An observational study of the discrediting of COX-2 NSAIDs in Australia: Vioxx or class effect?

BACKGROUND When a medicine such as rofecoxib (Vioxx) is withdrawn, or a whole class of medicines discredited such as the selective COX-2 inhibitors (COX-2s), follow-up of impacts at consumer level can be difficult and costly. The Australian Longitudinal Study on Women's Health provides a rare opportunity to examine individual consumer medicine use following a major discrediting event, the withdrawal of rofecoxib and issuing of safety warnings on the COX-2 class of medicines. The overall objective of this paper was to examine the impact of this discrediting event on dispensing of the COX-2 class of medicines, by describing medicine switching behaviours of older Australian women using rofecoxib in September 2004; the uptake of other COX-2s; and the characteristics of women who continued using a COX-2. METHODS Participants were concessional beneficiary status women from the Older cohort (born 1921-26) of the Australian Longitudinal Study on Women's Health who consented to linkage to Pharmaceutical Benefits Scheme data, with at least one rofecoxib prescription dispensed in the 12 months before rofecoxib withdrawal. A prescription was defined as one dispensing occasion. Women were grouped by rofecoxib pattern of use: continuous (nine or more prescriptions dispensed in the 12 months prior to rofecoxib withdrawal) or non-continuous (eight or less prescriptions dispensed in the 12 months prior to rofecoxib withdrawal) users. Incidence rate per 100,000 person days and incidence risk ratio described uptake of alternate medicines, following rofecoxib withdrawal. Kaplan-Meier curves described differences in uptake patterns by medicine and pattern of rofecoxib use. Patterns of use of COX-2s in the next 100 days after first COX-2 uptake were described. RESULTS Medicine switches and pattern of medicines uptake differed significantly depending upon whether a woman was a continuous or non-continuous rofecoxib user prior to rofecoxib discrediting. Continuous rofecoxib users overwhelmingly switched to another COX-2 and remained continuing COX-2 users for at least 100 days post-switch. CONCLUSIONS The typical switching behaviour of this group of women suggests that the issues leading to the discrediting of rofecoxib were not seen as a COX-2 class effect by prescribers to this high use group of consumers.