December 10th, 2010
Heart Failure Death Statistics: Don’t Believe What You Read on the Internet
Mary Knudson, Health Journalist
Well-known science journalist Mary Knudson is the author of HeartSense, a blog about heart failure, from which the following post is taken. In this post, she takes issue with the grim and outdated prognosis data presented to the public by a prominent heart failure website.
In its website section on heart failure facts, the Heart Failure Society of America directly addresses the question all people with heart failure and their loved ones desperately want to know:
“Q: What is the prognosis for a patient with heart failure?”
“A: Less than 50% of patients are living 5 years after their initial diagnosis, and less than 25% are alive at 10 years. Poor prognosis can be attributed to a limited understanding of how the heart weakens and insufficient private and government funding.”
I was startled to see those grim statistics on the HFSA website, given that clinical studies published in peer-reviewed journals have shown that ACE inhibitors and beta blockers prolong the lives of people with heart failure and that these medicines have become standard recommended therapy. Implanted defibrillators known as ICDs that prevent sudden death by shocking the heart when the heart goes into a chaotic rhythm; cardiac resynchronization therapy (CRT), which corrects abnormal beating of the left ventricle; and other effective treatments have also grown in use.
I also felt uneasy reading the HFSA answer that tied “poor prognosis” to “insufficient private and government funding.” That seemed to have a political tinge to it, out of place in an answer to worried patients and family members. Many conditions can cause heart failure, and many factors can contribute to a poor prognosis, including these: the doctor did not order the most effective medications that could have prevented progression; the patient didn’t correctly take appropriately prescribed medications because she couldn’t afford them or was unreliable; the patient didn’t observe a low-sodium, low-fat diet and get regular exercise; and, frequently, the patient has other significant health problems. Also, despite excellent care, the patient’s heart may be too damaged from a heart attack or a genetic malfunction to successfully pull out of heart failure. But I doubt any doctor ever tells a patient’s family, “Your husband and father is in late-stage heart failure and has only a few months to live because the government didn’t fund heart failure research.”
Working on the assumption that I could ask about the source of HFSA’s grim prognosis and get an answer, I contacted HFSA. I sent two e-mails to Cheryl Yano, HFSA’s longtime executive director, explaining that I was writing this report on heart failure death statistics but did not get a reply, so I called. She would not talk to me.
Loreen Anderza, the HFSA administrative assistant who answered the phone, said there is no specific source for the HFSA statement on how long people with heart failure can expect to live. It is “a consensus of experts in the field. They have no source for it,” she said, after putting me on hold while she spoke to Cheryl Yano. I asked if Ms. Yano would discuss whether heart failure is becoming a chronic condition that can be managed for most people, and Ms. Anderza said that Ms. Yano is not the right person to talk to because she is not an MD. I asked who at HFSA I could talk to, and she said Ms. Yano had no one to recommend. Ms. Anderza said that everyone uses the same numbers and suggested that I ask the American Heart Association if they know the source for the scary prognosis on the HFSA website.
Instead I contacted the president of HFSA, Barrie M. Massie, MD, Chief of the Cardiology Division at the San Francisco Veterans Affairs Medical Center, who responded in an e-mail:
“This is out of date. It is based on Framingham data and several trials, largely dating back 10 to 20 years.”
The Framingham Heart Study
The Framingham Heart Study supported by the National Heart Lung and Blood Institute, one of the National Institutes of Health, is an ongoing project begun in 1948 that has enrolled over 14,000 members of three generations and periodically issues reports on the risk factors for developing heart disease. The study has provided many important findings, including the risks of cigarette smoking, cholesterol, and high blood pressure, but it is designed to find information on all forms of heart disease, and its ability to track heart failure patients is quite limited. Original Framingham participants are seen at a clinical visit every two years, and their offspring are seen every four years. “Participants with heart failure often undergo treatment between clinic visits and before death, and these interventions are not captured in our clinic visits,” said Daniel Levy MD, director of the Framingham Heart Study, explaining the absence of information on treatments used by heart failure patients who died.
Many internet sites, including HFSA’s, base their prognoses on a Framingham heart failure study published October 31, 2002, in the New England Journal of Medicine. Even the American Heart Association’s Heart Disease and Stroke Statistics 2010 Update quotes the Framingham heart failure study death rates. I examined the framingham report on heart failure and found that the prognosis the study gives is based on a very small number of deaths.
The study followed 323 people who developed heart failure between 1990 and 1999. At 5 years, 86 of the 145 men (59%)and 80 of the 178 women (45%) were dead. The study did not determine whether these men and women died of heart failure or other causes, said Dr. Levy, the study’s lead author. The findings included data going back 15 or 20 years, and the study was conducted before ACE inhibitors and beta blockers, proven to prolong life in heart failure, were in use.
The authors also looked at deaths by decade going back to the 1950s. They note that there was an overall improvement of 12% per decade in survival rates after the onset of heart failure. In the decade since those data were reported, “there is optimistic evidence that we have improved treatment for people with heart failure,” Dr. Levy said in a telephone interview, though he would not estimate by how much.
Other Clues to Heart Failure Prognosis
In preparing this article, I talked to eight nationally known cardiologists to get a sense of where heart failure stands as a treatable versus a progressively fatal condition. Not all are quoted. One cardiologist who asked not to be identified because he knew what he was saying was “controversial,” commented on the annual AHA Heart Disease and Stroke Statistics Update: “These are not really current data. They are estimates extrapolated from the NHANES (the CDC’s National Health and Nutrition Examination Survey) . . . with changes based on changing size and age of the population. Hence, they are unlikely to be accurate and will not reflect real or measured changes. Consider them propaganda for those that thrive on high event rates. These data are useful for those seeking investment in development programs for heart failure treatment.”
NHANES annually surveys about 5,000 people in the U.S. and estimates results for the national population. The AHA Heart Disease and Stroke Statistics 2010 Update bases its estimated incidence of heart failure and prognosis of life expectancy largely on NHANES and the Framingham Heart Study of the 1990s.
One clue to how long people with heart failure live comes from clinical studies that try to prove a new drug or device is better than standard care at prolonging lives. Both Dr. Massie and Alice Macette, MD, chief of the National Heart Lung and Blood Institute’s Heart Failure and Arrhythmias Branch, point to improved life expectancy for the people in the placebo groups of these trials — those receiving the existing standard therapy against which the new treatment is being tested.
“For instance, in the SOLVD study of 1991, which first showed the benefit of ACE-inhibitor drugs, the 3-year survival rate was about 65% in the placebo group,” said Dr. Macette. And 3-year survival rates were approximately 80% (or greater) in two studies dealing with heart failure patients of varying degrees of severity (one on eplerenone and one on use of CRT for mild to moderate heart failure) reported this week at the American Heart Association. “In fact,” she said, “the improvement of heart failure outcomes has helped set the bar higher for any new therapy being tested.” Dr. Massie agreed. “If you compare the placebo groups over time, there is a substantial decline in the placebo group mortalities,” he said. “Used to be up to 20% per year and now is close to 8% per year. This low (death) event rate has made the conduct of clinical trials hugely expensive, which is why there are far fewer of these and even fewer positive ones.”
I also asked cardiologists to judge from their own experience how treatable heart failure has become. Edward K. Kasper, MD, Director of Clinical Cardiology at Johns Hopkins Hospital and a specialist in heart failure, (disclosure: Dr. Kasper was my co-author on Living Well with Heart Failure, the Misnamed, Misunderstood Condition) said “I expect most to improve with modern therapy for at least some period of time – say 75%.”
“Indeed there have been great advances and people do live longer, but progress has been slow and we need to do better,” said Dr. Massie.
I asked Mariell Jessup MD, chair of the American College of Cardiology/American Heart Association Guidelines for the Diagnosis and Management of Heart Failure in Adults this question:
“In your own experience, do you find that most people diagnosed with heart failure will be able to manage their condition, keeping it from advancing, or even improving it with the right treatments?”
“I agree,” she replied. She pointed to a study of 2,029 people in the general population in Olmsted County, Minnesota, who were classified according to how sick they were. Because this was a random sample, it included healthy people (stage 0). Stage A had risk factors for heart failure, stage B had cardiac structural or functional abnormalities found by testing but were not experiencing symptoms, stage C had symptoms of heart failure, and stage D had end-stage heart failure. Survival at 5 years was 99% in stage 0, 97% in stage A, 96% in stage B, 75% in stage C, but dropped to 20% in stage D, the smallest group with only 5 people. The study was published online March 12, 2007, in Circulation. “It is only those patients who present with intractable symptoms that do poorly,” Dr. Jessup said.
Needed: A Huge National Prospective Study or a National Registry
Although its numbers are small, the Minnesota study provides a window into more accurate prognosis for heart failure. But the only way doctors, patients, and families will get a really accurate picture of the prognosis with current therapies is from a huge prospective study or, at least, a national registry including many thousands of patients seen at academic centers and in the community by both cardiologists and general practitioners. The study or registry should include both sexes and a variety of races and ethnicities. Much could be learned from such a study, including:
- modern survival rates and deaths due to heart failure and not some other cause
- percentage of heart patients who experience sudden death due to ventricular fibrillation
- possible geographic differences in death rates
- treatment regimen up to and at time of death
- information on genetics of heart failure
Such a study or registry should have no funding from pharmaceutical companies.
Just before publishing this article, I checked the website of the Heart Failure Society of America. The unnecessarily scary prognosis for heart failure is still there, without even an asterisk explaining how old and outdated are the data on which it is based.
9 Responses to “Heart Failure Death Statistics: Don’t Believe What You Read on the Internet”
NEJM Journal Watch — Recent Cardiology Articles- Transcatheter Valve Replacement for Severe Tricuspid Regurgitation
- Time in Blood Pressure Target Range and Outcomes in Hypertension: Real-World Data
- When to Start Anticoagulation After Ischemic Stroke in Patients with Atrial Fibrillation
- Aortic-Valve Replacement vs. Clinical Surveillance for Asymptomatic Severe Aortic Stenosis
- Secondary Hypertension in Young Adults
I could not agree more. I now have to advise patients to be a little wary of the statistics quoted on many reputable websites. Although mortality for patients with intractable and unstable symptoms remains high, for many the annual mortality is much lower. I would hope that organisations such as Kaiser Permanente which collect a lot of data on patients electronically will be able to answer this question in time. In the UK more and more hospitals are routinely collecting data on all heart failure admissions and mortality data will follow from this.
So glad to hear you say this. Thank you for telling patients that much of the time heart failure is not what it sounds like and that life expectancy for people with heart failure is increasing. I recognize that mortality remains high for patients with intractable symptoms, and for some others there is the possibility of sudden death. But I wish all reputable websites would take the responsibility of publishing prognosis statistics they can gather from experienced cardiologists who see many heart failure patients or else just not address the subject at all. What is not right is to frighten newly diagnosed patients who are seeking information about their condition and read these terribly outdated death statistics.
I applaud what the UK hospitals are doing. I wish in the U.S. we could create a national registry or a huge prospective study in which information is collected uniformly and patients’ doctors would give out the information on cause of death for the registry rather than have a a third party try to ascertain from death certificates who died of heart failure. In addition to accurate death rates of people dying from heart failure itself and those experiencing sudden death, we could learn what medical regimens those who died were on and see if there is a pattern of people dying who didn’t get the best treatments. We not only want to know how many people die of heart failure. We want to know why they die. We could also gather a trove of genetic data that one day could be very helpful.
Thank you very much for commenting.
Mary
All death statistics are suspect as to the actual cause of death. As an intern, I once admitted a patient in a coma who died in the ICU within 12 hours. I was told that I could not use “coma of unknown cause” anywhere on the death certificate, because it then would automatically become a medical examiner’s case, and the burial would be delayed at least 3 days.
The larger issue here is that misinformation is rampant on the web. Of the two studies I came across recently, the accuracy, reliability and overall validity of most was horrifyingly low. I believe there is an unavoidable delay in translation of findings, and even more difficulty in getting the information out… I certainly do not fault the HFSA, since I am a member and fiercely loyal, but inadequate funding is likely responsible for lack of updates.
There is always a balance between divulging a truthful poor prognosis to patients and their families, and reluctance to overwhelm them with such despondency that it paralyzes and prevents partnership to help fight the illness.
My other point is that usually the media and reporting is woefully inadequate, largely lacking in perspective. So thanks, Mary, for your good work.
I remember talking to Mary Knudson some time ago. I made the point that there is no one mortality rate for heart failure patients. Patients hospitalized for their first episode of heart failure have a poor prognosis, especially if it is caused by a myocardial infarction, and the 50% 50-60% year mortality rates reported from the Framingham study in the late 1990s were astonishing. As noted by other commentators, these deaths often are not caused by heart failure or even related to cardiovascular disease. In a follow-up study from Framingham published in 2002, however, there was a these rates had declined by 31% and 32% in men and women, respectively. Most of this improvement occurred between 1980 and 2000 and probably reflected the impact of ACE inhibitor and beta-blocker therapy.
But numbers don’t tell the whole story. Framingham MA, a relatively homogenous, middle income town is not reflective of the United States. Nor is Olmstead County MN, mentioned by Ms. Knudson. By and large, residents of these communities have good access to health care. Another source of statistics is the data generated from large clinical trials. Compared to epidemiological studies, these patients uniformly have a much better prognosis. Why? They are carefully selected for the absence of other serious comorbid conditions and for their adherence to treatment. Their follow-up in the trial is frequent and rigorous. In these, mortality rates often fall below 5-8%/year, even in the control groups.
Despite what impresses me as a dramatic improvement in the outcomes of heart failure patients in the more than 30 years I have worked in this field, we still have a long way to go. Unfortunately, the attention focuses on the dramatic treatments that make it to TV—transplants, left ventricular assist devices, implantable defibrillators, stem cells. Yes, these work, but the costs are high, and the numbers of appropriate candidates for these are relatively small and will remain limited.
Hence, my plea would be that we shift our focus to prevention. Heart failure is relatively easy to prevent but will take a shift in our behavior. Early detection and effective treatment of hypertension can prevent up to 50% of call new heart failure cases. Prevention of heart attacks with changes in diet, life style, and smoking cessation will prevent another large number of heart failure cases. Changing diets can have a similar effect. A decrease of 1 to 3 grams of daily salt would save more lives than all of these high-tech interventions taken together.
Those of us who work in the field and keep up with the basic research advances in our understanding of the potential mechanisms causing heart failure and novel treatment targets are aware of the large number of potential therapies that have not been developed because the large pharmaceutical companies prefer to develop “me-too” drugs where the path to approval is more straightforward and the risk is less. Research is risky, but marketing works. In the end, they seem more interested in winning the marketing war than investing in “finding the cure”.
Using the usual sports adages, if we want to win the war on heart failure, we have to invest. We need to change behaviors, emphasize prevention, and build on the great success of the decades of neurohormonal directed interventions with novel and complementary approaches.
This should be the role of the NHLBI. It needs to invest on the next generation of novel therapies, perhaps in partnership with industry. It should use its prestige and resources to educate the public. As a recent seminal analysis by Bibbins-Domingo in the February 18 issue of the NEJM demonstrated, just a moderate reduction in salt intake (largely driven by intake of processed foods and fast food restaurants) would have a major impact on cardiovascular outcomes including heart failure. It would be much more cost-effective and save many more lives if we direct our research toward the prevention of heart failure or early diagnosis and modification of the process in its early stages. Wouldn’t this be a wiser way forward investing enormous resources on developing high tech interventions for the tip of the iceberg of patients with advanced heart failure?
Barry Massie
Competing interests pertaining specifically to this post, comment, or both:
None pertinent to this commentary
Thank you Richard Kones and Barry Massie for your thoughtful replies to my post about heart failure statistics. The reason I care so deeply about bad information on heart failure, especially outdated bad predictors of life expectancy, is that I have felt the fright it causes to read this bad information when you are affected and you don’t know enough to realize that it is bad information. I want to save others this experience.
You and I are in tune in trying to promote prevention of heart failure, Dr. Massie. As we start a new year I yesterday published the third in a series of posts on my HeartSense blog dealing with prevention. The first several posts are about exercise and then I will move to nutrition, starting with the deadliness of sodium. If any of you would like to contribute comments to the sodium post or any other heart failure prevention topic, will you please contact me? Also I would welcome a guest blog post on preventing heart failure or any one aspect of that.
As a journalist and as a person who had heart failure, you have my attention. You are the second significant leader in the field of heart disease, Dr. Massie, to tell me that heart failure would be easy to prevent. Daniel Levy, director of the Framingham Heart Study, said of heart failure: “in the majority of cases, it is preventable.”
But preventing heart failure must go beyond urging people to take personal responsibility for lifestyle changes. I would hope that the cardiology community will launch a take-no-prisoners campaign to drastically limit the sodium content in processed foods and restaurant and takeout food.
Thank you very much. I look forward to talking with you again.
Mary
There is an old adage that the way to a long life is to have a chronic disease and take care of it. Many people with CHF live for a long time, particularly if some of the underlying factors can be corrected. It is a very heterogeneous group so alarming web statistics should be taken with a “grain of salt”–so to speak.
These remarks are insightful indeed, and speak of several issues. Mary Knudson correctly points out that inaccurate older information, especially on sites of medical organizations should be updated, since they not only have professional functions and readers, but offer public information and hence operate in the public trust. The responsibility of such a site is significant. She further suggests that pooling data in a national registry would provide valuable current information.
The history of HF and coronary syndrome treatment, is instructive. In the days when I made house calls moonlighting as a resident, acute HF, then known as pulmonary edema, was treated with mercurial diuretics IV, rotating tourniquets, and morphine. Similarly, a confirmed MI was treated with a months’ bed rest as an inpatient, creating an open market on VTE! While this sounds embarrassing to me now, what will be said about our current treatments 75 years hence? Most likely, I suspect, equally barbaric and misdirected.
Thereafter, ventricular mechanics during ischemia and HF became better understood, with Braunwald, Sonnenblick, Parmley, Jay Cohn and many others as major contributors.
I brought up the extent of misinformation on the internet, which is a real concern. There are several forms: outright deception for sales, errors which are egregious in number or impact, significant ommissions, simple inaccuracies, or carelessness. However, given Google’s commercial orientation, the nonacademic entries usually get prioritized, leaving the extent of misinformation horrendous. I suspect there is a huge population that relies on web-based info for self-treatment of serious diseases. Put the two together and the potential for untold harm exists. There are patients buying colon cleanse kits to treat nearly every known codable illness, from acute appendicitis to MIs, to brain tumors, which can only result in tragedy.
Then there are investigational techniques. The unreliability of death certificate diagnoses, screening, the study populations, and run-in periods common in large clinical trials make one question the applicability of results, however generated, to real-life practice, and the ability to generalize. Tell someone nowadays that a meta-analysis is not gospel or law, and they recoil.
I could not agree more with Dr Massie, who himself has made significant contributions over the years. Prevention is a proven, cost-effective, and preferred method of handling chronic disease, but particularly CVD, where the incubation period occurs over decades, and the common causes in risk factors are widespread. Yet as Gerald Berenson MD (Bogolusa Heart Study) has said, prevention is a “hard sell.” Prevention trumps drugs, invasive procedures, devices, and surgery hands down in every way, yet continues to be ignored. It is not flashy in the sense that transplants, ICDs, LVADs, and stem cell research are, as Dr Massie mentioned above. We need to stop looking at the smoke and mirrors, but at the big picture, where the present inexorable increase in CV risk burden is leading. Flash, image, and marketing are more important in our society, not old stand-bys that may work better, at a fraction of the cost, without side effects, and be more humanitarian.
Less than 3% of our health budget is allocated to prevention. Cardiovascular prevention, despite the higher mortality rates from CVD, is actually afforded less attention for preventive monitoring by guidelines and grant committees than is cancer and HIV.
The benefits of prevention are not new, but have been known for over 30 years. However, if one simplistically asks the average person whether they would prefer to change their lifestyle or when faced with angina, have a stent implanted, they will overwhelmingly chose the stent. This choice tells the tale. Not only in health, but elsewhere, personal responsibility is on the wane, not on the upswing. Roots here are societal, not medical, although socially, wherever possible, it is medicalized as a denial mechanism for political and personal reasons.
Please see http://tak.sagepub.com/content/early/2010/12/21/1753944710391350.abstract, an abstract of an in-press personal “position statement” about prevention. My philosophy is also evident in three recent open-access papers, even though not concerning HF (1)http://www.dovepress.com/recent-advances-in-the-management-of-chronic-stable-angina-i-approach–peer-reviewed-article-VHRM (2)http://www.dovepress.com/recent-advances-in-the-management-of-chronic-stable-angina-ii-anti-isc-peer-reviewed-article-VHRM (3)http://www.dovepress.com/rosuvastatin-inflammation-c-reactive-protein-jupiter-and-primary-preve-peer-reviewed-article-DDDT.
Richard Kones
Wow! Mercurial diuretics, rotating tourniquets, and morphine! That is unbelievable.
But it is also unbelievable that we eat the foods we do, isn’t it. Another strong voice for prevention of heart disease. Thank you. I’m glad that CardioExchange exists to have these stimulating discussions. But the public needs to hear your voices. The public doesn’t hear you in this forum.
Will the community of cardiologists, perhaps in partnership with the American Medical Association which has money and lobbyists, please launch a public campaign to stress both personal and corporate responsibility for nutritional and other lifestyle changes? This would need to go beyond polite ads about benefits of eating right. Your campaign needs to squarely place sodium and trans fats in the same danger zone with cigarettes and house fires and drunk driving. Are you going to challenge food manufacturers and fast food chains and Congress? Are you going to collectively put to work the knowledge that you have from your training and your years of experience watching patients struggle with disease and die and go out there and change the eating habits of America and beyond?
Are you going to get angry about the state of our food and our eating habits and say so in full-page newspaper and magazine and internet statements and on talk shows and in congressional hearings and in posters placed in community barbershops and stores and churches and honkey tonk bars? Are you going to get pastors talking about personal and corporate responsibility to health and get singers singing about it and television programs writing it into their scripts?
What are you going to do, doctors?
Mary