"TheShinISS": an open-source tool for conducting distributed analyses within pharmacoepidemiological multi-database studies Introduction Healthcare databases represent useful source of data for conducting pharmacoepidemiological studies on drug and vaccine utilization, efficacy and safety. When the studies include diff erent geographical areas, it is appropriate to use a multi-database study approach, based on a Common Data Model (CDM), to conduct locally distributed analyses. This article describes the experience of the Istituto Superiore di Sanità (ISS) - Italian National Institute of Health in planning and implementing a statistical-informatics tool for conducting multi-database studies based on a Common Data Model (CDM). Materials and methods “TheShinISS” is an R open-source tool developed by ISS for conducting distributed analyses within the main epidemiological multi-database study designs: descriptive, cohort, case-control, case-cohort, self-controlledcase- series. “TheShinISS” allows elaborating and processing health archives at local level, performing data quality control, matching/sampling, record-linkage, and fi nally creating the anonymized dataset for the centralized data analyses. Results "TheShinISS" was used in pharmacoepidemiological studies conducted within the ITA-COVID19 network, which includes ISS, universities and regions. The network was set up during the pandemic emergency to promote the conduction of observational studies to provide prompt evidence on the role of drugs and vaccines on the prognosis of patients with COVID-19. Conclusions “TheShinISS” proved to be easy to use and adaptable to different research aims and designs. Through the reproducibility of every face of the analyses, it has provided satisfying quality control in the creation of analytical datasets. These features made it possible the inclusion of different regions/provinces in the ITA-COVID19 network, and above all reduced the time necessary for the conduction of studies. Key words: pharmacoepidemiology, electronic health records; multi-database study