Document downloaded from http://www.elsevier.es, day 16/06/2017. This copy is for personal use. Any transmission of this document by any media or format is strictly prohibited.
Rehabilitación (Madr). (2013);47(4):223---228
www.elsevier.es/rh
ORIGINAL
Reliability and validity of angular measures through the software
for postural assessment. Postural Assessment Software
R.M. Ruivo ∗ , P. Pezarat-Correia, A.I. Carita, J.R. Vaz
Faculdade de Motricidade Humana, Universidade de Lisboa, Portugal
Received 3 April 2013; accepted 5 July 2013
Available online 31 October 2013
KEYWORDS
Goniometry;
Photogrammetry;
Posture;
Reliability;
Validity
PALABRAS CLAVE
Fiabilidad;
Fotogrametría;
Goniometría;
Postura;
Validez
∗
Abstract
Introduction: Static posture analysis with the aid of specific software with images is beneficial
for large-scale studies and facilitates the diagnosis, planning and follow-up of any physiotherapy
treatment. Therefore, this study aimed to determine test---retest and inter-rater reliability
and validity of computerized photogrammetry using the Postural Assessment Software (PAS) in
relation to goniometry.
Materials and methods: Universal Goniometers were placed in a wall of the Lab, making it
possible to obtain four angular measures. A digital camera was positioned on a tripod at a
height of 147 cm and at a distance of three meters from the wall. A photo was taken with the
four goniometers. Two more photos were taken, and the goniometer angles were changed within
each photo, obtaining 12 different angular values. Each photograph was analyzed by the three
raters using the Postural Assessment Software. Bland---Altman method was used to document
the agreement between raters and between Goniometry and Postural Assessment Software.
Results: There were no statistical differences between the two evaluation methods: goniometry
and PAS software. The Bland---Altman graph supported the strong consistency between methods,
with an average difference equal to 0.06. The use of PAS software was also shown to be intrarater reliable, with the three raters (A, p = 0.17; B, p = 0.31; C, p = 0.17) with no statistical
significant differences between the first and the second measurement, and to have good interrater reliability with no statistical differences between raters.
Conclusions: Postural Assessment Software is an accurate and reliable method when compared
to the goniometry.
© 2013 Elsevier España, S.L. and SERMEF. All rights reserved.
Fiabilidad y validez de las medidas angulares con un programa para la evaluación
postural. Programa informático de evaluación postural
Resumen
Introducción: El análisis estático fotográfico de la postura con la ayuda de software específico es
práctico para estudios a gran escala y facilita el diagnóstico, la planificación y acompañamiento
de cualquier tratamiento fisioterapéutico. El objetivo fue determinar tanto test-retest e
Corresponding author.
E-mail address: [email protected] (R.M. Ruivo).
0048-7120/$ – see front matter © 2013 Elsevier España, S.L. and SERMEF. All rights reserved.
http://dx.doi.org/10.1016/j.rh.2013.07.002
Document downloaded from http://www.elsevier.es, day 16/06/2017. This copy is for personal use. Any transmission of this document by any media or format is strictly prohibited.
224
R.M. Ruivo et al.
inter-calificadores fiabilidad y validez de la fotogrametría computerizado con el software de
evaluación postural (SAP) en relación con goniometría
Materiales y métodos: Los goniómetros universales se pusieron en una pared del laboratorio,
permitiendo la obtención de 4 medidas angulares. Una cámara digital se coloca en un trípode a
una altura de 1,47 metros y a una distancia de 3 metros de la pared. Una fotografía fue tomada
con los cuatro goniómetros y dos fotos más fueron tomadas cambiando los valores angulares de
los goniometros entre cada foto, totalizando doce valores angulares diferentes. Cada fotografía
fue analizada por los tres evaluadores utilizando el Software de Evaluación postural. Método de
Bland---Altman se utilizó para documentar un acuerdo entre los evaluadores y entre Goniometría
y Software de Evaluación postural.
Resultados: No hubo diferencias estadísticamente significativas entre los dos métodos de
evaluación: goniometría y el software de SAP. La gráfica de Bland-Altman ratificó la
fuerte consistencia entre los métodos, con una diferencia media igual a 0,06. El uso de PAS
software también mostró ser fiable intra-calificadores (los tres evaluadores no presentaron
diferencias significativas entre la primera y la segunda medición) y tener una buena fiabilidad
entre evaluadores, sin presentar diferencias estadísticamente significativas entre ellos.
Conclusiones: Software de evaluación postural es un método preciso y fiable en comparación
con la goniometría.
© 2013 Elsevier España, S.L. y SERMEF. Todos los derechos reservados.
Introduction
Posture can be described as the positioning of all body
segments at a given point in time1 and it is an important
health indicator.2 Postural deviations and abnormalities are
potential etiological factors in the pathogenesis of regional
or widespread musculoskeletal disorders,3,4 with several
clinical pain syndromes such as neck pain5,6 or chronic
tension-type headache.7 In this context, postural realignment is a goal often sought by physicians, dentists and
physiotherapists and regular postural assessment is needed.
Most of the times the postural assessment is qualitative
and subjective, depending on past experiences and personal
interpretations, with a low intra- and inter-rater reliability, and abnormalities are identified only through visual
inspection.8 More strict procedures, with a quantitative postural assessment, are crucial and may help the physician to
monitor treatment outcomes.9 To perform these quantitative assessments there is an increasing number of tools,
methods and software, such as photogrammetry, with Postural Assessment Software (PAS/SAPO) and goniometry.
The goniometry is a very widespread method in physiotherapeutic clinic to measure joint angles10 and it is used to
assess range of motion.10,11
In addition to this method, nowadays, recent technological advances have paved the way for a development
of other highly reliable and applicable methods, such as
computer-assisted systems for the analysis of posture photographs --- photogrammetry.2,12 This method of obtaining
reliable information technology about physical objects and
the environment, by means of recording processes, measurement and interpretation of photographic images has
become widely used in the quantitative assessment of postural alignment, with the possibility of quantifying linear and
angular measurements8 and with the advantage of allowing
the record of subtle postural changes and the interrelation between different parts of the human body, which
are difficult to measure and register by other means.8,13
Some specific software has been developed to assist
posture assessment from digitalized pictures such as
PAS/SAPO (Postural Assessment Software).9
This software has been developed to assist posture
assessment from digitalized pictures and allows the measurement of distances and angles, it is easy to use and it is
accompanied by scientific tutorials. It includes features of
image calibration, zoom change, free point marking and corporal angles and distances measurements.9 To confidently
use all these advantages and measurement possibilities the
instrument used has to be reliable, accurate and valid.
To confirm these premises, we used the comparison with
goniometry, the most common method in physical therapy
practice with a good to excellent reliability.14 Therefore, the
present study aims to analyse the intra-rater and inter-rater
reliability and the validity of computerized photogrammetry
using the PAS in relation to goniometry. The hypothesis
tested in this study was that a photography-based method
with the aid of Postural Assessment Software would be more
accurate and reliable compared with conventional goniometry.
Materials and methods
This study was conducted in the Laboratory of Motor Behavior of the Interdisciplinary Centre for the Study of Human
Performance (CIPER), of the Faculty of Human Kinetics,
Technical University of Lisbon.
Four Universal Goniometers (Enraf Nonius) were placed
in a wall of the Lab, between two plumb lines. To mark
the points, three metal pieces were used, with 89.5 cm distances in the vertical axis and 82.5 cm in the horizontal
axis. One Canon Power Shot A4000 IS was placed on a tripod
Manfrotto, model 055 CLB (height of 147 cm) and were positioned 3 meters away from the wall. Three styrofoam balls
with 20 mm diameter were placed in Universal Goniometers
(Enraf Nonius), in the fulcrum, in the centre of the movable
ruler and in the centre of the fixed ruler. After this, an expert
who regularly uses a goniometer placed four goniometers
with the styrofoam balls in the panel, with different known
Document downloaded from http://www.elsevier.es, day 16/06/2017. This copy is for personal use. Any transmission of this document by any media or format is strictly prohibited.
Reliability and validity of angular measures through the software for postural assessment
Procedures
Three Exercise and Health graduates (three men, aged 32,
30 and 26 years old, identified as rater A, B and C) who were
not regular users of the PAS/SAPO were invited to participate
as raters; all of them had already used the software before
but they were not regular users. Raters were trained on how
to use the software. Each rater had 30 minutes to practice
and to ask questions during this training period. After training, photographs were taken and the angle values were
calculated in the PAS/SAPO, by the three evaluators, who did
not know the real angle values. Photographs were calibrated
according to distance and plumb line and the zoom was standardized at 200%. The raters worked on desktop computers
with the same dimensions. For reliability analyses, all pictures were given to the researchers randomly and no time
limit was established. After one week, to assess intra-rater
reliability, all procedures were repeated, such as styrofoam
balls and goniometers placement and photography shooting,
and tests were compared (test---retest). To assess inter-rater
reliability and photogrammetry validity the mean values of
the 24 angular measurements (12 in observation 1 and 12 in
observation 2) of each evaluator were compared with real
goniometer angles.
Statistical analysis
Data analysis was performed using SPSS Version 20 statistical software. The Shapiro---Wilk and the Mauchly’s tests
were used to assess normality and sphericity. In order to
analyse differences between the raters in the first and second PAS measurements, a paired-samples t-test was applied.
The Bland---Altman plot of difference against the mean was
used to compare the limits of agreement and verify the
existence of systematic bias in the reliability and validity
evaluations of PAS [30]. The 95% limits of agreement (LOA)
were calculated using the following formula: 95% limits of
agreement = mean difference ± 2 SD. We have also applied
the repeated measures ANOVA to test differences between
the PAS angular measurements of the three examiners. All
statistical tests were conducted at the ␣ = 0.05 level of significance.
Results
The study included 12 angular goniometric measures which
were determined by three raters.
0,2
Differences between examiners (1_2) with PAS
angular values. Then, a photograph was taken with the four
goniometers. These procedures were repeated two more
times. Between each photograph, goniometer angles were
changed. Therefore, a total of 12 angles were registered.
The goniometers were standardized and disposed in the wall
as follows: top left corner --- goniometer 1, top right corner
--- goniometer 2, low right corner --- goniometer 3, and low
right corner --- goniometer 4. In the first scenario the angles
were 90◦ , 120◦ , 40◦ and 150◦ . In the second, the angles were
110◦ , 33◦ , 170◦ and 80◦ , and finally in the third scenario, the
angles were 50◦ , 130◦ , 90◦ and 110◦ .
225
0,1
0,0000
–0,1
–0,2
–0,3
–0,4
A
Figure 1
B
C
Intra-examiner differences.
Part I: intra-rater reliability of the PAS angular
measurements
The intra-observer reliability of the photography-based PAS
method was determined from the test---retest comparisons
that are displayed in Table 1.
According to the results in Table 1, using the PAS, the
three examiners did not present statistical significant differences between the first and the second measurement,
one week later. We registered [t = −1.46; p = 0.17], for the
A and C rater, [t = −1.05; p = 0.32] and [t = −1.46; p = 0.17]
for the B rater.
Examiner B presented the lowest mean variation between
the measurements (−0.113) followed by examiner A
(−0.118).
Fig. 1 shows that the confidence limits for all raters
include the zero value, confirming that there are no significant intra-examiners differences. The differences between
the first and second angular measurement were negative
for the three examiners (−0.12, −0.11 and −0.14 for the A,
B and C rater, respectively).
If we examine Fig. 2 (Bland---Altman plots), we can
observe that there were no systematic biases in the
agreement of repeated measurements for the examiners A, B and C. All the measurements, except one from
examiner B, were in the acceptable range of variation
(±2SD).
Part II: inter-rater reliability of the PAS angular
measurements
The inter-examiner reliability of the photography-based
PAS method was determined with the repeated measures
ANOVA to test differences between the 24 PAS angular measurements of the three examiners as displayed in
Table 2.
As it can be observed, the results of repeated measures
ANOVA showed no statistical differences between the examiners (f = 3.38; p = 0.42).
Document downloaded from http://www.elsevier.es, day 16/06/2017. This copy is for personal use. Any transmission of this document by any media or format is strictly prohibited.
226
R.M. Ruivo et al.
Table 1
Mean values, SD and paired t-test results for the intra-examiner analysis (raters A, B and C).
Differences
Mean
SD
95% CI
T value
p
A1---A2
B1---B2
C1---C2
−0.118
−0.113
−0.135
0.282
0.376
0.321
(−0.297, 0.605)
(−0.352, 0.123)
(−0.339, 0.069)
−1.456
−1.045
−1.455
0.173
0.318
0.174
Significant difference between the first and second angular measurement (p < 0.05).
B
1,00
,50
0,444
,00
–0,118
–,50
–0,681
C
1,50
Differences in angular measures
obtained in PAS (C1_C2)
1,50
Differences in angular measures
obtained in PAS (B1_B2)
Differences in angular measures
obtained in PAS (A1_A2)
A
1,00
0,638
,50
,00
–0,113
–,50
–0,865
–1,00
–1,00
–1,50
–1,50
50,000
100,000
1,00
–0,135
,50
,00
0,508
–,50
–0,778
–1,00
–1,50
150,000
50,000
Mean angular measures obtained in PAS (A1_A2)
1,50
100,000
150,000
50,000
Mean angular measures obtained in PAS (B1_B2)
100,000
150,000
Mean angular measures obtained in PAS (C1_C2)
Figure 2 Plot of difference against mean angular values obtained with PAS in the first and second evaluation and registered by
the (A) A rater, (B) B rater and (C) C rater (in◦ ), with mean difference and 95% limits of agreement indicated (dashed lines).
Part III: PAS validity in comparison with goniometry
Discussion
,500
0,254
0.06
,000
–0,374
–,500
–1,500
50,0000
100,0000
150,0000
Mean angular measures goniometry and PAS
Figure 3 Bland---Altman plots are shown for the differences
in angular measurement between goniometry and PAS for the
expert examiners. The dashed line represents the 95% limits of
agreement.
conventional goniometry has shown some limitations,17 such
as the fact that due to the anatomic complexity some joints
are more difficult to assess,16 its accuracy largely depends on
the skills and experience of the examiner18 and there is the
Mean values, SD and ANOVA results of the twenty-four angular measurements of the three examiners (A, B and C).
A
PAS
1,000
–1,000
The quantitative posture assessment and the objective measurement of range of motion and joint angles are crucial
for the diagnosis, planning and follow-up of the progress
and results of a physiotherapeutic treatment15 and for the
creation of a postural database for normative comparison.
The present study confirmed the hypothesis that a
photography-based method with the aid of Postural Assessment Software for angular measurements is reliable and
valid. Conventional goniometry is not only a simple and lowcost method, it is also a manual evaluation technique that
has a high reliability to measure the joint angles of upper and
lower limbs and to measure range of motion of the different
joints such as elbow, shoulder and knee.14,16 However, since
Table 2
Differences goniometry and PAS
1,500
To compare the differences between the goniometry and the
PAS method, the Bland---Altman plot of difference against
the mean was used, as shown in Fig. 3.
As seen in Fig. 3, there are no systematic differences
between the two methods used, indicating that PAS has a
strong trend towards producing consistent results. The average change between procedures is 0.06.
In other words, comparing with the goniometry the PAS
method was virtually free of systematic error.
B
C
ANOVA
Mean
SD
Mean
SD
Mean
SD
f
P
97.858
42.70
98.970
42.70
97.850
42.74
3.370
0.415
Significant difference between the 24 angular measurements of the three examiners (p < 0.05).
Document downloaded from http://www.elsevier.es, day 16/06/2017. This copy is for personal use. Any transmission of this document by any media or format is strictly prohibited.
Reliability and validity of angular measures through the software for postural assessment
need to use both hands, making stabilization of the extremity more difficult.11 Thus, PAS should/could be a reliable and
accurate alternative to overcome goniometry limitations.
These findings have a direct effect in research investigations, in posture assessment and in clinical examination,
as there is the possibility to measure joint angles with a
precise and quantitative technique and without the need
for patients to submit for a direct goniometric evaluation.
Another advantage of this method is the easier and less timeconsuming possibility of archiving the captured images and
files.
Nowadays, with the advances in technology, more precisely Internet-based communication and e-email, as well as
nearly universal access to digital cameras (including video),
the potential to use the photogrammetry and specific software for angles and distances analyses is real, useful and it
is becoming more widespread.
In what concerns inter-rater reliability using ANOVA
results, in this study, it was not possible to observe statistical differences between the examiners (p < 0.05), and all the
raters tended to show similar results. The PAS also showed
to have good intra-rater reliability, with all angle measurements, using the Bland---Altman plots, in the acceptable
range of variation (±2SD), except for one of the examiner
B.
These results are in accordance with Ferreira (2010),12
that also found PAS to be a reliable tool for postural analysis because inter-rater and intra-rater agreement were very
good or excellent at 75% (22 variables) and 64.8% (20 variables), respectively. In this study we studied 29 variables,
consisting of angles or distances between different human
anatomic points.
These different aspects of reliability, such as intra-rater
agreement, inter-rater agreement and automatic versus
manual marker digitization have been addressed by other
studies with humans at different ages. Photographic reliability for standing posture assessment has so far been
performed in 7 to 10 year-old children,19,20 adolescents,21
university students22 and adults.16
One of the most recent studies on this topic examined
the reliability of a photographic method for assessment of
standing posture among elementary school children (10---13
years old) and proved this method to have a high intra-rater
reliability.20 Another study with children aged 5---12 years
old suggested that children’s standing posture (quantified
by five whole body or segmental angles) did not change
significantly on repeated testing proven to be reliable.2
In this study, authors assessed the intra-rater reliability
of head flexion, neck flexion and trunk angle in 38 children aged 5---12 and all postural angles were found to
have inter-trial ICCs of 0.93---0.99 suggesting that children have the ability to repeat the required positions. In
another study,21 four examiners measured, in a standing
position, eight postural angles from 22 healthy adolescent
subjects aged 13---17. Each examiner placed the markers
independently and it was ensured that there were no visible marks left on the skin from the removed marks. A
single experienced examiner, who did not participate in the
data collection, performed the digitalization process. As
a result, fair to good ICCs for consistency were observed
for all measures in standing except for pelvic tilt and sway
angle.
227
Besides proving to be reliable, the PAS also proved to
be valid to measure values in angular body segments in
this study and other studies. As seen in Fig. 3, there are
no systematic differences between the two methods and
the average difference between these two methods was
0.06. These results are in accordance with16 that also tested
the validity and the inter-rater and intra-rater reliability of
angular measurements using PAS/SAPO V0.68 in 15 different angular measures, obtained by goniometers displayed in
a panel. For validity, using the Bland---Altman plot of difference against the mean value, an average difference between
the PAS and the goniometry of 0.004 was reported, whereas
for the intra-rater and inter-rater reliability no statistical
differences were found and the ICC was of 0.99 for all analyses, showing that for the methodological situation applied,
the PAS was reliable and valid to measure corporal angles.
Nevertheless, authors did not describe the errors associated
with the measurements taken. Also Nery (2009)23 used the
PAS and studied the inter-rater and inter-rater reliability and
validity of the software to detect postural misalignments
in school children. For the intra-rater reliability, the photographs of 20 school children were digitized twice by only
one examiner, with a one-month interval and for the interrater analysis, three physiotherapists digitalized the school
photos. The author considered the PAS to be a reliable tool
since the majority of the measurements showed high rates
of correlation.
In this study, instead of the ICC, the analysis of the systematic error was used with the relative limits of agreement
as proposed by Bland and Altman (1995),24 because it is considered more significant in terms of statistics. Nevertheless,
we calculated the ICC for intra- and inter-rater reliability,
but, due to the nature of the data, the values were 1 or very
similar to 1.
Regarding the limitations of the study, although the PAS
and photogrammetry was found to be a valid and reliable
method when compared to the goniometry, the comparison
with other posture analysis systems could be useful. Another
limitation relates to the scarce number of measurements.
Conclusion
When compared to the goniometry, Postural Assessment
Software is an accurate and reliable method. This can
really be useful in a practical and clinical position, as it
offers a great opportunity to quantitatively assess posture,
facilitating the diagnosis, planning and follow-up of any
physiotherapeutic treatment and allowing for the creation
of a postural database for normative comparisons.
Ethical disclosures
Protection of human and animal subjects. The authors
declare that no experiments were performed on humans or
animals for this investigation.
Confidentiality of data. The authors declare that no patient
data appear in this article.
Right to privacy and informed consent. The authors
declare that no patient data appear in this article.
Document downloaded from http://www.elsevier.es, day 16/06/2017. This copy is for personal use. Any transmission of this document by any media or format is strictly prohibited.
228
Funding
This study has been supported by the Foundation for Science
and Technology Portugal (grant no. SFRH/BD/77633/2011).
Conflict of interest
The authors declare no conflict of interest.
References
1. Gangnet N, Pomero V, Dumas R, Skalli W, Vital J-M. Variability of the spine and pelvis location with respect to the gravity
line: a three-dimensional stereoradiographic study using a force
platform. Surg Radiol Anat: SRA. 2003;25:424---33.
2. McEvoy MP, Grimmer K. Reliability of upright posture measurements in primary school children. BMC Musculoskelet Disord.
2005;6:35.
3. McClure PW, Bialker J, Neff N, Williams G, Karduna A. Shoulder
function and 3-dimensional kinematics in people with shoulder impingement syndrome before and after a 6-week exercise
program. Phys Ther. 2004;84:832---48.
4. Cook M, Ludewig P. Alterations in shoulder kinematics and associated muscle activity in people with symptoms of shoulder
impingement. Phys Ther. 2000;80:276---91.
5. Silva AG, Punt TD, Sharples P, Vilas-Boas JP, Johnson MI. Head
posture and neck pain of chronic nontraumatic origin: a comparison between patients and pain-free persons. Arch Phys Med
Rehabil. 2009;90:669---74.
6. Yip CHT, Chiu TTW, Poon ATK. The relationship between head
posture and severity and disability of patients with neck pain.
Man Ther. 2008;13:148---54.
7. Fernández-de-Las-Peñas C, Cuadrado ML, Pareja JA. Myofascial trigger points, neck mobility, and forward head posture in
episodic tension-type headache. Headache. 2009;47:662---72.
8. Watson W, Donncha CA. A reliable technique for the assessment
of posture: assessment criteria for aspects of posture. J Sports
Med Phys Fitness. 2000;40:260---70.
9. Ferreira EA, Duarte M, Maldonado EP, Bersanetti AA, Marques AP.
Quantitative assessment of postural alignment in young adults
based on photographs of anterior, posterior, and lateral views.
J Manipulative Physiol Ther. 2010;34:371---80.
10. Sabari JS, Maltzev I, Lubarsky D, Liszkay E, Homel P. Goniometric assessment of shoulder range of motion: comparison of
testing in supine and sitting positions. Arch Phys Med Rehabil.
1998;79:647---51.
R.M. Ruivo et al.
11. Kolber MJ, Mdt C. The reliability and concurrent validity
of shoulder mobility measurements using a digital inclinometer and goniometer. Int J Sports Phys Ther. 2012;7:
306---13.
12. Ferreira EAG, Duarte M, Maldonado EP, Burke TN, Marques AP.
Postural Assessment Software (PAS/SAPO): validation and reliability. Clinics (São Paulo, Brazil). 2010;65:675---81.
13. Cowan N, Jones BH, Frykman PN, Polly Jr DW, Harman EA, Rosenstein RM, et al. Lower limb morphology and risk of overuse
injury among male infantry trainees. Med Sci Sports Exerc.
1996;28:945---52.
14. Blonna D, Zarkadas PC, Fitzsimmons JS, O’Driscoll SW.
Validation of a photography-based goniometry method for
measuring joint range of motion. J Shoulder Elbow Surg.
2012;21:29---35.
15. Sacco I. Confiabilidade da fotogrametria em relação a goniometria para avaliação postural de membros inferiores. Rev Bras de
Fisio. 2007;11:411---7.
16. Braz R, Pedroso F, Goes P, Carvalho G. Confiabilidade e validade de medidas angulares por meio de software para avaliação
postural. Fisioter Mov. 2008;21:117---26.
17. Sidaway B, Euloth T, Caron H, Piskura M, Clancy J, Aide A. Comparing the reliability of a trigonometric technique to goniometry
and inclinometry in measuring ankle dorsiflexion. Gait Posture.
2012;36:335---9.
18. Georgeu GA, Mayfield S, Logan AM. Lateral digital photography
with computer-aided goniometry versus standard goniometry for recording finger joint angles. J Hand Surg. 2002;27:
184---6.
19. Pausić J, Pedisić Z, Dizdar D. Reliability of a photographic
method for assessing standing posture of elementary school
students. J Manipulative Physiol Ther. 2010;33:425---31.
20. Penha P, Amado S, Casarotto R, Amino C, Penteado P. Postural
assessment of girls between 7 and 10 years of age. Clinics.
2005;60:9---16.
21. Perry M, Smith A, Straker L, Coleman J, O’Sullivan P. Reliability
of sagittal photographic spinal posture assessment in adolescents. Adv Physiother. 2008;10:66---75.
22. Normand MC, Descarreaux M, Harison DD, Harison DE, Perron DL,
Ferrantelli JR, et al. Three dimensional evaluation of posture
in standing with the PosturePrint: an intra- and inter-examiner
reliability study. Chiropr Osteopat. 2007;15:15.
23. Nery P. Análise da confiabilidade intra e interexaminador do
software de avaliação postural - SAPO em escolares do Ribeirão
Preto; 2009.
24. Bland JM, Altman DG. Comparing methods of measurement:
why plotting difference against standard method is misleading.
Lancet. 1995;346:1085---7.