Opportunities and Challenges of Using Big Data to Detect Drug‐Drug Interaction Risk

Comparison of potential psychiatric drug interactions in six drug interaction database programs: A replication study after 2 years of updates

“…All six drug interaction database programs include detailed terms of use and disclaimer statements on their websites, which state that the companies assume no responsibility or liability for the content provided including errors and omissions. Monteith et al, 2015;Monteith, Glenn, Geddes, et al, 2016;Quinney, 2019;Tornio et al, 2019;Vilar et al, 2018). There is also increasing recognition of the need to improve how DDI knowledge is standardized and presented for clinical decision making (Hochheiser et al, 2021;McEvoy et al, 2017;Payne et al, 2015;Scheife et al, 2015;Tilson et al, 2016).…”

“…These data resources include clinical data from electronic health records (EHR) and insurance claims, clinical trial data, post‐marketing surveillance systems, extractions from scientific literature, pharmacological knowledge databases, and patient generated social media posts. However, there are a myriad of challenges related to the resources and the analytical methods used in data mining including data quality, combining diverse data, sampling bias, algorithm transparency and bias, generalizability, rare occurrences, establishing temporal patterns, and the lack of standardization in defining DDI signals and events (Monteith & Glenn 2019; Monteith et al., 2015; Monteith, Glenn, Geddes, et al., 2016; Quinney, 2019; Tornio et al., 2019; Vilar et al., 2018). There is also increasing recognition of the need to improve how DDI knowledge is standardized and presented for clinical decision making (Hochheiser et al., 2021; McEvoy et al., 2017; Payne et al., 2015; Scheife et al., 2015; Tilson et al., 2016).…”

“…Physicians should be aware of the lack of consistency among drug interaction database programs, and that information about drug pairs may change over time.In recent decades, there has been progress in the field of DDI, including advances in the understanding of underlying mechanisms, new regulatory guidelines introducing methods to identify DDI risks during drug development, and expansion of the US Food and Drug Administration (FDA) Sentinel initiative to collect real world data on drug safety(FDA, 2019;Levy & Ragueneau-Majlessi, 2019;Tornio et al, 2019). Going forward, data mining of a wide variety of data resources is being emphasized to detect potential DDI for approved drugs, and help define research directions(Dal Pan, 2019;Quinney, 2019;Vilar et al, 2018;Zhang et al, 2020). These data resources include clinical data from electronic health records (EHR) and insurance claims, clinical trial data, post-marketing surveillance systems, extractions from scientific literature, pharmacological knowledge databases, and patient generated social media posts.However, there are a myriad of challenges related to the resources and the analytical methods used in data mining including data quality, combining diverse data, sampling bias, algorithm transparency and MONTEITH AND GLENN…”

What Does It Take to Transform Real‐World Data Into Real‐World Evidence?

“…However, during investigational product development, it is impractical to evaluate every drug combination in clinical trials. Quinney highlights the advantages and disadvantages of different RWD sources to predict potential drug‐drug interactions. However, at the moment, these bioinformatic assessments are hypothesis generating, although mechanistic understanding of drug action can improve confidence in these predictions.…”

ADDI: Recommending alternatives for drug–drug interactions with negative health effects

“…Understanding types of DDIs is essential to recommend alternatives that decrease unexpected adverse drug events (ADEs) and increase synergistic advantages [24][25][26].…”