Janssen’s claim that Xarelto benefits frail patients better than warfarin built on frail research

Does the news release adequately discuss the costs of the intervention?

Not Satisfactory

No costs were mentioned and it is a significant factor, seeing as warfarin is very inexpensive and produces an anticoagulant effect that is reversible, a characteristic not available with the other anticoagulants.

Does the news release adequately quantify the benefits of the treatment/test/product/procedure?

Not Satisfactory

The release relies solely on relative numbers in describing the benefits: “long-term XARELTO^® use reduced the risk of stroke and systemic embolism by 32 percent and ischemic stroke alone by 31 percent compared to warfarin, with no significant increase in major bleeding.”

It’s a major omission in the release to not report the absolute numbers. According to Table 3 in the published study, out of 2,635 total patients, 39 of those on rivaroxaban suffered a stroke or systemic embolism compared to 49 patients on warfarin. This makes the absolute differences in these combined events to be 1.48% (for rivaroxaban) and 1.86% (for warfarin). The difference of 0.43% means that the number needed to treat to show a benefit is 233. In other words: If you switch 233 patients from warfarin to rivaroxaban, then at 1 year you would prevent one stroke whereas the other 232 patients would remain stroke-free regardless. At 2 years, the absolute risk reduction was a bit higher at 0.98%.

Does the news release seem to grasp the quality of the evidence?

Not Satisfactory

The published study provides many appropriate cautions about the limitations of this research but unfortunately those details didn’t make it into the news release.

For example: “As a retrospective analysis of claims data, this study has limitations worthy of discussion. First, both misclassification (measurement error) and selection bias (selection of patients in a nonrandomized fashion) are always important limitations in claims database studies and may impact a study’s internal validity.”

Also: “propensity score matching can generate cohorts that are comparable in key characteristics, [yet] only those variables measured in MarketScan databases could be used for matching in this analysis. Therefore, regardless of the sophistication of the methodology and the number of variables used in developing propensity scores, residual confounding cannot be excluded.”

Also there were helpful notes about possibly underpowering the study due to the small sample size of the patient groups. All of these factors should urge a cautious interpretation of results.

Does the news release establish the true novelty of the approach?

Not Satisfactory

The release doesn’t make any specific claim about novelty, nor does it acknowledge that there are many other studies, even some randomized trials, comparing these newer anticoagulant agents agents to warfarin.

Does the news release include unjustifiable, sensational language, including in the quotes of researchers?

Not Satisfactory

It was unjustified use of language to use the term “significant” when it’s not clear that the results of the study proved the performance of Xarelto compared to other anticoagulants was “statistically significant.”