Friday Feedback: Fallout From Bad-News Statin Reports

A recent study found that many patients in Britain stopped taking statins after bouts of media coverage highlighting their adverse effects. The authors estimated that about 2,000 extra cardiovascular events would occur over 10 years in Britain as a result of those cessations.

We contacted cardiologists and preventive medicine experts via email to ask:

Do you think the media should be more restrained in reporting on research findings that question the safety of widely used drugs?

What could medical journals be doing differently to discourage sensational reporting of medical research?

Statins, with 5-year NNT values from 25 to upwards of 100, are often criticized for helping relatively few patients who take them. What is your response?

The participants this week are:

Steven E. Nissen, MD, chairman, cardiovascular medicine at Cleveland Clinic

James Underberg, MD, MS, clinical assistant professor of medicine, NYU Center for Prevention of Cardiovascular Disease and president-elect, National Lipid Association

Jennifer G. Robinson, MD, MPH, professor, epidemiology & medicine and director, Prevention Intervention Center at the University of Iowa

Michael Davidson, MD, clinical professor of medicine and director, preventive cardiology at The University of Chicago Medicine

Seth S. Martin, MD, MHS, associate director, Hopkins Lipid Clinic, Ciccarone Center for the Prevention of Heart Disease and assistant professor of medicine and cardiology at Johns Hopkins University School of Medicine

Paul D. Thompson, MD, chief of cardiology at Hartford Hospital in Connecticut.

John P. Erwin, III, MD, clinical professor and interim chair of internal medicine, Baylor Scott & White Health in Temple, Texas

Sensational Reporting

Nissen: The overly dramatic reporting of medical stories by the mass media is a major societal problem. There is often a failure to distinguish between legitimate, carefully conducted research, and poor quality observational studies. Unfortunately, scary or sensational findings are more likely to get attention ("if it bleeds, it leads"). This problem has been exacerbated by the instantaneous nature of contemporary news coverage with a rush to post stories before thoughtful physicians can offer a more temperate view. Responsible media need to be more careful. Statins seem to attract a cult of negative advocates who incessantly attack the value of these drugs. We need the media to communicate effectively about both the benefits and risks of therapies, which takes a commitment to careful, thoughtful, and balanced journalism.

Underberg: The media needs to be more restrained about reporting on research findings in general. Whether the story is positive or negative, the rush to sensationalize information that might not be clinically relevant, or may only be observational in nature, hence hypothesis generating at best, can often be misleading and have unintended consequences. This becomes more concerning when the topic is an intervention such as statins that prevents life threatening events. Often even if side effects discussed are an issue and knowledge of such may impact use, they may be tolerable based on the potential benefits of the questioned intervention. This balance is often not presented in an evenhanded way and may lead to unintended consequences.

Robinson: Reporting should include safety results reported in the top peer-reviewed journals. In the case of statins, we have a huge amount of evidence regarding their safety in a broad population of adults. There are a large number of randomized clinical trials with long-term follow-up that reveal no important safety signals of excess serious harm that would outweigh the benefit of preventing a heart attack, stroke or death. Encourage the reporters to have the patience to talk to a number of experts to get a more complete picture of the pros & cons (which there are for about anything).

Davidson: Media too often focuses on the "man bites dog" type of sensational stories that are contrary to conventional messages. If there is a legitimate study that questions the value of therapy of a subset of the population the media has a obligation to point out the patients in which a benefit has also been well established. Statins have tremendous data supporting the clinical value of reducing CHD events in patients at high risk and this message needs to be reinforced by the media rather than trying to defy conventional wisdom with a sensational stories. As this new data points out, this type of reporting has ramifications and the journalists should be held accountable.

Martin: When writing headlines and stories, I'd kindly ask the media to be mindful of the power that they have to influence patient behavior and the health of our society. As much as possible, please cover the most scientifically rigorous and credible studies so that patients are empowered with great data rather than flawed, misleading data. When reporting on the safety of widely used drugs, please ask: 1) How strong is the evidence that the finding is real? 2) How common does it seem to be? 3) Who is the finding relevant to? 4) How serious is it in comparison to benefits? 5) Is it reversible?

Thompson: There is no way that organized medicine will be able to restrain public reporting of the possible deleterious effects of widely used drugs. The public media has to sell papers and magazines; the public has a distrust of standard medicine so buys and reads the bad news. The best we can achieve is ensuring that putative experts provide a careful and balanced perspective on both the risks and benefits of these drugs. Many "experts" have not been blameless in providing misguided and incorrect opinions. I often wonder how many folks suffered cardiac events because they were not treated with ezetimibe, for example, when their LDL remained high despite statin treatment, because some "experts" doubted its efficacy because there was no definitive clinical trial. Absence of proof does not equal proof of absence.

Erwin: The media has an important role to perform in keeping people up to date with objective reporting of new information. That said, words are very powerful and it is inherent that reporting be objective and keep in balance all other data that has been collected on a particular subject to prevent harmful responses that may occur due to an article. Unfortunately, too many people read the bold headline and possibly the first few sentences of a report and may never see the objectivity that may come in the middle or end of an article. I think there needs to be some further training/certification in headline writing and for medical writing for media members who report on health issues. Perhaps new ethical standards should be proscribed for headlines and the introductory statements in articles that have the potential effect of altering people's health habits?

Role of Medical Journals

Martin: First off, it all comes back to ensuring that credible research is accepted for publication and that the conclusions are supported by the data. Once a study is ready for consumption by the public, some journals seem to do a better job than others interfacing with the media. I think journals can learn from each other on how to promote accurate reporting of science.

Davidson: If a controversial finding may have an adverse effect on the general population due to widespread media coverage it is imperative that medical journals provide either a balanced editorial by a respected expert or make sure the tone of the manuscript does not lead to unwarranted conclusions that can be taken out of context.

Erwin: I do not think that medical journals have any jurisdiction over reporters. The intended audience is the clinician and the scientist. People should be careful to get their health information from vetted sources and from their own healthcare teams that take care of them. I try to redirect my patients to credible sources that give better overviews of particular medical conditions and therapies.

Thompson: I also do not think medical journals can or will do much to discourage sensational reporting. Sensational reporting is also good for the medical journal business and anyone who doubts it's a business only needs to watch the proliferation of medical journals. I have made my reputation studying statin associated muscle side effects, but I think they are lifesaving drugs that have revolutionized cardiac care.

Underberg: Medical journals are often under the same pressures that the media is under to drive interest publicity about articles published. The utilization of a fair and unbiased editorial process with balanced counterpoint by accompanying editorials are important tools that can be utilized to assure a fair presentation of data and more importantly interpretation of that data. Despite these measures stories are all to often reported by others with agenda beyond the control of the journals.

Stats on Statins

Robinson: In the 2013 ACC/AHA cholesterol guideline we recommended statins for groups of patients where there was a clear margin of benefit [clinical cardiovascular disease, genetic high LDL-C (>190 mg/dl), or diabetes]. When the margin of benefit may be less clear, such as in primary prevention patients with >7.5% 10-year ASCVD risk, we encouraged shared decision-making by the patient and the clinician. Studies have shown that patients on average think an NNT of 30 would be reasonable for a preventive medication, while for physicians a reasonable NNT is 50. Of course the time horizon matters as well.

Underberg: Statins should be used in those who will benefit most from them, with the least likelihood of side effects and even then, a shared decision process between the healthcare practitioner and the patient will allow for an informed decision process that makes the most sense for every individual. We still take care of patients one at a time, and this means the use of medications needs to be addressed with personalized approach. What works for one patient may not always be true for others. When looking at the totality of evidence statins lower LDL-cholesterol and most importantly they reduce cardiovascular events in patients at risk for cardiovascular disease. Assessing risk prior to use, and a complete discussion of potential side effects is intrinsic to any pharmacologic intervention in the shared decision making paradigm.

Davidson: The statin trials were only 2-5 years in duration and yet there were major clinical benefits demonstrated. There are very few therapies that can reverse many years of exposure to a causal risk factor such as elevated LDL-C. Once a lifetime of benefits can be factored in as well, statins have a much more favorable number needed to treat.

Martin: The NNT will tend to decrease over a longer time horizon. The typical clinical trial duration of 3-5 years for statin therapy is considerably shorter than the anticipated treatment period of decades for many patients. Therefore, the often-quoted NNT values are not reflective of the expected NNT in clinical practice.

Erwin: Statins have clearly positively impacted the trajectory of coronary artery disease in both the primary and secondary prevention arenas. It is important, however, to treat each case separately and use these medications only in the people who are going to obtain the most benefit from them.

Thompson: The criticism that they help relatively few subjects in 5-year studies is misguided. Who wants to live only 5 years? Five years studies are only proof of concept studies and the proved concept is that statins prevent cardiovascular events. I know few cardiologists not on these drugs. Finally, who can say that stopping statins is such a terrible thing? The West of Scotland follow up demonstrated that only 5 years of statin treatment provided benefit that extended for 20 years after the study, and coronary regression studies show that much of the regression of atherosclerosis occurs in the first two years of statin treatment. Consequently, estimates of the risk of stopping statins are simply estimates because we have not actually studied that risk.

Comment