Read Original Story

STAT’s cheerleading coverage of prescription fish oil capsule lacks crucial context

Amarin fish oil capsule shows dramatic benefit for cardiovascular patients, potentially upending market

Our Review Summary

A 25% reduction in cardiovascular events, including death, heart attack or stroke sounds like a big deal… and maybe it is. But this story about the results of a trial of a prescription drug derived from fish oil (brand name Vascepa) relies on a news release and company statements that don’t provide basic information about the precise number or type of cardiovascular events observed by researchers. Without absolute numbers, it is impossible to judge the importance of these results. It is also premature to claim that the results represent “heart-medicine history,” especially given that the findings haven’t been peer-reviewed and are at odds with numerous previous studies.

The story takes a cheerleading tone and fails to supply key context. For example, it highlights a dramatic quote from a researcher without noting that he was paid over $120,000 by the drug company, including payments for speeches and other appearances on its behalf, in 2017. The story headline and text do make clear that only patients with cardiovascular disease and high levels of triglycerides or diabetes were in the trial, so it should be clear to readers that this drug was not shown to prevent disease in healthy people. The story does not give any details about potential harms of the drug. It reports key details of the trial, but without access to the full trial report, it is impossible to judge the quality of the evidence or its real significance.


Why This Matters

This story is based on the news release claim of a 25% reduction in the risk of major cardiovascular events. But what does that mean? It’s difficult to say, because that outcome consists of a combination of cardiovascular disease death, heart attack, stroke, coronary revascularization procedures (placing a stent in a blocked artery) and hospitalization for unstable angina (chest pain). Based on what little we know, it is possible that the 25% reduction was driven mainly by a lower rate of hospitalizations for chest pain — an outcome that’s less important and also more subject to bias on the part of investigators who decide which patients get hospitalized and which ones don’t. It’s possible there was little or no reduction in deaths, heart attacks or strokes. Until the full trial results are released, we simply don’t know.


Does the story adequately discuss the costs of the intervention?


The story reports that Vascepa costs about $2,400 per year. However, that price is much lower than indicated by online searches (here and here) that show prices around $3,400 per year at the 4 gram per day dose used in this study.

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

Not Satisfactory

The story repeats the relative risk reduction of major cardiovascular events of 25% that the company provided. But it doesn’t grapple with the vital question: 25% of what? The true size of the benefit depends on the baseline risk of these participants, which is unknown and probably won’t be known until the results are presented at a conference in November. (Read our primer on absolute risk for more information on why this is important.)

The story does note that the company is withholding details about the study results for now, but readers should have been clearly alerted to the fact that without knowing how many events occurred or what they were, it is impossible to judge the importance of the results. The study outcome was a combination of cardiovascular disease death, heart attack, stroke, coronary revascularization procedures and hospitalization for unstable angina. The story would have been better if it had alerted readers that based on the sketchy information released by the company, it is impossible to say whether there was an important reduction in deaths, heart attacks or strokes. It could be that much or even all of the reduction was in the number of times patients were hospitalized for chest pain. (Read our primer Why you should be careful with composite outcomes in clinical trials for more information on why this is important.)

Does the story adequately explain/quantify the harms of the intervention?

Not Satisfactory

Although the story says the company reported a “side effect profile consistent with the drug’s currently approval label”, it did not tell readers what that means. The most important adverse event reported in other studies of Vascepa is joint pain, affecting up to about one in 25 patients. The story does not mention warnings about possible allergic reactions in people allergic to seafood or interactions with anti-clotting drugs, which many heart patients take.

Does the story seem to grasp the quality of the evidence?

Not Satisfactory

It is difficult to rate the story on this criterion. The story reports key features of the REDUCE-IT trial, noting that it included 8,179 patients, compared Vascepa treatment to placebo, and followed patients for a median duration of almost five years. However, since the company is withholding key details of the trial, including critical information about the absolute numbers of cardiovascular events, adverse events and any limitations of the study, it is impossible to judge the quality of the evidence. The story does quote sources who say they are waiting to see detailed data before reaching a firm judgment about the results, but it doesn’t clarify what’s missing. And these faint notes of caution are buried under a mountain of premature praise about the study’s importance and significance.

Does the story commit disease-mongering?


The headline notes that this trial included only people with cardiovascular disease. The story reports that the patients were taking statins to reduce their cholesterol, but they still had high triglyceride levels or diabetes (which increases the risk of heart attacks and strokes).

Does the story use independent sources and identify conflicts of interest?

Not Satisfactory

The story includes comments from experts who were not involved in this trial and it identified others as being researchers for this trial or working for Amarin. However, when the first quote calls this drug “absolutely the most significant study in the field of cardiovascular risk reduction since the statins were introduced”, it is not enough to identify the source as “a Vascepa study investigator” when that researcher, Matthew Budoff, received over $120,000 in 2017 from Amarin, including payments of more than $4000 per appearance for speaking and teaching engagements on behalf of the company.

[Note: The text has been corrected to indicate that the quoted source’s name is Matthew Budoff, not Martin Budoff as it initially appeared.]

Does the story compare the new approach with existing alternatives?


The story reports that another fish oil derivative, Lovaza, did not demonstrate benefits in trial results released in August. It also mentions another similar drug, Epanova, that is currently being studied.

The story compares the results of the trial of Vascepa to those from much costlier cholesterol drugs known as PCSK9 inhibitors, but it’s hard to know whether the comparison is useful given the unknowns about the what sort and how many cardiovascular events occurred in patients taking Vascepa.

Does the story establish the availability of the treatment/test/product/procedure?


The story reports that Vascepa is available by prescription to treat high triglyceride levels.

Does the story establish the true novelty of the approach?


The story reports that this is the first trial to show a reduction in major cardiovascular events among patients taking this type of fish oil derivative.

Does the story appear to rely solely or largely on a news release?

Not Satisfactory

The story relies on a news release from the company that produces Vascepa. Although some cautionary comments and quotes are included, this story is not really about the results of a trial of Vascepa (which the company has yet to fully release), but only about a news release and company statements that make claims about trial results. This case illustrates how it is impossible to do a good job reporting on a drug trial when the sponsors do not release critical details of the trial.

Total Score: 5 of 10 Satisfactory

Comments (7)

Please note, comments are no longer published through this website. All previously made comments are still archived and available for viewing through select posts.

Gary Deyana

September 26, 2018 at 12:46 pm

The skepticism here is reflective of a serious absence of understanding regarding statistical matters for clinical trials. The result on the primary endpoint here is so overwhelming convincing, and the powering of the trial so strong, that even if certain “hard MACE” endpoints don’t show the full 25% risk reduction, they will still carry low p-values — certainly lower than p < .05, and likely much lower.

This kind of sour grapes approach makes one question the motives in play here.


    Kevin Lomangino

    September 26, 2018 at 2:50 pm


    We hope just as much as anyone that the drug is effective for preventing serious cardiovascular events. But until we receive corroboration of that effect, we think reporters should refrain from parroting dramatic claims from company PR. That’s not sour grapes — it’s just good journalism.

    Kevin Lomangino
    Managing Editor


glenn dzidowski

September 26, 2018 at 1:15 pm

Amarin is not allowed to give more detailed information because it is subject to an
embargo until their presentation to the American Health Association on Nov. 10. I
can appreciate how frustrating that is as I have waited for years to see this medically
valuable information. Keep in mind that Vascepa has been prescribed for over six years and is proven to be very safe with no negative drug interactions including blood
thinners. The trial itself was conducted with patients with high triglyceride levels who
were already on statins and best standard of care. Amarin did say that many of the
secondairy endpoints as well as primairy endpoints were robust! (We’ll find out exactly
how robust in November.) This study was not conducted to see how long it could
keep healthy well well, it was conducted to find out if those patients with advanced
CVD would benefit from Vascepa beyond statins or best standard of care. When the
embargo is lifted I think the data will confirm that Vascepa will lead to a new paradigm
in health care for those patients with CVD. Understand that Vascepa is pure EPA (it
does not contain DHA which may cause AFIB or detract from the anti-inflammatory
benefits of EPA.) Vascepa was dosed at 4 grams per day (4 1gram pills) which you
will not get from supplements or from fish. Also, fish and fish oil supplements may
become rancid and then do more harm than good. Vascepa, having no DHA does
not smell like fish nor will you get fishy burps like you might with Lovasa. The Epanova trial may not fare well as that medication contains DHA and has an especially
“messy” side effect. (Though it might make a good laxative.) I do hope you report
again on Vascepa once the embargo is lifted. Good article.


    Kevin Lomangino

    September 26, 2018 at 2:45 pm


    We’re well aware of the rules surrounding embargos and have written about them extensively. The issue here is what journalists should do when they can’t access the full study data to corroborate a company’s claims about results. In our view, to simply parrot those claims without sufficient context is acting in a promotional/public relations capacity and not in a journalistic capacity. That’s the substance of our critique.

    Best regards,

    Kevin Lomangino
    Managing Editor


Win Tang

September 27, 2018 at 7:45 am

I feel the same way as the article when I heard about the primary endpoint and the 25% reduction for the 8,000 pt. 5/10 score is actually not bad… this is science so we’ll wait for the conclusions and data before all doctors start prescribing and all the pts start using this with their statins.


Andrew Holtz

September 27, 2018 at 12:29 pm

The motive at play in this review is highlighting the importance of absolute numbers to understanding what this trial means to patients. Is a 25% reduction in relative risk a welcome advance or a phenomenal advance? We don’t know. Trumpeting relative risk reduction without knowing the absolute numbers is like reporting the score of a football game, but only for one team. Stanford scored 74 points! Hurray? The fan reaction depends on whether Cal scored 25, 73 or 75. What’s more, since the company didn’t break down the relative risk reduction by specific events, we don’t even know yet which teams were playing. If there were a lot of deaths, heart attacks or strokes… and an important reduction in them… that will be tremendous news for patients with similar profiles. But the primary endpoint also included revascularization procedures and hospitalizations for unstable angina. If it turns out that most of the reduction was in hospitalizations… that’s good news, but not of the magnitude of a reduction in deaths. Again, the point is that the information the company has yet to release makes all the difference in what this trial could mean to patients. Will this drug benefit millions of people? Will it be prescribed to millions, but offer only a modest benefit to a relative few of them? I don’t know. Until we see the absolute numbers, all we can say for sure is that the company is excited.


Joshua Durham DO

October 10, 2018 at 4:00 pm

I want to know the absolute risk reduction. The relative risk reduction is a statistical manipulation. The absolute risk reduction of statins is very poor. No “cholesterol” med has good absolute and relative risk reductions. When a placebo group has 0.6 events and the group has 0.3 they say it has a relative risk reduction of 50%. That makes everyone want to go take it. But it is just a statistical manipulation. Absolute risk shows these medications are not helpful.