How Widely Should Statins Be Used - Primary Prevention for All?
The medical community and public health experts are having a long-term discussion about how broadly to prescribe statin medications. The success of statins in high-risk patients has caused an interest in using them even more widely to prevent as many heart attacks as possible. The most enthusiastic promoters have proposed that everyone past age 30 (or even 20) be put on statins, regardless of cholesterol levels, cardiovascular health, or overall health. This enthusiasm is part of the drive behind the conception of the polypill, which would include a statin and be disturbed very widely, if not indiscriminately.
Should people in their 20s and 30s without big risks, be prescribed statins, and so begin a regimen that would presumably last for the rest of their lives? The proximate goal of this pharmacotherapy is reduction in overall cholesterol levels, but the ultimate goal of course is reduction in mortality due to heart disease and stroke.
Should people in their 60s with low blood pressure and low LDL-C levels get statins? Do the benefits outweigh the costs?
A metaphor you see a lot is the fruit-picking one. The low-hanging fruit are easy to pick off the tree because you don’t have to climb up to get them. The higher the fruit on the tree the harder they are to get. The cost-benefit calculation may work in favor of picking low-hanging fruit, but the cost to get the higher-hanging fruit is higher and the benefit from getting that fruit does not go up, so often would-be fruit pickers ignore the higher-hanging fruit.
In the medical world, the low-hanging fruit are patients that the cost-benefit estimate says will really benefit from therapy. A person with very high LDL-C levels or who has had a heart attack or embolism is low-hanging fruit. The benefits of treating this person with statins is high and worth the downsides of increased risk of diabetes or muscle pain. People at lower risk of heart disease are higher hanging fruit. The benefit from giving those people statins is lower.
Since their introduction decades ago, statins have been prescribed to a bigger and bigger population. We are climbing the tree and getting lower risk patients in the pool of people using statins. This means the average efficiency of statin treatment is declining. It must decline as we expand usage through lower risk patients. When usage increases, there is of mathematical necessity less benefit per prescription on average. Just because efficiency is declining is not a reason to stop the expansion of statin usage.
Epidemiologist have a metric called Number Needed to Treat (NNT). This is an estimate (an informed estimate) of how many patients must get a treatment to generate a certain benefit. For instance, a group of patients ages 30 to 50 with no significant health problems might have an NNT of 45 for the flu vaccine. That means for every 45 people in this group who gets the vaccine, one case of flu will be prevented. Sometimes complications from the flu can be fatal, so it is also possible to speak of an NNT to prevent a death. This number will be higher – let’s say 2000 for this group. For another group, let’s say ages 70 to 80 the NNT to prevent death is lower. Older people are more likely to die as a consequence of the flu. The NNT for this group may be 400. So the older group are the lower-hanging fruit here. The costs to administer the flu shot to each group is the same, but the derived benefit is higher for the older patients.
It’s the same for statin therapy to prevent cardiovascular disease. One way to divide patients is between those who have had specific CVD events and those who have not. If someone has experienced a stroke or heart attack, we say the statins are given for secondary prevention. The goal is to prevent another event. People who have never had such an event can be given statins as primary prevention. Primary prevention patients are the higher-hanging fruit. There is a benefit from treating them, but it is not as great on average as the benefit from treating people who have already had heart attacks.
The NNT is higher for primary prevention than for secondary prevention. (how much higher is hard to say, and estimates of NNT are always crouched in terms of patient populations and definition of benefits). Of course, dividing patients into those needing primary prevention and those needed secondary prevention is a simplification. There is a range of risks among patients, just as there are many different places on the tree.
When the authorities who promulgate guidelines made changes in the first two decades of the 21st Century they vastly increased the number of patients who would be eligible for statin therapy. This provoked a vigorous discussion about the appropriateness of giving medicine to so many people with no symptoms.
Tradeoff between Cardiovascular Risk and Metabolic Risk
Cost/benefit calculations are difficult in medicine in general, but they are particularly difficult when considering therapy that will go on for decades. Academics who look at clinical trials and observational studies of statin users generally agree that the drugs lower cholesterol and that the short-term side effects are not worth worrying about. (Doctors in clinical practice are more likely to worry about myopathy and memory loss than their ivory tower counterparts.) However, one big question about statins is whether they cause diabetes in the long run. If they do increase the risk for diabetes, the numbers are not clear (it is hard to disentangle the causes of diabetes, which are many, over the years.) Without a firm handle on the costs (which include the risk of side effects), the people who try to make broad recommendations for the large population have a hard time telling low-risk patients to use statins.
Progress in figuring stuff out
You might run into skeptics or skeptical writing that claims statins haven’t been studied enough to warrant wide use. Some of this writing may be old (check the publication date) because over the decades the types and numbers of people who have been studied has expanded considerably.
The earliest randomized controlled trials used to support approval from the regulatory agencies focused on high-risk patients. These were often male patients who had previously had heart attacks or strokes. The trials showed benefits but the skeptics said they were relevant only to the population studied: high-risk middle-aged men. The elderly, women, young adults, and people with no history of cardiovascular problems were not represented in the early studies so it was irresponsible to promote statins for those groups.
However, doctors prescribed statins for those people and over the years and many observational studies of tens of thousands of patients we have enough data on those groups to show statins work for those populations, too.
Studies of studies
Dozens of studies have been done on the efficacy and side effects of statins over the years, worldwide, and one way of aggregating the results with statistical methods results in a meta-study. The meta-studies, or meta-analyses, can include clinical trials and observational studies of different sizes, different protocols, and different measured outcomes. They require care in the analysis, but if done right, they can yield valuable insights.
The Cholesterol Treatment Trialists' Collaboration (CTTC) published a meta-analysis in prestigious journal Lancet in 2012 and provoked more debate. Looking at 22 statin vs. placebo studies and 5 studies on different levels of statins, the researchers found that statins reduced risk of adverse cardiovascular events in people of both sexes, different ages, various starting cholesterol levels, and whether or not the subject had experienced cardiovascular problems. This was not surprising to anyone who had been keeping up with the science, but it functioned as a capstone to recent findings confirming the benefits of statins.
The CTTC study found has reported that relative risk reduction is independent of the level of baseline risk; although subjects at higher risk obtain more absolute risk reduction, significant risk reduction still occurs when the 5-year risk of a major vascular event is 5% to 10%,.
The controversial part of their findings regarded low-risk individuals. For those with a projected 5-year risk of major vascular events above 10%, the benefits of reduction in LDL levels achievable from statin therapy was said to far exceed the downsides. The authors recommended vast expansion in use of statins.
Writing at the blog cardiobrief,org Dr David Neumann criticized the method used in the Lancet article.(see his thoughts at
http://cardiobrief.org/2012/05/27/guest-post-data-drugs-and-deception-a-true-story/)
The researchers tried to the answer to the wrong question. Neumann writes "this new information tells us little or nothing about the benefits someone might expect if they take a statin. Instead it tells us the average benefits among those who had a 40-point drop in LDL." It was reasoning that is orthogonal to the question that needed to be answered. There isn’t much dispute that lower LDL levels are beneficial, but not all patients get such a big drop. Some bodies respond better to statins and some people are better at adherence than others.
The Lancet report showed benefits exceeding risks, but critics pointed out that the researchers were playing with a stacked deck to some extent. Individuals invited to the studies analyzed were given a pretrial of statins. If the patients reported immediate side effects, they were not included in the main multi-year studies.
A previous meta-analysis of largely the same studies published in the Journal of the American Medical Association in 2010 “did not find evidence for the benefit of statin therapy on all-cause mortality in a high-risk primary prevention set-up”. The authors said "we observed that statin therapy for an average period of 3.7 years had no benefit on all-cause mortality in a high-risk primary prevention population."
Another meta-analysis by the Cochrane Collaboration specifically looked at statins for primary prevention. It concluded "all-cause mortality was reduced by statins as was combined fatal and non-fatal CVD endpoints" but that "there was evidence of selective reporting of outcomes, failure to report adverse events and inclusion of people with cardiovascular disease. Only limited evidence showed that primary prevention with statins may be cost effective and improve patient quality of life."
The Institute of Medicine (IOM) created a framework to measure public health initiatives. Each initiative is evaluated whether it is safe, effective, patient-centered, timely, efficient, and equitable. This is one method to judge whether the effort to expand statin usage is warranted.
What can we conclude?
The cost/benefit calculation is difficult because of the long time periods in question and because it appears that neither the costs nor the benefits are very strong. Minor blips in the data and twists in how the data is interpreted can tilt the balance in one direction or another.
The bang-for-the-buck from statins is small for low-risk individuals, and although the costs (monetary and side effects) are small, they may outweigh the benefits.
Skeptics say: there are many risk factors for heart attacks, and LDL levels are one, but by itself, LDL is a fairly weak predictor of problems. For people with those other risk factors, LDL reductions with statins are certainly worth it. For people without other risk factors, the benefit from LDL reduction is minimal.
Today’s guidelines recognize there are many risk factors and LDL-C is only one; they incorporate age, blood pressure, and more into the risk calculation.
NIH Research Matters: Who to Treat with Statins
BMJ Editorial: Statins for All?
USA Today: Heart experts debate who should take statins
