Coeficiente de correlación intraclase / Intraclass correlation coefficient

El coeficiente de correlación intraclase permite medir la concordancia general entre dos o más mediciones que implican variables de carácter cuantitativo, obtenidas estas con diferentes instrumentos de medida o evaluadores. Se basa en un modelo de análisis de varianza con medidas repetidas. Su uso solo es posible si existe normalidad de las distribuciones de las variables, igualdad de varianzas e independencia entre los errores producidos por los observadores (AU)

The intraclass correlation coefficient makes it possible to measure the general concordance between two or more measurements that involve variables of a quantitative nature, obtained with different measuring instruments or evaluators. It is based on an analysis of variance model with repeated measures. Its use is only possible if there is normality of the distributions of the variables, equality of variances and independence between the errors produced by the observers (AU)

[Intraclass correlation coefficient]. / Coeficiente de correlación intraclase.

The intraclass correlation coefficient makes it possible to measure the general concordance between two or more measurements that involve variables of a quantitative nature, obtained with different measuring instruments or evaluators. It is based on an analysis of variance model with repeated measures. Its use is only possible if there is normality of the distributions of the variables, equality of variances and independence between the errors produced by the observers.

Confiabilidad de una escala de valoración de resúmenes para congresos

Introducción: En 2013, desarrollamos una escala, para evaluar resúmenes de congresos de la Sociedad de Cirujanos de Chile (SOCICH). Objetivo: Determinar consistencia interna y confiabilidad interobservador de una escala para evaluar resúmenes de congresos. Material y Método: Estudio de confiabilidad. Doce cirujanos fueron capacitados de forma virtual durante 8 horas, para aplicar la escala. Una vez finalizado el entrenamiento, se les envió un cuestionario para evaluar contenidos de la capacitación, y varios resúmenescasos para ser evaluados con la escala antes señalada. Se aplicó estadística descriptiva, luego se estimó el grado de acuerdo entre observadores para cada ítem de la escala. Posteriormente, se evaluó el coeficiente de correlación (CCI), utilizando un modelo de dos factores mixtos en el que los efectos de los evaluadores son aleatorios y los ítems fijos; utilizando una definición de acuerdo absoluto. Además, se evaluó la consistencia interna de los ítems utilizando alfa de Cronbach, considerando intérvalos de confianza del 95% (IC 95%). Resultados: Luego de analizar las mediciones de los 9 ítems por los 12 observadores, se verificó que el CCI fue de 0,871; con un IC 95% de 0,700; 0,965. El valor de la consistencia interna fue de 0,7 considerando los 9 ítems, no se recomienda eliminar ningún ítem. Conclusión: La escala tiene buena confiabilidad interobservador y los ítems son consistentes entre sí; por lo que puede ser considerada como un instrumento confiable para la valoración de resúmenes de congresos.

Background: In 2013, we developed a scale to evaluate the abstracts of the congresses of the Society of Surgeons of Chile (SOCICH). Objective: To determine internal consistency and interobserver reliability of a scale to evaluate conference abstracts. Material and Methods: Reliability study. Twelve surgeons were trained virtually for 8 hours, to apply the scale. Once the training was finished, they were sent a questionnaire to evaluate the contents of the training, and several summaries-cases to be evaluated with the aforementioned scale. Descriptive statistics were applied, then the degree of agreement between observers was estimated for each item of the scale. Subsequently, intraclass correlation coefficient (ICC) was evaluated, using a mixed two-factor model where the effects of the evaluators are random and the items are fixed, using a definition of absolute agreement. In addition, the internal consistency of the items was evaluated using Cronbach's alpha, considering 95% confidence intervals (95% CI). Results: After analyzing the measurements of the 9 items by the 12 observers, it was verified that the ICC was 0.871; with a 95% CI of 0.700; 0.965. The internal consistency value was 0.7 considering the 9 items, it is not recommended to delete any item. Conclusions: The scale has good internal consistency and interobserver reliability. Therefore, it can be considered as reliable instrument to be used in the evaluation of abstracts for congresses.

Reproducibility of manual choroidal thickness measurements using optical coherence tomography. / Reproducibilidad de la medición manual del grosor coroideo utilizando la tomografía de coherencia óptica.

PURPOSE: Spectral-domain optical coherence tomography (SD-OCT) is the most useful tool to measure choroidal thickness (CT). CT may be increased in ocular and systemic diseases. However, there are concerns relating reproducibility and external validity of OCT. The aim of this study was to assess the inter-observer and intra-observer variability of manual OCT measurements. METHODS: CT was manually measured in the central choroid of 40 eyes from 21 subjects (11 healthy and 10 with ankylosing spondylitis) using RTVue-100 OCT (Optovue Inc., Fremont, CA, EE.UU.). Measurements were performed by 9 independent ophthalmologists from 6 different centers. To assess the inter-observer variability, the intra-class correlation coefficient (ICC) method was calculated. Also, intra-observer variability was assessed in 2 of the ophthalmologists. RESULTS: The mean subfoveal CT was 364.9±85.1µm (range, 170 to 572). The inter-observer ICC was 0.823 (CI 95%, 0.749 to 0.888, p<0.001). The intra-observer ICCs were 0.885 (CI 95%, 0.783 to 0.939, p<0.001) and 0.925 (CI 95%, 0.859 to 0.960. p<0.001). CONCLUSIONS: In this study, manual measurements of CT with OCT showed a good concordance. These results suggest that manual OCT is a valid tool for multicenter studies.

Concordancia entre los diferentes conteos celulares según los parámetros morfológicos y morfométricos del endotelio corneal / Concordance between different cell counts according to morphologic and morphometric parameters of the corneal endothelium

RESUMEN Objetivo: Estimar los valores morfológicos y morfométricos del endotelio corneal según la cantidad de células y evaluar la concordancia interobservadores para los diferentes parámetros, considerados según los diferentes conteos celulares en adultos sin alteraciones corneales. Métodos: Se realizó una investigación observacional, descriptiva y transversal de serie de casos en el Servicio de Cirugía Refractiva del Instituto Cubano de Oftalmología "Ramón Pando Ferrer" en dos años de estudio. Después de aplicar los criterios de exclusión, la muestra quedó conformada por 200 ojos de 100 pacientes adultos sin alteraciones corneales. Se realizó microscopia endotelial de no contacto SP-3000P, para identificar los valores morfológicos (hexagonalidad y coeficiente de variabilidad) y morfométricos (densidad celular), así como el promedio del tamaño celular corneal según cantidad de células evaluadas. Resultados: Según la cantidad de células endoteliales evaluadas, no existieron diferencias significativas de las variables morfológicas y morfométricas (p> 0,05) en ambos ojos. La concordancia entre los diferentes conteos celulares según los valores de los coeficientes de correlación intraclase fueron todos altos. La concordancia interobservadores (excepto para la hexagonalidad) y los coeficientes de correlación intraclase fueron altos. Conclusiones: Los valores morfológicos y morfométricos del endotelio corneal según cantidad de células evaluadas, son similares en todos los conteos celulares. Se demuestra una buena concordancia entre los diferentes conteos celulares estudiados para los diferentes parámetros estimados(AU)

ABSTRACT Objective: To estimate the morphological and morphometric values of the corneal endothelium according to the number of cells and evaluate the interobserver concordance for the different parameters, estimated according to the different cell counts in adults without corneal alterations. Methods: An observational, descriptive and cross-sectional case series research was carried out in the Refractive Surgery Service of Ramón Pando Ferrer Cuban Institute of Ophthalmology during two years of study. After applying the exclusion criteria, the sample was made up of 200 eyes of 100 adult patients without corneal alterations. Non-contact endothelial microscopy SP-3000P was performed to identify morphological values (hexagonality and coefficient of variability) and morphometric values (cell density), as well as the average corneal cell size according to the number of cells evaluated. Results: According to the amount of endothelial cells evaluated, there were no significant differences between morphological and morphometric variables (p>0.05) in both eyes. The agreement between the different cell counts according to the values of the interclass correlation coefficients (ICC) were all high. The interobserver concordance and ICCs were also high, except for hexagonality. Conclusions: The morphological and morphometric values of the corneal endothelium, according to the number of cells evaluated, are similar in all cell counts. Good concordance between the different cell counts studied for the different estimated parameters is demonstrated(AU)