In order to address food safety problems, abundant research is conducted every year by academic institutions, governmental agencies and food industries. Although investigations aiming to provide answers to the same research question have in some instances comparable outcomes, and in others, contradictory outcomes, it is highly unlikely that they lead to the same results. A series of regular factors, such as the variability in composition according to production regions, or even differences in laboratorial procedures, bring about such divergences. Nevertheless, the outcomes from different experimental set-ups can still be combined, allowing our current knowledge on a food safety matter to be updated to facilitate microbial modelling and assessing food safety risk that requires the integration of data from multiple studies and sources. Although the interest in integrating existing knowledge on food microbial modelling is relatively recent, so far many applications have been done by using Bayesian statistics and meta-analysis. Bayesian inference mathematically describes the learning process: we may start off with a vague opinion, and then modify our opinion when presented with new evidence, and this can be done repeatedly. On the other hand, meta-analysis refers to the statistical analysis of a large collection of results from primary studies investigating the same research question. Apart from the capability to produce a more precise estimate of the effect of a particular intervention or treatment, meta-analysis can also pinpoint some of the study characteristics responsible for the heterogeneity in the outcomes from the different primary studies. Thus, the objective of this symposium is to review some applications of Bayesian statistics and meta-analysis in food microbial safety, as two powerful tools capable of updating, integrating and summarising research outcomes from different independent studies.

Presentations