SCI NTIFIC/CU NICAL A RTICL S

The Use of the Rapid Exchange Grip Test inDetecting Sincerity of Effort, Part I:

Administration of the Test

Catherine Taylor, MHS, OTR/LOccupational TherapistThe Hand Therapy CenterGainesville, Florida

Orit Shechtman, PhD, OTR/LDepartment of Occupational TherapyCollege of Health ProfessionsUniversityof FloridaGainesville

The Jamar dynamometer is the most commoninstrument used to evaluate grip strength. The gripstrength test measures the amount of static muscularforce applied by the hand when gripping. The reliabil­ity and validity of the Jamar dynamometer in assess­ing grip strength has been established.' However, theability of the Jamar to provide an objective and accu­rate measure of grip strength is compromised whenthe person being assessed intentionally does not exertfull effort during testing.

There are many consequences of low effort duringassessment. Intentional insincere effort on the part of apatient can obviously lead to frustration on the part ofthe therapist, because of prolonged and seeminglyineffectual treatment as well as numerous and unnec­essary therapeutic procedures. Another outcome is thesubstantial and unnecessary increase in the cost of the

Correspondence and reprint requests to Orit Shechtman, PhD,OTR/L, Department of Occupational Therapy, University ofFlorida, Gainesville, FL 32610;e-mail : <oshechtm@hp.ufl.edu>.

ABSTRACT: A review of studies that utilize the rapid exchangegrip (REG) test revealed that there is no standardized test proto­col for administering the test. The purpose of this study was toinvestigate three factors that affect the result of the REG test: thehand switch rate, the number of grips performed during the test,and the comparative tests used in the interpretation of the REGtest. The 146 uninjured subjects performed a series of randomizedgrip strength tests including the REG test, the maximal static griptest (MSGT), and the five-rung (5R) test while making maximaland submaximal efforts . Results revealed no significant differ­ences in peak REG scores between hand switch rates of 45 and 60rpm . Significant differences were found for peak REG scoresobtained from three vs. five trials. Peak scores from the 5R test vs,the MSGT differed sign ificantly for maximal efforts but not forsubmaximal efforts . These results led to the following recomm en­dati ons for a standardized protocol for administering the REGtest: 1) the REG maneuver may be adm inistered at a rate of either45 or 60 rpm; 2) at least five hand-grip exchanges mu st be per­formed; and 3) only one comparative test, either the MSGT or the5R test, should be used consistently for comparison with the REGmaneuver. The findings of the present study, however, did notverify which static grip test should be used for comparison withREG scores.JHAND THER. 2000;13:195-202.

health care services provided to such persona/ Theissue of sincerity of effort is also crucial in determininga person's readiness to return to work and in makingdecisions concerning any financial award for loss offunction due to a job-related injury," Understandably,considerable effort has been expended to develop away to accurately determine whether a person beingtested is exerting full effort. The effectiveness of sin­cerity-of-effort tests, however, remains undetermined.Contradictory findings are present in the literature.t"particularly with regard to the rapid exchange grip(REG) test. Further examination of these contradictoryfindings reveals an underlying lack of consensusregarding the REG test protocol.

The use of some terms in this paper needs clarifica­tion : "REG maneuver" or "REG scores" refers torapid hand exchanges during the REG test, while"REG test" refers to the procedure of comparing REGscores with scores of a static grip (SG) test. "MSGT"refers to the maximal static grip test using only one(the second or third) handle setting of the Jamar

July-September 2000 195

dynamometer, and "SG" refers to a static grip test,either the five-rung (5R) grip test or the MSGT.

The REG test was first introduced by Lister," whoasserted that submaximal efforts could not be repli­cated during rapid gripping and that, consequently,an increase in REG scores compared with peak 5Rscores could be observed in those patients whofeigned weakness. When peak REG scores exceededpeak SG scores, the REG test was considered "posi­tive" and submaximal effort was suspected.Conversely, a REG test was deemed "negative" whenpeak SG scores were higher than peak REG scores.According to Lister," a negative REG test indicated asincere effort.

The REG test was first used in a scientific study byCzitrom and Lister,S who described it as a test inwhich "the patient, gripping strongly the dynamo­meter set at the optimal span grasp position, is askedto alternate rapidly from one hand to the other." Theauthors deemed this test to be reliable in distinguish­ing sincere from insincere effort, although no statisti­cal analysis regarding reliability or validity was per­formed. This article provided no details regardingthe majority of the variables of the REG test protocol(e.g., number of grips performed, hand switch rate,length of rest periods, and handling of the instru­ment). No additional information about the develop­ment, rationale, or interpretation of the REG test wasprovided.

Hildreth et a1.4 were the first to empirically exam­ine the effectiveness of the REG test in determiningsincerity of effort. They investigated the ability of theREG test to differentiate subjects exerting sincereeffort from those exerting insincere effort. Theauthors first administered the 5R test, followed bythe REG maneuver. The REG maneuver was admin­istered using the dynamometer handle setting thatrendered the highest grip score during the 5R griptest. Their description of the protocol used for admin­istering the REG maneuver was vague, at best. Thehand switch rate was not predetermined but ratherwas established by asking the subject to alternatehands "as rapidly as possible." The number ofrequired grips was not set but rather was in the rangeof "five to ten times." It was also not specifiedwhether this number referred to grips per hand ortotal grips performed. Termination of the testoccurred when the examiner determined the test tobe "negative" or "positive." Although the criteria forthis determination were not specified, it was presum­ably based on how the REG scores, in general, com­pared with those of the 5R grip test."

The usefulness of the REG test in determining sin­cerity of effort was then examined by [oughin et al.s Acomparatively thorough description of the REG testprotocol was provided in this study. A static grip testconsisting of one trial and using the third handle posi­tion of the Jamar dynamometer was administered

196 JOURNAL OF HAND THERAPY

first, followed by the REG maneuver. These authorswere the first to specify a rate for switching hands dur­ing testing and the first to examine the implications ofhand switch rate on the results of the REG test. Thetwo hand switch rates used for testing were 80 and 100repetitions per minute (rpm), which were establishedand maintained by a metronome during testing. Norationale for the selection of these rates for testing wasprovided.

Since increasing the hand switch rate to 100 rpmdid not improve the efficacy of the REG test, theauthors concluded that the REG should be adminis­tered at 80 rpm.s In addition, testing duration wasdescribed as continuing "for five to ten grips until apattern in grip strength became clear." However, itwas not specified whether this number referred togrips per hand or total grips performed. The posi­tioning of the subject during testing was thoroughlydescribed. Standardized verbal instructions wereused, although these instructions were not madeavailable in the article. One difference between thisstudy and the one by Hildreth et a1.4 was that thisstudy compared peak REG scores with peak MSGTscores rather than with peak 5R scores. The MSGTscores were obtained at the third handle level withonly one repetition.f

Stokes et al.6 provided the most thorough descrip­tion of the REG test protocol. Aspects of this protocoldiffered notably from protocols used in the earlierstudies.4•

5 For instance, the hand switch rate usedwas 1.5 seconds per grip, or 45 rpm, which is a muchslower rate than the switch rate of 80 rpm recom­mended by Joughin et a1.5 Again, no rationale for theestablishment of the hand switch rate was provided.Standardized instructions to the subject were usedand were provided in the study. Also, the positioningof subjects was described in detail. Other testing vari­ables such as cueing, the number of grips, and therest periods were also described in detail. A principaldifference in this study was the instrumentation.While the other studies investigatins the REG testused a regular Jamar dynamometer, .5 Stokes et a1.6

used a modified Jamar dynamometer in conjunctionwith the Greenleaf Evaluation System (GreenleafMedical, Palo Alto, California). The use of this com­puterized system ameliorated the difficulty of manu­ally recording all trials during the REG maneuver.This system is expensive, however, and is not widelyused in clinics; therefore, the clinical applicability ofthis study is somewhat limited.

Problems associated with insufficient descriptionof testing protocols were addressed in an article byKeating and Marras." The authors stressed the needto provide a thorough description of key variablesin testing, to allow for accurate replication of stud­ies. Without the ability to replicate a study's proto­col, it is impossible to determine whether the proto­col rendered reliable or valid results. A summary of

TABLE 1. Summary of Rapid Exchange Grip (REG) Test Protocols

Czitro m et al. (1988) Hildr eth et al. (1989) [oughin et al (1993) Stokes et al. (1995)

Handle setting: "The optimal span Handle setting Third handle settinggrasp position" rendering highest

reading on 5R

Switch rate: Not ment ioned Not menti oned 80 and 100 rpm

Cueing: Not addressed No cuei ng system Audi tory cues :specified metronome

No . of grips: Not addressed 5-10 grips 10-15 grips

Mani pulation Not addressed Dynamometer held Dynamometer heldof instrument: in place in place

Rest period : Not addressed Mentioned but not Some 2 min ; othersspecified not specified

Com parative test: Not addressed 5R MSGT

Peak or mean Not addressed Unclear Mean REG scoresscores:

Position ing of Not addressed Not addressed ASHT recomm endedsubject: position

Instructions to Not addressed Not specified Standardizedsubjects (not pro vided )

Instrument used: Jamar dynamometer Jamar dynamometer Jamar dynamometer Modified Jamardynamometer withGreenleaf MedicalEvalua tion System

Handl e settingrenderin g highestread ing on 5R

45rpm

Aud itory cues:computer

8 grips each hand,16 total

Unclear whetherdynamometer wasmoved or held in place

No res t period allow ed

5R

Peak REG scores

Not addressed

Standa rdized(provided in articl e)

NOTE: 5R indicates five-rung test; MSGT, maximal static grip test.

the various protocols reported in these studies(Table 1) shows th e variation among them in suchfactors as handling of the ins trument, rest periods,hand switch rate, and the number of grips per­formed.

The variations in the testing protocols used in thesestudies clearly demonstrate the need for a standard­ized testing protocol for the administration and inter­pretation of the REG test. Attempts to deem the REGtest a useful, reliable, and valid measure of sincerityof effort seem premature without a standardized pro­tocol that would allow for an accurate replication ofthe test. Certain variables of grip testing have beenresearched, and recommendations regarding theiruse during testing have been made. For instance, theASHT lO has specific positioning recommendationsfor subjects to follow during grip testing. These rec­ommendations may be applied in the REG protocolon the assumption that they will increase the abilityof the test to be accurately replicated.

Other, nonstandardized variables of the REG testmay have profound implica tions on its results. Non­standardized hand switch rates may greatly affectREG scores. The basic premise behind the REG man­euver is that requiring the subject to rapidly alternategrips allows him or her less time to concentrate andto voluntarily control weakness of grip.?,8 The physi-

ologic explanation for this premise relies on themotor unit recruitment model.l! According to thismodel, a submaximal effort is difficult to repli catebecause it requires grading of the muscular contrac-. d . ft' I 11tion an constant corrections 0 mo or SIgna s.

Therefore, replicating a submaximal, or insincere,effort will require more time to execute than replicat­ing a maximal effort. Theoretically, limiting theamount of time allowed for executing a repeated sub­maximal muscular contraction should interfere withthe subject's ability to replicate that contraction (orgripping force ) consistently. The REG maneuver pro­tocol imposes a rapid exchange of hands, therebydecreasing the time allowed for performing the mus­cular contraction of gripping. Consequently, theaccuracy of replicating a submaximal effort is expect­ed to decrease. However, if the hand switch rate isnot rapid enough, the individual's ability to accu­rately replicate the submaximal effort will not be hin­dered . Therefore, the speed of alternating grips (thehand swi tch rate) may play a significan t role in theeffectiveness of the REG test to detect submaximal(insincere) effort.

In addition, it is not known how many grip repe­titions must be performed at th is "rapid " rate to ren­der meaningful results. Therefore, further investiga­tion into how the number and rate of hand

July-September 2000 197

exchanges affect the results of the REG test is war­ranted. Another variability in reports of REG testingis that some studies compare the REG maneuverwith the MSGT, whereas others compare the REGmaneuver with the 5R test. 4

--8 Both the MSGT andthe 5R test are static grip strength tests, but they dif­fer in which handle setting on the dynamometer isused. In the 5R test, only one repetition (or trial) isperformed on each of the five handle settings,whereas in the MSGT, three repetitions are per­formed on the second10 or third" handle setting. Asthis comparative relationship determines test posi­tivity, further investigation is warranted to establishexactly which test should be compared with theREG maneuver.

This variation and lack of standardization of theREG maneuver protocol have considerable profes­sional implications for therapists who use gripstrength testing. Grip strength testing is a valuableclinical assessment that therapists often use to sup­plement other assessment tools (e.g., goniometryand volumetry) in making recommendations anddecisions regarding patient treatment.V The varioustests that measure sincerity of effort of grip strengthmust be standardized and deemed valid and reliable.Without the standardization of clinical tests, profes­sional decisions based on these tests become ques­tionable.

The ASHT recognized that the comparisonbetween the studies investigating the REG test is lim­ited because of variations in test protocol, patientposture, and dynamometer calibration." Therefore,the ASHT made the following recommendation:"Before the issue of voluntary submaxirnal effortmay be resolved, further study is needed usinginstruments and test protocols that meet consistentstandards and criteria.,,10

The purpose of this study was to examine threetesting protocol variables of the REG test and to rec­ommend, on the basis of the findings, a standardizedtesting protocol. First, the effect of hand switch rates(45 vs. 60 rpm) on peak scores was studied. Then, theeffect, on peak scores, of the number of grip repeti­tions (three vs, five) performed during the REGmaneuver was investigated. Finally, the differencesbetween the comparative assessments (the MSGT vs,the 5R test) were explored.

MATERIALS AND METHODS

Subjects

The participants in this study were 146 volunteers,mostly occupational therapy students, who had noprior knowledge of or experience with grip strengthtests and had no upper extremity injuries. The ages ofsubjects ranged from 18 to 64 years, with a mean age

198 JOURNAL OF HAND THERAPY

of 23 ± 7.33 years. The sample included 17 men and129 women; 90% of the subjects were right-handdominant.

Materials

All grip strength data were collected using theJamar Dynamometer (Asimov Engineering Co., LosAngeles, California). Calibration of the dynamometerwas performed by the manufacturer two weeks priorto data collection. Calibration was tested daily bysuspending known weights across the dynamometerand checking for the accuracy of the measurement.Weights were added gradually in 5-lb. increments.This procedure for daily calibration was based on adescription in a technical monograph provided bySammons Preston.13

An electronic metronome (Franz ManufacturingCo., New Haven, Connecticut; model number LM4)was used to set a cadence at 45 and 60 rpm in order toestablish the hand switch rates for the two REG tests.All data were collected in a laboratory room. Prior totesting, all subjects read and signed an informed con­sent form that was approved by the InstitutionalReview Board at the University of Florida .

Design

Each subject completed a series of four randomlyassigned grip tests , which included the 5R test, theMSGT, and two REG maneuvers (at rates of 45 and 60rpm). The series of tests was completed once withboth hands utilizing full effort and once with onehand exerting ma ximal effort and one hand exertingsubmaximal (50% of maximal) effort. This allowedfor within-subject comparison of maximal and sub­maximal trials of the same hand. Each subject per­formed five grips per hand for each test and a total of40 grip repetitions per hand. The hand that per­formed maximally (dominant vs. nondominanthand) was randomly assigned.

To define the order of testing, some facts weretaken into consideration. Since therapists in clinicalsituations usually administer a static grip strengthtest first (either the 5R test or the MSGT) and the REGmaneuver only after that, we attempted to keep thisorder of testing. To do so, we created two testing pro­tocols. In the first protocol, the 5R test was adminis­tered first, followed by the REG maneuver at a rate of60 rpm. In the second protocol, the MSGT (at the sec­ond handle level) was administered first, followed bythe REG maneuver at a rate of 45 rpm. The subjectswere randomly assigned to begin testing with one ofthese two protocols. In addition, the subjects wererandomly assigned to begin testing with either amaximal or a submaximal effort.

Testing always began with the right hand and pro­ceeded in an alternating fashion. The investigator

held the dynamometer and presented it to the subjectduring all tests. Two-minute rest periods were pro­vided after each set of two tests (ten repetitions perhand). This duration of rest was based on the find­ings of Trossman and Li14 and was used in anattempt to preclude any effects of fatigue during test­ing. To replicate clinical situations, rest periods werenot allowed between the static grip test and the REGmaneuver, except for the time needed to adjust thedynamometer handle and instruct the subject.

Procedure

For all tests, subjects were positioned according tothe recommendations of the ASHT. lO Each subjectwas seated in an adjustable chair without arm rests,which allowed for proper positioning. The subject'sfeet were resting flat on the floor, and the hips werepositioned as far back in the chair as possible. Bothhips and knees were positioned at approximately900

• The tested arm was positioned with the shoul­der adducted and neutrally rotated, the elbowflexed at 900

, and the forearm and wrist in neutralposition. Reminders were given to each subject asneeded to ensure that proper positioning was main­tained.

Two examiners tested each subject. One adminis­tered the tests, and the other recorded the dynamo­meter readings. The dial of the dynamometer facedaway from the subject so that he or she was not ableto view it during testing. No visual or verbal feed­back about the subject's performance was providedduring the testing session. A blue foam strap wasplaced on the wrist of the hand that was selected toexert submaximal effort, to serve as a reminder forboth the subject and the examiner.

Standardized verbal instructions for performingeach of the grip tests were provided. The sameinstructions were given for the maximal and sub­maximal trials, with the exception that, for the sub­maximal trials, the subject was reminded which handshould perform at 50% of maximal effort. Subjectswere instructed to hold each static grip until a dropin strength was noted on the dynamometer needle;each grip lasted approximately three to five seconds.

The REG maneuver was administered at rates of 45and 60 rpm. The second handle setting was used forthis test. The verbal instructions were as follows: "Thisis a rapid exchange grip test. You will grip right, left,right, left, and so on until I tell you to stop. Thismetronome will be keeping a beat during the test; Iwant you to switch hands in time with themetronome. Remember to begin with your righthand." After providing the verbal instructions, theexaminer demonstrated the REG maneuver. Next, theexaminer held the dynamometer in the midline infront of the subject at a level that best allowed the sub­ject to maintain proper positioning. The metronome

emitted an auditory cue for the hand switch rate of thetest. For practicing the hand switch rate, the subjectwas asked to move the hands back and forth towardthe dynamometer in time with the beat of themetronome but without gripping. This hand switch­ing practice continued until the examiner determinedthat the subject had established the correct rhythm.Next, the examiner gave the verbal instructions,speaking in time with the rhythm, "Ready, set, grip ...grip ... grip .... " This continued until five measure­ments had been recorded for each hand. The outcomescore derived from the REG maneuver was the peakscore of all repeated grip trials.

The 5R test began with the first (and most narrow)setting on the dynamometer. Grip strength wasrecorded for the right and left hands on this setting.The handle was then moved to the next setting andso on until measurements had been collected for bothhands at all five handle settings. No rest period wasgiven during this test, other than that provided whenthe dynamometer handle was being adjusted. Theverbal instructions for the 5R test were as follows:"On this test, you will squeeze the dynamometer ashard as you can on each of the five settings. We willbegin with the right hand and the smallest setting.When I say grip, I want you to grip as hard as youcan. Keep gripping until I tell you to stop. Are youready? Grip." The outcome score derived from the 5Rtest was the peak score of all five grip trials at the fivehandle settings.

The MSGT test was administered with thedynamometer handle at the second setting, which isthe handle settin~ recommended for use in static gripstrength testing. ,15,16 Grip measurements were col­lected first on the right hand, then the left, and so onuntil five measurements were recorded for eachhand. The examiner cued the subject when to beginand end each grip. The verbal instructions for theMSGT test were as follows: "When I say grip, I wantyou to grip as hard as you can. Keep gripping until Itell you to stop. Are you ready? Grip." The examinertold the subject to stop when the dial on thedynamometer leveled off and started to drop, afterapproximately 5 seconds of gripping. The outcomescore derived from the SG test was the peak score ofall repeated grip trials.

Statistical Analysis

The peak scores of all five trials and of the firstthree of five trials were entered for each hand for theREG maneuvers and the MSGT. For peak scores onthe 5R test, the highest recorded score from all han­dle settings was entered for each hand. Althoughdata collection was performed for all questionssimultaneously, the data analysis followed a sequen­tial order of interest. Therefore, two-tailed~aired t­tests were conducted for all comparisons.1 First, to

July-September 2000 199

TABLE 2. Results of Pai red t Tests, Observed Power, and Effect Sizes

Scores Compared Mean Difference (kg) t Value Probability (p) Power Effect Size

REG-4s vs REG-60:Maxi mal effort 000%) -0.007 -0.003 0.974 <0.21 0.0038Submaximal effort (50%) -0.343 -0.850 0.397 0.21 0.0999

REG-60: thr ee vs five grips :Maximal effort (100%) -0.315 -3.676 <0.0005" 0.95 0.3888Submaximal effort (50%) -0.829 -5.367 <0.0005" 0.995 0.6252

REG-4s : three vs five gri ps :Maximal effort 000%) -0.267 -3.428 <0.001" 0.95 0.3974Sub maximal effort (50%) -0.795 -5.438 <0.0005" 0.995 0.6332

sn vsMSGT:Maxima l effor t (100%) -1.445 -5.761 <0.0005" 0.995 0.6904Submaximal effort (50%) -0.596 -1.545 0.13 0.51 0.1825

NOTE: REG-45 and REG-60, rap id exchange grip test performed at 45 and 60 rpm , respectively; 5R, five-rung test; MSGT, maximal static griptest. An asterisk (") ind icate s significant d ifferences. .

determine the differences in REG scores betweenhand switch rates of 45 rpm and 60 rpm, paired t­tests were conducted on peak scores, both at 50% and100% of maximal strength. Then, to determine thedifferences in the number of repetitions (three vs.five), the peak scores of the first three repetitions ofthe REG maneuver were compared with the peakscores of all five repetitions of the REG maneuver,both at 50% and 100% of maximal strength. Last, toinvestigate the differences between the comparativeassessments, paired t-tests were conducted to com­pare peak scores on the SR with peak scores on theMSGT, both at 50% and 100% of maximal strength.Table 2 provides a complete list of all investigatedcomparisons . Comparisons were significant at the0.05 alpha level. Bonferroni adjustments for the num­ber of contrasts (eight) corrected the alpha level to0.006.17 In addition, power and effect size were calcu­lated for each of the paired comparisons.V

RESULTS

The first research question addressed the effect ofhand switch rates (45 rpm vs. 60 rpm) on peak scoresof the REG maneuver. There were no significant dif­ferences between the scores obtained from the REGmaneuver performed at 45 rpm (REG-45) and at 60rpm (REG-60), when the subject was exerting eithermaximal or submaximal effort (Figure 1). The secondresearch question explored the effect of the numberof trials (three vs. five) on peak REG scores. PeakREG scores derived from five trial s were significant­ly greater than peak REG scores derived from threetrials, for both maximal and submaximal efforts andfor both REG-45 and REG-60 (Figure 2; because ofsimilarities in REG results at 45 and 60 rpm, only theREG 60 data are shown). The third research questionexamined the differences between the comparativeassessments (the MSGT vs. the 5R test). During max-

*

35

1

35

1

3030 T T. .Ci Ci.>< .><-;; 25 - 25..E EE ECl Cl

g 20 oS! 20ll:

rTIi2

15 15

10 1050"10 100"10

OJ50"10

*T ...

100"10

Percentage of Maximal Ellort

FIGURE 1. Comparison of peak rapid exchange grip (REG) testscoresfor maximal (100%) and submaximal (50%) efforts at 45rpm (wh ite columns) and 60 rpm (gray columns). There wereno significantdifferences.

200 JOURNAL OF HAND THERAPY

Percentage of Maximal Ellort

FIGURE2. Comparison of peak rapid exchangegrip (REG) testscores at 60 rpm, for maximal (100%) and submaximal (50% )efforts in three trials (white columns) and five trials (g raycolumns). Asterisk (*) indicates significant differences.

*35

DISCUSSION

the effects on peak REG scores of using three vs. fivetrials during the REG maneuver. Our findingsrevealed significant differences between peak scoresobtained from three trials and peak scores obtainedfrom five trials. Five trials rendered higher peakscores, during both maximal and submaximal trials.Since the REG maneuver is designed to reveal the truemaximal grip, we concluded that five grip exchanges,which rendered higher peak scores than three repeti­tions, should be used.

Finally, we investigated the differences betweenthe two comparative tests that are commonly used inconjunction with the REG maneuver, namely, theMSGT and the 5R test. In the literature, the REGmaneuver was compared with the MSGT test in onestudy'' and with the 5R test in two studies.v" Usingthe MSGT instead of the 5R test could shorten theoverall time of administration of the REG test. Thus,this inquiry was also made in an attempt to make theREG test a more time-efficient clinical tool. Our find­ings showed that peak MSGT scores were signifi­cantly greater than peak 5R scores only for maximalefforts, not for submaximal efforts. Our findings didnot verify which static grip test scores should be rec­ommended for comparison with REG scores.Therefore, we can merely suggest that only one com­parative test, either the MSGT or the 5R test, be usedconsistently for comparison with the REG test.

The issue of comparative tests for the REG test wasfurther addressed by Shechtman and Taylor." Theyfound that the concept of a "positive REG" (indicating a~ubmaximal effort) was supported only when compar­mg peak REGscores with peak 5R scores, as opposed tocomparing peak REG scores with peak MSGT scores.However, the authors also found that the best combina­tion of specificity and sensitivity values for the REG testwas obtained when using the REG-45 vs, the MSGT.Consequently, the authors stated that it was difficult toclearly recommend which comparative test should beused in administering the REG test. Further studies areneeded to answer this question.

The present study has a few limitations, which mayaffect the interpretation and application of its results.First, a large percentage of the subjects were women(88%),a fact that may limit the generalizability of thefindings. Second, the subjects were normal controls,with no known upper extremity injury; thus, theresults of this study may not be applicable to patientswith hand and upper extremity injuries. In addition,to mimic insincere efforts, the subjects were asked toexert 50% of their maximal effort. Exerting this typeof sub maximal effort "on request" may render differ­ent results from those gathered in clinical situationsin which an individual independently and intention­ally exerts an insincere effort. We recommend thatfuture studies exploring the effectiveness of the REGtest in detecting sincerity of effort be conducted onspecific patient populations.

_CD50% 100o/;~

Percentage 01 Maximal Ellort

10

15

c; 30~l/l

~ 25Cl

.!252 20

imal effort, peak MSGT scores were significantlygreater than peak 5R scores. For submaximal effort,however, peak MSGT scores and peak 5R scores didnot differ significantly (Figure 3). The effect size ofeach of the nonsignificant comparisons was less thana "small effect size" (d = 0.20), indicating that thesedifferences are truly nonsignificant and are not dueto an insufficient number of subjects.

The present study investigated the differencesbetween some aspects of the testing protocol of theREG test, including the hand switch rate during theREG maneuver, the number of repeated strength tri­als, and the two comparative tests used in conjunc­tion with the REG test. First, we examined how thehand switch rate of the REG maneuver affected theresults rendered from this test. The hand switch ratescited in the literature ranged from 45 rpm" to 100rpm.s The authors of the present study were unableto maintain and record hand switch rates of 80 and100 rpm, which led to the assumption that these handswitch rates were not likely to be used in clinical set­tings without the benefit of special equipment.

To increase the applicability of our findings to thecommon clinical setting, we investigated hand switchrates of 45 and 60 rpm. Paired comparisons revealedno significant differences between peak scores of theREG maneuver administered at 45 and 60 rpm. SinceREG scores were not affected by the hand switchrates of either 45 or 60 rpm, we concluded that eitherrate may be used to administer the REG test as longas one rate is used consistently.

The number of grips performed during the REGma~euver was examined next. The number of gripsvanes considerably in the Iiteraturev" and is notalways clearly specified4

,7,8 (see Table 1).We examined

FIGURE 3. Comparison of peak five-rung scores (whitecolu?"lns) and peak static grip scores (gray columns) duringmaximal (100%) and submaximal (50%) efforts. Asterisk (*)indicates significant differences.

July-September 2000 201

An additional limitation was that we used the secondhandle position of the Jamar for the REG maneuver,rather than the handle setting that yielded the maximalscore on the 5R test (which could be any of the handlesettings). However, the majority of our subjects (120subjects, or 82%of all subjects) obtained their peak scoreon the 5R test while gripping the dynamometer at thesecond handle position. (The other 18% of subjectsobtained their peak score as follows: 12% at the thirdhandle position, 4% at the first handle position, and 3%at the fourth handle position). We analyzed the dataseparately for the 120subjects whose peak score on the5R test was at the second handle position and found thesame results for each of the three research questions asfor all 146subjects.Therefore, this limitation seems to beof no consequence.

CONCLUSIONS

Based on the findings of the present study, we madeseveral recommendations for standardizing the testingprotocol of the REG test. First, since there was no sig­nificant difference between the peak scores obtainedfrom the REG test at 45 and 60 rpm, we conclude thatthe REGtest may be administered at either rate. Second,since peak scores obtained from five trials are signifi­cantly higher than peak scores obtained from three tri­als, we recommend that at least five hand-gripexchanges be performed during the REG test. Finally,we suggest that only one comparative test, either theMSGT or the 5R test, be used consistently for compari­son with the REG test. The findings of the presentstudy, however, did not give rise to a clear-cut prefer­ence for one comparative test over the other.

REFERENCES

1. Mathiowetz v, Weber K Volland G, Kashman N. Reliabilityand validity of hand strength evaluation. J Hand Surg. 1984;9A:222-6.

202 JOURNAL OF HAND THERAPY

2. Robinson ME, Geisser ME, Hanson CS, O'Connor PD.Detecting submaximal efforts in grip strength testing with thecoefficient of variation. J Occup Rehab. 1993;3(1):45-50.

3. Ashford RF, Nagelburg S, Adkins R. Sensitivity of the Jamardynamometer in detecting submaximal grip effort. J HandSurg. 1996;21A:402-5.

4. Hildreth DH, Breidenbach We, Lister GD, Hodges AD.Detection of submaximal effort by use of the rapid exchangegrip. J Hand Surg. 1989;14A:742-5.

5. Joughin K, Gulati P, Mackinnon SE, et aL An evaluation ofrapid exchange and simultaneous grip tests. J Hand Surg.1993;18A:245-52.

6. Stokes HM, Landrieu KW, Domangue B, Kunen S. Identi­fication of low-effort patients through dynamometry. J HandSurg. 1995;20A:1047-56.

7. Lister G. The Hand: Diagnosis and Indications. 2nd ed. NewYork: Churchill Livingstone, 1984:112.

8. Czitrom AA, Lister GD. Measurement of grip strength in thediagnosis of wrist pain. J Hand Surg. 1988;13A(1):16-9.

9. Keating JL, Marras TA. The influence of subject and test designon dynamometric measurements of extremity muscles. PhysTher.1996;76(8):866-89.

10. Fess EE. Grip strength. In: Clinical Assessment Recommend­ations. Chicago, ill: American Societyof Hand Therapists, 1992.

11. Kroemer KHE, Marras WS. Towards an objective assessmentof the "maximal voluntary contraction" component in routinemuscle strength measurements. Eur J Appl PhysioL1980;45:1-9.

12. Dutton R. Biomechanical frame of reference. In: Hopkins HL,Smith HD (eds), Willard and Spackman's OccupationalTherapy. Philadelphia, Pa: Lippincott, 1993:66-7.

13. Carruth WL. A comparison of the performance characteristicsof the hydraulic and electronic Jamar hand dynamometers: atechnical monograph. Control Systems Engineering, Inc.

14. Trossman PB, Li PW. The effect of the duration of intertrial restperiods on isometric grip strength performance in youngadults. Occup Ther J Res. 1989;9(6):362-78.

15. Trombly CA. Evaluation of biomechanical and physiologicalaspects of motor performance. In: Trombly CA (ed).Occupational Therapy for Physical Dysfunction. 4th ed.Baltimore, Md: Williams & Wilkins, 1995:73-156.

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

17. Portney LG, Watkins MP. Foundations of Clinical Research,Applications to Practice. Norwalk, Conn: Appleton & Lange,1993.

18. Shechtman 0, Taylor C. The use of the rapid exchange grip testin detecting sincerity of effort, part II: validity of the test. JHand Ther. 2000;13:203-10.