Un sistema de clasificación de los diseños de investigación en psicología / A classification system for research designs in psychology

En este trabajo se elabora un marco conceptual y se desarrollan unos principios básicos para fundamentar un sistema de clasificación de los diseños de investigación más usuales en psicología basado en tres estrategias (manipulativa, asociativa y descriptiva) de donde emanan varios tipos de estudios, tres para la estrategia manipulativa (experimentales, cuasiexperimentales y de caso único), tres para la asociativa (comparativos, predictivos y explicativos) y dos para la descriptiva (observacionales y selectivos) (AU)

In this work we devise a conceptual framework and develop some basic principles to promove a classification system for the most usual research designs in psychology based on three strategies (manipulative, associative and descriptive) from which emerge different types of studies, three for manipulative strategy (experimental, quasi-experimental and single-case), three for associative strategy (comparative, predictive and explanatory) and two for descriptive strategy (observational and selective) (AU)

Place of residence before hospital admission and mortality at 12-months in Spanish patients aged 70 years or older.

AIM: Hospitalization of elderly people is often followed by high mortality rates. The aim of this study was to analyze the influence of prior residence on 1-year mortality after hospital discharge in patients aged 70 years and over. METHODS: This was a prospective observational cohort study. Participants were 426 patients discharged from the Internal Medicine Department at a Spanish Hospital who were followed for a 12-month period. Data collection was carried out during hospitalization and included sociodemographic characteristics, comorbidity (Charlson index), functional (Barthel index and Lawton scale) and cognitive conditions (Short Portable Mental Status Questionnaire), together with parameters related to the disease causing admission (diagnosis related group, laboratory tests, length of hospital stay). Mortality was carried out using telephone interviews. RESULTS: A total of 420 (98.6%) patients were located at the end of follow up. Of these, 95 patients had died, giving an overall 1-year mortality of 22.6%. The mortality rate for patients living in their private homes was 15.6% versus 24.7% for those living with relatives and 60% for those living in institutions. After adjustment for potential confounders, prior residence was associated with mortality with a hazard ratio of 3.98 (95% CI 1.94-8.17) for those institutionalized and a hazard ration of 1.68 (95% CI 0.99-2.16) for those living with relatives, as compared with patients living in their private homes. CONCLUSIONS: Prior residence is associated with 1-year-mortality following discharge after controlling for several multidimensional factors.

A simulation study of rater agreement measures with 2x2 contingency tables

Mediante un estudio de simulación se aborda una comparación entre seis medidas obtenidas usando tres enfoques diferentes para la evaluación del acuerdo. Los coeficientes de acuerdo elegidos fueron Sigma de Bennet (1954), Pi de Scott (1955), Kappa de Cohen (1960) y Gamma de Gwet (2001; 2008) para representar el enfoque clásico descriptivo, el coeficiente Alfa de Aickin (1990), para representar el enfoque de los modelos loglineal y mixtura (“mixture models”) y la medida Delta de Martín and Femia (2004) para representar el enfoque de los test de elección multiple. Los resultados obtenidos confirman que los coeficientes Pi y Kappa presentan diferencias notables en relación a los restantes coeficientes particularmente en presencia de valores extremos de prevalencia y sesgo entre observadores. El mejor comportamento fue observado con los coeficientes Sigma de Bennet y Delta de Martín and Femia para todos los valores de prevalencia y sesgo entre observadores(AU)

A comparison between six rater agreement measures obtained using three different approaches was achieved by means of a simulation study. Rater coefficients suggested by Bennet’s Sigma (1954), Scott’s Pi (1955), Cohen’s Kappa (1960) and Gwet’s Gamma (2008) were selected to represent the classical, descriptive approach, Alpha agreement parameter from Aickin (1990) to represent loglinear and mixture model approaches and Delta measure from Martín and Femia (2004) to represent multiple-choice test. Main results confirm that Pi and Kappa descriptive measures present high levels of mean bias in presence of extreme values of prevalence and rater bias but small to null levels with moderate values. The best behavior was observed with Bennet and Martín and Femia agreement measures for all levels of prevalence(AU)

Trombosis en aorta torácica aparentemente normal: nuevo caso y nuevos interrogantes / Thrombosis of an apparently normal thoracic aorta: another case, another conundrum

No disponible

No disponible

Análisis comparativo de tres enfoques para evaluar el acuerdo entre observadores / Comparative analysis of three approaches for rater agreement

En este trabajo se realiza un análisis comparativo de tres enfoques generales para evaluar el acuerdo entre observadores que originan tres tipos diferentes de medidas: descriptivas (coeficientes sigma, pi y kapa), basadas en modelos loglineales y basadas en modelos con mezcla (mixture) de distribuciones. Aunque los enfoques loglineal y mixture asumen un concepto de corrección del azar diferente, la flexibilidad de ambos enfoques permite utilizar sus parámetros para realizar un análisis en profundidad de las pautas de acuerdo y desacuerdo. Se demuestra que las medidas descriptivas pueden obtenerse introduciendo restricciones al modelo loglineal de cuasi-independencia. Y que aunque los enfoques loglineal y mixture tienen una interpretación similar cuando todos los parámetros diagonales son positivos, su interpretación difiere en caso contrario debido a la diferente escala de medida utilizada

This work is a comparative analysis of three general approaches for rater agreement measurement which produce three different statistical coefficients: descriptive (sigma,pi y kappa coefficients), based on log-linear models and on mixture models. Although loglinear and mixture approaches assume a different concept of chance correction, their flexibility allow using its parameters to analyze agreement and disagreement patterns. We show that all descriptive measures can also be obtained by introducing constraints to a basic loglinear (or mixture) quasi-independence model. Due to the different measurement scale used, both loglinear and mixture approaches have a similar interpretation when diagonal parameters are positive but a different interpretation otherwise

Procedimientos para detectar y medir el sesgo entre observadores / Methods for detecting and assessing the interrater bias

En este trabajo se realiza un análisis de los distintos métodos para detectar y medir el sesgo entre observadores desde dos perspectivas básicas: el enfoque clásico, fundamentado en índices descriptivos y el enfoque del modelado, fundamentado en modelos loglineales. Se demuestra que estas medidas no son satisfactorias para detectar y medir el sesgo de forma unívoca porque presentan resultados contradictorios y se proponen nuevas alternativas a desarrollar que permitan descomponer correctamente error de medida y sesgo

In this paper we present an analysis of the different methods commonly used to detect and assess interrater bias from two basic perspectives: classical approach based on descriptive-type criteria and loglinear model approach. We show that all these procedures are not satisfactory for the detection and measurement of observer bias in a univoque way due to contradictory results. We propose some new alternatives to develop which allow the correct separation of bias and measurement error

[Comparative analysis of three approaches for rater agreement]. / Análisis comparativo de tres enfoques para evaluar el acuerdo entre observadores.

This work is a comparative analysis of three general approaches for rater agreement measurement which produce three different statistical coefficients: descriptive (sigma, pi y kappa coefficients), based on log-linear models and on mixture models. Although loglinear and mixture approaches assume a different concept of chance correction, their flexibility allow using its parameters to analyze agreement and disagreement patterns. We show that all descriptive measures can also be obtained by introducing constraints to a basic loglinear (or mixture) quasi-independence model. Due to the different measurement scale used, both loglinear and mixture approaches have a similar interpretation when diagonal parameters are positive but a different interpretation otherwise.

Assessment of disability in spinal cord injury.

PURPOSE: Spinal cord injury (SCI) implies a drastic change in the quality of life of an individual, and this fact must be considered by the physiatrist. The aim of this study was to assess the degree of disability shown by patients with spinal cord injury at 5 years post-SCI. For this purpose we used the Craig Handicap Assessment and Reporting Technique (CHART) scale which included the evaluation of occupation, physical independence, mobility, social integration and economic self-sufficiency. METHOD: A questionnaire was mailed to 243 patients who were admitted over a one-year period to the Hospital Nacional de Parapléjicos in Toledo, Spain. From these, we received 90 answers. No significant differences were observed when they were compared to the 90 patients who answered the questionnaire. RESULTS: 30% of the patients had less than 333 points on the scale, meaning that they could be considered severely disabled, while a score higher than 438 points was seen in 46.7% of the patients: most of these individuals were younger than 18 and had less severe neurologic injuries. CONCLUSIONS: Despite the enormous impact of SCI, according to the scale used most patients have scores above the disability level.