The research questions are important. The scope of the study is large. The funders are serious. And the journal is one of the best read in the field. So why does this news release give reporters — and the rest of the world — so little in the way of details about how and whether this study is important? No quantification of benefits. No mention of harms. No mention of costs and no comparison to alternatives. The release started with a useful explanation of the study’s design, but neglected to mention this study was an extended observation of a small subgroup of patients studied in an earlier trial. The initial trial covered 4.7 years, during which no benefit was evident; this study followed a subset of patients for another 5 years.
The use of cholesterol lowering medicines of the “statin” class is widely regarded as the first-line therapy to prevent heart disease in patients with prior heart attacks or at increased risk of having one. However, many patients cannot tolerate statins due to side effects and while beneficial, these medicines do not entirely eliminate the risk. This study examined whether adding fenofibrate to a statin could reduce the risk of heart events more than just the statin alone.
The results essentially confirm a previous report from these same researchers showing that in a subgroup of patients with high levels of fat in the blood (triglycerides) and low “good” (HDL) cholesterol levels, fenofibrate may have been beneficial. But the overall results of the original study were negative — something this release never touches upon. And the release never cautions readers as to the many limitations of this subgroup analysis, particularly the fact that the findings in this small group might reflect factors other than the effects of the medication. The release hints at this concern when it says “a randomized study is needed to confirm these findings,” but a more direct acknowledgment was needed.
The drug being studied, fenofibrate, goes by the brand name Tricor. Its costs are well known — about $100 for 90 pills — and could easily have been dropped into the release, making it clear how much it would cost for a typical person to replicate the regimen necessary to see any of the benefits seen in the study.
To actually see what the benefits were, one would have to read the full study in JAMA Cardiology. While we always recommend that reporters read the actual study that is the subject of their story, we live in an age where releases like this quickly find their way into the news cycle without the journal ever being consulted. So one of the most basic elements of any good release is a quantification of the benefits found in the study it’s based on. Instead, the release offers vague language like this: “The findings suggest that fenofibrate therapy may be beneficial in the way the researchers hoped: by reducing cardiovascular events in patients with type 2 diabetes who take statins but still have especially high triglycerides levels and low HDL cholesterol levels.” Even the abstract, a much abbreviated summary of the study, provides data that could have easily been used in the news release.
All drugs come with some kind of risk profile. And, in this case, we are talking about treating patients with diabetes who, presumably, already are managing that disease with prescription medication. At a minimum, the release should mention any side effects or harms observed in the study.
We like the format that the release used, beginning with the question that the researchers were trying to answer and then explaining in clear terms what the researchers did to answer that question. But the release focuses on a small subgroup of the much larger, initial trial. First, the study was a follow-up of a previous report that overall showed that this medicine did not provide additional benefit. The only benefit was seen in this small subgroup, both in the original study and this one that followed patients for a longer period. The second point was that only 4.3% of patients randomized initially to fenofibrate continued to take it after the randomized part of the study was over. This may suggest that the benefit in this subgroup may be due to factors other than the proposed benefit of the medicine. The release should have included these important distinctions, which the study itself calls ample attention to:
“It is also important to note that these prespecified subgroup analyses can only be considered hypothesis-generating and in some cases are based on a relatively small number of events.”
There is no disease mongering in this release. But there’s very little context about the disease either.
The release makes the funding sources for the study clear.
The release makes no mention of alternatives. Not only are there many statins on the market, but there are other approaches to managing high cholesterol levels and triglycerides. While statins are clearly the standard of care for patients with specific characteristics, the release should at least make mention of alternatives.There have been a number of medicines used in patients with diabetes to examine their role in decreasing cardiovascular events. The fact that there are many alternatives — from changes in diet and exercise to supplements (niacin) and medications (such as those derived from Omega-3 fish oils and fibrate drugs which lower triglycerides) — should have been mentioned.
The availability of the drug is not clear in the release. Although this is a previously approved drug that can be prescribed by doctors, some insurers may not cover it or cover it with higher co-payments. This could have been mentioned in the release.
The release does not establish the novelty of the findings and really can’t do so because there are so few details in the release. As noted above, this is a follow-up report that evaluated longer term outcomes. The reported results are the same as the initial. While longer follow-up is important, the prior information about this drug and this study could have easily been added.
We do not believe there is any unjustifiable language in the release.