Reliability of photogrammetric evaluation of the craniovertebral angle, swayback posture, and knee hyperextension in university students

Abstract

[Purpose] Forward head posture (FHP), swayback posture, and knee hyperextension are musculoskeletal conditions that can lead to pain, dysfunction, and reduced quality of life. Accurate assessment of these postural deviations is critical for prevention and treatment. This study aimed to assess the inter- and intra-examiner reliability of photogrammetry in evaluating the craniovertebral angle (CVA), swayback posture, and knee hyperextension in a sample of university students. [Participants and Methods] Thirty participants were evaluated by four physical therapists (PTs) on the same day, with each PT performing three repetitions of the measurements. One PT repeated the measurements on a subsequent day to determine the retest reliability. [Results] The results showed excellent reliability, with intraclass correlation coefficients (ICC) of 0.98 for both inter- and intra-examiner reliability. [Conclusion] Photogrammetry is a reliable noninvasive method for evaluating these postural deviations, making it suitable for use in clinical and research settings.

INTRODUCTION

Postural deviations such as FHP, swayback posture, and knee hyperextension have risen in prevalence, especially among young adults engaged in prolonged sedentary activities and with poor postural habits. According to Ruivo et al.¹⁾, FHP is increasingly observed in young adults and is associated with neck pain, reduced cervical range of motion, and respiratory dysfunction. Muscle tension, particularly in the neck region, and headaches are frequently associated with forward head posture (FHP) due to increased strain on cervical muscles and altered biomechanics.

Swayback posture and knee hyperextension contribute to musculoskeletal dysfunction in a similar way. Excessive lumbar lordosis and hyperextension in swayback posture can result in chronic lower back pain and structural imbalance. Knee hyperextension leads to joint instability and increases the risk of ligamentous injuries and degenerative changes in the knee²⁾.

FHP is characterized by anterior displacement of the head relative to the spine, which can lead to symptoms such as neck pain, upper back discomfort, and frequent headaches³⁾. Swayback posture involves excessive posterior pelvic tilt and lumbar hyperextension, which increases strain on the lower back, contributing to lumbar discomfort and biomechanical imbalance. Knee hyperextension, defined as excessive backward bending of the knee joint, often results from muscular imbalance or joint instability and can lead to injury.

Accurate assessment of these deviations is vital for diagnosing and treating musculoskeletal problems and evaluating treatment effectiveness. Photogrammetry, which involves the use of photographs to measure body angles, is a recognized noninvasive and reliable method for postural analysis. Studies have validated the reliability of photogrammetry as a method for assessing postural deviations⁴⁾. Gadotti and Biasotto-Gonzalez⁵⁾ demonstrated the sensitivity of photogrammetry in measuring FHP, while Falla et al.⁶⁾ confirmed its reliability in evaluating cervical posture and range of motion. However, the literature lacks comprehensive data on the reliability of combined postural measurements across multiple raters and over time, particularly in relation to the CVA, swayback posture, and knee hyperextension.

PARTICIPANTS AND METHODS

Thirty university students (mean [SD]: age=22.03 [1.89]; weight=77.42 [5.32]; height=1.78 [.15]) from the University of Patras participated in this study. The inclusion criteria for this study were as follows: participants had to be university students aged 18–25, willing to provide informed consent and participate in the full duration of the study, and free from significant musculoskeletal or neurological conditions affecting posture, neck, spine, or knee functionality, such as scoliosis, herniated discs, severe kyphosis or lordosis, chronic neck or back pain, or neurological disorders like multiple sclerosis, cerebral palsy, or peripheral neuropathy. These conditions were assessed through participant self-reports via a medical history questionnaire and a brief physical examination conducted by a trained physiotherapist. The physical examination included posture and any observable signs of neuromusculoskeletal impairments. Any participant presenting with symptoms or a history of these conditions was excluded from the study to ensure a homogenous sample of healthy individuals. Furthermore, individuals unable or unwilling to follow study protocols or provide informed consent were not eligible for participation.

This criteria set ensured a focus on a healthy student population without pre-existing issues related to head projection or neck posture, aligning with the study’s objectives.

Ethical approval was obtained from the University of Patras Research Ethics Committee (14750/2021), and informed consent was collected from all participants prior to the study. The study size was calculated based on the need to detect moderate reliability with an intraclass correlation coefficient (ICC) of at least 0.7, as recommended by de Vet et al⁷⁾. With four PTs serving as raters and each performing three repetitions on 30 participants, the formula determined that a minimum of 25 participants would be necessary. To account for potential dropouts, the final sample size was increased to 30 participants.

The photogrammetric assessments were based on photographs captured using an iPhone 15Pro mounted on a tripod to ensure a consistent height and angle for all measurements. The camera was positioned 1.4 meters away from the participants, with the height adjusted to 1.8 meters to align with the shoulder level. Markers were placed at key anatomical landmarks for each measurement to ensure accuracy. For the craniovertebral angle (CVA), markers were positioned on the tragus of the ear and the C7 vertebra. In the case of swayback posture assessment, markers were placed on the acromion, the greater trochanter, and the lateral malleolus. For evaluating knee hyperextension, markers were positioned on the lateral malleolus, the center of the knee joint, and the greater trochanter. All raters underwent structured training in anatomical landmark identification, marker placement, and photogrammetric techniques to ensure standardisation. Calibration sessions aligned methods among raters by comparing marker placement on the same model and addressing inconsistencies. A detailed written protocol outlining procedures for marker placement, participant positioning, and camera settings was followed, ensuring consistent data collection. Participants were instructed to maintain a relaxed, neutral standing position for each measurement. For knee hyperextension, they stood with their knees naturally aligned⁸⁾.

The photogrammetric analysis was performed using ImageJ software (NIH, Bethesda, MD, USA), a validated tool for postural assessment. The angles for the postural deviations were calculated based on the anatomical landmarks. Four physiotherapists with specialized training in photogrammetry conducted the assessments, each performing three repetitions of the postural measurements (CVA, swayback posture, and knee hyperextension) on the same day. One physical therapist (KM) repeated the measurements on the same participants 24 hours later to assess intra-examiner (test–retest) reliability. The raters were trained to follow a standardized protocol for conducting the photogrammetric assessments and analyzing the images using ImageJ software.

The sequence of assessments was randomized using an online randomization tool (http://www.randomization.com) managed by an independent researcher. The randomization results were concealed in sealed, sequentially numbered opaque envelopes, which were opened only on the first day of the assessment in the presence of the participants. The same randomized sequence was used for both days of assessment.

The reliability of the postural measurements was quantified using the intraclass correlation coefficient (ICC) across three models to evaluate agreement and consistency. ICC1 measured absolute agreement without bias, assessing the degree to which raters or repetitions agree without systematic bias. ICC2 evaluated absolute agreement with bias, accounting for systematic biases that may affect agreement. Lastly, ICC3 assessed consistency in the presence of bias, evaluating the consistency of measurements while allowing for systematic biases. The key differences between the ICC models lie in rater selection, generalizability, and the type of ratings assessed. ICC1 treats raters as random and evaluates single-rater reliability without assuming consistent raters across participants. ICC2 also considers raters as random but focuses on the reliability of averaged measurements across multiple raters, allowing generalization to other raters. ICC3 treats raters as fixed, assessing the reliability of specific raters or tools without generalizing to others. These ICC models follow the methodology outlined by Lijequist et al.⁹⁾, to evaluate reliability in postural assessments. ICC values greater than 0.9 were considered indicative of excellent reliability based on strict literature standards.

Analyses were conducted for three scenarios: a single physical therapist (PT) on a single day, a single PT on different days, and multiple PTs on a single day. All statistical analyses were performed using the IBM SPSS Statistics (version 29, IBM Corp., Armonk, NY, USA).

RESULTS

Inter-examiner reliability, which evaluates agreement across the four trained PTs on a single day, yielded an ICC of 0.98 (95% CI: 0.96–0.99) for all three postural measurements (Table 1). This value indicates excellent reliability, suggesting that photogrammetry produces stable results across different examiners, minimizing variability due to examiner technique or subjective judgment. Each PT demonstrated consistent placement of anatomical markers and precise image capture, which contributed to the high reliability scores.

Table 1. Intraclass correlation coefficients (ICC) for postural measurements.

Variable	Condition	Single rater ICC (95% CI)	Average rater ICC (95% CI)
Craniovertebral angle (CVA)	Single PT, Same day	0.98 (0.97–0.99)	0.99 (0.99–1.00)
	Single PT, Different day	0.95 (0.92–0.97)	0.99 (0.99–1.00)
	Multiple PTs, Same day	0.98 (0.97–0.99)	1.00 (0.99–1.00)
Swayback posture	Single PT, Same day	0.99 (0.99–1.00)	1.00 (1.00–1.00)
	Single PT, Different day	0.99 (0.99–1.00)	1.00 (1.00–1.00)
	Multiple PTs, Same day	0.99 (0.99–1.00)	1.00 (1.00–1.00)
Knee hyperextension	Single PT, Same day	0.99 (0.99–1.00)	1.00 (1.00–1.00)
	Single PT, Different day	0.99 (0.99–1.00)	1.00 (1.00–1.00)
	Multiple PTs, Same day	0.99 (0.99–1.00)	1.00 (1.00–1.00)

PT: physical therapist.

The ICC for FHP measurement among the four PTs was 0.98, indicating nearly perfect agreement in assessing the craniovertebral angle (CVA), which is a critical indicator of FHP severity. The swayback posture measurements also yielded a high ICC of 0.97, demonstrating that examiners consistently captured the relevant angles and markers to reflect the degree of pelvic tilt and lumbar hyperextension. Similarly, the measurement of knee hyperextension exhibited an ICC of 0.98 across examiners, highlighting photogrammetry’s sensitivity to small changes in knee angle. The ICC for intra-examiner reliability across all three postural assessments was also excellent, with a value of 0.98 (95% CI: 0.97–0.99). This consistency indicates that photogrammetric evaluations of CVA, swayback posture, and knee hyperextension remain stable even with time gaps between assessments, providing reliable data for longitudinal studies and treatment monitoring.

DISCUSSION

The findings of this study demonstrate excellent inter- and intra-examiner reliability for photogrammetric assessment of the craniovertebral angle (CVA), swayback posture, and knee hyperextension. The observed ICC values of 0.98 across both the inter-examiner and intra-examiner conditions suggest that photogrammetry is a consistent and highly reliable method for evaluating these postural deviations, regardless of the examiner or time of measurement. These results are consistent with the existing literature on the reliability of photogrammetry in various postural assessments.

Other studies have explored the reliability of photogrammetry for individual postural deviations, particularly FHP. For example, Gadotti and Biasotto-Gonzalez⁵⁾ reported an ICC of 0.93 when assessing FHP using photogrammetry. This consistency across studies underscores the reliability of photogrammetry for measuring the craniovertebral angle, making it a dependable tool for evaluating FHP in both clinical and research settings.

A significant contribution of this study lies in the inclusion of swayback posture and knee hyperextension in the analysis. Previous studies, such as that by Falla et al.⁶⁾, primarily focused on cervical spine posture and range of motion, with limited exploration of lower body postural deviations. By incorporating swayback posture and knee hyperextension, this study offers a more comprehensive evaluation of postural assessments, particularly in the lower body, where postural issues can often go unassessed compared to the cervical and thoracic regions.

The measurement of swayback posture in this study yielded similarly high ICC values, reflecting excellent reliability in the assessments. While fewer studies have explored the reliability of photogrammetry for swayback posture, the results align with the general consensus in the literature that photogrammetry is a valuable tool for measuring spinal deviations. A study by Ruivo et al.¹⁾ on adolescents’ posture found photogrammetry to be effective in evaluating spinal curvatures and posture-related pain, particularly in the thoracic and lumbar regions. Our findings further support the use of photogrammetry for assessing postural deviations in the lumbar spine, providing clinicians with a noninvasive and reliable method for evaluating swayback posture.

Knee hyperextension has not been extensively studied using photogrammetry. However, the results of this study demonstrate that photogrammetry is equally reliable in assessing this lower-limb postural issue, with ICC values consistently above 0.9. These findings are consistent with research by Kelln et al.¹⁰⁾, who used goniometric measurements to assess knee hyperextension and reported similar reliability levels. The addition of photogrammetry as an alternative assessment method offers clinicians a noninvasive, time-efficient tool that can provide consistent and accurate measurements of knee joint posture. Furthermore, the fact that multiple raters achieved high levels of agreement emphasizes the method’s robustness.

The test–retest reliability (intra-examiner reliability) was also found to be excellent, with an ICC of 0.98 across all postural measurements. This aligns with the findings of Falla et al.⁶⁾, who reported strong test–retest reliability for cervical range of motion using photogrammetry. Similarly, Motiallah et al.²⁾ found that repeated assessments of head posture using photogrammetry yielded reliable results over time, confirming the stability of the method. The high level of consistency observed in this study further validates photogrammetry as a reliable tool for longitudinal assessments, where repeated measurements are necessary to monitor changes over time, such as in rehabilitation programs.

This study’s results have several practical implications for both clinicians and researchers. The high reliability observed across different raters and time points suggests that photogrammetry can be confidently used to assess postural deviations in various settings. In clinical practice, reliable postural assessments are essential for diagnosing musculoskeletal conditions and tracking the effectiveness of treatment interventions. Photogrammetry’s noninvasive nature, ease of use, and low cost make it an attractive alternative to more invasive methods, such as radiography or motion capture systems, particularly for routine postural evaluations.

In research settings, the excellent reliability observed in this study provides a solid foundation for future studies investigating the relationship between postural deviations and functional outcomes. For example, longitudinal studies evaluating the effectiveness of specific interventions, such as corrective exercises or physical therapy protocols, would benefit from using photogrammetry as a reliable and noninvasive assessment tool. Furthermore, the ability to evaluate multiple postural deviations simultaneously will allow researchers to investigate the interplay between different postural anomalies and their combined impact on pain, functionality, and overall health.

The ICC values found in this study exceeded the commonly accepted threshold of 0.9 for excellent reliability, as outlined in previous literature^{8, 11)}. The results were consistent across all postural measurements, further confirming the robustness of the method. This consistency between different raters and across different days reinforces the value of photogrammetry as a tool for ensuring reliable postural assessments in both cross-sectional and longitudinal studies.

Despite this study’s strengths, using samples of university students in postural research has important limitations that affect the generalization and ecological validity of results within clinical populations. University students, often characterized by their age, state of health and socio-economic history, may not adequately represent the various demographic data in clinical populations. Such homogeneity limits the applicability of results, such as the postural dynamics observed in this demographic group may not result in older adults or individuals with specific pathologies. In addition, controlled and often artificial contexts in which postural studies are carried out do not summarize the complex environments of the real world that clinical populations face. Consequently, the results derived from student samples can ignore the crucial environmental factors influencing posture in various contexts, thus compromising ecological validity. Thus, dependence on samples of university students in postural research underlines the need for more inclusive methodologies which adopt a broader demographic range to improve the applicability of research results in clinical contexts. Future studies should aim to include a more diverse sample to determine whether the reliability of photogrammetry holds across different age groups and clinical populations.

Additionally, while this study focused on the reliability of three specific postural measurements (CVA, swayback posture, and knee hyperextension), future research should explore the reliability of photogrammetry for other postural deviations, such as thoracic kyphosis, lumbar lordosis, and pelvic tilt. This would provide a more comprehensive understanding of the method’s applicability to a wider range of postural issues.

Future research should explore the validity of this method by comparing photogrammetric measurements to those obtained from more established techniques, such as radiographic imaging or 3D motion analysis. This would help to establish photogrammetry as the gold standard for postural assessments in clinical and research settings.

This study provides robust evidence that photogrammetry is a highly reliable method for assessing CVA, swayback posture, and knee hyperextension. The high inter- and intra-examiner reliability values observed across all postural measurements underscore the method’s consistency and accuracy. Given its noninvasive nature, ease of use, and cost-effectiveness, photogrammetry is well suited for both clinical and research applications, particularly for longitudinal assessments of postural deviations. Future research should be conducted to validate these findings in broader populations and to investigate the method’s validity in comparison to more established postural assessment tools.

Conflict of interest

The author declares no conflict of interest