapter, tests of isometricstrength are the only strength tests that have shown theability to predict individuals with a high risk of futureinjury or illness on physically stressful jobs [89] Theaccuracy of this prediction appears to be dependent onthe quality of the job evaluation on which the strengthtests are based, and on the care with which the tests areadministered
Tests of isoinertial strength are de®ned as those inwhich the mass properties of an object are held con-stant, as in lifting a given weight (mass) over a prede-termined distance Several strength tests reviewed inthis chapter possess the attribute in this de®nition.However, there are signi®cant philosophical and pro-cedural dierences among the dierent isoinertialprocedures in use, and the authors have subdividedisoinertial strength tests into maximal isoinertiastrength tests [19,21,25], and psychophysical strength
Trang 18tests [12] The following distinctions are made between
these techniques:
1 In maximal isoinertial strength tests, the
amount of weight lifted by the subject is
system-atically adjusted by the experimenter In
con-trast, in psychophysical tests, weight
adjustment is freely controlled by the subject
2 Maximal isoinertial strength tests are designed
to quickly establish an individual's maximal
strength using a limited number of lifting
repeti-tions, whereas psychophysical strength
assess-ments are typically performed over a longer
duration of time (usually at least 20 min), and
instructions are that the subject select an
accep-table (submaximal) weight of lift, not a maximal
one
3 Maximal isoinertial strength tests have
tradi-tionally been used as a worker selection tool (a
method of matching physically capable
indivi-duals to demanding tasks) A primary focus of
psychophysical methods has been to establish
data that can be used for the purpose of
ergo-nomic job design [12]
Two primary maximum isoinertial strength tests have
been described One involves the use of a modi®ed
weightlifting machine where a subject lifts a rack of
unseen weights to various prescribed heights (often
termed the LIFTEST) The other, the progressive
isoinertial lifting evaluation (PILE), uses a standard
lifting box, into which weights are placed incrementally
until the lifting limit is reached Both procedures
appear to be safe to administer and remarkably free
of injury These techniques also appear to compare
favorably to other strength tests in terms of test±retest
reliability Both tests are practical in that they require
relatively inexpensive hardware, and can be
adminis-tered quickly with minimal time needed for subject
instruction and learning The dynamic nature of the
LIFTEST gives the procedure a similarity to certain
industrial lifting tasks, and has correlated well with
psychophysical test results [41]
A vast and expanding base of literature is devoted
to psychophysical strength assessment The
psycho-physical method, as applied to strength, has been
used to determine maximum acceptable weights and
forces associated with manual materials handling
tasks for healthy adult male and female industrial
workers [33,35] The focus of this approach is to
estab-lish data that can be used to improve the design of
manual materials handling activities Psychophysical
strength tests appear very safe, with isolated reports
of muscle strain Psychophysical results are very ducible and seem to be related to low back pain [34].The cost of the procedure is extremely low, except inthe time that it takes to administer the tests Of all thestrength techniques reviewed in this chapter, the psy-chophysical approach is the one best suited to simulat-ing speci®c industrial work tasks However, it should
repro-be noted that at high lifting frequencies, test subjectsmay select weights and forces that exceed manualmaterials handling limits based on metabolic or diskcompression criteria Furthermore, there is some ques-tion as to whether psychophysical lifting tests are sen-sitive to bending and twisting motions, which are oftenassociated with the onset of low-back pain At thistime, the use of psychophysical methods of strengthassessment for the prediction of future risk of injury,illness, impairment, or disability for an individual hasnot been validated
The characteristics of isokinetic strength tests arevariable displacement and constant velocity of motion[71] The majority of isokinetic devices focus on quan-tifying torques about isolated joints or body segments.Isolated joint testing may be most useful in rehabilita-tion or in clinical use, but is more limited in terms ofevaluating occupational job demands However,devices that measure isokinetic whole-body liftingstrength, consisting of a handle connected by rope to
a winch (which rotates at a speci®ed isokinetic velocity)have also been developed Studies using this type ofdevice have shown good correlations between anisokinetic lift from ¯oor to chest height and psycho-physically acceptable weights for infrequent liftingtasks [72,74] Given proper procedures and super-vision, isokinetic strength tests appear to be a reason-ably safe method of evaluating muscular strength andendurance However, some investigators have indi-cated that low velocity isokinetic exertions may bepainful [75] There are numerous isokinetic devices
on the market, and all appear to possess high ity The practicality of isokinetic strength testing maywell hinge on the considerable cost associated withpurchase of the equipment Another issue in terms ofpracticality is the ability of isokinetic devices to easilysimulate a variety of occupational tasks Many indus-trial tasks are clearly too complicated to be evaluatedusing current isokinetic technologies Thus far, pro-spective studies have shown that generic isokinetic lift-ing tests are poor predictors of future low backdisorders [88] Whether isokinetic tests can be used topredict injury or illness when careful comparisons ofjob demands and individual strength capabilities areperformed has not yet been investigated
Trang 19reliabil-A ®nal point on strength assessment should be
made An individual's strength capability cannot be
considered a ®xed human attribute Strength training
regimens can increase an individual's strength
capabil-ity by 30±40% Whether such changes have a
preven-tive eect when a person performs heavy physical
work has yet to be established in epidemiologic
studies; however, some anecdotal evidence supports
the possibility [89]
REFERENCES
1 LS Caldwel, DB Chan, FN Dukes-Dobos, KHE
Kroemer, LL Laubach, SH Snook, et al A proposed
standard procedure for static muscle strength testing
Am Ind Hyg Assoc J 35:201±206, 1974
2 DB Chan Ergonomics guide for the assessment of
human static strength Am Ind Hyg Assoc J 36:505±
511, 1975
3 M Ikai, AH Steinhaus Some factors modifying the
expression of strength J Appl Physiol 16:157±163, 1991
4 KHE Kroemer, WS Marras, JD McGlothlin, DR
McIntyre, M Nordin On the measurement of human
strength Int J Indust Ergon, 6:199±210, 1990
5 TJ Stobbe The development of a practical strength
testing program in industry Unpublished PhD
disserta-tion, University of Michigan, Ann Arbor, MI, 1982
6 DB Chan, GBJ Andersson Occupational
Biomechanics 2nd ed New York: John Wiley and
Sons, 464±466, 1991
7 Troup, JDG, JW Martin, DCEF Lloyd, Back pain in
industry A prospective study Spine 6:61±69, 1981
8 MC Battie, SJ Bigos, LD Fisher, TH Hansson, ME
Jones, MD Wortley Isometric lifting strength as a
pre-dictor of industrial back pain Spine 14:851±856, 1989
9 RA Mostardi, DA Noe, MW Kovacik, JA Porter®eld
Isokinetic lifting strength and occupational injury: A
prospective study Spine 17(2):189±193, 1992
10 DB Chan Ergonomic basis for job-related strength
testing In: Disability Evaluation SL Demeter, GBJ
Anderson, GM Smith, eds Louis, MO: Mosby, 1996,
159±167
11 V Mooney, K Kenney, S Leggett, B Holmes
Relationship of Lumbar Strength in Shipyard
Workers to Workplace Injury Claims Spine 21:2001±
2005, 1996
12 SH Snook The design of manual handling tasks
Ergonomics 21 (12):963±985, 1978
13 DB Chan, GBJ Andersson Occupational
Biomechanics 2nd ed New York: John Wiley and
Sons, pp 105±106, 1991
14 FT Schanne Three dimensional hand force capability
model for a seated person Unpublished PhD
disserta-tion, University of Michigan, Ann Arbor, MI, 1992
15 TJ Stobbe, RW Plummer A test±retest criterion forisometric strength testing Proceedings of the HumanFactors Society 28th Annual Meeting, Oct 22±26,
1984, San Antonio, TX, pp 455±459, 1984
16 DB Chan, GD Herrin, WM Keyserline ment strength testing: an updated position J OccupatMed 20(6): 403±408, 1978
Pre-employ-17 WM Keyserling, GD Herrin, DB Chan Isometricstrength testing as a means of controlling medical inci-dents on strenuous jobs J Occupat Med 22(5):332±366,1980
18 KHE Kroemer Development of LIFTEST: A dynamictechnique to assess the individual capability to lift mate-rial Final Report, NIOSH Contract 210-79-0041.Blacksburg, VA: Ergonomics Laboratory, IEORDepartment, Virginia Polytechnic Institute and StateUniversity, 1982
19 KHE Kroemer An isoinertial technique to assess vidual lifting capability Hum Factors 25(5):493±506,1983
indi-20 KHE Kroemer Testing individual capability to liftmaterial: repeatability of a dynamic test comparedwith static testing J Safety Res 16(1):1±7, 1985
21 JW McDaniel, RJ Shandis, SW Madole Weight liftingcapabilities of Air Force basic trainees AFAMRL-TR-83-0001 Wright-Patterson AFBDH, Air ForceAerospace Medical Research Laboratory, 1983
22 M Parnianpour, M Nordin, N Kahanovitz, VFrankel The triaxial coupling of torque generation
of trunk muscles during isometric exertions and theeect of fatiguing isoinertial movements on the motoroutput and movement patterns Spine 13(9):982±992,1988
23 MM Ayoub, A Mital Manual Materials Handling.London: Taylor and Francis, 1989, pp 241±242
24 DB Chan, GBJ Andersson OccupationalBiomechanics New York: John Wiley and Sons,
1991, pp 152±153
25 TG Mayer, D Barnes, ND Kishino, G Nichols, RJGatchell, H Mayer, V Mooney Progressive isoinertiallifting evaluationÐI A standardized protocol and nor-mative database Spine 13(9):993±997, 1988
26 BC Jiang, JL Smith, MM Ayoub Psychophysical elling of manual materials-handling capacities using iso-inertial strength variables Hum Factors 28(6):691±702,1986
mod-27 LT Ostrom, JL Smith, MM Ayoub The eects of ing on the results of the isoinertial 6-foot incrementallift strength test Int J Indust Ergon 6:225±229, 1990
train-28 JM Stevenson, JT Bryant, SL French, DR Greenhorn,
GM Andrew, JM Thomson Dynamic analysis of inertial lifting technique Ergonomics 33(2): 161±172,1990
iso-29 DO Myers, DL Gebhardt, CE Crump, EA Fleishman.Validation of the Military Entrance Physical StrengthCapacity Test (MEPSCAT) U.S Army Research
Trang 20Institute Technical Report 610, NTIS No AD-A142
169, 1984
30 TG Mayer, D Barnes, G Nichols, ND Kishino, K
Coval, B Piel, D Hoshino, RJ Gatchell Progressive
isoinertial lifting evaluationÐII A comparison with
isokinetic lifting in a chronic low-back pain industrial
population Spine 13(8):998±1002, 1988
31 SS Stevens On the psychophysical law Psychol Rev
64:153±181, 1957
32 LA Jones Perception of force and weight: Theory and
research Psychol Bull 100(1):29±42, 1986
33 SH Snook Psychophysical acceptability as a constraint
in manual workbility of the psychophysical approach to
manual materials handling activities Ergonomics
29:237±248, 1986
34 SH Snook Psychophysical considerations in
permissi-ble loads Ergonomics 28(1):327±330, 1985
35 SH Snook, VM Ciriello The design of manual handling
tasks: revised tables of maximum acceptable weights
and forces Ergonomics 34(9):1197±1213, 1991
36 MM Ayoub, NJ Bethea, S Devanayagam, SS Asfour,
GM Bakken, D Liles, A Mital, M Sherif
Determination and modeling of lifting capacity, ®nal
report HEW (NIOSH) Grant No
5-RO1-OH-00545-02
37 A Mital Comprehensive maximum acceptable weight
of lift database for regular 8 h shifts Ergonomics
27:1127±1138, 1978
38 DD Thompson, DB Chan Can biomechanically
determined stress be perceived? Human Factors and
Ergonomics Society, Proceedings of the 37th Annual
Meeting, Seattle WA 1993, pp 789±792
39 SH Snook Approaches to the control of back pain in
industry: Job design, job placement, and education/
training Spine: State Art Rev 2:45±59, 1987
40 VM Ciriello, SH Snook, AC Blick, PL Wilkinson The
eects of task duration on psychophysically determined
maximum acceptable weights and forces Ergonomics
33:187±200, 1990
41 BC Jiang, JL Smith, MM Ayoub Psychophysical
mod-elling for combined manual materials-handling
activ-ities Ergonomics 29(10):1173±1190, 1986
42 W Karwowski, JW Yates Reliability of the
psychophy-sical approach to manual materials handling activities
Ergonomics 29:237±248, 1986
43 A Mital The psychophysical approach-in-manual
lift-ingÐa veri®cation study Hum Factors 25(5):485±491,
1983
44 SH Snook, RA Campanelli, JW Hart A study of three
preventive approaches to low back injury J Occup Med
20(7):478±481, 1978
45 MM Ayoub, JL Selan, DH Liles An ergonomics
approach for the design of manual materials-handling
tasks Hum Factors 25(5):507±515, 1983
46 GD Herrin, M Jaraiedi, CK Anderson Prediction of
overexertion injuries using biomechanical and
psycho-physical models Am Ind Hyg Assoc J 47(6):322±330,1986
47 SH Snook Assessment of human strength:Psychophysical methods Roundtable presentation atthe American Industrial Hygiene Conference andExposition, Boston, 1992
48 SJ Legg, WS Myles Metabolic and cardiovascular cost,and perceived eort over an 8 hour day when liftingloads selected by the psychophysical method.Ergonomics 28(1):337±343, 1985
49 S Gallagher Acceptable weights and psychophysicalcosts of performing combined manual handling tasks
in restricted postures Ergonomics 34(7):939±952, 1991
50 VM Ciriello, SH Snook A study of size, distance,height, and frequency eects on manual handlingtasks Hum Factors 25(5):473±483, 1983
51 A Mital, MM Ayoub Eect of task variables and theirinteractions in lifting and lowering loads Am Ind HygAssoc J 42:134±142, 1981
52 SS Asfour, MM Ayoub, AM Genaidy A cal study of the eect of task variables on lifting andlowering tasks J Hum Ergol 13:3±14, 1984
psychophysi-53 A Garg, A Mital, SS Asfour A comparison of isometricand dynamic lifting capability Ergonomics 23(1):13±27,1980
54 A Mital, I Manivasagan Maximum acceptable weight
of lift as a function of material density, center of gravitylocation, hand preference, and frequency Hum Factors25(1):33±42, 1983
55 SJ Legg, DR Haslam Eect of sleep deprivation on selfselected workload Ergonomics 27(4):389±396, 1984
56 JL Smith, BC Jiang A manual materials handling study
of bag lifting Am Ind Hyg Assoc J 45(8):505±508, 1984
57 A Mital, HF Fard Psychophysical and physiologicalresponses to lifting symmetrical and asymmetricalloads symmetrically and asymmetrically Ergonomics29(10):1263±1272, 1986
58 A Mital Maximum weights of asymmetrical loadsacceptable to industrial workers for symmetrical lifting
Am Ind Hyg Assoc J 48(6):539±544, 1987
59 A Mital Psychophysical capacity of industrial workersfor lifting symmetrical loads and asymmetrical loadssymmetrically and asymmetrically for 8 hour workshifts Ergonomics 35(718):745±754, 1992
60 CG Drury, JM Deeb, B Hartman, S Wooley, CEDrury, S Gallagher Symmetric and asymmetric manualmaterials handling Part 1 Physiology and psychophy-sics Ergonomics 32(5):467±489, 1989
61 SL Legg, CM Pateman Human capabilities in tive lifting Ergonomics 28(1):309±321, 1985
repeti-62 CG Drury, JM Deeb Handle positions and angles in adynamic lifting task Part 2 Psychophysical measuresand heart rate Ergonomics 29(6):769±777, 1986
63 A Mital Maximum acceptable weights of lift acceptable
to male and female industrial workers for extendedwork shifts Ergonomics 27(11):1115±1126, 1984