During the post-marketing period, when medicines are used by large population contingents and for more extended periods, adverse events (AE) can occur that can alter the drug’s risk-benefit ratio enough to require regulatory action. AE are defined as health problems that can emerge in a user or patient during treatment with a pharmaceutical product, potentially resulting from medication errors, deviation in the drugs’ quality, adverse drug reactions (ADR), drug-drug interactions, intoxications [1].

According to the World Health Organization (WHO), pharmacovigilance is defined as “as the science and activities relating to the detection, assessment, understanding, and prevention of adverse effects or any other drug-related problem” [2]. Pharmacovigilance is responsible for identifying, assessing, and monitoring drug-related AE’s occurrence to guarantee that the benefits outweigh the risks caused by them [1]. The main instrument in pharmacovigilance is spontaneous reporting, informing government agencies on AE that have occurred with the drugs’ use to achieve this objective.

In Brazil, pharmacovigilance activities are shared by the state and municipal health surveillance agencies and the Brazilian Health Regulatory Agency (Anvisa) [2,3]. The rate of AE reports received by Anvisa is low [4], often far lower than the target proposed by the international literature, which suggests 300 reports per million inhabitants [5]. It is thus necessary to use other sources to detect AE.

AE can be identified during the drug’s study phase before marketing, known as the clinical phase. Clinical tests occur in three distinct phases, known as phases I, II, and III, conducted with healthy volunteers and a limited number of patients. Also, patient selection and treatment generally differ from actual clinical practice [6,7]. AE detected later, in the post-marketing period (also known as phase IV), may require a significant increase in health care and result in unnecessary and often fatal harm to patients [8]. Therefore, the discovery of AE as soon as possible in the post-marketing period is a key objective for health systems and especially for pharmacovigilance systems.

Computational methods commonly referred to as “signal detection” allow drug safety evaluators to analyze large data volumes to identify risk signals for potential AE and serve as an essential component of pharmacovigilance. For example, the U.S. Food and Drug Administration (FDA) routinely uses a signal tracking process to calculate statistics, reporting associations for all the millions of drug combinations and events in its system for communicating AE [8]. These signals alone are not sufficient to establish a causal relationship, but they are considered early warnings that require in-depth assessment by specialists to establish causality.

Dedicated resources for subsequent assessment of each of the multiple signals usually generated by detection algorithms are not feasible. Resources deployed for false leads can undermine a pharmacovigilance system [9]. Automated strategies are thus imperative to reduce the amounts of false-positives and set priorities to assess only the most promising signals. 

Text mining consists of techniques to retrieve textual information, extract information, and process natural language with algorithms and methods for discovering knowledge, data mining, and machine learning. Most of the studies on text mining in pharmacovigilance have focused on electronic health records and medical case reports. Social networks have also been used recently for this purpose. 

At Physiotarget, we integrate data from multiple sources, such as spontaneous reports and social networks, to provide Real-time information on AE’s relevant signals. This set of information is available to users in an easy to understand graphical interface.



1 – Mendes M, Pinheiro R, Avelar K, Teixeira J, Silva G. História da farmacovigilância no Brasil. Rev Bras Farm 2008; 89:246-51.
2 – World Health Organization. Pharmacovigilance. 
3 – Balbino EE, Dias MF. Farmacovigilância: um passo em direção ao uso racional de plantas medicinais e fitoterápicos. Rev Bras Farmacogn 2010; 20:992-1000.
4 – Mota DM. Evolução e resultados do sistema de farmacovigilância do Brasil [Dissertação de Mestrado]. Porto Alegre: Faculdade de Medicina, Universidade Federal do Rio Grande do Sul; 2017.
5 – Meyboom RH, Egberts AC, Gribnau FW, Hekster YA. Pharmacovigilance in perspective. Drug Saf 1999; 21:429-47.
6 – Venulet J, ten Ham M. Methods for monitoring and documenting adverse drug reactions. Int J Clin Pharmacol Ther 1996; 34:112-29.
7 – Cardoso MA, Amorim MAL. A farmacovigilância e sua importância no monitoramento das reações adversas a medicamentos. Revista Saúde e Desenvolvimento 2013; 4:33-56.
8 – Harpaz R, Vilar S, DuMouchel W, Salmasian H, Haerian K, Shah NH, et al. Combining signals from spontaneous reports and electronic health records for detection of adverse drug reactions. J Am Med Inform Assoc 2013; 20:413-9.
9 – Hauben M, Bate A. Data mining in drug safety: side effects of drugs essay. Side Effects of Drugs Annual 2007; 29:xxxiii-xlvi.