This cost-effectiveness analysis is based on the results of a meta-analysis and costs of events obtained from Sunnybrook Health Science Centre. Costs for major and minor events in this analytis were obtained from one hospital in Totonto, which may not necessarily be the same at other hospitals across Canada. A more general description of costs would involve cost data (including participation from an expert panel) from several hospitals in different regions.
Also, the cost of stroke and major events was based on first time admission to hospital. Intuitively, for a patient who has had a previous stroke or major bleeding complication, the probabiHty of a second event is likely greater than the first event and an alternate intervention may be more appropriate. flovent inhaler
It is import ant to be critical of data used in the metaanalysis because results depend directly on them. All ofthe trials used very similar criteria. Of concern, however, was the percentage of time patients were in the target inr when prothrombin time was measured. These values ranged from 42 to 83% (Table 7), a wide range, which may further complicate the analysis by increasing or decreasing reported bleeding complications. This, however, is unclear and no apparent pattern emerges.