2016 The Reliability and Validity of The Saliba Postural Classification System
2016 The Reliability and Validity of The Saliba Postural Classification System
2016 The Reliability and Validity of The Saliba Postural Classification System
Cristiana Kahl Collins, Vicky Saliba Johnson, Ellen M. Godwin & Evangelos
Pappas
To cite this article: Cristiana Kahl Collins, Vicky Saliba Johnson, Ellen M. Godwin & Evangelos
Pappas (2016): The reliability and validity of the Saliba Postural Classification System, Journal of
Manual & Manipulative Therapy, DOI: 10.1080/10669817.2016.1138599
Article views: 37
of Physical Art, 43449 Elk Run, Steamboat Springs 80487, CO, USA, 3Faculty of Health Sciences, Discipline of
Physiotherapy, University of Sydney, 75 East St, Lidcombe 2141, NSW, Australia
Objectives: To determine the reliability and validity of the Saliba Postural Classification System (SPCS).
Methods: Two physical therapists classified pictures of 100 volunteer participants standing in their habitual
posture for inter and intra-tester reliability. For validity, 54 participants stood on a force plate in a habitual and a
corrected posture, while a vertical force was applied through the shoulders until the clinician felt a postural give.
Data were extracted at the time the give was felt and at a time in the corrected posture that matched the peak
vertical ground reaction force (VGRF) in the habitual posture.
Results: Inter-tester reliability demonstrated 75% agreement with a Kappa = 0.64 (95% CI = 0.524–0.756,
SE = 0.059). Intra-tester reliability demonstrated 87% agreement with a Kappa = 0.8, (95% CI = 0.702–0.898,
Journal of Manual & Manipulative Therapy
SE = 0.05) and 80% agreement with a Kappa = 0.706, (95% CI = 0.594–0818, SE = 0.057). The examiner
applied a significantly higher (p < 0.001) peak vertical force in the corrected posture prior to a postural give when
compared to the habitual posture. Within the corrected posture, the %VGRF was higher when the test was
ongoing vs. when a postural give was felt (p < 0.001). The %VGRF was not different between the two postures
when comparing the peaks (p = 0.214).
Discussion: The SPCS has substantial agreement for inter- and intra-tester reliability and is largely a valid postural
classification system as determined by the larger vertical forces in the corrected postures. Further studies on the
correlation between the SPCS and diagnostic classifications are indicated.
Keywords: Posture, Postural alignment, Postural classification, Postural stability
DOI 10.1080/10669817.2016.1138599 Journal of Manual and Manipulative Therapy 2016 VOL. XX NO. X 1
© Taylor & Francis 2016
Collins et al. The reliability and validity of the Saliba Postural Classification System
(continued)
absence of a standardized and clinically useful postural these tools are rarely available in the clinical setting.8
classification system may contribute to this discrepancy Postural assessment in the clinical setting is most often
and the absence of conclusive studies in this area. performed through a visual analysis of standing and/or
The use of a valid, reliable and clinically useful meas- sitting postures based on typical posture types commonly
urement tool is essential to a clinician’s ability to properly labelled in rehabilitation.3 It is often limited to a static
assess posture and the effect of treatment on posture.8 While analysis of alignment against an imaginary plumb line or
motion analysis systems and 3-D posture analysis sys- the use of unreliable tools such as goniometric measures
tems may provide valid and reliable measures of posture, and measures of distances between bony landmarks.8,20
a
The classification is named based on the vertical position of the thoracic block in relationship to the pelvic block (1st name) and the
angulation of the thoracic block in the sagittal plane (2nd word).
b
The position of the head will vary and is not used in this classification system. The diagram uses a head simply to indicate vertical and a
sideview as a point of reference.
c
The position of the pelvis is not a component of the classification as it may vary although it is depicted in the diagrams as neutral.
Sagittal balance has been described as a posture where a the ribcage in relationship to the pelvic girdle. This is an
vertical line falls through the centre of mass allowing for extremely important component of posture as supported
horizontal gaze and the maintenance of standing with- by the cylinder model of postural control where the rela-
out external support and with minimal muscular effort.21 tionship between posture and respiration is described.22
Ideal alignment is traditionally described as the sagittal In this model, efficient intra-thoracic and intra-abdominal
view alignment of a plumb line at prescribed landmarks: pressures (lid and floor for the cylinder of the trunk) are
posterior to the apex of the coronal suture, through the described as necessary for optimal respiration and postural
external meatus, through the odontoid process of the axis, stability.23
midway through the shoulder, through the bodies of the The Saliba Postural Classification System (SPCS),
lumbar vertebrae, through the sacral promontory, posterior developed by one of the authors, is a classification sys-
to the centre of the hip joint, anterior to the axis of the tem that aims to provide clinicians with an effective and
knee joint, anterior to the lateral malleolus and through the clinically useful tool for assessing posture and its rela-
calcaneaocuboid joint.3 In addition to this ideal alignment, tionship to pain and function. The objective of this initial
Kendall et al.3 outlined three static postures that have pro- study was to examine the reliability and validity of the
vided clinicians with a common language for a descriptive classification system. The SPCS consists of six postural
assessment of posture: the kyphotic-lordotic, the flat-back classifications based on the visual examination of the
and the sway-back postures. These guidelines for postural angulation of the thoracic block (thoracic spine, rib cage
assessment have provided clinicians with a method and a and sternum) in the sagittal plane and its vertical align-
common language for visually assessing and documenting ment in relationship to the pelvic block (pelvis, sacrum
static posture, however, they do not take into account how and coccyx; Table 1). While the SPCS recognizes that the
posture affects the transfer of forces through the body and lower extremity and the head/neck are important aspects
the connection between posture and function, an important of posture, it focuses on the trunk as the central component
component of the clinical examination. It is important to of postural alignment. It defines efficient alignment as one
note that these guidelines also do not provide information where the centre of mass of the thoracic block is verti-
on the position of the thoracic segment or the alignment of cally aligned over the centre of mass of the pelvic block,
Methods
Participants
The study was conducted in two phases: reliability and
validity. Participants in both phases were healthy male and
female volunteers from the local community ranging in
age from 18 to 60. Participants were excluded if they had a
history of vestibular, neurological or musculoskeletal dis-
orders or injury/surgery within 6 months of the study. The
Figure 1 PT 1 passively positioning a participant in the
vertical/vertical SPCS, or the corrected posture, for testing.
study was approved by the primary author’s Institutional
Review Board and written informed consent was obtained
from all participants prior to participation in the study.
Journal of Manual & Manipulative Therapy
Reliability phase
One hundred volunteer participants were instructed to stand
in their habitual posture and were photographed from a
side view wearing shorts and a sports bra (females) or only
shorts (males). Participants were 45 males and 55 females
(mean [SD]; age = 27.69 [6.8] years; height = 170.47
[9.73] cm; weight = 68.57 [14.01] kg).
Validity phase
Fifty-four different participants volunteered for the valid-
ity phase. Seven participants were positioned in the same
posture for both tests in an effort to further blind the testing
PT to the postures being tested. Validity was tested by one’s
ability to transfer higher forces through the spine and into
the lower extremities without any signs of instability or
postural give (operationally defined as a shear, a buckling
or a rotation at or across a segment or segments). Data were
collected as part of a larger study that included the assign-
Figure 2 PT 2 applying a vertical force through both
shoulders to test the participant in the corrected posture. ment of a subjective test grade to the vertical force applied
at the time a postural shear, buckling or rotation was per-
ceived. Subjective data were not analysed in this study. To
through a relaxed abdomen, with both segments in a neu- further improve blinding of the PT assessing the postures,
tral tilt. This alignment is proposed to allow for efficient it was decided a priori that seven participants would be
distribution of vertical forces through the musculoskeletal tested in the same posture twice and that the results would
system thereby supporting pain-free function of the neu- be excluded from data analysis. As parts of the data analysis
romuscular and motor control systems. This classification required comparisons between the habitual and the cor-
system was developed from the observation that the posi- rected postures, it was also decided a priori that participants
tion and alignment of the thoracic and pelvic blocks alters who received the same score would be excluded from data
the mechanics of weight transference through the system, analysis (n = 10). In addition, baseline force plate data from
affecting postural stability and function. Consistent with one participant had incomplete data leaving 36 participants
the cylinder model of postural control,22,23 this efficient for data analysis. Participants were 22 males and 14 females
alignment assures that the lid of the abdominal cavity (the (mean [SD]; age = 28.2 [9.7] years; height = 167.78 [9.60]
thoracic block) is vertically aligned with the bottom of cm; weight = 67.34 [14.53] kg).
Table 2. Inter-tester and intra-tester reliability for the Saliba Postural Classification System.
Photographs were labelled numerically in the order they order of postural alignment for each participant. Privacy
were taken. A black bar was added across the participants’ screens were used to assure that the testing PT was unable
faces to block their identity and minimize chances of the to view the participant’s sagittal view while entering the
PT recognizing a photograph. Classification forms were room. A gradually increasing vertical force was applied
mailed to two PTs for classification of each participant. through both shoulders until the PT felt a shear, a buckle
Both PTs have used this classification system in clinical or rotation indicating that the force was no longer trans-
practice for over 15 years. Based on a visual analysis of the ferring vertically through the system. Force plate values
photograph, each PT classified each participant’s posture in all three planes were extracted at peak vertical ground
according to the SPCS. The completed documents were reaction force (VGRF) for all participants in both postures
mailed back to the primary investigator for analysis. One and at the time point when the VGRF of the corrected
week following the initial classification, the PTs completed posture matched the peak VGRF of the habitual posture.
classification on a second set of the same photographs,
arranged in a different and randomly determined order. Statistical analysis
Upon completion of the second set, the forms were mailed Data analysis was performed using SPSS Version 21
to the primary investigator for analysis. The investigators (SPSS, Inc, Chicago, IL, USA) for both the reliability and
felt that it was important for reliability of the SPCS to be the validity phases.
determined based on an assessment of the same postures,
hence the use of still photographs. Reliability phase
Inter-tester and intra-tester agreement were determined
Validity phase using kappa coefficients (K) with 95% CI and SE.
Each participant stood on a AMTI OR-6 force-plate Additional information includes frequency of observed
(AMTI, Watertown MA, USA) collecting data at 100 Hz agreement (PO), frequency of change agreement (PC),
with a two-pole, low-pass, 1000 Hz Butterworth filter prevalence- and bias-adjusted Kappa (PABAK), maximal
while a PT (CKC) applied a vertically directed force attainable Kappa (Kmax), prevalence index and bias index.
through both shoulders. Participants were tested while Kappa results were interpreted according to Landis and
standing in their habitual posture and in a corrected pos- Koch.24
ture. The SPCS efficient posture (Vertical/Vertical) was
used as the corrected posture. One PT (VSJ) led the par- Validity phase
ticipant to the force plate for testing in their habitual pos- Paired t-tests were used to test if the total force applied
ture or positioned them in a corrected posture through by the PT prior to the posture giving way was higher in
passive re-alignment. Another PT (CKC), blinded to the the corrected posture when compared to the habitual pos-
participant’s postural alignment, applied the vertical force ture; if the peak VGRF measured as a % of the total force
(Figures 1 and 2). In order to assure that the PT applying (%VGRF) was different between the corrected posture
the vertical force was blinded to the posture being tested, when compared to the %VGRF in the habitual posture at
the time the posture gave way and the test was terminated; classification system.8 The SPCS provides clinicians and
and if the %VGRF in the corrected posture was higher researchers with a reliable assessment and classification
when the test was ongoing compared to when the examiner system for postural alignment and its ability to sustain
felt a “postural give”. Pearson’s correlations were used to vertical force requiring no equipment beyond well-trained
test if the PT applied greater force when testing heavier eyes and hands.
participants. The results on the validity of the SPCS were positive
but warrant further research. The authors set out to deter-
Power analysis mine face and construct validity based on the hypothesis
For the validity phase, an a priori sample size calculation that an efficient alignment would allow for greater vertical
was performed using G* Power 3.1.7.25 With 80% power force translation through the system as measured by the
and 0.05 α level, it was determined that 36 participants force plate. This is indeed what the results demonstrated
were needed to detect a moderate correlation. Fifty-four thereby supporting the construct and face validity of the
participants were recruited to account for any possible classification system. The vertical alignment between the
missing data and technical problems. thoracic and the pelvic blocks with the thoracic block in
neutral angulation was shown to allow for higher verti-
Results cal forces to be transferred through the musculoskeletal
Reliability phase system prior to the posture giving way, or the clinician
The reliability phase was aimed at establishing perceiving a shear, a buckle or a rotation in response to
inter-rater and intra-rater reliability for the SPCS. Inter- the increased force travelling through the system. These
tester and intra-tester agreement were determined using results demonstrate that the SPCS is a valid measure of
kappa coefficients (Table 2). Substantial agreement was postural alignment and efficiency. The results also demon-
found for inter-tester reliability (Kappa = 0.64) and intra- strate that when the examiner perceived a “postural give”,
Journal of Manual & Manipulative Therapy
tester reliability (Kappa = 0.80 and 0.71 for testers 1 and postural efficiency dropped as noted by the lower %VGRF
2, respectively). when compared to when the test was ongoing. This is
encouraging as it demonstrates that a trained PT can accu-
Validity phase rately perceive the inefficiency in the postural system, and
The findings of the validity phase were that the PT exerted a postural giving way, as confirmed by the high fidelity
a greater total force in the corrected posture compared to measurements of the force plate.
the habitual posture (p < 0.001); the %VGRF was not dif- The results also demonstrate that the PT exerted a
ferent between the two postures when comparing the peaks greater total force in the corrected position suggesting
(p = 0.214); and within the corrected posture, %VGRF that the alignment of the corrected posture allowed greater
was lower at the end of the test when the examiner felt forces to be translated to the force plate and that an expe-
the “postural give” compared to when the test was ongo- rienced clinician can perceive when the system can tol-
ing (p < 0.001). The examiner exerted greater force when erate a greater compression force. While the clinicians
testing heavier participants (r = 0.72, p < 0.001). involved in this study had several years of experience with
the SPCS, the authors believe that the SPCS requires min-
Discussion imal training. The SPCS has been taught for several years
The ability to speak the same examination and classifica- in entry-level classroom and continuing education settings.
tion language within a profession has been shown to lead Clinicians, novice and advanced, appear to easily grasp an
to improved clinical practice, increased collaboration and understanding of this classification system. Nonetheless,
improved understanding of research results.26 The need for further studies are indicated to assess the role of clinical
reliable measures or classification systems has long been experience on the accuracy of the clinical classification.
established as a requirement for assessing changes and The findings did not confirm the hypothesis that the
documenting patient progress. Despite the accepted impor- corrected posture would have less dispersion of forces
tance of postural assessment and the need for rehabilita- outside the vertical axis when compared to the habitual
tion of efficient posture, the various measures of posture posture. This may be due to a “ceiling effect” as the ver-
currently available and accessible to clinicians have been tical component was 98.9 and 98.5% of the total force for
shown to have poor reliability and validity.8 the habitual and corrected postures, respectively. It is also
This study comprehensively examined the reliability possible that the greater dispersion of forces in the habit-
and several validity aspects of SPCS. The intra-rater and ual posture occurs following the postural break instead of
inter-rater reliability of the SPCS, through visual exam- at the time of the break. This possibility was not tested
ination of one’s standing sagittal posture, was consist- as the force was terminated when the PT felt a postural
ently high. This finding has important implications for give. An alternative explanation for these results may be
clinical research and practice as it represents a significant that the participants need acclimatization time in the cor-
step forward in this field considering the mixed results rected posture to allow the neuromusculoskeletal system
of previous attempts to create a postural assessment and to adjust in the new posture. In addition to examining these
Notes on contributors
with an objective, reliable and valid method for assessing
Cristiana K Collins is a faculty with the Institute of
structural alignment that can help direct the plan of care.
Physical Art, which offers professional development
With the goal of re-establishing one’s efficient alignment
courses on Functional Manual Therapy, including con-
for function, based on the ability of the system to transfer
cepts of the SPCS.
force, the classification system can direct intervention. The
Vicky S Johnson is the developer of the Saliba Postural
clinician aims to re-establish a neutral angulation of the
Classification System. She was not involved in developing
thoracic block and the vertical alignment of the COM of
the study design, data analysis and data interpretation. She
the thoracic and pelvic blocks. For example, if a patient is
is also a co-director of the Institute of Physical Art, which
classified as a vertical/posterior alignment, the emphasis
offers professional development courses on Functional
is the alteration of the angulation of the thoracic block as
Manual Therapy, including concepts of the SPCS.
the COM of both components is already assessed to be
vertical. However, if the patient presents with a posterior/
Conflict of interest statement
posterior alignment, the initial correction must be at the
The authors declare they have no conflict of interest but
vertical alignment of the blocks prior to correcting the
believe the above information should be disclosed.
angulation of the thoracic block.
In summary, this study proposed a postural classifica-
References
tion system based on the inter-relationship of alignment
1 American Academy of Orthopaedic Surgeons. Posture and its
and position of the thoracic and pelvic blocks. The SPCS relationship to orthopaedic disabilities. A report of the posture
is an important addition to the examination of patients in committee. 1947. American Academy of Orthopedic Surgeons.
Evanston, Illinois.
clinical settings. It provides clinicians with a quick and 2 American Physical Therapy Association. Minimum required skills of
portable method of postural assessment based on the visual physical therapists graduates at entry-level. 2013. Available from http://
www.apta.org/uploadedFiles/APTAorg/About_Us/Policies/Education/
observation of the vertical relationship of the thoracic and MinimumRequiredSkillsPTGrads.pdf-search=%22minimum%20
pelvic blocks and the angulation of the thoracic block in required%20skills%20pt%22 (accessed 5 April 2015).
3 Kendall FP, McCreary EK, Provance PG, Rodgers MM, Romani
the sagittal plane. The results of this study provide pre- WA. Muscles: testing and function with posture and pain. 5th ed.
liminary data supporting the SPCS as a reliable and valid Baltimore, MD: Lippincott Williams & Wilkins; 2005.
measure of postural alignment as it relates to one’s ability 4 Sahrmann S. Diagnosis and treatment of movement impairment
syndromes. St Louis, MO: Mosby; 2002.
to accept vertical force as a measure of dynamic stability 5 Sahrmann SA. Does postural assessment contribute to patient care?
within the system. The reliability and validity of the SPCS J Orthop Sports Phys Ther. 2002;32(8):376–9.
6 Dolphens M, Cagnie B, Coorevits P, Vanderstraeten G, Cardon G,
appear to have been established by this investigation in a D’Hooge R, et al. Sagittal standing posture and its association with
population of healthy adults. Future studies on the SPCS spinal pain: a school-based epidemiological study of 1196 flemish
adolescents before age at peak height velocity. Spine (Phila Pa 1976).
should include populations with various pathologies and Sep 1 2012;37(19):1657–66.
pain presentations and those with diminished functional 7 Lewis CL, Sahrmann SA. Effect of posture on hip angles and moments
during gait. Man Ther. Feb 2015;20(1):176–82.
activity and participation levels.
8 Fortin C, Feldman DE, Cheriet F, Labelle H. Clinical methods patients with low back pain during symptomatic tests. Man Ther.
for quantifying body segment posture: a literature review. Disabil Feb 2009;14(1):52–60.
Rehabil. 2011;33(5):367–83. 18 Sorensen CJ, Norton BJ, Callaghan JP, Hwang CT, Van Dillen LR.
9 Nilsen TI, Holtermann A, Mork PJ. Physical exercise, body mass Is lumbar lordosis related to low back pain development during
index, and risk of chronic pain in the low back and neck/shoulders: prolonged standing? Man Ther. Aug 2015;20(4):553–7.
longitudinal data from the nord-trondelag health study. Am J 19 Raine S, Twomey L. Attributes and qualities of human posture and
Epidemiol. 2011;174(3):267–73. their relationship to dysfunction or musculoskeletal pain. Crit Rev
10 Teichtahl AJ, Urquhart DM, Wang Y, Wluka AE, O’Sullivan R, Phys Rehabil Med. 1994;6:409–37.
Jones G, et al. Physical inactivity is associated with narrower lumbar 20 Fortin C, Feldman DE, Cheriet F, Gravel D, Gauthier F, Labelle H.
intervertebral discs, high fat content of paraspinal muscles and low Reliability of a quantitative clinical posture assessment tool among
back pain and disability. Arthritis Res Ther. 2015;17:114–120. persons with idiopathic scoliosis. Physiotherapy. Mar 2012;98(1):
11 Buckley JP, Hedge A, Yates T, Copeland RJ, Loosemore M, Hamer 64–75.
M, et al. The sedentary office: an expert statement on the growing 21 Lamartina C, Berjano P. Classification of sagittal imbalance based
case for change towards better health and productivity. Br J Sports on spinal alignment and compensatory mechanisms. Eur Spine J. Jun
Med. 2015;49(21):1357–62. 2014;23(6):1177–1189.
12 Chau JY, Grunseit A, Midthjell K, Holmen J, Holmen TL, Bauman 22 Massery M. Musculoskeletal and neuromuscular interventions: a
AE, et al. Sedentary behaviour and risk of mortality from all-causes physical approach to cystic fibrosis. J R Soc Med. 2005;98(Suppl
and cardiometabolic diseases in adults: evidence from the hunt3 45):55–66.
population cohort. Br J Sports Med. Jun 2015;49(11):737–42. 23 Massery M, Hagins M, Stafford R, Moerchen V, Hodges PW. Effect
13 Kim Y-G, Kang M-H, Kim J-W, Jang J-H, Oh J-S. Influence of the of airway control by glottal structures on postural stability. J Appl
duration of smartphone usage on flexion angles of the cervical and Physiol (1985). Aug 15 2013;115(4):483–90.
lumbar spine and on reposition error in the cervical spine. Phys Ther 24 Landis JR, Koch GG. The measurement of observer agreement for
Korea. 2013;20(1):10–17. categorical data. Biometrics. Mar 1977;33(1):159–74.
14 Lee S, Kang H, Shin G. Head flexion angle while using a smartphone. 25 Faul F, Erdfelder E, Buchner A, Lang AG. Statistical power analyses
Ergonomics. 2015;58(2):220–6. using g*power 3.1: tests for correlation and regression analyses.
15 Claus AP, Hides JA, Moseley GL, Hodges PW. Different ways Behav Res Methods. Nov 2009;41(4):1149–1160.
to balance the spine: subtle changes in sagittal spinal curves 26 Rosenbaum P, Eliasson AC, Hidecker MJ, Palisano RJ. Classification
affect regional muscle activity. Spine (Phila Pa 1976). Mar 15 in childhood disability: focusing on function in the 21st century. J
2009;34(6):E208–14. Child Neurol. Aug 2014;29(8):1036–1045.
16 O’Sullivan PB, Dankaerts W, Burnett AF, Farrell GT, Jefford E, 27 Reeves NP, Narendra KS, Cholewicki J. Spine stability: the six
Naylor CS, et al. Effect of different upright sitting postures on spinal- blind men and the elephant. Clin Biomech (Bristol, Avon). Mar
Journal of Manual & Manipulative Therapy