Bayesian Meta-Analysis for Binary Data and Prior Distribution on Models.

In meta-analysis, the structure of the between-sample heterogeneity plays a crucial role in estimating the meta-parameter. A Bayesian meta-analysis for binary data has recently been proposed that measures this heterogeneity by clustering the samples and then determining the posterior probability of the cluster models through model selection. The meta-parameter is then estimated using Bayesian model averaging techniques. Although an objective Bayesian meta-analysis is proposed for each type of heterogeneity, we concentrate the attention of this paper on priors over the models. We consider four alternative priors which are motivated by reasonable but different assumptions. A frequentist validation with simulated data has been carried out to analyze the properties of each prior distribution for a set of different number of studies and sample sizes. The results show the importance of choosing an adequate model prior as the posterior probabilities for the models are very sensitive to it. The hierarchical Poisson prior and the hierarchical uniform prior show a good performance when the real model is the homogeneity, or when the sample sizes are high enough. However, the uniform prior can detect the true model when it is an intermediate model (neither homogeneity nor heterogeneity) even for small sample sizes and few studies. An illustrative example with real data is also given, showing the sensitivity of the estimation of the meta-parameter to the model prior.

Vídeos como herramientas de diseminación y transferencia de conocimiento / Videos as tools for scientific dissemination and knowledge transfer

No disponible

La respuesta de Gaceta Sanitaria a la pandemia de COVID-19. Gestión rápida y transferencia / Gaceta Sanitaria’s response to the COVID-19 pandemic. Rapid management and transfer

No disponible

Gaceta Sanitaria en 2016. Una nueva etapa, fortalecimiento de eGaceta e internacionalización / Gaceta Sanitaria in 2016. A new stage; the strengthening of e-Gaceta and its internationalisation

No disponible

Efecto de una capacitación sobre los puntajes de la prueba de admisión de la Universidad de Costa Rica: una aproximación bayesiana

Se usó un diseño cuasi-experimental con pre y post-test para estimar el efecto de una capacitación para la prueba de admisión de la Universidad Costa Rica, un test estandarizado que mide habilidades de razonamiento en contextos verbales y matemáticos. Cuatro colegios públicos del área metropolitana central del país participaron en el estudio, asignándose dos de ellos aleatoriamente al grupo de intervención y los otros dos al grupo de control, con 61 estudiantes en el primer grupo y 80 en el segundo. La intervención consistió de 5 sesiones de capacitación de 3 horas, utilizando como guía un manual desarrollado por una experta pedagoga, con enfoque constructivista. Las medidas antes y después fueron formas reducidas de la prueba de admisión 2014. La variable dependiente fue la diferencia entre ambas mediciones. El efecto de la capacitación fue de 3.5 puntos porcentuales y significativo, y se estimó utilizando un modelo bayesiano de regresión multinivel.

A quasi-experimental design with pre and post- test was used to estimate training effects for the University of Costa Rica's admission test, a standardized exam that measures reasoning abilities in mathematical and verbal contexts. Four secondary public schools from the metropolitan central area of the country participated in the study; two of them were randomly assigned to the intervention group and the other two to the control group, with 61 students in the first group and 80 in the second. The intervention consisted of five three-hour training sessions, using a written guide developed by a pedagogy expert with a constructivist approach. Before and after measures were reduced test forms of the real admission test from the year 2014. The dependent variable was the difference between the two measures. The effect of the training was estimated using a multilevel Bayesian regression model with a significant magnitude of 3.5 percentage points.

Time trends in socio-economic inequalities in the lack of access to dental services among children in Spain 1987-2011.

INTRODUCTION: Adult oral health is predicted by oral health in childhood. Prevention improves oral health in childhood and, consequently in adulthood, so substantial cost savings can be derived from prevention. The burden of oral disease is particularly high for disadvantaged and poor population groups in both developing and developed countries. Therefore, an appropriate and egalitarian access to dental care becomes a desirable objective if children's dental health is to be promoted irrespective of socioeconomic status. The aim of this research is to analyse inequalities in the lack of access to dental care services for children in the Spanish National Health System by socio-economic group over the period 1987-2011. METHODS: Pooled data from eight editions of the Spanish National Health Survey for the years 1987-2011, as well as contextual data on state dental programmes are used. Logistic regressions are used to examine the related factors to the probability of not having ever visited the dentist among children between 6 and 14 years old. Our lack of access variable pays particular attention to the socioeconomic level of children's household. RESULTS: The mean probability of having never been to the dentist falls considerably from 49.5% in 1987 to 8.4% in 2011. Analysis by socioeconomic level indicates that, in 1987, the probability of not having ever gone to the dentist is more than two times higher for children in the unskilled manual social class than for those in the upper non-manual social class (odds ratio 2.35). And this difference is not reduced significantly throughout the period analysed, rather it increases as in 1993 (odds of 2.39) and 2006 (odds of 3.03) to end in 2011 slightly below than in 1987 (odds ratio 1.80). CONCLUSION: There has been a reduction in children's lack of access to dentists in Spain over the period 1987-2011. However, this reduction has not corrected the socioeconomic inequalities in children's access to dentists in Spain.

Equity in specialist waiting times by socioeconomic groups: evidence from Spain.

In countries with publicly financed health care systems, waiting time--rather than price--is the rationing mechanism for access to health care services. The normative statement underlying such a rationing device is that patients should wait according to need and irrespective of socioeconomic status or other non-need characteristics. The aim of this paper is to test empirically that waiting times for publicly funded specialist care do not depend on patients' socioeconomic status. Waiting times for specialist care can vary according to the type of medical specialty, type of consultation (review or diagnosis) and the region where patients' reside. In order to take into account such variability, we use Bayesian random parameter models to explain waiting times for specialist care in terms of need and non-need variables. We find that individuals with lower education and income levels wait significantly more time than their counterparts.

Bayesian analysis of nosocomial infection risk and length of stay in a department of general and digestive surgery.

OBJECTIVE: Nosocomial infection is one of the main causes of morbidity and mortality in patients admitted to hospital. One aim of this study is to determine its intrinsic and extrinsic risk factors. Nosocomial infection also increases the duration of hospital stay. We quantify, in relative terms, the increased duration of the hospital stay when a patient has the infection. METHODS: We propose the use of logistic regression models with an asymmetric link to estimate the probability of a patient suffering a nosocomial infection. We use Poisson-Gamma regression models as a multivariate technique to detect the factors that really influence the average hospital stay of infected and noninfected patients. For both models, frequentist and Bayesian estimations were carried out and compared. RESULTS: The models are applied to data from 1039 patients operated on in a Spanish hospital. Length of stay, the existance of a preoperative stay and obesity were found the main risk factors for a nosomial infection. The existence of a nosocomial infection multiplies the length of stay in the hospital by a factor of 2.87. CONCLUSION: The results show that the asymmetric logit improves the predictive capacity of conventional logistic regressions.