Pub

lishe

d by

Man

ey P

ublis

hing

(c)

W S

Man

ey amp

Son

Lim

ited

Intratester and Intertester Reliability ofthe Palpation Meter (PALM)in Measuring Pelvic Position

Marshall Hagins PT Martha Brown PT Clare Cook PT Karen Gstalder PT Michael Kam PTGene Kominer PT Katesel Strimbeck PT

Abstract This study determined the intra- and intertester reliability of the Palpation Meter(PALM) in measuring frontal and sagittal plane pelvic positions among asymptomatic adultsduring static standing Four examiners measured 24 physical therapy students in two trialsThe sagittal plane measurement was taken as the angle formed by a line connecting the ASISand PSIS versus the horizontal The frontal plane measurement was taken as the angle formedby a line connecting the superior border of the iliac crests versus the horizontal Unlike previ-ous studies this study attempted to replicate the realities of clinical practice by using the PALMto perform measurements over clothing without applying adhesive markers for landmarks andwithout controls for postural sway Intraclass correlation coefficients suggest intratesterreliability was high for both frontal (084) and sagittal plane measures (098) and intertesterreliability was high for sagittal plane measures (089) but moderate for frontal plane measures(065) Standard error of the means for frontal and sagittal plane measures are presented andclinicians are cautioned to observe the limitations of precision inherent in this device

Key Words Reliability Posture PALM

rontal plane and sagittal plane measurements ofpelvicsymmetry are standard components ofpostural evalu-

ationsl-9bull Frontal plane asymmetry as measured by rela-tive iliac crest height is thought to generally reflect leglength discrepancy (LLD) which has been associated withhip10-13 knee13 and low back pain10 and lower extremitystress fractures in runnersll Sagittal plane asymmetrycommonly referred to as anterior or posterior rotationofan innominate is most often considered a sign of iliosacraldysfunction1415 or as a component of LLD 16-18

Address all correspondence and request for reprints toMarshall HaginsDivision of Physical TherapyLong Island UniversityBrooklyn CampusOne University PlazaBrooklyn NY 11201

130 The Journal of Manual amp Manipulative Therapy 1998

Currently clinicians use palpation and visualizationto measure frontal and sagittal plane asymmetry in thepelvis Despite evidence that this method has poor re-liability 19-24 findings of postural asymmetry in the pel-vis are commonly used to identify dysfunctions and todirect treatment Asclinicians are increasingly challengedto validate their treatments using reliable measures andto provide quantification for interdisciplinary commu-nication and third-party reimbursement the need for asimple quick and reliable clinical measurement tool isbecoming more acute

Several studies have examined various tools that haveattempted to address the problems of reliability andquantification ofpelvic asymmetries Radiographic analysistypically has high reliability25 but has been suggested astoo expensive as inaccessible and as potentially harm-ful for routine clinical assessment26bull Although the IowaAnatomical Position System (lAPS) composed of an elec-

The Journal of Manual amp Manipulative TherapyVol 6 No3 (1998) 130 - 136

Pub

lishe

d by

Man

ey P

ublis

hing

(c)

W S

Man

ey amp

Son

Lim

ited

tromechanical probe and precision potentiometers hasbeen shown to be highly reliable and non-invasive it isnot generally available27 A simple measuring stick wasutilized in several studies that measured from floor toASIS and PSIS and calculated differences utilizing trigo-nometry262829this method was highly reliable but requiredapproximately 10 minutes to administer for a single sideSeveral studies have examined the use of a pelvic-level-ing device that measures if the iliac crests are levepo-32If an unlevel posture exists wooden shims are placed underone leg until a level reading is reached and LLD is de-termined by the height of the shims This tool is limitedto use in the frontal plane and was found in one study tohave unacceptable levels of reliability and validity32 Threestudies similar to the current study have examined theuse of caliper-inclinometer tools to measure sagittal planepelvic positions253334High levels of reliability were re-ported but the tools used in these studies were eitherhandmade or modified

A commercially produced caliper-inclinometer in-strument the Palpation Meter referred to as the PALMhas recently been introduced for clinical postural evalu-ation The potential advantages of this instrument includeuniform construction general availability direct palpa-tion of bony landmarks during testing short time periodrequired for measurement and direct measurement outputin degrees Reliability of this instrument has not previ-ously been reported

The purpose of this study was to examine the reli-ability of the PALMin measuring frontal and sagittal planeposition of the pelvis in asymptomatic subjects An at-tempt was made to design the methodology so as to mirrorthe realities of current clinical practice by avoiding con-straints that may increase reliability but decreasegeneralizability such as (1) requiring subjects to disrobeadequately to palpate bony landmarks directly 25-2931-3335(2) applying adhesive markers to bony landmarks 252628333(3) restricting postural sway253334

None of the previous studies examined asymmetrybetween innominates considered by many orthopedic andmanual therapists to be a critical measure3-69 This studymeasured both sides of the pelvis in the sagittal planeand examined relative innominate tilt

PALM Performance Attainment Associates 3550LaBore Road Suite 8 St Paul MN 55110

Methods

Subjects24 subjects were selected as a sample of convenience

from the Physical Therapy Program at Long Island Uni-versity Exclusion criteria included a self-reported inabilityto stand for 20 minutes See Table 1 for demographicdata

InstrumentationThe PALM combines the features of a caliper and an

inclinometer (Figure 1) The analog caliper dial used todetermine distance between the caliper arms was not usedin this study and will not be described further The bodyof the PALM contains a bubble level in a semi-circulararc with one degree gradations that range from zero degreesto thirty degrees on either side of the midline The PALMis suspended from around the examiners neck by anadjustable cord and held with both hands directly in frontof the body The caliper arms are placed on bony land-marks and the degree of deviation from horizontal isread from the inclinometer Three types of removabletips are available for the ends of the caliper arms The0 tips used in this study were designed for palpationconcurrent with measurement by allowing the fingertipsto protrude through a flexible rubber circle and comeinto direct contact with the landmark

ProcedureAll subjects read and signed an informed consent

Subjects were required to wear gym shorts T-shirts andno shoes Subjects were asked to march in place ten timesand stand in a normal static position with equal weightbearing on both lower extremities The width of theirstance was self-selected but the distal tip of their halluxwas adjusted to be touching a floor tile border lying inthe frontal plane To prevent their arms from being inthe examiners path subjects were asked to fold their armsacross their chest The subjects were instructed to notlet the examiner push them forward or backward duringthe measurement procedure

Four examiners measured all subjects in two trialsin one day Two subjects were measured in the same time

Table 1 Mean and range for subjects age (years) height (feetinches) weight (lbs)

1-~-l~IITIl~f iIJ~ I~J~rfif~-)J il~l~~)J

male 9 286 (21-43) 510 (56-63) 174 (155-210)

female 15 27 (21-40) 55 (52-595) 131 (112-174)

Intratester and Intertester Reliability of the Palpation Meter(PALM) in Measuring Pelvic Position I 131

Pub

lishe

d by

Man

ey P

ublis

hing

(c)

W S

Man

ey amp

Son

Lim

ited

Fig 1 Palpation Meter

period utilizing two PALMs to speed data collection Theorder of examiners for each pair of subjects was randomlyselected As examiners completed a measurement theyreported the value to an assistant who documented thevalue After the initial pair of examiners completed Trial1 the second pair of examiners completed Trial 1 Thisprocess was repeated for Trial 2 immediately followingthe completion of Trial 1

The same procedure for measurement was used byall examiners and was discussed in detail prior to datacollection Palpation occurred over clothing Palpationwas first performed without the PALM in order to derivean initial kinesthetic sense of the location of the land-marks and to determine how much adipose tissue existedand how much force would be needed to find a firm endfeel The examiners then placed their index or middlefinger into the O-tips and repeated the initial palpa-tion and determined the value from the inclinometer Theexaminers then released the PALMand repeated the exactprocess again providing two independent values to therecorder for averaging prior to analysis

The order of measurement was arbitrarily chosen andstandardized frontal plane left sagittal plane right sag-ittal plane Examiners first stood behind each subjectand palpated bilaterally moving from the lateral aspectsof the abdomen to the superior border of the iliac crestsAttempts were made to move adipose tissue as necessaryand to apply firm pressure to determine the superiorborder of the iliac crest For sagittal plane measuresexaminers moved to a lateral position relative to the subjectand were instructed to scoop under the ASIS and PSIS

132 The Journal of Manual amp Manipulative Therapy 1998

and to move superiorly to the most outwardly projectingpoint for location of the landmark In order to reduceerror attempts were made to maintain a parallel posi-tion of the PALM to the horizontal line connecting theexaminer to the subject This often required examinersto adjust their height by bending their knees or to ad-just the length of string supporting the PALM By con-vention an inferior position of the left caliper arm wasassigned a negative value and an inferior position of theright caliper arm was assigned a positive value Thisconvention was taken into account during analysis

Data AnalysisMeasurements obtained from each subject were sum-

marized using descriptive statisticsIntraclass Correlation Coefficients (21) were computedto indicate the agreement within testers and between testersSagittal plane measures within individuals for Trial 1 wereexamined for relative differences in innominate rotationand described

ResultsA summary of descriptive statistics is presented in

Table 2 Although men and women had similar mean frontalplane values the mean values for sagittal plane were quitedifferent with women having approximately twice as muchdeviation from the horizontal as men Mean sagittal planevalues for left and right sides were 734 degrees and 693degrees respectively Table 3 presents information on asubset of the data (Trial 1 examiners 1 and 2) relatingdegree of innominate tilt (1deg_7deg) to percentage of sub-jects A separate analysis using the same subset of databut not shown in Table 3 determined that the left in-nominate was anterior 43 of the time the right innominatewas anterior 52 of the time and 4 (1 person) hadlevel innominates

Intraclass Correlation Coefficients and their corre-sponding lower confidence intervals are presented in Table4 Intratester reliability was high for both frontal (084)and sagittal (098) plane measures 36 Intertester reli-ability was high for sagittal (089) plane measures butmoderate for frontal (065) plane measures36bull

Discussion

Frontal PlaneNo studies were found that utilized an inclinometer

with a direct output in degrees to measure relative iliaccrest heights Studies have been performed using a pelvicleveling device that indicates the presence or absence ofa horizontal position3031 In these studies examiners placedshims with known values ranging from 32-5 mm thickunder the low iliac crest until a level position is reached

Pub

lishe

d by

Man

ey P

ublis

hing

(c)

W S

Man

ey amp

Son

Lim

ited

Table 2 Descriptive statistics All values in degrees Standard Deviation (SD) Standard Error of the Mean (SEM)

1~~li)jJ~(~]fiLtrfff[fID~nJ

Frontal

SagittalLeft

SagittalRight

i~il~Iij

-077t

734

693

Ilft3rlmlill11UVltftIiJmale -075female -08

male -48female -87

male 46female 83

~]J

145

52

sectsectfl)755

3658

(~~~

2 - (-65)t

15 - (-19)t

t Negative values represent inferior position of left arm of caliper in relation to examiner Combined left and right sagittal values

Table 3 Degree of difference in sagittal plane measures computed by averaging values of examiner 1 and 2 duringTrial 1 and comparing right and left innominates

Differenceright vs leftinnominate

Percentageof subjects

400 000

Table 4 ICC and lower confidence limits (CI) for Intra- and Intertester reliability

[~lIlil9J rJijJfu1J~~1~f IJjftmll~lIff IIH~ln~lt~Ii1mJIfnmill [f3 11l~lJ21rftlaj1emiddotJ f

lIn~l~ml~lr11E)413YillY~4 O

Frontal

Combinedright and leftsagittal

084

098

070

095

065

089

047

070

and then count the shims to determine height differencesIf the average distance between femoral heads as sug-gested by the literature is 20 cm37 approximate com-parisons to these studies can be performed using trigo-nometry (Table 5) Each degree of relative tilt read fromthe PALM is equal to approximately 35 mm of iliac crestheight difference Consequently from Table 2 the meandifference found in this study in iliac crest height was27 mm and the SD was 51 mm The mean values werealmost identical to those found by Jonson and Gross31

and Woerman et apo at 22 mm and 32 mm respectivelyJonson and Gross31 found a SD of 26 mm and Woermanet apo found a SD of 39 mm The variability in this studywas higher than previous studies and may be explained

by the methodology which avoided constraints not nor-mally encountered in clinical practice

Sagittal PlaneOther studies25263338 examining the angle formed be-

tween a line connecting the ASIS to PSIS and the hori-zontal have found mean values ranging from 835 de-grees26 to 113 degrees38bull This relatively narrow range wasfound despite various measurement methods and the useof both men and women as subjects These mean valuescorrespond well to those found in this study for womenbut differ markedly from those found in men It is un-clear why this distinction between men and women occured

Intratester and Intertester Reliability of the Palpation Meter(PALM) in Measuring Pelvic Position 133

Pub

lishe

d by

Man

ey P

ublis

hing

(c)

W S

Man

ey amp

Son

Lim

ited

Table 5 Translation of PALM measures in frontal plane to trigonometrically derived height difference in the iliaccrests based on average of approximately 8 inches37 between femoral heads

~-li _I ~ ~IFl~ ~fr~~~(~1 n~ = Iil middoti)ii~ - iHIPii ~ lnlJrIisect~ -_ __-

1deg 349 349 1372deg 698 698 27483deg 1048 1048 41254deg 1398 1398 55045deg 1749 1749 68896deg 2102 2102 82687deg 2455 2455 9645

The standard deviations reported in three of the previ-ously mentioned studies263338 for sagittal plane measurementvaried between 34 degrees and 43 degrees these valuesagree reasonably well with the value in this study of 52degrees The deliberate approach in this study to avoidconstraints on the measurement process beyond what islikely to occur in the clinic (placing adhesive markerson landmarks etc) may be responsible for the greaterdegree of variability

The apparent symmetry in the sagittal plane betweenright and left innominates in Table 2 appears to be acharacteristic of summary statistics rather than a typi-cal finding on an individual Approximately half of thesubjects were anterior on the right the other half wereanterior on the left Importantly the degree of asym-metry was relatively small between innominates with96 of the subjects having less than or equal to 4 de-grees of asymmetry (Table 3)

Although it is possible in clinical practice to ask patientsto disrobe sufficiently to expose the ASIS and PSIS andto mark landmarks with adhesive markers and to havethem brace themselves against a thoracic or femoral supportthe authors felt that constraints were both impracticaland improbable in a busy clinic Consequently the de-cision was made to utilize the PALM in a manner thoughtto replicate probable clinical practice with this new in-strument so that the observed reliability could be au-thentically generalizable The high reliability found inthe sagittal plane for both intra- and intertester mea-sures and in the frontal plane for intratester measures isremarkable given such a lack of controls

The ICC is based on an analysis of variance that par-titions variance into categories for comparison The mod-erate intertester frontal plane ICC value (065) possiblyreflects the lack of variance in the true values of frontalplane position in this asymptomatic population It is probablethat including a number of subjects with genuine leg lengthdiscrepancies would have increased the variance of thetrue values and provided higher ICC values for reliabil-

134 The Journal of Manual amp Manipulative Therapy 1998

ity This can be appreciated by comparing the SD valuesbetween frontal (SD = 145) and sagittal (SD = 52) planemeasures Future studies should address reliability insymptomatic populations

Although the ICC values found in this study gener-ally suggest that measured values are consistent the clinicaluse of the PALM must also be evaluated in terms of theprecision of the instrument The SEM for sagittal planemeasures was approximately 37 degrees Clinically thisvalue indicates that the examiner cannot be certain thatone measurement in the sagittal plane is different (romanother unless that difference is more than two timesthe SEM or 74 degrees A necessary change of 74 de-grees is relatively large when one considers that someauthors suggest that the sacro-iliac joint has a range ofmotion of 1-11 degrees1739 We found that 96 of all rightversus left innominate differences were 4 degrees or lessIf a clinician measures an anterior innominate of 14 degreesprior to an intervention and measures a value of 9 de-grees after an intervention he or she cannot be certainthat a change has occurred However if initial measure-ments determine that the right innominate is tilted 8degrees and the left is tilted 0 degrees and an interven-tion succeeds in creating an 8 degree tilt on the left thenthe clinician can state that a real change has occurred

Similar reasoning can be followed for frontal planemeasurements but in this case the SEM is 755 degreesindicating a necessary difference of 15 degrees beforethe clinician can be certain a difference exists A 15 degreedifference translates to 52 mm height difference betweeniliac crests The amount of LLD necessary for symptomgeneration is controversial but a recent survey of theliterature suggests that 10 mm is r equired40bull The PALMtherefore appears to have sufficient precision to deter-mine through the indirect method of iliac crest heightmeasurement if a significant LLD exists

A few points regarding methodology should be men-tioned Examiners were not blinded to their own mea-surements as it was felt that the instrument required a

Pub

lishe

d by

Man

ey P

ublis

hing

(c)

W S

Man

ey amp

Son

Lim

ited

direct vertical view for accurate reading of the inclinom-eter At worst this may have inflated the intratester reliabilitydespite attempts to direct the examiners to perform eachmeasure as an isolated event The intertester reliabilityvalues were not subject to this source of error and tendedto fall as expected slightly lower than the intratester valuessuggesting little effect from the lack of blinding on theintratester values Another source of error was the po-tential differences between the two PALMs used in thisstudy Again the authors felt the value of this study wasin creating a clinically realistic methodology using aninstrument that is commercially produced and readilyavailable It was decided that any error created by mul-tiple versions of the instrument although potentially usefulto know should not be separated out in this study

ConclusionsIn summary this study examined standing static pelvic

REFERENCES1 Hoppenfeld S Physical Examination of the Spine and Extremi-

ties Norwalk Appleton- Century-Crofts 1976 pp 239-2422 Kendall F McCreary E Muscles Testing and Function 3rd ed

BaltimoreWilliams and WIlkins 1983 pp 2213 Denslow JS Chace JA Mechanical Stresses in the Human Lumbar

Spine and Pelvis J Am Osteopath Assoc 1962 61706-7124 DonTigny RL Measuring PSIS movement Clinical Management

1990 1043-445 DonTigny RL Dysfunction of the Sacroiliac Joint and its

Treatment Journal Orthopedic and Sports Physical Therapy 1979123-25

6 Grieve GP The Sacro-iliac Joint Physiotherapy 1976 62384-4007 Donatelli RAWooden MJ Orthopaedic Physical Therapy New York

Churchill Livingstone 1993 pp 521-5238 Greenman PE Principles of Manual Medicine 20th ed Baltimore

Williams and Wilkins 1989 pp 20-229 Subotnick SI Limb Length Discrepancies of the Lower

Extremity Journal Orthopedic and Sports Physical Therapy 1985311-16

10 Friberg O Clinical Symptoms and Biomechanics of Lumbar Spineand Hip Joint in Leg Length Inequality Spine 1983 8643-651

11 McCaw ST Leg length inequality Implications for Runninginjury Prevention Sports Med 1992 14422-429

12 Rothenberg RJ Rheumatic Disease Aspects of Leg Length InequalitySemin Arthritis Rheum 1988 17196-205

13 Tjernstrom B Olerud S Karlstrom G Direct Leg Lengthening JOrthop Trauma 1993 6543-551

14 DonTigny RL Anterior Dysfunction of the Sacroiliac Joint as aMajor Factor in the Etiology of Idiopathic Low Back PainSyndrome Phys Ther 1990 70250-265

15 Cibulka MT Koldehoff R Leg Length Disparity and its Effect onSacroiliac Joint Dysfunction Clinical Management 1986 610-11

posture in asymptomatic subjects using the PALM andfound that intratester reliability was high for both frontaland sagittal plane measures and that intertester reli-ability was high for sagittal plane measures but moder-ate for frontal plane measures Mean values agreedreasonably well with other studies while variability wasslightly higher potentially due to a methodology thatattempted to replicate the lack of constraints typicallyfound in clinical practice The clinical use of the PALMshould always be performed with consideration givento the limitations inherent in the precision of the in-strument

AcknowledgmentsWe thank Bill Susman PhD for his generous assis-

tance in reviewing initial drafts of this study We alsothank Rudi Hiebert for his invaluable guidance with thestatistical analysis 0

16 Cummings G Scholz Jp Barnes K The Effect ofImposed Leg LengthDifference on Plevic Bone Symmetry Spine 1993 18368-373

17 Pitkin H Pheasant HC A Study of Sacral Mobility Journal of Boneand Joint Surgery 1936 18365-374

18 Chamberlain E The Symphsis Pubis in the Roentgen Examina-tion of the Sacroiliac Joint Radium Therapy and Nuclear Medi-cine 1930 24621-625

19 Potter NA Rothstein JM Intertester Reliability for Selected Clini-cal Tests of the Sacroiliac Joint Phys Ther 1985 111671-1675

20 Mann M Glasheen-Wray M Nyberg R Therapist Agreement forPalpation and Observation of Iliac Crest Heights Phys Ther 198464334-338

21 Clarke GR Unequal Leg Length An Accurate method of Detectionand Some Clinical Results Rheumatology and Physical Medicine1972 11385-390

22 Bailey HW Beckwith CG Short Leg and Spinal Anomalies TheirIncidence and Effects on Spinal Mechanics J Am Osteopath Assoc1937 36319-327

23 Clark GR Unequal Leg Length An Accurate Method of Detectionand Some Clinical Results Rheum Phys Med 1972 11385-390

24 Kerr HE Grant JH MacBain RN Some Observations on theAnatomical Short Leg in a Series of Patients Presenting Them-selves for Treatment of Low-Back Pain J Am Osteopath Assoc 194342437 -440

25 Gilliam J Brunt D MacMillian M Kinard RE Montgomery WJRelationship of the Pelvic Angle to the Sacral Angle Meaurementof Clinical Reliability and Validity Journal Orthopedic and SportsPhysical Therapy 1994 20193-199

26 Gajdosik R Simpson R Smith R DonTigny RL IntratesterReliability of Measuring the Standing Position and Range of MotionPhys Ther 1985 65169-173

27 Day JW Smidt GL Lehman T Effect of Pelvic Tilt on Standing

Intratester and Intertester Reliability of the Palpation Meter(PALM) in Measuring Pelvic Position I 135

Pub

lishe

d by

Man

ey P

ublis

hing

(c)

W S

Man

ey amp

Son

Lim

ited

tromechanical probe and precision potentiometers hasbeen shown to be highly reliable and non-invasive it isnot generally available27 A simple measuring stick wasutilized in several studies that measured from floor toASIS and PSIS and calculated differences utilizing trigo-nometry262829this method was highly reliable but requiredapproximately 10 minutes to administer for a single sideSeveral studies have examined the use of a pelvic-level-ing device that measures if the iliac crests are levepo-32If an unlevel posture exists wooden shims are placed underone leg until a level reading is reached and LLD is de-termined by the height of the shims This tool is limitedto use in the frontal plane and was found in one study tohave unacceptable levels of reliability and validity32 Threestudies similar to the current study have examined theuse of caliper-inclinometer tools to measure sagittal planepelvic positions253334High levels of reliability were re-ported but the tools used in these studies were eitherhandmade or modified

A commercially produced caliper-inclinometer in-strument the Palpation Meter referred to as the PALMhas recently been introduced for clinical postural evalu-ation The potential advantages of this instrument includeuniform construction general availability direct palpa-tion of bony landmarks during testing short time periodrequired for measurement and direct measurement outputin degrees Reliability of this instrument has not previ-ously been reported

The purpose of this study was to examine the reli-ability of the PALMin measuring frontal and sagittal planeposition of the pelvis in asymptomatic subjects An at-tempt was made to design the methodology so as to mirrorthe realities of current clinical practice by avoiding con-straints that may increase reliability but decreasegeneralizability such as (1) requiring subjects to disrobeadequately to palpate bony landmarks directly 25-2931-3335(2) applying adhesive markers to bony landmarks 252628333(3) restricting postural sway253334

None of the previous studies examined asymmetrybetween innominates considered by many orthopedic andmanual therapists to be a critical measure3-69 This studymeasured both sides of the pelvis in the sagittal planeand examined relative innominate tilt

PALM Performance Attainment Associates 3550LaBore Road Suite 8 St Paul MN 55110

Methods

Subjects24 subjects were selected as a sample of convenience

from the Physical Therapy Program at Long Island Uni-versity Exclusion criteria included a self-reported inabilityto stand for 20 minutes See Table 1 for demographicdata

InstrumentationThe PALM combines the features of a caliper and an

inclinometer (Figure 1) The analog caliper dial used todetermine distance between the caliper arms was not usedin this study and will not be described further The bodyof the PALM contains a bubble level in a semi-circulararc with one degree gradations that range from zero degreesto thirty degrees on either side of the midline The PALMis suspended from around the examiners neck by anadjustable cord and held with both hands directly in frontof the body The caliper arms are placed on bony land-marks and the degree of deviation from horizontal isread from the inclinometer Three types of removabletips are available for the ends of the caliper arms The0 tips used in this study were designed for palpationconcurrent with measurement by allowing the fingertipsto protrude through a flexible rubber circle and comeinto direct contact with the landmark

ProcedureAll subjects read and signed an informed consent

Subjects were required to wear gym shorts T-shirts andno shoes Subjects were asked to march in place ten timesand stand in a normal static position with equal weightbearing on both lower extremities The width of theirstance was self-selected but the distal tip of their halluxwas adjusted to be touching a floor tile border lying inthe frontal plane To prevent their arms from being inthe examiners path subjects were asked to fold their armsacross their chest The subjects were instructed to notlet the examiner push them forward or backward duringthe measurement procedure

Four examiners measured all subjects in two trialsin one day Two subjects were measured in the same time

Table 1 Mean and range for subjects age (years) height (feetinches) weight (lbs)

1-~-l~IITIl~f iIJ~ I~J~rfif~-)J il~l~~)J

male 9 286 (21-43) 510 (56-63) 174 (155-210)

female 15 27 (21-40) 55 (52-595) 131 (112-174)

Intratester and Intertester Reliability of the Palpation Meter(PALM) in Measuring Pelvic Position I 131

Pub

lishe

d by

Man

ey P

ublis

hing

(c)

W S

Man

ey amp

Son

Lim

ited

Fig 1 Palpation Meter

period utilizing two PALMs to speed data collection Theorder of examiners for each pair of subjects was randomlyselected As examiners completed a measurement theyreported the value to an assistant who documented thevalue After the initial pair of examiners completed Trial1 the second pair of examiners completed Trial 1 Thisprocess was repeated for Trial 2 immediately followingthe completion of Trial 1

The same procedure for measurement was used byall examiners and was discussed in detail prior to datacollection Palpation occurred over clothing Palpationwas first performed without the PALM in order to derivean initial kinesthetic sense of the location of the land-marks and to determine how much adipose tissue existedand how much force would be needed to find a firm endfeel The examiners then placed their index or middlefinger into the O-tips and repeated the initial palpa-tion and determined the value from the inclinometer Theexaminers then released the PALMand repeated the exactprocess again providing two independent values to therecorder for averaging prior to analysis

The order of measurement was arbitrarily chosen andstandardized frontal plane left sagittal plane right sag-ittal plane Examiners first stood behind each subjectand palpated bilaterally moving from the lateral aspectsof the abdomen to the superior border of the iliac crestsAttempts were made to move adipose tissue as necessaryand to apply firm pressure to determine the superiorborder of the iliac crest For sagittal plane measuresexaminers moved to a lateral position relative to the subjectand were instructed to scoop under the ASIS and PSIS

132 The Journal of Manual amp Manipulative Therapy 1998

and to move superiorly to the most outwardly projectingpoint for location of the landmark In order to reduceerror attempts were made to maintain a parallel posi-tion of the PALM to the horizontal line connecting theexaminer to the subject This often required examinersto adjust their height by bending their knees or to ad-just the length of string supporting the PALM By con-vention an inferior position of the left caliper arm wasassigned a negative value and an inferior position of theright caliper arm was assigned a positive value Thisconvention was taken into account during analysis

Data AnalysisMeasurements obtained from each subject were sum-

marized using descriptive statisticsIntraclass Correlation Coefficients (21) were computedto indicate the agreement within testers and between testersSagittal plane measures within individuals for Trial 1 wereexamined for relative differences in innominate rotationand described

ResultsA summary of descriptive statistics is presented in

Table 2 Although men and women had similar mean frontalplane values the mean values for sagittal plane were quitedifferent with women having approximately twice as muchdeviation from the horizontal as men Mean sagittal planevalues for left and right sides were 734 degrees and 693degrees respectively Table 3 presents information on asubset of the data (Trial 1 examiners 1 and 2) relatingdegree of innominate tilt (1deg_7deg) to percentage of sub-jects A separate analysis using the same subset of databut not shown in Table 3 determined that the left in-nominate was anterior 43 of the time the right innominatewas anterior 52 of the time and 4 (1 person) hadlevel innominates

Intraclass Correlation Coefficients and their corre-sponding lower confidence intervals are presented in Table4 Intratester reliability was high for both frontal (084)and sagittal (098) plane measures 36 Intertester reli-ability was high for sagittal (089) plane measures butmoderate for frontal (065) plane measures36bull

Discussion

Frontal PlaneNo studies were found that utilized an inclinometer

with a direct output in degrees to measure relative iliaccrest heights Studies have been performed using a pelvicleveling device that indicates the presence or absence ofa horizontal position3031 In these studies examiners placedshims with known values ranging from 32-5 mm thickunder the low iliac crest until a level position is reached

Pub

lishe

d by

Man

ey P

ublis

hing

(c)

W S

Man

ey amp

Son

Lim

ited

Table 2 Descriptive statistics All values in degrees Standard Deviation (SD) Standard Error of the Mean (SEM)

1~~li)jJ~(~]fiLtrfff[fID~nJ

Frontal

SagittalLeft

SagittalRight

i~il~Iij

-077t

734

693

Ilft3rlmlill11UVltftIiJmale -075female -08

male -48female -87

male 46female 83

~]J

145

52

sectsectfl)755

3658

(~~~

2 - (-65)t

15 - (-19)t

t Negative values represent inferior position of left arm of caliper in relation to examiner Combined left and right sagittal values

Table 3 Degree of difference in sagittal plane measures computed by averaging values of examiner 1 and 2 duringTrial 1 and comparing right and left innominates

Differenceright vs leftinnominate

Percentageof subjects

400 000

Table 4 ICC and lower confidence limits (CI) for Intra- and Intertester reliability

[~lIlil9J rJijJfu1J~~1~f IJjftmll~lIff IIH~ln~lt~Ii1mJIfnmill [f3 11l~lJ21rftlaj1emiddotJ f

lIn~l~ml~lr11E)413YillY~4 O

Frontal

Combinedright and leftsagittal

084

098

070

095

065

089

047

070

and then count the shims to determine height differencesIf the average distance between femoral heads as sug-gested by the literature is 20 cm37 approximate com-parisons to these studies can be performed using trigo-nometry (Table 5) Each degree of relative tilt read fromthe PALM is equal to approximately 35 mm of iliac crestheight difference Consequently from Table 2 the meandifference found in this study in iliac crest height was27 mm and the SD was 51 mm The mean values werealmost identical to those found by Jonson and Gross31

and Woerman et apo at 22 mm and 32 mm respectivelyJonson and Gross31 found a SD of 26 mm and Woermanet apo found a SD of 39 mm The variability in this studywas higher than previous studies and may be explained

by the methodology which avoided constraints not nor-mally encountered in clinical practice

Sagittal PlaneOther studies25263338 examining the angle formed be-

tween a line connecting the ASIS to PSIS and the hori-zontal have found mean values ranging from 835 de-grees26 to 113 degrees38bull This relatively narrow range wasfound despite various measurement methods and the useof both men and women as subjects These mean valuescorrespond well to those found in this study for womenbut differ markedly from those found in men It is un-clear why this distinction between men and women occured

Intratester and Intertester Reliability of the Palpation Meter(PALM) in Measuring Pelvic Position 133

Pub

lishe

d by

Man

ey P

ublis

hing

(c)

W S

Man

ey amp

Son

Lim

ited

Table 5 Translation of PALM measures in frontal plane to trigonometrically derived height difference in the iliaccrests based on average of approximately 8 inches37 between femoral heads

~-li _I ~ ~IFl~ ~fr~~~(~1 n~ = Iil middoti)ii~ - iHIPii ~ lnlJrIisect~ -_ __-

1deg 349 349 1372deg 698 698 27483deg 1048 1048 41254deg 1398 1398 55045deg 1749 1749 68896deg 2102 2102 82687deg 2455 2455 9645

The standard deviations reported in three of the previ-ously mentioned studies263338 for sagittal plane measurementvaried between 34 degrees and 43 degrees these valuesagree reasonably well with the value in this study of 52degrees The deliberate approach in this study to avoidconstraints on the measurement process beyond what islikely to occur in the clinic (placing adhesive markerson landmarks etc) may be responsible for the greaterdegree of variability

The apparent symmetry in the sagittal plane betweenright and left innominates in Table 2 appears to be acharacteristic of summary statistics rather than a typi-cal finding on an individual Approximately half of thesubjects were anterior on the right the other half wereanterior on the left Importantly the degree of asym-metry was relatively small between innominates with96 of the subjects having less than or equal to 4 de-grees of asymmetry (Table 3)

Although it is possible in clinical practice to ask patientsto disrobe sufficiently to expose the ASIS and PSIS andto mark landmarks with adhesive markers and to havethem brace themselves against a thoracic or femoral supportthe authors felt that constraints were both impracticaland improbable in a busy clinic Consequently the de-cision was made to utilize the PALM in a manner thoughtto replicate probable clinical practice with this new in-strument so that the observed reliability could be au-thentically generalizable The high reliability found inthe sagittal plane for both intra- and intertester mea-sures and in the frontal plane for intratester measures isremarkable given such a lack of controls

The ICC is based on an analysis of variance that par-titions variance into categories for comparison The mod-erate intertester frontal plane ICC value (065) possiblyreflects the lack of variance in the true values of frontalplane position in this asymptomatic population It is probablethat including a number of subjects with genuine leg lengthdiscrepancies would have increased the variance of thetrue values and provided higher ICC values for reliabil-

134 The Journal of Manual amp Manipulative Therapy 1998

ity This can be appreciated by comparing the SD valuesbetween frontal (SD = 145) and sagittal (SD = 52) planemeasures Future studies should address reliability insymptomatic populations

Although the ICC values found in this study gener-ally suggest that measured values are consistent the clinicaluse of the PALM must also be evaluated in terms of theprecision of the instrument The SEM for sagittal planemeasures was approximately 37 degrees Clinically thisvalue indicates that the examiner cannot be certain thatone measurement in the sagittal plane is different (romanother unless that difference is more than two timesthe SEM or 74 degrees A necessary change of 74 de-grees is relatively large when one considers that someauthors suggest that the sacro-iliac joint has a range ofmotion of 1-11 degrees1739 We found that 96 of all rightversus left innominate differences were 4 degrees or lessIf a clinician measures an anterior innominate of 14 degreesprior to an intervention and measures a value of 9 de-grees after an intervention he or she cannot be certainthat a change has occurred However if initial measure-ments determine that the right innominate is tilted 8degrees and the left is tilted 0 degrees and an interven-tion succeeds in creating an 8 degree tilt on the left thenthe clinician can state that a real change has occurred

Similar reasoning can be followed for frontal planemeasurements but in this case the SEM is 755 degreesindicating a necessary difference of 15 degrees beforethe clinician can be certain a difference exists A 15 degreedifference translates to 52 mm height difference betweeniliac crests The amount of LLD necessary for symptomgeneration is controversial but a recent survey of theliterature suggests that 10 mm is r equired40bull The PALMtherefore appears to have sufficient precision to deter-mine through the indirect method of iliac crest heightmeasurement if a significant LLD exists

A few points regarding methodology should be men-tioned Examiners were not blinded to their own mea-surements as it was felt that the instrument required a

Pub

lishe

d by

Man

ey P

ublis

hing

(c)

W S

Man

ey amp

Son

Lim

ited

direct vertical view for accurate reading of the inclinom-eter At worst this may have inflated the intratester reliabilitydespite attempts to direct the examiners to perform eachmeasure as an isolated event The intertester reliabilityvalues were not subject to this source of error and tendedto fall as expected slightly lower than the intratester valuessuggesting little effect from the lack of blinding on theintratester values Another source of error was the po-tential differences between the two PALMs used in thisstudy Again the authors felt the value of this study wasin creating a clinically realistic methodology using aninstrument that is commercially produced and readilyavailable It was decided that any error created by mul-tiple versions of the instrument although potentially usefulto know should not be separated out in this study

ConclusionsIn summary this study examined standing static pelvic

REFERENCES1 Hoppenfeld S Physical Examination of the Spine and Extremi-

ties Norwalk Appleton- Century-Crofts 1976 pp 239-2422 Kendall F McCreary E Muscles Testing and Function 3rd ed

BaltimoreWilliams and WIlkins 1983 pp 2213 Denslow JS Chace JA Mechanical Stresses in the Human Lumbar

Spine and Pelvis J Am Osteopath Assoc 1962 61706-7124 DonTigny RL Measuring PSIS movement Clinical Management

1990 1043-445 DonTigny RL Dysfunction of the Sacroiliac Joint and its

Treatment Journal Orthopedic and Sports Physical Therapy 1979123-25

6 Grieve GP The Sacro-iliac Joint Physiotherapy 1976 62384-4007 Donatelli RAWooden MJ Orthopaedic Physical Therapy New York

Churchill Livingstone 1993 pp 521-5238 Greenman PE Principles of Manual Medicine 20th ed Baltimore

Williams and Wilkins 1989 pp 20-229 Subotnick SI Limb Length Discrepancies of the Lower

Extremity Journal Orthopedic and Sports Physical Therapy 1985311-16

10 Friberg O Clinical Symptoms and Biomechanics of Lumbar Spineand Hip Joint in Leg Length Inequality Spine 1983 8643-651

11 McCaw ST Leg length inequality Implications for Runninginjury Prevention Sports Med 1992 14422-429

12 Rothenberg RJ Rheumatic Disease Aspects of Leg Length InequalitySemin Arthritis Rheum 1988 17196-205

13 Tjernstrom B Olerud S Karlstrom G Direct Leg Lengthening JOrthop Trauma 1993 6543-551

14 DonTigny RL Anterior Dysfunction of the Sacroiliac Joint as aMajor Factor in the Etiology of Idiopathic Low Back PainSyndrome Phys Ther 1990 70250-265

15 Cibulka MT Koldehoff R Leg Length Disparity and its Effect onSacroiliac Joint Dysfunction Clinical Management 1986 610-11

posture in asymptomatic subjects using the PALM andfound that intratester reliability was high for both frontaland sagittal plane measures and that intertester reli-ability was high for sagittal plane measures but moder-ate for frontal plane measures Mean values agreedreasonably well with other studies while variability wasslightly higher potentially due to a methodology thatattempted to replicate the lack of constraints typicallyfound in clinical practice The clinical use of the PALMshould always be performed with consideration givento the limitations inherent in the precision of the in-strument

AcknowledgmentsWe thank Bill Susman PhD for his generous assis-

tance in reviewing initial drafts of this study We alsothank Rudi Hiebert for his invaluable guidance with thestatistical analysis 0

16 Cummings G Scholz Jp Barnes K The Effect ofImposed Leg LengthDifference on Plevic Bone Symmetry Spine 1993 18368-373

17 Pitkin H Pheasant HC A Study of Sacral Mobility Journal of Boneand Joint Surgery 1936 18365-374

18 Chamberlain E The Symphsis Pubis in the Roentgen Examina-tion of the Sacroiliac Joint Radium Therapy and Nuclear Medi-cine 1930 24621-625

19 Potter NA Rothstein JM Intertester Reliability for Selected Clini-cal Tests of the Sacroiliac Joint Phys Ther 1985 111671-1675

20 Mann M Glasheen-Wray M Nyberg R Therapist Agreement forPalpation and Observation of Iliac Crest Heights Phys Ther 198464334-338

21 Clarke GR Unequal Leg Length An Accurate method of Detectionand Some Clinical Results Rheumatology and Physical Medicine1972 11385-390

22 Bailey HW Beckwith CG Short Leg and Spinal Anomalies TheirIncidence and Effects on Spinal Mechanics J Am Osteopath Assoc1937 36319-327

23 Clark GR Unequal Leg Length An Accurate Method of Detectionand Some Clinical Results Rheum Phys Med 1972 11385-390

24 Kerr HE Grant JH MacBain RN Some Observations on theAnatomical Short Leg in a Series of Patients Presenting Them-selves for Treatment of Low-Back Pain J Am Osteopath Assoc 194342437 -440

25 Gilliam J Brunt D MacMillian M Kinard RE Montgomery WJRelationship of the Pelvic Angle to the Sacral Angle Meaurementof Clinical Reliability and Validity Journal Orthopedic and SportsPhysical Therapy 1994 20193-199

26 Gajdosik R Simpson R Smith R DonTigny RL IntratesterReliability of Measuring the Standing Position and Range of MotionPhys Ther 1985 65169-173

27 Day JW Smidt GL Lehman T Effect of Pelvic Tilt on Standing

Intratester and Intertester Reliability of the Palpation Meter(PALM) in Measuring Pelvic Position I 135

Pub

lishe

d by

Man

ey P

ublis

hing

(c)

W S

Man

ey amp

Son

Lim

ited

Fig 1 Palpation Meter

period utilizing two PALMs to speed data collection Theorder of examiners for each pair of subjects was randomlyselected As examiners completed a measurement theyreported the value to an assistant who documented thevalue After the initial pair of examiners completed Trial1 the second pair of examiners completed Trial 1 Thisprocess was repeated for Trial 2 immediately followingthe completion of Trial 1

The same procedure for measurement was used byall examiners and was discussed in detail prior to datacollection Palpation occurred over clothing Palpationwas first performed without the PALM in order to derivean initial kinesthetic sense of the location of the land-marks and to determine how much adipose tissue existedand how much force would be needed to find a firm endfeel The examiners then placed their index or middlefinger into the O-tips and repeated the initial palpa-tion and determined the value from the inclinometer Theexaminers then released the PALMand repeated the exactprocess again providing two independent values to therecorder for averaging prior to analysis

The order of measurement was arbitrarily chosen andstandardized frontal plane left sagittal plane right sag-ittal plane Examiners first stood behind each subjectand palpated bilaterally moving from the lateral aspectsof the abdomen to the superior border of the iliac crestsAttempts were made to move adipose tissue as necessaryand to apply firm pressure to determine the superiorborder of the iliac crest For sagittal plane measuresexaminers moved to a lateral position relative to the subjectand were instructed to scoop under the ASIS and PSIS

132 The Journal of Manual amp Manipulative Therapy 1998

and to move superiorly to the most outwardly projectingpoint for location of the landmark In order to reduceerror attempts were made to maintain a parallel posi-tion of the PALM to the horizontal line connecting theexaminer to the subject This often required examinersto adjust their height by bending their knees or to ad-just the length of string supporting the PALM By con-vention an inferior position of the left caliper arm wasassigned a negative value and an inferior position of theright caliper arm was assigned a positive value Thisconvention was taken into account during analysis

Data AnalysisMeasurements obtained from each subject were sum-

marized using descriptive statisticsIntraclass Correlation Coefficients (21) were computedto indicate the agreement within testers and between testersSagittal plane measures within individuals for Trial 1 wereexamined for relative differences in innominate rotationand described

ResultsA summary of descriptive statistics is presented in

Table 2 Although men and women had similar mean frontalplane values the mean values for sagittal plane were quitedifferent with women having approximately twice as muchdeviation from the horizontal as men Mean sagittal planevalues for left and right sides were 734 degrees and 693degrees respectively Table 3 presents information on asubset of the data (Trial 1 examiners 1 and 2) relatingdegree of innominate tilt (1deg_7deg) to percentage of sub-jects A separate analysis using the same subset of databut not shown in Table 3 determined that the left in-nominate was anterior 43 of the time the right innominatewas anterior 52 of the time and 4 (1 person) hadlevel innominates

Intraclass Correlation Coefficients and their corre-sponding lower confidence intervals are presented in Table4 Intratester reliability was high for both frontal (084)and sagittal (098) plane measures 36 Intertester reli-ability was high for sagittal (089) plane measures butmoderate for frontal (065) plane measures36bull

Discussion

Frontal PlaneNo studies were found that utilized an inclinometer

with a direct output in degrees to measure relative iliaccrest heights Studies have been performed using a pelvicleveling device that indicates the presence or absence ofa horizontal position3031 In these studies examiners placedshims with known values ranging from 32-5 mm thickunder the low iliac crest until a level position is reached

Pub

lishe

d by

Man

ey P

ublis

hing

(c)

W S

Man

ey amp

Son

Lim

ited

Table 2 Descriptive statistics All values in degrees Standard Deviation (SD) Standard Error of the Mean (SEM)

1~~li)jJ~(~]fiLtrfff[fID~nJ

Frontal

SagittalLeft

SagittalRight

i~il~Iij

-077t

734

693

Ilft3rlmlill11UVltftIiJmale -075female -08

male -48female -87

male 46female 83

~]J

145

52

sectsectfl)755

3658

(~~~

2 - (-65)t

15 - (-19)t

t Negative values represent inferior position of left arm of caliper in relation to examiner Combined left and right sagittal values

Table 3 Degree of difference in sagittal plane measures computed by averaging values of examiner 1 and 2 duringTrial 1 and comparing right and left innominates

Differenceright vs leftinnominate

Percentageof subjects

400 000

Table 4 ICC and lower confidence limits (CI) for Intra- and Intertester reliability

[~lIlil9J rJijJfu1J~~1~f IJjftmll~lIff IIH~ln~lt~Ii1mJIfnmill [f3 11l~lJ21rftlaj1emiddotJ f

lIn~l~ml~lr11E)413YillY~4 O

Frontal

Combinedright and leftsagittal

084

098

070

095

065

089

047

070

and then count the shims to determine height differencesIf the average distance between femoral heads as sug-gested by the literature is 20 cm37 approximate com-parisons to these studies can be performed using trigo-nometry (Table 5) Each degree of relative tilt read fromthe PALM is equal to approximately 35 mm of iliac crestheight difference Consequently from Table 2 the meandifference found in this study in iliac crest height was27 mm and the SD was 51 mm The mean values werealmost identical to those found by Jonson and Gross31

and Woerman et apo at 22 mm and 32 mm respectivelyJonson and Gross31 found a SD of 26 mm and Woermanet apo found a SD of 39 mm The variability in this studywas higher than previous studies and may be explained

by the methodology which avoided constraints not nor-mally encountered in clinical practice

Sagittal PlaneOther studies25263338 examining the angle formed be-

tween a line connecting the ASIS to PSIS and the hori-zontal have found mean values ranging from 835 de-grees26 to 113 degrees38bull This relatively narrow range wasfound despite various measurement methods and the useof both men and women as subjects These mean valuescorrespond well to those found in this study for womenbut differ markedly from those found in men It is un-clear why this distinction between men and women occured

Intratester and Intertester Reliability of the Palpation Meter(PALM) in Measuring Pelvic Position 133

Pub

lishe

d by

Man

ey P

ublis

hing

(c)

W S

Man

ey amp

Son

Lim

ited

Table 5 Translation of PALM measures in frontal plane to trigonometrically derived height difference in the iliaccrests based on average of approximately 8 inches37 between femoral heads

~-li _I ~ ~IFl~ ~fr~~~(~1 n~ = Iil middoti)ii~ - iHIPii ~ lnlJrIisect~ -_ __-

1deg 349 349 1372deg 698 698 27483deg 1048 1048 41254deg 1398 1398 55045deg 1749 1749 68896deg 2102 2102 82687deg 2455 2455 9645

The standard deviations reported in three of the previ-ously mentioned studies263338 for sagittal plane measurementvaried between 34 degrees and 43 degrees these valuesagree reasonably well with the value in this study of 52degrees The deliberate approach in this study to avoidconstraints on the measurement process beyond what islikely to occur in the clinic (placing adhesive markerson landmarks etc) may be responsible for the greaterdegree of variability

The apparent symmetry in the sagittal plane betweenright and left innominates in Table 2 appears to be acharacteristic of summary statistics rather than a typi-cal finding on an individual Approximately half of thesubjects were anterior on the right the other half wereanterior on the left Importantly the degree of asym-metry was relatively small between innominates with96 of the subjects having less than or equal to 4 de-grees of asymmetry (Table 3)

Although it is possible in clinical practice to ask patientsto disrobe sufficiently to expose the ASIS and PSIS andto mark landmarks with adhesive markers and to havethem brace themselves against a thoracic or femoral supportthe authors felt that constraints were both impracticaland improbable in a busy clinic Consequently the de-cision was made to utilize the PALM in a manner thoughtto replicate probable clinical practice with this new in-strument so that the observed reliability could be au-thentically generalizable The high reliability found inthe sagittal plane for both intra- and intertester mea-sures and in the frontal plane for intratester measures isremarkable given such a lack of controls

The ICC is based on an analysis of variance that par-titions variance into categories for comparison The mod-erate intertester frontal plane ICC value (065) possiblyreflects the lack of variance in the true values of frontalplane position in this asymptomatic population It is probablethat including a number of subjects with genuine leg lengthdiscrepancies would have increased the variance of thetrue values and provided higher ICC values for reliabil-

134 The Journal of Manual amp Manipulative Therapy 1998

ity This can be appreciated by comparing the SD valuesbetween frontal (SD = 145) and sagittal (SD = 52) planemeasures Future studies should address reliability insymptomatic populations

Although the ICC values found in this study gener-ally suggest that measured values are consistent the clinicaluse of the PALM must also be evaluated in terms of theprecision of the instrument The SEM for sagittal planemeasures was approximately 37 degrees Clinically thisvalue indicates that the examiner cannot be certain thatone measurement in the sagittal plane is different (romanother unless that difference is more than two timesthe SEM or 74 degrees A necessary change of 74 de-grees is relatively large when one considers that someauthors suggest that the sacro-iliac joint has a range ofmotion of 1-11 degrees1739 We found that 96 of all rightversus left innominate differences were 4 degrees or lessIf a clinician measures an anterior innominate of 14 degreesprior to an intervention and measures a value of 9 de-grees after an intervention he or she cannot be certainthat a change has occurred However if initial measure-ments determine that the right innominate is tilted 8degrees and the left is tilted 0 degrees and an interven-tion succeeds in creating an 8 degree tilt on the left thenthe clinician can state that a real change has occurred

Similar reasoning can be followed for frontal planemeasurements but in this case the SEM is 755 degreesindicating a necessary difference of 15 degrees beforethe clinician can be certain a difference exists A 15 degreedifference translates to 52 mm height difference betweeniliac crests The amount of LLD necessary for symptomgeneration is controversial but a recent survey of theliterature suggests that 10 mm is r equired40bull The PALMtherefore appears to have sufficient precision to deter-mine through the indirect method of iliac crest heightmeasurement if a significant LLD exists

A few points regarding methodology should be men-tioned Examiners were not blinded to their own mea-surements as it was felt that the instrument required a

Pub

lishe

d by

Man

ey P

ublis

hing

(c)

W S

Man

ey amp

Son

Lim

ited

direct vertical view for accurate reading of the inclinom-eter At worst this may have inflated the intratester reliabilitydespite attempts to direct the examiners to perform eachmeasure as an isolated event The intertester reliabilityvalues were not subject to this source of error and tendedto fall as expected slightly lower than the intratester valuessuggesting little effect from the lack of blinding on theintratester values Another source of error was the po-tential differences between the two PALMs used in thisstudy Again the authors felt the value of this study wasin creating a clinically realistic methodology using aninstrument that is commercially produced and readilyavailable It was decided that any error created by mul-tiple versions of the instrument although potentially usefulto know should not be separated out in this study

ConclusionsIn summary this study examined standing static pelvic

REFERENCES1 Hoppenfeld S Physical Examination of the Spine and Extremi-

ties Norwalk Appleton- Century-Crofts 1976 pp 239-2422 Kendall F McCreary E Muscles Testing and Function 3rd ed

BaltimoreWilliams and WIlkins 1983 pp 2213 Denslow JS Chace JA Mechanical Stresses in the Human Lumbar

Spine and Pelvis J Am Osteopath Assoc 1962 61706-7124 DonTigny RL Measuring PSIS movement Clinical Management

1990 1043-445 DonTigny RL Dysfunction of the Sacroiliac Joint and its

Treatment Journal Orthopedic and Sports Physical Therapy 1979123-25

6 Grieve GP The Sacro-iliac Joint Physiotherapy 1976 62384-4007 Donatelli RAWooden MJ Orthopaedic Physical Therapy New York

Churchill Livingstone 1993 pp 521-5238 Greenman PE Principles of Manual Medicine 20th ed Baltimore

Williams and Wilkins 1989 pp 20-229 Subotnick SI Limb Length Discrepancies of the Lower

Extremity Journal Orthopedic and Sports Physical Therapy 1985311-16

10 Friberg O Clinical Symptoms and Biomechanics of Lumbar Spineand Hip Joint in Leg Length Inequality Spine 1983 8643-651

11 McCaw ST Leg length inequality Implications for Runninginjury Prevention Sports Med 1992 14422-429

12 Rothenberg RJ Rheumatic Disease Aspects of Leg Length InequalitySemin Arthritis Rheum 1988 17196-205

13 Tjernstrom B Olerud S Karlstrom G Direct Leg Lengthening JOrthop Trauma 1993 6543-551

14 DonTigny RL Anterior Dysfunction of the Sacroiliac Joint as aMajor Factor in the Etiology of Idiopathic Low Back PainSyndrome Phys Ther 1990 70250-265

15 Cibulka MT Koldehoff R Leg Length Disparity and its Effect onSacroiliac Joint Dysfunction Clinical Management 1986 610-11

posture in asymptomatic subjects using the PALM andfound that intratester reliability was high for both frontaland sagittal plane measures and that intertester reli-ability was high for sagittal plane measures but moder-ate for frontal plane measures Mean values agreedreasonably well with other studies while variability wasslightly higher potentially due to a methodology thatattempted to replicate the lack of constraints typicallyfound in clinical practice The clinical use of the PALMshould always be performed with consideration givento the limitations inherent in the precision of the in-strument

AcknowledgmentsWe thank Bill Susman PhD for his generous assis-

tance in reviewing initial drafts of this study We alsothank Rudi Hiebert for his invaluable guidance with thestatistical analysis 0

16 Cummings G Scholz Jp Barnes K The Effect ofImposed Leg LengthDifference on Plevic Bone Symmetry Spine 1993 18368-373

17 Pitkin H Pheasant HC A Study of Sacral Mobility Journal of Boneand Joint Surgery 1936 18365-374

18 Chamberlain E The Symphsis Pubis in the Roentgen Examina-tion of the Sacroiliac Joint Radium Therapy and Nuclear Medi-cine 1930 24621-625

19 Potter NA Rothstein JM Intertester Reliability for Selected Clini-cal Tests of the Sacroiliac Joint Phys Ther 1985 111671-1675

20 Mann M Glasheen-Wray M Nyberg R Therapist Agreement forPalpation and Observation of Iliac Crest Heights Phys Ther 198464334-338

21 Clarke GR Unequal Leg Length An Accurate method of Detectionand Some Clinical Results Rheumatology and Physical Medicine1972 11385-390

22 Bailey HW Beckwith CG Short Leg and Spinal Anomalies TheirIncidence and Effects on Spinal Mechanics J Am Osteopath Assoc1937 36319-327

23 Clark GR Unequal Leg Length An Accurate Method of Detectionand Some Clinical Results Rheum Phys Med 1972 11385-390

24 Kerr HE Grant JH MacBain RN Some Observations on theAnatomical Short Leg in a Series of Patients Presenting Them-selves for Treatment of Low-Back Pain J Am Osteopath Assoc 194342437 -440

25 Gilliam J Brunt D MacMillian M Kinard RE Montgomery WJRelationship of the Pelvic Angle to the Sacral Angle Meaurementof Clinical Reliability and Validity Journal Orthopedic and SportsPhysical Therapy 1994 20193-199

26 Gajdosik R Simpson R Smith R DonTigny RL IntratesterReliability of Measuring the Standing Position and Range of MotionPhys Ther 1985 65169-173

27 Day JW Smidt GL Lehman T Effect of Pelvic Tilt on Standing

Intratester and Intertester Reliability of the Palpation Meter(PALM) in Measuring Pelvic Position I 135

Pub

lishe

d by

Man

ey P

ublis

hing

(c)

W S

Man

ey amp

Son

Lim

ited

Table 2 Descriptive statistics All values in degrees Standard Deviation (SD) Standard Error of the Mean (SEM)

1~~li)jJ~(~]fiLtrfff[fID~nJ

Frontal

SagittalLeft

SagittalRight

i~il~Iij

-077t

734

693

Ilft3rlmlill11UVltftIiJmale -075female -08

male -48female -87

male 46female 83

~]J

145

52

sectsectfl)755

3658

(~~~

2 - (-65)t

15 - (-19)t

t Negative values represent inferior position of left arm of caliper in relation to examiner Combined left and right sagittal values

Table 3 Degree of difference in sagittal plane measures computed by averaging values of examiner 1 and 2 duringTrial 1 and comparing right and left innominates

Differenceright vs leftinnominate

Percentageof subjects

400 000

Table 4 ICC and lower confidence limits (CI) for Intra- and Intertester reliability

[~lIlil9J rJijJfu1J~~1~f IJjftmll~lIff IIH~ln~lt~Ii1mJIfnmill [f3 11l~lJ21rftlaj1emiddotJ f

lIn~l~ml~lr11E)413YillY~4 O

Frontal

Combinedright and leftsagittal

084

098

070

095

065

089

047

070

and then count the shims to determine height differencesIf the average distance between femoral heads as sug-gested by the literature is 20 cm37 approximate com-parisons to these studies can be performed using trigo-nometry (Table 5) Each degree of relative tilt read fromthe PALM is equal to approximately 35 mm of iliac crestheight difference Consequently from Table 2 the meandifference found in this study in iliac crest height was27 mm and the SD was 51 mm The mean values werealmost identical to those found by Jonson and Gross31

and Woerman et apo at 22 mm and 32 mm respectivelyJonson and Gross31 found a SD of 26 mm and Woermanet apo found a SD of 39 mm The variability in this studywas higher than previous studies and may be explained

by the methodology which avoided constraints not nor-mally encountered in clinical practice

Sagittal PlaneOther studies25263338 examining the angle formed be-

tween a line connecting the ASIS to PSIS and the hori-zontal have found mean values ranging from 835 de-grees26 to 113 degrees38bull This relatively narrow range wasfound despite various measurement methods and the useof both men and women as subjects These mean valuescorrespond well to those found in this study for womenbut differ markedly from those found in men It is un-clear why this distinction between men and women occured

Intratester and Intertester Reliability of the Palpation Meter(PALM) in Measuring Pelvic Position 133

Pub

lishe

d by

Man

ey P

ublis

hing

(c)

W S

Man

ey amp

Son

Lim

ited

Table 5 Translation of PALM measures in frontal plane to trigonometrically derived height difference in the iliaccrests based on average of approximately 8 inches37 between femoral heads

~-li _I ~ ~IFl~ ~fr~~~(~1 n~ = Iil middoti)ii~ - iHIPii ~ lnlJrIisect~ -_ __-

1deg 349 349 1372deg 698 698 27483deg 1048 1048 41254deg 1398 1398 55045deg 1749 1749 68896deg 2102 2102 82687deg 2455 2455 9645

The standard deviations reported in three of the previ-ously mentioned studies263338 for sagittal plane measurementvaried between 34 degrees and 43 degrees these valuesagree reasonably well with the value in this study of 52degrees The deliberate approach in this study to avoidconstraints on the measurement process beyond what islikely to occur in the clinic (placing adhesive markerson landmarks etc) may be responsible for the greaterdegree of variability

The apparent symmetry in the sagittal plane betweenright and left innominates in Table 2 appears to be acharacteristic of summary statistics rather than a typi-cal finding on an individual Approximately half of thesubjects were anterior on the right the other half wereanterior on the left Importantly the degree of asym-metry was relatively small between innominates with96 of the subjects having less than or equal to 4 de-grees of asymmetry (Table 3)

Although it is possible in clinical practice to ask patientsto disrobe sufficiently to expose the ASIS and PSIS andto mark landmarks with adhesive markers and to havethem brace themselves against a thoracic or femoral supportthe authors felt that constraints were both impracticaland improbable in a busy clinic Consequently the de-cision was made to utilize the PALM in a manner thoughtto replicate probable clinical practice with this new in-strument so that the observed reliability could be au-thentically generalizable The high reliability found inthe sagittal plane for both intra- and intertester mea-sures and in the frontal plane for intratester measures isremarkable given such a lack of controls

The ICC is based on an analysis of variance that par-titions variance into categories for comparison The mod-erate intertester frontal plane ICC value (065) possiblyreflects the lack of variance in the true values of frontalplane position in this asymptomatic population It is probablethat including a number of subjects with genuine leg lengthdiscrepancies would have increased the variance of thetrue values and provided higher ICC values for reliabil-

134 The Journal of Manual amp Manipulative Therapy 1998

ity This can be appreciated by comparing the SD valuesbetween frontal (SD = 145) and sagittal (SD = 52) planemeasures Future studies should address reliability insymptomatic populations

Although the ICC values found in this study gener-ally suggest that measured values are consistent the clinicaluse of the PALM must also be evaluated in terms of theprecision of the instrument The SEM for sagittal planemeasures was approximately 37 degrees Clinically thisvalue indicates that the examiner cannot be certain thatone measurement in the sagittal plane is different (romanother unless that difference is more than two timesthe SEM or 74 degrees A necessary change of 74 de-grees is relatively large when one considers that someauthors suggest that the sacro-iliac joint has a range ofmotion of 1-11 degrees1739 We found that 96 of all rightversus left innominate differences were 4 degrees or lessIf a clinician measures an anterior innominate of 14 degreesprior to an intervention and measures a value of 9 de-grees after an intervention he or she cannot be certainthat a change has occurred However if initial measure-ments determine that the right innominate is tilted 8degrees and the left is tilted 0 degrees and an interven-tion succeeds in creating an 8 degree tilt on the left thenthe clinician can state that a real change has occurred

Similar reasoning can be followed for frontal planemeasurements but in this case the SEM is 755 degreesindicating a necessary difference of 15 degrees beforethe clinician can be certain a difference exists A 15 degreedifference translates to 52 mm height difference betweeniliac crests The amount of LLD necessary for symptomgeneration is controversial but a recent survey of theliterature suggests that 10 mm is r equired40bull The PALMtherefore appears to have sufficient precision to deter-mine through the indirect method of iliac crest heightmeasurement if a significant LLD exists

A few points regarding methodology should be men-tioned Examiners were not blinded to their own mea-surements as it was felt that the instrument required a

Pub

lishe

d by

Man

ey P

ublis

hing

(c)

W S

Man

ey amp

Son

Lim

ited

direct vertical view for accurate reading of the inclinom-eter At worst this may have inflated the intratester reliabilitydespite attempts to direct the examiners to perform eachmeasure as an isolated event The intertester reliabilityvalues were not subject to this source of error and tendedto fall as expected slightly lower than the intratester valuessuggesting little effect from the lack of blinding on theintratester values Another source of error was the po-tential differences between the two PALMs used in thisstudy Again the authors felt the value of this study wasin creating a clinically realistic methodology using aninstrument that is commercially produced and readilyavailable It was decided that any error created by mul-tiple versions of the instrument although potentially usefulto know should not be separated out in this study

ConclusionsIn summary this study examined standing static pelvic

REFERENCES1 Hoppenfeld S Physical Examination of the Spine and Extremi-

ties Norwalk Appleton- Century-Crofts 1976 pp 239-2422 Kendall F McCreary E Muscles Testing and Function 3rd ed

BaltimoreWilliams and WIlkins 1983 pp 2213 Denslow JS Chace JA Mechanical Stresses in the Human Lumbar

Spine and Pelvis J Am Osteopath Assoc 1962 61706-7124 DonTigny RL Measuring PSIS movement Clinical Management

1990 1043-445 DonTigny RL Dysfunction of the Sacroiliac Joint and its

Treatment Journal Orthopedic and Sports Physical Therapy 1979123-25

6 Grieve GP The Sacro-iliac Joint Physiotherapy 1976 62384-4007 Donatelli RAWooden MJ Orthopaedic Physical Therapy New York

Churchill Livingstone 1993 pp 521-5238 Greenman PE Principles of Manual Medicine 20th ed Baltimore

Williams and Wilkins 1989 pp 20-229 Subotnick SI Limb Length Discrepancies of the Lower

Extremity Journal Orthopedic and Sports Physical Therapy 1985311-16

10 Friberg O Clinical Symptoms and Biomechanics of Lumbar Spineand Hip Joint in Leg Length Inequality Spine 1983 8643-651

11 McCaw ST Leg length inequality Implications for Runninginjury Prevention Sports Med 1992 14422-429

12 Rothenberg RJ Rheumatic Disease Aspects of Leg Length InequalitySemin Arthritis Rheum 1988 17196-205

13 Tjernstrom B Olerud S Karlstrom G Direct Leg Lengthening JOrthop Trauma 1993 6543-551

14 DonTigny RL Anterior Dysfunction of the Sacroiliac Joint as aMajor Factor in the Etiology of Idiopathic Low Back PainSyndrome Phys Ther 1990 70250-265

15 Cibulka MT Koldehoff R Leg Length Disparity and its Effect onSacroiliac Joint Dysfunction Clinical Management 1986 610-11

posture in asymptomatic subjects using the PALM andfound that intratester reliability was high for both frontaland sagittal plane measures and that intertester reli-ability was high for sagittal plane measures but moder-ate for frontal plane measures Mean values agreedreasonably well with other studies while variability wasslightly higher potentially due to a methodology thatattempted to replicate the lack of constraints typicallyfound in clinical practice The clinical use of the PALMshould always be performed with consideration givento the limitations inherent in the precision of the in-strument

AcknowledgmentsWe thank Bill Susman PhD for his generous assis-

tance in reviewing initial drafts of this study We alsothank Rudi Hiebert for his invaluable guidance with thestatistical analysis 0

16 Cummings G Scholz Jp Barnes K The Effect ofImposed Leg LengthDifference on Plevic Bone Symmetry Spine 1993 18368-373

17 Pitkin H Pheasant HC A Study of Sacral Mobility Journal of Boneand Joint Surgery 1936 18365-374

18 Chamberlain E The Symphsis Pubis in the Roentgen Examina-tion of the Sacroiliac Joint Radium Therapy and Nuclear Medi-cine 1930 24621-625

19 Potter NA Rothstein JM Intertester Reliability for Selected Clini-cal Tests of the Sacroiliac Joint Phys Ther 1985 111671-1675

20 Mann M Glasheen-Wray M Nyberg R Therapist Agreement forPalpation and Observation of Iliac Crest Heights Phys Ther 198464334-338

21 Clarke GR Unequal Leg Length An Accurate method of Detectionand Some Clinical Results Rheumatology and Physical Medicine1972 11385-390

22 Bailey HW Beckwith CG Short Leg and Spinal Anomalies TheirIncidence and Effects on Spinal Mechanics J Am Osteopath Assoc1937 36319-327

23 Clark GR Unequal Leg Length An Accurate Method of Detectionand Some Clinical Results Rheum Phys Med 1972 11385-390

24 Kerr HE Grant JH MacBain RN Some Observations on theAnatomical Short Leg in a Series of Patients Presenting Them-selves for Treatment of Low-Back Pain J Am Osteopath Assoc 194342437 -440

25 Gilliam J Brunt D MacMillian M Kinard RE Montgomery WJRelationship of the Pelvic Angle to the Sacral Angle Meaurementof Clinical Reliability and Validity Journal Orthopedic and SportsPhysical Therapy 1994 20193-199

26 Gajdosik R Simpson R Smith R DonTigny RL IntratesterReliability of Measuring the Standing Position and Range of MotionPhys Ther 1985 65169-173

27 Day JW Smidt GL Lehman T Effect of Pelvic Tilt on Standing

Intratester and Intertester Reliability of the Palpation Meter(PALM) in Measuring Pelvic Position I 135

Pub

lishe

d by

Man

ey P

ublis

hing

(c)

W S

Man

ey amp

Son

Lim

ited

Table 5 Translation of PALM measures in frontal plane to trigonometrically derived height difference in the iliaccrests based on average of approximately 8 inches37 between femoral heads

~-li _I ~ ~IFl~ ~fr~~~(~1 n~ = Iil middoti)ii~ - iHIPii ~ lnlJrIisect~ -_ __-

1deg 349 349 1372deg 698 698 27483deg 1048 1048 41254deg 1398 1398 55045deg 1749 1749 68896deg 2102 2102 82687deg 2455 2455 9645

The standard deviations reported in three of the previ-ously mentioned studies263338 for sagittal plane measurementvaried between 34 degrees and 43 degrees these valuesagree reasonably well with the value in this study of 52degrees The deliberate approach in this study to avoidconstraints on the measurement process beyond what islikely to occur in the clinic (placing adhesive markerson landmarks etc) may be responsible for the greaterdegree of variability

The apparent symmetry in the sagittal plane betweenright and left innominates in Table 2 appears to be acharacteristic of summary statistics rather than a typi-cal finding on an individual Approximately half of thesubjects were anterior on the right the other half wereanterior on the left Importantly the degree of asym-metry was relatively small between innominates with96 of the subjects having less than or equal to 4 de-grees of asymmetry (Table 3)

Although it is possible in clinical practice to ask patientsto disrobe sufficiently to expose the ASIS and PSIS andto mark landmarks with adhesive markers and to havethem brace themselves against a thoracic or femoral supportthe authors felt that constraints were both impracticaland improbable in a busy clinic Consequently the de-cision was made to utilize the PALM in a manner thoughtto replicate probable clinical practice with this new in-strument so that the observed reliability could be au-thentically generalizable The high reliability found inthe sagittal plane for both intra- and intertester mea-sures and in the frontal plane for intratester measures isremarkable given such a lack of controls

The ICC is based on an analysis of variance that par-titions variance into categories for comparison The mod-erate intertester frontal plane ICC value (065) possiblyreflects the lack of variance in the true values of frontalplane position in this asymptomatic population It is probablethat including a number of subjects with genuine leg lengthdiscrepancies would have increased the variance of thetrue values and provided higher ICC values for reliabil-

134 The Journal of Manual amp Manipulative Therapy 1998

ity This can be appreciated by comparing the SD valuesbetween frontal (SD = 145) and sagittal (SD = 52) planemeasures Future studies should address reliability insymptomatic populations

Although the ICC values found in this study gener-ally suggest that measured values are consistent the clinicaluse of the PALM must also be evaluated in terms of theprecision of the instrument The SEM for sagittal planemeasures was approximately 37 degrees Clinically thisvalue indicates that the examiner cannot be certain thatone measurement in the sagittal plane is different (romanother unless that difference is more than two timesthe SEM or 74 degrees A necessary change of 74 de-grees is relatively large when one considers that someauthors suggest that the sacro-iliac joint has a range ofmotion of 1-11 degrees1739 We found that 96 of all rightversus left innominate differences were 4 degrees or lessIf a clinician measures an anterior innominate of 14 degreesprior to an intervention and measures a value of 9 de-grees after an intervention he or she cannot be certainthat a change has occurred However if initial measure-ments determine that the right innominate is tilted 8degrees and the left is tilted 0 degrees and an interven-tion succeeds in creating an 8 degree tilt on the left thenthe clinician can state that a real change has occurred

Similar reasoning can be followed for frontal planemeasurements but in this case the SEM is 755 degreesindicating a necessary difference of 15 degrees beforethe clinician can be certain a difference exists A 15 degreedifference translates to 52 mm height difference betweeniliac crests The amount of LLD necessary for symptomgeneration is controversial but a recent survey of theliterature suggests that 10 mm is r equired40bull The PALMtherefore appears to have sufficient precision to deter-mine through the indirect method of iliac crest heightmeasurement if a significant LLD exists

A few points regarding methodology should be men-tioned Examiners were not blinded to their own mea-surements as it was felt that the instrument required a

Pub

lishe

d by

Man

ey P

ublis

hing

(c)

W S

Man

ey amp

Son

Lim

ited

direct vertical view for accurate reading of the inclinom-eter At worst this may have inflated the intratester reliabilitydespite attempts to direct the examiners to perform eachmeasure as an isolated event The intertester reliabilityvalues were not subject to this source of error and tendedto fall as expected slightly lower than the intratester valuessuggesting little effect from the lack of blinding on theintratester values Another source of error was the po-tential differences between the two PALMs used in thisstudy Again the authors felt the value of this study wasin creating a clinically realistic methodology using aninstrument that is commercially produced and readilyavailable It was decided that any error created by mul-tiple versions of the instrument although potentially usefulto know should not be separated out in this study

ConclusionsIn summary this study examined standing static pelvic

REFERENCES1 Hoppenfeld S Physical Examination of the Spine and Extremi-

ties Norwalk Appleton- Century-Crofts 1976 pp 239-2422 Kendall F McCreary E Muscles Testing and Function 3rd ed

BaltimoreWilliams and WIlkins 1983 pp 2213 Denslow JS Chace JA Mechanical Stresses in the Human Lumbar

Spine and Pelvis J Am Osteopath Assoc 1962 61706-7124 DonTigny RL Measuring PSIS movement Clinical Management

1990 1043-445 DonTigny RL Dysfunction of the Sacroiliac Joint and its

Treatment Journal Orthopedic and Sports Physical Therapy 1979123-25

6 Grieve GP The Sacro-iliac Joint Physiotherapy 1976 62384-4007 Donatelli RAWooden MJ Orthopaedic Physical Therapy New York

Churchill Livingstone 1993 pp 521-5238 Greenman PE Principles of Manual Medicine 20th ed Baltimore

Williams and Wilkins 1989 pp 20-229 Subotnick SI Limb Length Discrepancies of the Lower

Extremity Journal Orthopedic and Sports Physical Therapy 1985311-16

10 Friberg O Clinical Symptoms and Biomechanics of Lumbar Spineand Hip Joint in Leg Length Inequality Spine 1983 8643-651

11 McCaw ST Leg length inequality Implications for Runninginjury Prevention Sports Med 1992 14422-429

12 Rothenberg RJ Rheumatic Disease Aspects of Leg Length InequalitySemin Arthritis Rheum 1988 17196-205

13 Tjernstrom B Olerud S Karlstrom G Direct Leg Lengthening JOrthop Trauma 1993 6543-551

14 DonTigny RL Anterior Dysfunction of the Sacroiliac Joint as aMajor Factor in the Etiology of Idiopathic Low Back PainSyndrome Phys Ther 1990 70250-265

15 Cibulka MT Koldehoff R Leg Length Disparity and its Effect onSacroiliac Joint Dysfunction Clinical Management 1986 610-11

posture in asymptomatic subjects using the PALM andfound that intratester reliability was high for both frontaland sagittal plane measures and that intertester reli-ability was high for sagittal plane measures but moder-ate for frontal plane measures Mean values agreedreasonably well with other studies while variability wasslightly higher potentially due to a methodology thatattempted to replicate the lack of constraints typicallyfound in clinical practice The clinical use of the PALMshould always be performed with consideration givento the limitations inherent in the precision of the in-strument

AcknowledgmentsWe thank Bill Susman PhD for his generous assis-

tance in reviewing initial drafts of this study We alsothank Rudi Hiebert for his invaluable guidance with thestatistical analysis 0

16 Cummings G Scholz Jp Barnes K The Effect ofImposed Leg LengthDifference on Plevic Bone Symmetry Spine 1993 18368-373

17 Pitkin H Pheasant HC A Study of Sacral Mobility Journal of Boneand Joint Surgery 1936 18365-374

18 Chamberlain E The Symphsis Pubis in the Roentgen Examina-tion of the Sacroiliac Joint Radium Therapy and Nuclear Medi-cine 1930 24621-625

19 Potter NA Rothstein JM Intertester Reliability for Selected Clini-cal Tests of the Sacroiliac Joint Phys Ther 1985 111671-1675

20 Mann M Glasheen-Wray M Nyberg R Therapist Agreement forPalpation and Observation of Iliac Crest Heights Phys Ther 198464334-338

21 Clarke GR Unequal Leg Length An Accurate method of Detectionand Some Clinical Results Rheumatology and Physical Medicine1972 11385-390

22 Bailey HW Beckwith CG Short Leg and Spinal Anomalies TheirIncidence and Effects on Spinal Mechanics J Am Osteopath Assoc1937 36319-327

23 Clark GR Unequal Leg Length An Accurate Method of Detectionand Some Clinical Results Rheum Phys Med 1972 11385-390

24 Kerr HE Grant JH MacBain RN Some Observations on theAnatomical Short Leg in a Series of Patients Presenting Them-selves for Treatment of Low-Back Pain J Am Osteopath Assoc 194342437 -440

25 Gilliam J Brunt D MacMillian M Kinard RE Montgomery WJRelationship of the Pelvic Angle to the Sacral Angle Meaurementof Clinical Reliability and Validity Journal Orthopedic and SportsPhysical Therapy 1994 20193-199

26 Gajdosik R Simpson R Smith R DonTigny RL IntratesterReliability of Measuring the Standing Position and Range of MotionPhys Ther 1985 65169-173

27 Day JW Smidt GL Lehman T Effect of Pelvic Tilt on Standing

Intratester and Intertester Reliability of the Palpation Meter(PALM) in Measuring Pelvic Position I 135

Pub

lishe

d by

Man

ey P

ublis

hing

(c)

W S

Man

ey amp

Son

Lim

ited

direct vertical view for accurate reading of the inclinom-eter At worst this may have inflated the intratester reliabilitydespite attempts to direct the examiners to perform eachmeasure as an isolated event The intertester reliabilityvalues were not subject to this source of error and tendedto fall as expected slightly lower than the intratester valuessuggesting little effect from the lack of blinding on theintratester values Another source of error was the po-tential differences between the two PALMs used in thisstudy Again the authors felt the value of this study wasin creating a clinically realistic methodology using aninstrument that is commercially produced and readilyavailable It was decided that any error created by mul-tiple versions of the instrument although potentially usefulto know should not be separated out in this study

ConclusionsIn summary this study examined standing static pelvic

REFERENCES1 Hoppenfeld S Physical Examination of the Spine and Extremi-

ties Norwalk Appleton- Century-Crofts 1976 pp 239-2422 Kendall F McCreary E Muscles Testing and Function 3rd ed

BaltimoreWilliams and WIlkins 1983 pp 2213 Denslow JS Chace JA Mechanical Stresses in the Human Lumbar

Spine and Pelvis J Am Osteopath Assoc 1962 61706-7124 DonTigny RL Measuring PSIS movement Clinical Management

1990 1043-445 DonTigny RL Dysfunction of the Sacroiliac Joint and its

Treatment Journal Orthopedic and Sports Physical Therapy 1979123-25

6 Grieve GP The Sacro-iliac Joint Physiotherapy 1976 62384-4007 Donatelli RAWooden MJ Orthopaedic Physical Therapy New York

Churchill Livingstone 1993 pp 521-5238 Greenman PE Principles of Manual Medicine 20th ed Baltimore

Williams and Wilkins 1989 pp 20-229 Subotnick SI Limb Length Discrepancies of the Lower

Extremity Journal Orthopedic and Sports Physical Therapy 1985311-16

10 Friberg O Clinical Symptoms and Biomechanics of Lumbar Spineand Hip Joint in Leg Length Inequality Spine 1983 8643-651

11 McCaw ST Leg length inequality Implications for Runninginjury Prevention Sports Med 1992 14422-429

12 Rothenberg RJ Rheumatic Disease Aspects of Leg Length InequalitySemin Arthritis Rheum 1988 17196-205

13 Tjernstrom B Olerud S Karlstrom G Direct Leg Lengthening JOrthop Trauma 1993 6543-551

14 DonTigny RL Anterior Dysfunction of the Sacroiliac Joint as aMajor Factor in the Etiology of Idiopathic Low Back PainSyndrome Phys Ther 1990 70250-265

15 Cibulka MT Koldehoff R Leg Length Disparity and its Effect onSacroiliac Joint Dysfunction Clinical Management 1986 610-11

posture in asymptomatic subjects using the PALM andfound that intratester reliability was high for both frontaland sagittal plane measures and that intertester reli-ability was high for sagittal plane measures but moder-ate for frontal plane measures Mean values agreedreasonably well with other studies while variability wasslightly higher potentially due to a methodology thatattempted to replicate the lack of constraints typicallyfound in clinical practice The clinical use of the PALMshould always be performed with consideration givento the limitations inherent in the precision of the in-strument

AcknowledgmentsWe thank Bill Susman PhD for his generous assis-

tance in reviewing initial drafts of this study We alsothank Rudi Hiebert for his invaluable guidance with thestatistical analysis 0

16 Cummings G Scholz Jp Barnes K The Effect ofImposed Leg LengthDifference on Plevic Bone Symmetry Spine 1993 18368-373

17 Pitkin H Pheasant HC A Study of Sacral Mobility Journal of Boneand Joint Surgery 1936 18365-374

18 Chamberlain E The Symphsis Pubis in the Roentgen Examina-tion of the Sacroiliac Joint Radium Therapy and Nuclear Medi-cine 1930 24621-625

19 Potter NA Rothstein JM Intertester Reliability for Selected Clini-cal Tests of the Sacroiliac Joint Phys Ther 1985 111671-1675

20 Mann M Glasheen-Wray M Nyberg R Therapist Agreement forPalpation and Observation of Iliac Crest Heights Phys Ther 198464334-338

21 Clarke GR Unequal Leg Length An Accurate method of Detectionand Some Clinical Results Rheumatology and Physical Medicine1972 11385-390

22 Bailey HW Beckwith CG Short Leg and Spinal Anomalies TheirIncidence and Effects on Spinal Mechanics J Am Osteopath Assoc1937 36319-327

23 Clark GR Unequal Leg Length An Accurate Method of Detectionand Some Clinical Results Rheum Phys Med 1972 11385-390

24 Kerr HE Grant JH MacBain RN Some Observations on theAnatomical Short Leg in a Series of Patients Presenting Them-selves for Treatment of Low-Back Pain J Am Osteopath Assoc 194342437 -440

25 Gilliam J Brunt D MacMillian M Kinard RE Montgomery WJRelationship of the Pelvic Angle to the Sacral Angle Meaurementof Clinical Reliability and Validity Journal Orthopedic and SportsPhysical Therapy 1994 20193-199

26 Gajdosik R Simpson R Smith R DonTigny RL IntratesterReliability of Measuring the Standing Position and Range of MotionPhys Ther 1985 65169-173

27 Day JW Smidt GL Lehman T Effect of Pelvic Tilt on Standing

Intratester and Intertester Reliability of the Palpation Meter(PALM) in Measuring Pelvic Position I 135

Pub

lishe

d by

Man

ey P

ublis

hing

(c)

W S

Man

ey amp

Son

Lim

ited

Posture Phys Ther 1984 64510-51628 Alviso DJ Dong GT Lentell GL Intertester Reliability for Measur-

ing Pelvic Tilt in Standing Phys Ther 1988 681347-135129 Sanders G Stavrakas PA Technique for Measuring Pelvic Tilt Phys

Ther 1981 6149-5030 WoermanAL Binder-Macleod SALeg Length Discrepancy Assessment

Accuracy and Precision in Five Clinical Methods of EvaluationThe Journal of Orthopedic and Sports Physical Therapy 19845230-239

31 Jonson SR Gross MT Intraexaminer Reliability InterexaminerReliability and Normal Values for Nine Lower Extremity SkeletalMeasures Journal Orthopedic and Sports Physical Therapy 199625253-263

32 Gross MT Burns CB Shane WC et al Reliability and Validity ofRigid Lift and Pelvic Leveling Device Method in AssessingFunctional Leg Length Inequality Journal Orthopedic and SportsPhysical Therapy 1998 27285-294

33 Crowell RD Cummings GS Walker JR Tillman LJ Intratester andIntertester Reliability and Validity of Measures of Innominate BoneInclination Journal Orthopedic and Sports Physical Therapy 19942088-97

136 The Journal of Manual amp Manipulative Therapy 1998

34 Walker ML Rothstein JM Finucane SD Lamb RL RelationshipsBetween Lumbar Lordosis Pelvic Tilt and Abdominal MusclePerformance Phys Ther 1987 67512-516

35 Burdett RG Brown KE Fall MP Reliability and Validity of FourInstruments for Measuring Lumbar Spine and Pelvic PositionsPhys Ther 1986 66677-684

36 Richman J Madrides L Prince B Research Methodology and AppliedStatistics Part 3 Measurement Procedures in Research PhysiotherCanada 1980 32253-257

37 Gofton Jp Trueman GE Studies in Osteoarthritis of the Hip PartII Osteoarthritis of the Hip and Leg Length Disparity CMA Jour-na11971 104791-799

38 Levine D Whittle MW The Effects of Pelvic Movement on LumbarLordosis in the Standing Position Journal Orthopedic and SportsPhysical Therapy 1996 3130-135

39 DonTigny RL Function and Pathomechanics of the Sacroiliac JointA Review Phys Ther 1985 6535-44

40 Fischer P Clinical Measurement and Significance of teg Lengthand Iliac Crest Height Discrepancies The Journal of Manual andManipulative Therapy 1997 557-60