On the normativity of evidence - Lessons from philosophy of science and the "VALIDATE" project.

"Evidence" is a key term in medicine and health services research, including Health Technology Assessment (HTA). Randomized clinical trials (RCTs) have undoubtedly dominated the scene of generating evidence for a long period of time, becoming the hallmark of evidence-based medicine (EBM). However, due to a number of misunderstandings, the lay audience and some researchers have sometimes placed too much trust in RCTs compared to other methods of investigation. One of the principal misunderstandings is to consider RCTs findings as isolated and self-apparent pieces of information. In other words, what has been essentially lacking was the awareness of the value-context of the evidence and, in particular, the value- and theory-ladenness (normativity) of scientific knowledge. This paper aims to emphasize the normativity that exists in the production of scientific knowledge, and in particular in the conduct of RCTs as well as in the performance of HTA. The work is based on some lessons learned from Philosophy of Science and the European project "VALIDATE" (VALues In Doing Assessments of healthcare TEchnologies"). VALIDATE was a three-year EU Erasmus+ strategic partnerships project (2018-2021), in which training in the field of HTA was further optimized by using insights from political science and ethics (in accordance with the recent definition of HTA). Our analysis may reveal useful insights for addressing some challenges that HTA is going to face in the future.

Regularized MANOVA (rMANOVA) in untargeted metabolomics.

Many advanced metabolomics experiments currently lead to data where a large number of response variables were measured while one or several factors were changed. Often the number of response variables vastly exceeds the sample size and well-established techniques such as multivariate analysis of variance (MANOVA) cannot be used to analyze the data. ANOVA simultaneous component analysis (ASCA) is an alternative to MANOVA for analysis of metabolomics data from an experimental design. In this paper, we show that ASCA assumes that none of the metabolites are correlated and that they all have the same variance. Because of these assumptions, ASCA may relate the wrong variables to a factor. This reduces the power of the method and hampers interpretation. We propose an improved model that is essentially a weighted average of the ASCA and MANOVA models. The optimal weight is determined in a data-driven fashion. Compared to ASCA, this method assumes that variables can correlate, leading to a more realistic view of the data. Compared to MANOVA, the model is also applicable when the number of samples is (much) smaller than the number of variables. These advantages are demonstrated by means of simulated and real data examples. The source code of the method is available from the first author upon request, and at the following github repository: https://github.com/JasperE/regularized-MANOVA.