This story, like the Health Day story we reviewed on this study, reports that aspirin may cut the risk of death for some prostate cancer patients; a claim not supported by the study, which simply reported an association, not any cause-and-effect relationship. And the headline and lead don’t tell readers that the study is relevant only to certain men with localized prostate cancer. But at least this story alerts readers right away that the findings are preliminary and should not prompt men to start taking aspirin in hopes it could reduce their risk of prostate cancer death. It also includes a standard caution about medical meeting presentations not having gone through the type of peer review used by leading medical journals.
An observational study, especially this sort of analysis of past medical records, can only suggest associations. It takes an experiment that follows patients (some who get the drug in question and some who don’t) to try to demonstrate cause-and-effect and largely rule out other potential explanations. News stories should avoid wording that implies cause-and-effect before such experiments are done.
Cost was not mentioned in this story. Now you might say that the focus is on aspirin, which readers are likely to already know is relatively inexpensive. However, the story should have highlighted the cost differential between aspirin and some of the other drugs included in the study, such as warfarin and Plavix, since those on the inexpensive aspirin derived the most benefit. And it should be noted that even drugs that cost little for each pill may cost the nation a substantial amount if long-term use is recommended for many millions of individuals
This story would have met this criterion if not for the headline and lead claiming that this study suggested aspirin may cut the risk of cancer death. Other than muddling the distinction between an association and a cause-and-effect relationship, the story does a generally good job of giving readers details about the study and its results. Although the lead highlights the difference in relative risk, the body of the story includes the absolute risk figures (10-year prostate cancer death rates of 10 percent vs. 4 percent). That said, the story should not have implied that the study proved the anticlotting drugs could clearly claim credit for the observed differences.
The story mentions that the drugs carry risks, including bleeding. However, the story should have been clear that such bleeding can be lethal in some cases and that these drugs also raise the risk of prostate cancer patients suffering complications during treatments.
On most counts, this story does a good job describing the study, including an alert near the top that the findings are preliminary and shouldn’t prompt men to stat taking aspirin. It also alerts readers that the results are being announced at a medical meeting and so the findings have not been scrutinized by peer reviewers to the degree required by leading medical journals.
However, by saying in the headline and the lead sentence that aspirin may cut cancer deaths, the story implies evidence of a cause-and-effect relationship when all this study could do is identify an association between the use of aspirin or other anticlotting drugs and prostate cancer death rates. Readers should have been told about the limitations of this kind of retrospective analysis of medical records, including unknown differences between the patients who took aspirin and those who didn’t… and the possibility that some unidentified third factor plays a role in explaining the findings.
The story did include the now-boilerplate (and laudable) WebMD language for stories coming from presentations at conferences – “This study was presented at a medical conference. The findings should be considered preliminary as they have not yet undergone the “peer review” process, in which outside experts scrutinize the data prior to publication in a medical journal.” But that caveat doesn’t cover the issue we raise above.
The body of story is clear about who these results are relevant to. It points out that the study involved the records of men who had been diagnosed with prostate cancer that had not spread. However, the headline and lead imply that aspirin could lower the risk of prostate cancer death for all men, not just those diagnosed with localized tumors of a type likely to spread.
It was easy to meet the basic requirement of this criterion, because the news briefing was moderated by an expert who was not involved in this study. Also, the researchers did not declare any conflicts of interest. However, it is possible that the moderator (who is also the president of the medical association sponsoring the meeting) may have had some role in selecting or shaping this study presentation and so it is hard to know for sure how independent he is.
The researchers say it would be premature to recommend the use of anticoagulants in prostate cancer patients, because the benefits and risks have yet to be clearly identified… and so it is also premature to expect comparisons to alternatives.
The study looked at the use of available anticoagulant medications.
The story notes that this study builds on clues from clinical experience as well as lab and animal experiments suggesting there may be a relationship between blood clotting and how cancer spreads.
The story includes quotes from a news conference.