Perspectives on the use of data mining in pharmaco-vigilance.

In the last 5 years, regulatory agencies and drug monitoring centres have been developing computerised data-mining methods to better identify reporting relationships in spontaneous reporting databases that could signal possible adverse drug reactions. At present, there are no guidelines or standards for the use of these methods in routine pharmaco-vigilance. In 2003, a group of statisticians, pharmaco-epidemiologists and pharmaco-vigilance professionals from the pharmaceutical industry and the US FDA formed the Pharmaceutical Research and Manufacturers of America-FDA Collaborative Working Group on Safety Evaluation Tools to review best practices for the use of these methods.In this paper, we provide an overview of: (i) the statistical and operational attributes of several currently used methods and their strengths and limitations; (ii) information about the characteristics of various postmarketing safety databases with which these tools can be deployed; (iii) analytical considerations for using safety data-mining methods and interpreting the results; and (iv) points to consider in integration of safety data mining with traditional pharmaco-vigilance methods. Perspectives from both the FDA and the industry are provided. Data mining is a potentially useful adjunct to traditional pharmaco-vigilance methods. The results of data mining should be viewed as hypothesis generating and should be evaluated in the context of other relevant data. The availability of a publicly accessible global safety database, which is updated on a frequent basis, would further enhance detection and communication about safety issues.

The reporting odds ratio and its advantages over the proportional reporting ratio.

PURPOSE: The proportional reporting ratio (PRR) is the proportion of spontaneous reports for a given drug that are linked to a specific adverse outcome, divided by the corresponding proportion for all or several other drugs. The PRR is similar to the proportional mortality ratio (PMR), an old epidemiologic measure calculated from death registries and constructed in similar fashion to the PRR. The PMR has important deficiencies, however, which the PRR shares. Miettinen and Wang demonstrated that the PMR could be improved by reformulating it as an odds ratio and applying the principles of a case-control study to the measure. In this paper, we review the problem with the PRR and show how the corresponding odds ratio represents an improvement over the PRR. METHODS: The method used is discussion and illustration by way of a hypothetical example. RESULTS: The PRR does not estimate relative risk. If, however, a spontaneous report database is viewed as source data for a case-control study, the reporting odds ratio (ROR) can be used to estimate relative risk. Treating the data as source data for a case-control study allows for further reduction of bias by the judicious choice of controls. CONCLUSIONS: Calculating the ROR in spontaneous report databases offers advantages over the PRR. It allows for estimation of the relative risk, and focuses attention on which people or reports should be included or excluded from the control series, permitting more deliberate elimination of biases. It also highlights the inherent weaknesses in spontaneous report data, which become more evident in light of the usual principles of control selection in case-control studies.

Skin reactions in patients with influenza treated with oseltamivir: a retrospective cohort study.

Oseltamivir phosphate is an FDA-approved treatment for influenza that has been available for prescription use in the USA since 1999. The present report describes findings from a post-marketing safety study of skin reactions associated with oseltamivir use. All patients in the claims-derived Ingenix Research Database with a physician diagnosis of influenza and/or a dispensing of oseltamivir between 1 December 1999 and 31 March 2002 were identified. Cohort eligibility criteria included minimum baseline enrolment duration of 3 months, age of at least 1 year and no influenza vaccination on the date of influenza diagnosis or oseltamivir dispensing. Patients were classified into two primary cohorts, influenza diagnosis and oseltamivir dispensing on the same day, and influenza diagnosis but no oseltamivir at any time, and a cohort included for secondary analyses comprising patients who received an oseltamivir dispensing without an influenza diagnosis on the same day. Outcomes included general skin reactions and several specific skin reactions. Events occurring during the 30 days following the date of influenza diagnosis or oseltamivir dispensing were examined using Cox proportional hazards models. Model covariates included age, use of another influenza drug, month and year of index date, and use of antitussives. Adjusted rate ratios for the general class of skin reactions among the primary cohort of oseltamivir users versus non-users were 1.05 (95% CI: 0.88-1.24) for incident cases and 0.98 (95% CI: 0.77-1.24) among patients with a history of a skin reaction. Similar results were seen for the other skin reaction categories, and secondary analyses investigating the oseltamivir users without influenza revealed no elevation in risk. It is concluded that oseltamivir use does not appear to be associated with an increased risk of skin reactions.