Dial-a-headline on NSAIDs & cardiac disease gives us a heart attack

What are we supposed to do with headlines like these? Depending on which headline you came across your take home-home message — and level of concern — could have swung 180 degrees.

Especially when most of us have taken nonsteroidal anti-inflammatory drugs (NSAIDs) like Aleve (naproxen), Voltaren (diclofenac), Celebrex (celecoxib), or Motrin or Advil (ibuprofen) at some point in our life.

Heart disease is the leading cause of death in this country. And, on any given day, over 30 million Americans take NSAIDs for a variety of acute and chronic problems. So suggesting a link between the two is bound to trigger of flood of news coverage.

Although these headlines may sell papers, or get clicks, they will leave many of us confused–even terrified–and asking our doctors: “Hey doc! I saw that my painkillers could kill me … should I stop taking them?”

Before you rush off to your doctor, or abruptly stop taking your pills, you need to know that the BMJ study cited was an observational study–more specifically, a meta-analysis of four previously published studies designed to look for trends–so, given that study design, here’s the critical information that wasn’t in the headlines:

• this study can not show cause and effect, i.e., that NSAIDs cause heart attacks
• this study did NOT include aspirin (also an NSAID), which in low doses has been shown to lower the risk of heart attack
• according to the authors of the study, the absolute risk of heart attacks associated with NSAIDs is actually quite small: about 1 percent a year

But were these three bullet points included in most of the news coverage? No. (Kudos to CBS News who did include them!)

Time certainly did not. A USA Today video overtly fear-mongered with “a heart attack will likely take place within the first month” of taking ibuprofen. And the New York Times published a paltry 237-word piece in their “Well” section — entitled, Pain Relievers Tied to Immediate Heart Risks — that only did well in enraging its readers:

Most of the coverage we saw latched on to the relative risk numbers and implied that people who took NSAIDs would have a “20 to 50 percent increased risk of having a heart attack.” But it was HealthDay — with their more cautious “slight rise in heart attack risk” headline — who got the researchers to express this in more useful, absolute terms:

“For most people, however, this represents only a small increased risk — about 1 percent a year.”

This is in stark contrast to the sensationalistic Guardian headline implying a “100%” increase in risk of a heart attack. This was a strange percentage to highlight since it refers to the relative increase in heart attacks associated with rofecoxib (Vioxx), a Merck drug pulled off the US market in 2004 because of an increased risk of heart attacks and strokes after prolonged use at high dosages.

It seems almost paradoxical that the Guardian would allow such a misleading headline and then turn around and include one of the most important limitations of this study:

“But the lack of absolute risks of heart attack – for people using NSAIDs and those who are not – in the paper, and the fact that the researchers were unable to exclude other possible influencing factors, led some independent commentators to conclude that it was difficult to assess its significance.”

Cardiologist Dr. Deepak Bhatt of Brigham and Women’s Hospital, was one of those independent commentators. He told CBS:

“Someone who is young and healthy and takes an NSAID for a sprained ankle, the risk of heart attack is already very low so it raises the risk from very low to very low … on the flip side, an 80-year-old who has high blood pressure, who’s had bypass surgery, for that patient, the risk could be substantial.”

What the study does suggest is that the risk of heart attack associated with using NSAIDS — although quite small — increased with increasing dosages, and was most pronounced during the first month of taking the medications.

Because the headlines generated by this study varied so wildly in both accuracy and level of hype, it’s likely doctors will be bombarded with questions about NSAIDs. So I turned to Dr. Adam Cifu who is an internist at the University of Chicago with a special interest in evidence-based medicine. What advice would he have for patients who might be concerned about the relationship between NSAIDs and heart disease?

“We have known that using NSAIDs slightly increases the risk for cardiovascular events (mainly heart attacks). This is probably true in all patients, but really not important for people without a history of heart disease. In people with heart disease the increased risk is still small, but high enough to try to limit the exposure to these meds, and only use them at low dosages and for short periods of time. These drugs also seem to reduce the beneficial effect of aspirin. So people with cardiovascular disease, who take aspirin, should keep taking their aspirin and try to take their aspirin a couple of hours before the NSAID.”

Maybe the headline should have been this: “BMJ study regarding painkillers and heart disease adds little new information.” But who would click on that?

---

[UPDATE: May 16, 2017]

We have some sharp readers — like Dr. Mirjam Jenny, a research scientist at the Max Planck Institute for Human Development in Berlin — who brought our attention to this:

https://medium.com/wintoncentre/is-the-media-now-giving-scientists-lessons-in-research-integrity-51edcd970adb

Seems several people were confused, as we were, by the lack of absolute risk numbers in the published BMJ paper. Likewise, it was hard to find the source for the authors’ subsequent reference in interviews and a news release to a mysterious (?absolute risk) number of “1 percent a year.”

Michael Freeman and Michael Blastland — both of the Winton Centre for Risk & Evidence Communication at the University of Cambridge — ask the following:

“Are they seriously saying that every year 1 in every 100 people who take NSAIDs will have a heart attack because of them? Or 1 percent of people who take them for a year?”

In not publishing the absolute risk numbers associated with this study, the BMJ has gone against it’s own policy. Even if they have a good reason for not doing so, it seems that with a study that generated as much global news coverage as this one did, they have an obligation to follow-up with a clarification.