Answer to the Letter to the Editor of L. P. Ardigò et al. concerning 'The validity and reliability of 'Spinal Mouse' assessment of spinal curvatures in the frontal plane in pediatric adolescent idiopathic thoracolumbar curves' by Livanelioglu A, Kaya F, Nabiyev V, Demirkiran G, Fırat T (2015) Eur Spine J Apr 22 [Epub ahead of print]

Autor: Ayşe Livanelioğlu
Rok vydání: 2015
Předmět:
Zdroj: European spine journal : official publication of the European Spine Society, the European Spinal Deformity Society, and the European Section of the Cervical Spine Research Society. 25(2)
ISSN: 1432-0932
Popis: 1. ‘‘Sample size design’’ was not used in the starting of our study. Later, necessity for reaching 52 cases was calculated with G-Power 3.17 program based on the presuppositions which include 95 % confidence interval, 80 % power and lowest correlation value between the measurements of two evaluators as 0.50 and impact factor as 0.20. Owing to achieving better correlation we think that the power of our study is sufficient. 2. The aim of our study was to investigate the validity and reliability of SM based on the X-ray measurement that accepted as golden standard as we mentioned in method section. In this manner ICC and Bland–Altman Plot are the most preferred statistical analysis methods in the literature and both tests have different advantages and disadvantages [2–4]. It was shown that ICC analyses were utilized in all 27 articles presented in which is pointing the validity and reliability of the nonradiological methods to evaluate the spine [5]. 3. Distribution of the data was tested with Kolmogorov– Smirnov analysis, but this information was not expressed due to usage of non-parametric tests such as Wilcoxon and Kruskal–Wallis. 4. Only 7 articles contain standard error measurement (SEM) values associated with ICC and 95 % interval values of 18 (38.8 %) high quality articles that were scored based on Brink and Louw’s Quality Appraisal of Diagnostic Reliability Studies (QAREL) criteria [5]. This is the point to be considered. From this point of view it is clear to express SEM values together with ICC values will make the result more strong and reliable [6]. In accordance with this important contribution, SEM values which accompanying to our ICC values were found between 3.3 and 6.1 (Cobb– interobserver ICC = 0.962, SEM = 3.8 ; Cobb 1–SM 1 ICC = 0.933, SEM = 4.5 ; Cobb 1–SM 2 ICC = 0.872, SEM = 6.1 ; SM–interobserver ICC = 0.928, SEM = 4.2 ; SM 1–intraobserver ICC = 0.961, SEM = 3.3 ; SM 2–intraobserver ICC = 0.944, SEM = 3.5 ).
Databáze: OpenAIRE
Externí odkaz: