It is certainly welcome news when a study reports that a cheaper, over-the-counter drug produces results about the same as more expensive prescription drugs that are generally trickier to use. However, this release about aspirin performing about the same as anticoagulant drugs for preventing blood clots in patients undergoing knee replacement surgery implies that “good old aspirin” is essentially without risks. The release neglects to tell readers about important limitations of the study, and its description of key aspects of the study is a bit muddled. The release does clearly point out the price differences between the drugs studied.
News releases should be clear about when they are referring to the specific design or results of a study and when they are providing general context for readers. This release makes several references to blood clot prevention following knee replacement surgery and it lists the price of a 30-day supply of some of the drugs that were studied. Most readers would probably conclude that researchers studied long-term use of the drugs, when actually the only variable studied was what, if any, drug was used the day before surgery, the day of surgery, and the day after. The release also uses a quote from a researcher that improperly implies that the study compared prescribing practices before and after participating surgeons largely shifted from prescribing anticoagulants to preferring aspirin. The imprecision muddies the message to readers.
The cost differences between aspirin and anticoagulant drugs are listed and the potential savings by using aspirin are highlighted.
The release lists the percentages of patients who had a blood clot within three months of their knee replacement and emphasized that there was no statistical difference between those who took only aspirin and those who took anticoagulants. However, the release is remiss in not pointing out that the majority of patients were also treated with pneumatic compression boots.
“Good old aspirin” implies that aspirin is benign and without potential harms, which is not accurate, even if the harms of aspirin are typically less than those of anticoagulants. There are two potential harms that were examined: bleeding and formation of clot. While the report notes that aspirin was “not inferior” to other anticoagulants, it was silent on the incidence of bleeding.
In several places, the release mischaracterizes the study design and muddles distinctions between the results of this study and information from other sources. It neglects to point out key limitations noted by the study authors, including that the results seen in this network of Michigan hospitals may not reflect the experiences of patients elsewhere and that “this is an observational study and can only demonstrate association and not causation.”
The study variable was the choice of a clot-prevention drug prescribed to patients the day before surgery, the day of surgery and the day after. The release muddles things by stating, “it’s routine for patients to take clot-preventing drugs for some time afterward.” The headline, subhead and lead sentence all refer to patients taking aspirin “after” or “following” a knee replacement. Readers are likely to mistakenly believe that the study looked at long-term use of clot-preventing drugs, rather than what it actually tracked: the choice of drug for the brief 3-day period around the procedure.
The release also quotes a researcher saying, “This study is truly a real-world experience of what happened in Michigan when the majority of surgeons switched to aspirin,” Hallstrom says. “The incidence of blood clots, pulmonary embolus and death did not increase despite this dramatic change in practice.” However, this study was not set up to compare outcomes before and after a change in practice, so this prominent description of the study is not faithful to the actual study design.
The release refers to “the two-year study period,” which readers could misinterpret to mean patients were followed for two-years, not just 90 days; though elsewhere the release does refer to the percentage of blood clots “over three months.”
The release does not push knee replacement or clot-preventing drugs, though it does somewhat overstate the number of total knee replacements. The release cites a figure of “nearly 1 million Americans” each year, while other sources estimate the rate is closer to 700,000 per year.
The disclosure statements included with the journal article reporting on this study note that Blue Cross and Blue Shield of Michigan support the quality improvement group that did this study and pay some of the salaries of key researchers. Even though the disclosure statements say Blue Cross and Blue Shield did not specifically fund the work on this study, the release should have noted the funding relationship.
The release highlights the comparison between alternatives for minimizing blood clot risk due to knee replacement surgery.
The release notes that many knee replacements surgeries are performed every year.
The release notes that the question of the best clot-prevention strategy for knee replacement patients has been debated for years. It cites an advantage over one recent study. Comparing the language of the release to the more-specific description in the journal article, it does seem that the release appropriately portrays this study as adding to the evidence that surgeons can use when prescribing drugs.
Although the release paints a rather rosy and benign image of “good old aspirin,” the release is clear that this study found that there was no statistical difference in outcomes between patients taking aspirin versus anticoagulant drugs…and it pointed out that doctors need to consider a patient’s history of blood clots, obesity and other factors.