Chris Kelman

Mining Unexpected Temporal Associations: Applications in Detecting Adverse Drug Reactions

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.

Mining risk patterns in medical data

In this paper, we discuss a problem of finding risk patterns in medical data. We define risk patterns by a statistical metric, relative risk, which has been widely used in epidemiological research. We characterise the problem of mining risk patterns as an optimal rule discovery problem. We study an anti-monotone property for mining optimal risk pattern sets and present an algorithm to make use of the property in risk pattern discovery. The method has been applied to a real world data set to find patterns associated with an allergic event for ACE inhibitors. The algorithm has generated some useful results for medical researchers.

Estimating the Cost of Complications of Diabetes in Australia Using Administrative Health‐Care Data

OBJECTIVES To estimate Australian health-care costs in the year of first occurrence and subsequent years for major diabetes-related complications using administrative health-care data. METHODS The costs were estimated using administrative information on hospital services and primary health-care services financed through Australias national health insurance system Medicare. Data were available for 70,340 patients with diabetes in Western Australia (mean duration of 4.5 years of follow-up). Multiple regression analysis was used to estimate inpatient and primary care costs. RESULTS For a man aged 60 years, the average costs in the year the event first occurred were: amputation

Signaling Potential Adverse Drug Reactions from Administrative Health Databases

20,416 (95% CI 18,670-22,411); nonfatal myocardial infarction (MI)

Mining unexpected associations for signalling potential adverse drug reactions from administrative health databases

11,660 (10,931-12,450); nonfatal stroke

Association Rule Discovery with Unbalanced Class Distributions

14,012 (12,849-15,183); ischaemic heart disease

Representing association classification rules mined from health data

12,577 (12,026-13,123); heart failure

It's time: record linkage — the vision and the reality

15,530 (13,965-17,009); renal failure

Risk equations to predict life expectancy of people with Type 2 diabetes mellitus following major complications: a study from Western Australia

28,661 (22,989-34,202); and chronic leg ulcer

Patterns of analgesic and anti-inflammatory medicine use by Australian veterans

15,413 (13,089-18,123). The costs in subsequent years for a man aged 60 years range from 14% for nonfatal MI to 106% for renal failure, of event costs. CONCLUSIONS Estimates of the health-care costs associated with diabetes-related complications can be used in modeling the long-term costs of diabetes and in evaluating the cost-effectiveness of improving care.