Introduction: According to the requirements of Evidence Based Medicine, every research tool and method should be subjected to reliability evaluation. A variety of inclinometers are used for the assessment of sagittal plane spinal curvatures in clinical practice. The aim of the study was to evaluate the intra- and interobserver repeatability of measurements of the anterior-posterior spinal curvatures taken with Saunders digital inclinometer.
Material and methods: The study involved 30 healthy subjects (5 males, 25 females) aged 20 to 35 years. The anteroposterior curvatures of the spine were measured in all subjects using a Saunders digital inclinometer. In order to evaluate the intraobserver repeatability of measurements, the results obtained by one investigator were compared. To evaluate interobserver repeatability, the results obtained by three investigators were compared. Additionally, the levels of reliability and measurement error were determined.
Results: The measurements conducted by one investigator showed good repeatability for sacral slope, lumbar lordosis, and thoracic kyphosis, including the angle of proximal (Th1-Th6) and distal (Th6-Th12) thoracic kyphosis (p>0.05). For measurements taken by three investigators, interobserver repeatability was achieved for thoracic kyphosis (p>0.05). The reliability of the measurements was good (Cronbach's alpha was 0.9> α ≥0.8), and the measurement error was between 2.8°-3.8°.
Conclusions: The assessment of anteroposterior curvatures of the spine by one investigator provided good repeatability and reliability of measurements. Measurements performed by more than one investigator displayed partial repeatability. The value of the observational error should be taken into account in the interpretation of results of measurements performed with the Saunders inclinometer. The assessment of anteroposterior curvatures of the spine should be preceded by a verification of measurement reliability.