Sciensano Publications Repository
Not a member yet
11597 research outputs found
Sort by
Methods for meta-analysis and meta-regression of binomial data: concepts and tutorial with Stata command metapreg
No full textBackground
Despite the widespread interest in meta-analysis of proportions, its rationale, certain theoretical and methodological concepts are poorly understood. The generalized linear models framework is well-established and provides a natural and optimal model for meta-analysis, network meta-analysis, and meta-regression of proportions. Nonetheless, generic methods for meta-analysis of proportions based on the approximation to the normal distribution continue to dominate.
Methods
We developed metapreg, a tool with advanced statistical procedures to perform a meta-analysis, network meta-analysis, and meta-regression of binomial proportions in Stata using binomial, logistic and logistic-normal models. First, we explain the rationale and concepts essential in understanding statistical methods for meta-analysis of binomial proportions and describe the models implemented in metapreg. We then describe and demonstrate the models in metapreg using data from seven published meta-analyses. We also conducted a simulation study to compare the performance of metapreg estimators with the existing estimators of the population-averaged proportion in metaprop and metan under a broad range of conditions including, high over-dispersion and small meta-analysis.
Conclusion
metapreg is a flexible, robust and user-friendly tool employing a rigorous approach to evidence synthesis of binomial data that makes the most efficient use of all available data and does not require ad-hoc continuity correction or data imputation. We expect its use to yield higher-quality meta-analysis of binomial proportions.</p
Surveillance épidémiologique des infections invasives causées par les streptocoques du groupe A S. pyogenes - 2017 à 2023
No full textMessages clés:
• Toutes les sources de données (laboratoires vigies, Centre National de Référence et déclaration obligatoires) indiquent les mêmes tendances générales : la fréquence de l’iGAS a été faible pendant les années pandémiques 2020-2021 et a fortement augmenté à partir de la fin de 2022, avec des chiffres très élevés en 2023.
• Le pic du nombre de cas signalés a été atteint au tournant de l’année 2022-2023 : décembre ‘22 selon les chiffres du Centre national de référence (CNR) et les laboratoires vigies, janvier ‘23 selon les chiffres de la déclaration obligatoire (DO).
• Il est difficile de déterminer l’ampleur exacte de l’augmentation relative par rapport aux années précédentes en raison des limites des différentes sources de données.
• Le génotype le plus courant pendant le pic 2022-2023 était M1.
• On ne sait pas exactement ce qui a provoqué ce pic, mais il est probable qu’une combinaison de facteurs joue un rôle : une faible circulation pendant la pandémie entraînant une réduction de l’immunité chez les jeunes enfants, une augmentation soudaine des contacts étroits et des autres infections virales (qui sont un facteur de risque pour l’iGAS) après la levée des mesures d’hygiène, et peut-être aussi un génotype plus virulent.
• Les groupes d’âge les plus touchés sont les enfants de moins de 5 ans et les adultes de plus de 65 ans.
• Il s’agit d’infections graves qui nécessitent presque toujours une hospitalisation et sont associées à une mortalité élevée. Toutefois, les chiffres exacts relatifs à la mortalité ne sont pas disponibles en raison de problèmes d’encodage des données (pour les certificats de décès et les données du Résumé Hospitalier Minimal), de l’enregistrement au moment de l’infection aiguë, sans suivi dans le temps (données du CNR), ou d’une couverture limitée (déclaration obligatoire).</p
General characteristics of the Sentinel General Practitioners (SGP) network in Belgium: short report on 2023.
No full textMaintaining the quality of vaccines through the use of standards: Current challenges and future opportunities.
No full textAn international hybrid meeting held 21-22 June 2023 in Ottawa, Canada brought together regulators, scientists, and industry experts to discuss a set of principles and best practices in the development and implementation of standards. Although the use of international standards (ISs) and international units (IUs) has been an essential part of ensuring human and animal vaccine quality in the past decades, the types and uses of standards have expanded with technological advances in manufacture and testing of vaccines. The needs of stakeholders are evolving in response to the ever-increasing complexity, diversity, and number of vaccine products as well as increasing efforts to replace animal-based potency tests with in vitro assays that measure relevant quality attributes. As such, there must be a concomitant evolution in the design and implementation of both international and in-house standards. Concomitantly, greater harmonization of regulatory expectations must be achieved through collaboration with standard-setting organizations, national control laboratories and manufacturers. Stakeholders provided perspectives on challenges and several recommendations emerged as essential to advancing agreed upon objectives.</p
Unlocking the genomic landscape: Results of the Beyond 1 Million Genomes (B1MG) pilot in Belgium towards Genomic Data Infrastructure (GDI)
No full textAbstract
Genomic medicine has great potential to offer insights into how humans’ genetic variation can affect their health, prevention options and treatment responses. The Beyond 1 Million Genomes (B1MG) project was kicked off in 2020 with the aim of building a federated network of genomic data in Europe, in which Belgium took part as a piloting country. B1MG developed a framework to enable all interested countries to self-evaluate the level of maturity of national genomic medicine practices following a common matrix, called Maturity Level Model (MLM), that contained 49 indicators across eight domains: I. Governance and strategy; II. Investment and economic model; III. Ethics, legislation and policy; IV. Public awareness and acceptance; V. Workforce skills and organisation; VI. Clinical organisation, infrastructure and tools; VII. Clinical genomics guidelines and infrastructure; and VIII. Data management, standards and infrastructure. The ongoing Genomic Data Infrastructure (GDI) project aims to capitalise on the experience of B1MG piloting countries and their MLM results. In this paper, we present the qualitative and quantitative outcomes of B1MG MLM assessment in Belgium and discuss their relevance to GDI. The insights gained from this study can be helpful for steering future policy directions and interventions on genomics in Belgium and beyond.</p
Two cases of congenital lipomatosis in calves.
No full textFetal tumours in domestic animals are considered rare. Congenital lipomatous changes have been sporadically reported in fetal, neonatal and young bovids. Their prognosis is often guarded compared with benign lipocytic tumours and depends on their location and degree of infiltration. They can infiltrate the subcutis, deeper musculature and even the skeleton. Four cases of congenital infiltrative lipomatosis have been described in bovine fetuses and young calves. In this report we describe an additional two cases in a neonatal Belgian Blue White calf and a fetal dairy calf. A potential role in bovine abortion needs further investigation. Furthermore, a suggestion is made to adapt the nomenclature from infiltrative lipoma to lipomatosis in order to emphasize the poor prognosis due to the local infiltrative behaviour of these lesions.</p
