Accurate recording of spinal posture with simple and accessible measurement devices in clinical practice may lead to spinal loading optimization in occupations related to prolonged sitting and standing postures. Therefore, the purpose of this study was to establish the level of reliability of sagittal lumbosacral posture in quiet standing and the validity of the method in differentiating between male and female subjects, establishing in parallel a normative database. 183 participants (83 males and 100 females), with no current low back or pelvic pain, were assessed using the “iHandy Level” smartphone application. Intrarater reliability (3 same-day sequential measurements) was high for both the lumbar curve (
Sagittal lumbopelvic alignment is an important physical assessment parameter for the orthopaedic [
Postural deviations from the “neutral” range [
For screening purposes of lumbosacral posture in large populations, several more easily applied, low-cost, accurate, and valid assessment methods, such as body contour and inclinometric posture measurements, have been developed. Youdas et al. (1996) in a reliability study of 90 healthy participants, using the flexible curve method (tangential calculation of angles) for the lumbar curve and an inclinometer for the pelvic inclination (sacral slope), reported intrarater
Lately, digital inclinometers in the form of smartphone applications have emerged as alternative measurement methods [
The purpose of our study was to examine the intrarater reliability of the method in a sample larger than previously reported [
One hundred and eighty-three healthy adults (83 males and 100 females) volunteered to participate. All participants were students or employees from various faculties at Akmi Metropolitan College and were verbally invited to participate to the research by team members. The mean (SD) age, height, body mass, and BMI of participants were 26.1 (10.04) years, 172 (9.4) cm, 68.7 (14.5) kg, and 22.8 (3.6) kg/m2, respectively. Exclusion criteria were to have active low back pain and/or trauma and for female participants’ menstruation, in order to ensure the confounding effects of current pain presence in body posture of the participants included in the investigation. All participants included in the study were approached via notice-board and e-mail announcements. They all signed a written informed consent, presenting the exclusion criteria and the aims and purposes of the study prior to their participation. The study protocol was approved by the Ethics Committee of the Athens Metropolitan College. All rights of participants were protected at all times, according to the declaration of Helsinki.
Smartphones with the Android operating system were used, with the “iHandy Level” application installed. The “iHandy Level” application is a tool that has already been validated for the measurement of spinal range of movement [
The pilot study was conducted between the members of the team (10 participants), to accustom themselves with the measurement procedure. Participants were asked to stand in a comfortable position with their arms at their side and their lower limbs parallel to each other. Three sequential measurements of pelvic and low back posture were performed, to establish the reliability of the procedure. Between measurements, the participants were asked to perform 10 steps, to alter their posture between the measurements.
Each team member was assigned a task prior to the commencement of measurements: one member was responsible for verbally informing the participants of the purposes of this research and the measurement procedures involved and for collecting the participants demographic details, two members were involved in the palpation and skin-marking, two were involved in the measurement process, and one member recorded the posture data obtained for the two members involved in the actual measurement procedure to be blinded to the recording of results.
Participants stood with their low back area from T12-top of sacrum exposed for the skin-marking and palpation procedures. Our measurement method took into account that the sacral slope (SS) corresponds to the angle between the upper sacral endplate and the horizontal plane [
The last step of the investigation was to conduct the measurement with the “iHandy Level” application tool. The researchers calibrated (zeroed) the application’s indication on a stable level surface before obtaining the 3 sequential measurements. This calibration procedure was followed for each participant separately. The angle readings from the smartphone, when placed with its upper vertical side at T12-L1 and S1-S2 spinous processes, were recorded (Figure
Sacral slope and lumbar curve measurement technique with the smartphone placed at (a) S1-S2 interspace and (b) T12-L1 interspace.
The methodology followed in our study is the double inclinometer method, followed by Waddell et al. 1992, used to assess spinal movement, by placing an electronic goniometer at T12-L1 and S1-S2 landmarks in upright standing and at end-range spinal movements [
All statistical analyses were performed using the Statistical Package for Social Sciences (SPSS), version 20. Descriptive data including mean measurement angles with standard deviations were calculated for each series of measurements. All data were tested with the Kolmogorov-Smirnov test, and it was verified that they were normally distributed.
Repeated measures ANOVAs were performed between the 3 measurements of sacral slope and lumbar curve separately and the level of significance was set at
Construct validity represents a postulated attribute of participants like gender and its associated characteristics or relationships between attributes that are assumed to be reflected in test performance [
According to the descriptive categorical statistics, 54.6% of the participants were females and 45.4% were males. Most of the participants were students 77.6% (142/183) and a smaller percentage 22.4% (41/183) were administrative personnel.
Data for the 3 testing occasions for both lumbar curves and sacral slopes are analytically presented in Table
Lumbar curve and sacral slope mean (SD) values for the 3 measurements, the mean of 3 measurements, and repeated measures ANOVA results.
Measurement | Lumbar curve° | Sacral slope° | ||
---|---|---|---|---|
Mean (SD) | ANOVA | Mean (SD) | ANOVA | |
1 | 31.29 (10.00) |
|
20.40 (8.55) |
|
2 | 31.76 (10.30) | 20.80 (8.77) | ||
3 | 31.63 (10.18) | 20.70 (8.74) | ||
Mean | 31.56 (10.01) | 20.64 (8.59) |
Comparison of results of this with one previously performed similar study using the smartphone application “iHandy Level” in healthy populations.
Angle data in degrees [mean (SD)]. Lumbar curves and sacral slopes differed significantly between male and female participants
Salamh and Kolber [ |
Current study | ||||
---|---|---|---|---|---|
Gender | Mixed ( |
Male |
Female ( |
Total ( |
Male-female mean difference |
|
|||||
Lumbar curve° | 32-33 | 28.37 (7.19) | 34.21 |
31.56 (10.01) | 5.84 |
Sacral slope° | — | 17.80 (6.82) | 22.99 |
20.64 (8.59) | 5.18 |
Intrarater reliability statistics
Salamh and Kolber [ |
Current study | |||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
Gender | Mixed ( |
Male ( |
Female ( |
Total ( | ||||||||
|
||||||||||||
|
SEM° |
|
|
SEM° |
|
|
SEM° |
|
|
SEM° |
| |
|
||||||||||||
Lumbar curve | 0.81 (0.61–0.91) | 3 | 7 | 0.93 (0.89–0.95) | 2.02 | 5.59 | 0.96 (0.95–0.97) | 2.23 | 6.17 | 0.96 (0.94–0.97) | 2.13 | 5.9 |
Sacral slope | — | — | — | 0.95 (0.93–0.96) | 1.57 | 4.35 | 0.97 (0.96–0.98) | 1.63 | 4.51 | 0.97 (0.96–0.97) | 1.61 | 4.46 |
ICC: Intraclass Correlation Coefficient; SEM: Standard Error of the Measurement; MDC: Minimum Detectable Change.
Repeated measures of ANOVA analysis for both angles were almost at or above
Based on the results shown in Table
Agreement between the 3 testing occasions was presented through a graphical representation of mean differences and 95% LOAs (±2SDs of the differences) of all participants’ values (Figures
Bland and Altman plots for lumbar curve measurements, indicating mean differences and 95% limits of agreement between (a) measurements 1 and 2, (b) measurements 2 and 3, and (c) measurements 1 and 3.
Bland and Altman plots for sacral slope measurements, indicating mean differences and 95% limits of agreement between (a) measurements 1 and 2, (b) measurements 2 and 3, and (c) measurements 1 and 3.
Pearson’s correlation coefficient between the 2 angles was very high (
Scatterplot of the association between lumbar curve and sacral slope.
The method used was able to differentiate between male and female participants’ data for both angle values measured. Between-gender means (SD) and mean differences are reported in Table
Between-gender differences (mean ± 1SD) separately for lumbar curve and sacral slope.
One aim of our research was to examine the reliability of low back and pelvic posture with three sequential measurements on the same day and time of day, and another aim was to examine whether there is a correlation between lumbar curve and sacral slope and whether posture measurements differ between male and female participants. All subjects had no low back or pelvic pain for at least 2 years prior to the measurement, with the majority of the subjects never having a serious lumbopelvic pain incident in their life. Posture measurements were conducted with the use of the “iHandy Level” application for smartphones. The hypotheses of our research were that (a) low back curvature and pelvic tilt sagittal posture measurements could be measured with adequate intrarater reliability, which would be equally high for male and female participants, (b) sacral slope and lumbar curve angle data would be significantly associated, and (c) there would be significant between-gender standing low back and pelvic posture angle data, as measured by the method employed.
The study examined only spinal sagittal alignment, as it was easier to record with a common electronic goniometric smartphone application. Nevertheless, the risk factor of sustained loading positions in standing for low back pain development [
High BMI was not an exclusion criterion in our study; however, it can make palpation of landmarks and sometimes placing of external measuring instruments difficult. Indeed, a very small percentage of our population (6/183 = 3.27%) had BMI over 30 kg/m2. Analytically, these were 3 females (with BMI’s of 30.44, 30.48, and 31.24 kg/m2) and 3 males (with BMI’s of 30.04, 30.61, and 32.85 kg/m2). This very small percentage of subjects with BMI > 30 kg/m2 is unlikely to have affected significantly the results of our overall population.
Our research examined test-retest, intrarater reliability with more than one raters. Each of the raters (2 involved) measured half of the subjects and no subject was measured by both of the raters involved. Our findings indicated that the reliability of “iHandy Level” application is high, based on the ICC, SEM, and MDC results (Table
The correlation between the lumbar curve and sacral slope in the sagittal plane was high (Pearson’s
We have additionally demonstrated the ability of measuring spinopelvic alignment differences in standing between male and female participants. Several studies have confirmed there are between-gender differences in spinal posture [
Therefore, both of the hypotheses of our research have been confirmed, although there are two limitations. The first is based on the limited age range of our participants (mean age 26), because the majority of them were college students, not to be able to generalize the results in younger or older populations. Secondly, future research could enhance the measurements of standing posture with habitual sitting postures, include low back and/or pelvic pain populations, and examine how their data differ from those of their healthy counterparts, in order to examine the effect of back pain on spinal posture reliability measurements. Finally, it has to be stated that the measurement method used does not take into account the shape of the lumbar curve and therefore a uniform curve is assumed for all participants. Differences in the shape of the curve as well as differences in intersegmental movement differences could potentially be important for spinal pathology and its evaluation.
The intrarater reliability of sagittal low back and pelvic alignment in standing on the same day and time of day has been confirmed and the construct validity of the spinal posture measurements was established through intercorrelations of both angles measured and between-gender comparison of relaxed upright lumbar and pelvic posture in standing.
The authors declare that there are no competing interests regarding the publication of this paper.
This work was completed as a research project towards the fulfillment for a “BSc (Hons) Physiotherapy” degree at the Athens Metropolitan College Greece, an affiliated programme of Queen Margaret University, Edinburgh, UK.