The World Health Organization has declared that antimicrobial resistance is one of the top 10 global public health threats facing humanity. Among the factors that cause the dissemination of multidrug-resistant bacteria is the overuse of antimicrobials in hospitals. This thesis is based on the premise that it is necessary to use historical data to improve antimicrobial prescription and thus reduce the burden of antimicrobial resistance in hospital settings. Its specific goals include analyzing data to provide information that can support antimicrobial prescription, thus avoiding antimicrobial resistance rates remaining high after the COVID-19 pandemic and preventing future similar protocol breakdowns. It also investigates the differences in outcomes between presenting resistant vs. non-resistant bacteria in community-acquired infections. To achieve these objectives, the methods include data analysis tools such as descriptive and inferential statistics, Logistic Regression, Process Mining, and Text Mining. The data includes information on patients admitted to Intensive Care Units in hospitals from a private network located in Rio de Janeiro, Brazil. The thesis comprises three articles and describes a CDSS developed to support antimicrobial prescription in hospitals. The thesis s findings revealed a significant increase in antimicrobial consumption and high variability in treatments for COVID-19 patients. Specifically, meropenem, a carbapenem-class antimicrobial, presented the highest adjusted number of doses prescribed for COVID-19 patients in the analyzed hospitals. The escalation in carbapenem prescription probably explains the observed increase in carbapenem resistance during the COVID-19 surge. In the post-surge, the carbapenem resistance rate decreased, following the decrease pattern we found in carbapenem consumption after the first months of the pandemic. Even though there was a decrease in carbapenem resistance, the post-surge levels remained higher than before the surge. Besides, this thesis did not find an association between presenting with antimicrobial-resistant bacteria and higher chances of hospital mortality or sepsis in patients with community-acquired infections.