Mining Unexpected Temporal Associations: Applications in Detecting Adverse Drug Reactions

Abstract

Abstract-In various real-world applications, it is very useful mining unanticipated episodes where certain event patterns unexpectedly lead to outcomes, e.g., taking two medicines together sometimes causing an adverse reaction. These unanticipated episodes are usually unexpected and infrequent, which makes existing data mining techniques, mainly designed to find frequent patterns, ineffective. In this paper, we propose unexpected temporal association rules (UTARs) to describe them. To handle the unexpectedness, we introduce a new interestingness measure, residual-leverage, and develop a novel case-based exclusion technique for its calculation. Combining it with an event-oriented data preparation technique to handle the infrequency, we develop a new algorithm MUTARC to find pairwise UTARs. The MUTARC is applied to generate adverse drug reaction (ADR) signals from real-world healthcare administrative databases. It reliably shortlists not only six known ADRs, but also another ADR, flucloxacillin possibly causing hepatitis, which our algorithm designers and experiment runners have not known before the experiments. The MUTARC performs much more effectively than existing techniques. This paper clearly illustrates the great potential along the new direction of ADR signal generation from healthcare administrative databases.

iADRs: towards online adverse drug reaction analysis

“…These studies can generally be divided into two categories. The first category uses statistical or data-mining methods to identify the signals of adverse drug reactions (Beta et al 1998; Dumouchel 1999; Evans et al 2001; Huang et al 2007; Jin et al 2007; Orre et al 2000; Szarfman et al 2002). These stand-alone methods, without any integration with knowledge discovery systems, are tedious and inconvenient for users to use to identify possible adverse drug reactions.…”

“…An algorithm that adopts the temporal association rule technology, called MUTARC, was proposed by Jin et al (Jin et al 2007; Jin et al 2010). Using a database provided by the Queensland Department of Health, called the Queensland Linked Data Set, an Unexpected Temporal Association Rule (UTAR) was defined to identify ADRs.…”

Identification of adverse drug-drug interactions through causal association rule discovery from spontaneous adverse event reports

“…Association rule mining is a well-established data mining method for discovering relationships in data and its algorithmic variations have been developed for ADR detection [24, 40, 46]. An association rule is an implication expression of the form A → B , where A and B are two event sets that do not share any common events.…”

Good Signal Detection Practices: Evidence from IMI PROTECT

“…At present, signal detection is predominantly based on individual case reports, but the use of electronic health records and insurance claims to detect ADRs is an area of active research [ 61 – 64 ]. While some work has focussed on traditional sequential approaches and looking to use them for signal detection, new methods have also been adapted or proposed [ 65 – 68 ]. Other publications have highlighted the challenges and limitations of using longitudinal observational data for signal detection [ 69 , 70 ].…”