Dr Dingle's Blog / data
Tens of millions of unsuspecting people are prescribed cholesterol-lowering drugs—statins like Pravachol®, Zocor® and Lipitor®—each year at a cost of more than billions of pounds with very little, if any, benefit. In the U.S., some 40 million people currently take statins at a cost of more than $3.00 per pill, more than $1,000 per year, totaling more than $40 billion a year.
While there are many exaggerated claims and a lot of hype about the benefits of statins, there are also many studies showing no benefits at all. The pro-statin hype is based on the misuse and abuse of statistics. Various independent studies in prestigious, peer-reviewed journals have shown that statin use in primary prevention—that is, to save lives—has minimal or no value in reducing mortality and certainly nothing that is considered anywhere near clinically significant to warrant their widespread use. It does not matter how one manipulates the statistics, the results just aren’t there. In data gathered in 2009 from six trials, a review of the efficacy in lowering the risk of death with statins found virtually no difference between the treatment group and the control group.[i] There are many more of these studies.
In an independent meta-analysis (when a number of studies are put together to achieve more statistical power) of randomized controlled trials in patients without CVD, statin therapy decreased the incidence of major coronary and cerebrovascular events and revascularizations but not coronary heart disease or overall mortality.[ii] Taking statins for a number of years will not reduce mortality: “Primary prevention with statins provides only small and clinically hardly relevant improvement of cardiovascular morbidity/mortality.”[iii] “Hardly relevant” means there is virtually no clinical benefit; as the authors of these particular studies are independent, they gain nothing by stating this. Another review found that “current clinical evidence does not demonstrate that titrating lipid therapy (trying to lower cholesterol with statins) to achieve proposed low LDL cholesterol levels is beneficial or safe.”[iv] In other words, lowering lipids has no real benefit and has the potential for adverse effects. Following up on this, in a major independent review of studies funded by the Ministry of Health of British Columbia (Canada) on statins and primary prevention, researchers reported that “statins have not been shown to provide an overall health benefit in primary prevention trials.”[v] This is a government report carried out by an independent university yet its findings are still ignored.
The problem really comes down to vested interests and the abuse of statistics. To overcome the limitations of small studies, vested parties combine many studies into a meta-analysis. The researchers themselves select the studies used in the meta-analysis. A fundamental problem is that researchers with direct links to drug companies have the authority to select the most positive studies and ignore the rest—including independent studies not funded by pharmaceutical companies. Despite this, they have still not been able to show any clinically significant findings.
With few new studies being conducted, the drug company researchers look at the data every year or so and conduct another meta-analysis to show virtually the same results as the earlier studies. This makes it appear that another big study has been conducted, but frequently the data is simply reused from older studies. There is very little clinical significance in these studies, yet the media outlets pick up the drug companies’ PR and repeat, word for word, the pharmaceutical industry’s party line.
As readers of the scientific journals, we should not be confused between statistical significance and clinical significance. For an outcome to be “statistically significant” means that the outcome was likely result of the treatment—whether the result was 100% effective or less than 0.1% effective. That is, if you treat 1,000 people to save one life (0.1%) it may be statistically significant but it is not clinically significant. “Clinical significance” means 20% to 30% or more. The drug companies’ most positive studies on statins for prevention of CVD report statistical significance, mostly 1% or less, and none have found any clinical significance.
Busy medical professionals don’t have time to review the statistics; few of them may be aware of the different ways the statistics are manipulated. A study found that a high number of psychiatrists did not understand the difference between statistical and clinical significance.[vi] Consultants, that is the senior psychiatrists, did less well than senior trainees. So if the experienced professionals don’t understand the results of these studies, how do we expect the media or public to understand? The entire situation is also complicated by the comments in abstracts of the studies, in which researchers state the evidence is significant and don’t mention that it is only statistically significant but not clinically significant. The drug companies fund the researchers and authors of most of these studies.
The studies on statins also report “relative risk,” not “absolute risk” or “real risk.” The relative risk reduction is highly misleading[vii],[viii],[ix],[x].[xi] if not deceptive. An example of relative risk is: if you have four people in a study who die in the placebo group (no drug) compared to three people who die in the drug treatment group—that is, four were expected to die but with the drug only three did—then there is a 25% relative risk reduction. However, to get this effect of saving one life you would have to treat 1,000 people and the real risk reduction is 0.1%. Relative risk is like adding 1+1 to get 11 or 2+5 to get 25 or more. How can the pharmaceutical companies and the researchers working for them get away with this? This is probably because (at least in my experience) most people are afraid of statistics.
In studies by the Medical Research Council dating back to the late 1980s, researchers found that of 1,000 men ranging in age from 35 to 64 who received treatment for mild hypertension over five years, there were six fewer strokes and two fewer cardiovascular events than would be expected.[xii],[xiii] The real risk reduction over five years was 0.9%. Ten years later, a study of Pravachol® was released in the media, with much fanfare, as having a 22% drop (relative risk, not real risk) in mortality. However, when one looks at the numbers and statistics behind the calculations, treating 1,000 middle-aged men who had hypercholesterolemia (high cholesterol) and no evidence of a previous heart attack with pravastatin for five years resulted in seven fewer deaths from cardiovascular causes, and two fewer deaths from other causes than would be expected in the absence of treatment.[xiv] The real risk reduction, however, was a mere 0.9%, less than 1% or nine lives out of 1,000 when treated for five years. The research was sponsored by Bristol-Myers Squibb Pharmaceutical (West of Scotland Coronary Prevention Study). Conservatively, put another way, researchers treated 1,000 people for five years at a total cost of over $5 million to save seven people from CVD. One might wish to compare this to the cost and efficacy of adopting healthy lifestyle choices.
In the Heart Protection Study in the United Kingdom, more than 20,000 participants aged 40 to 80 years with high risk of cardiovascular disease but average-to-low levels of total cholesterol and LDL cholesterol were treated with 40mg daily of simvastatin (marketed under several trade names including Zocor®). Of 20,500+ study participants, 577 on statins died from a heart attack, 701 not treated died from a heart attack. That is a 25% relative risk reduction over five years.[xv] Sounds good, doesn’t it? An interesting point is that using their statistical method, the researchers got their own claims wrong.[xvi] The real percentage improvement is actually 1.7%. Over the five-year study, they saved 25 people per year in a high-risk population with previous cerebrovascular disease, peripheral artery disease, renal impairment or diabetes. These are seriously ill people and the researchers still achieved a benefit of only 1.7%. Researchers neglected to mention that around 30,000 people were not allowed in or dropped from the study and not counted in the percentage of people with side effects. There were 10,269 people on statins and 10,267 people on a placebo.[xvii]
A study of 90,056 participants combining 14 randomised trials looked at the best outcome for people who had pre-existing conditions: 47% had pre-existing chronic heart disease, 21% had a history of diabetes and 55% a history of hypertension. The death rate was 8.5% among the statin group compared to 9.7% in the control group. This difference represents 1.2%. That is, if you treat 100 people who already have serious pre-existing diseases that predispose them to a heart attack or stroke, you are likely to save 1.2 people.[xviii] The study made no mention of side effects for all participants.
The well-known JUPITER study compared a placebo group to a statin-taking group. The study found that there were 68 heart attacks in the placebo group and 31 heart attacks in the drug treatment group—a 58% relative risk reduction. There were 64 strokes in the placebo group, compared to 33 strokes in the treatment group, a relative risk reduction of 48%.[xix] Sounds good, doesn’t it? However, the drug treatment group had 8,901 participants in it. In real terms, the heart attack risk went from a very low 0.76% to 0.35% and the risk of stroke went from 0.72% to 0.37%. Effectively, if you treat 300 people with expensive and dangerous drugs you might save one life. Under the best possible scenario, the real risk reduction was well under one half of one percent. The real risk reduction of consuming a handful of raw mixed nuts is much higher. It is interesting to note that one of the risk factors used to select the participants in the study was C- Reactive Protein (CRP), an indicator of inflammation and the real cause of the problem.
In an independent assessment of the same statistics in 2010 titled “Cholesterol Lowering, Cardiovascular Diseases, and the Rosuvastatin-JUPITER Controversy. A Critical Reappraisal “ by Michel de Lorgeril and her 8 colleagues found that “the JUPITER Study” was severely flawed.[xx] This recent analysis did a careful and independent review of both results and methods used in the Jupiter Study and reported that the “trial was flawed”. In an unprecedented attack on the study they (scientist other than myself usually don’t say boo even when it is serious) stated that “The possibility that bias entered the trial is particularly concerning because of the strong commercial interest in the study.” In other words the big pharmaceutical money influenced the study. And concluded "The results of the trial do not support the use of statin treatment for primary prevention of cardiovascular diseases and raise troubling questions concerning the role of commercial sponsors.” This is a scathing attack in scientific terms of the earlier drug company sponsored study. Scientist do not go out of their way to create waves but these ones have not just found different results but also criticised the earlier studies link with pharmaceutical industry. It highlights not only that the studies don’t show any significant results but these studies and the education of our doctors is strongly influenced by the drug companies.[xxi]
More recently, a study reported in the BMJ was a meta-analysis of 10 randomized clinical trials of about 70,000 people followed for an average of four years.[xxii] In these trials, people with risk factors for cardiovascular disease but no history of existing disease were randomized to receive statins or no treatment. The relative risk reduction was 12% for total mortality, 30% for coronary event and 19% for a cerebrovascular event (stroke). However, the real risk reduction was 0.6%, 1.3% and 0.4% respectively. The actual number needed to treat to save one life was 167. Despite this outcome the authors of the study concluded, “In patients without established cardiovascular disease but with cardiovascular risk factors, statin use was associated with significantly (statistical not clinical) improved survival and large (statistical) reductions in the risk of major cardiovascular events.” (“emphasis added.”). In fact, the authors had significant associations with the drug companies and failed to mention it was statistically significant but not clinically significant. Again, busy medical professionals tend to read only the abstracts; claims like this are pretty convincing, though very misleading.
More telling however, is the latest findings in June 2010 where two major independent studies, one the re analysis of the Jupiter Study reported above and the other “A Meta-analysis of 11 Randomized Controlled Trials Involving 65 229 Participants” (don’t worry about the title) by Ray Kausik and 6 other independent researchers. The study, wait for it, found the use of statins in high-risk individuals was not associated with a statistically significant reduction in mortality. That is, they don’t save lives. Their data combined from 11 studies with 65 229 participants followed for approximately 244 000 person-years, a very big study, reported that this “meta-analysis did not find evidence for the benefit of statin therapy on all-cause mortality in a high-risk primary prevention set-up.” In other words they don’t save lives even in a high risk group. Even if you have all the elevated risk factors these drugs don’t work.
How many more studies to we need to do to show these drugs don’t work?
Even the economists agree with me
It is not only the scientists jumping up and down but also the economists are continually questioning the reason for so much statin treatment. In an economic review of statin use, authors reported that it is not cost-effective to treat low-risk people and also stated, “Economic analyses need to increase their transparency to reduce their vulnerability to bias and increase their reproducibility.”[xxiii] A recent study in the U.K. on reducing the cost of coronary heart disease mortality[xxiv] found statins in primary prevention cost £27 828 per life-years gained (LYG), reaching £69 373 per LYG in men aged 35 to 44. That is, to add one year to a person’s life requires spending £69 373 per year. The authors reported, “Amounts of NHS funding are being spent on relatively less cost-effective interventions, such as statins for primary prevention.”[xxv] There are now dozens of studies like this that pose the same question I am asking: Why are we using these drugs when they are not effective, but you can also add, especially when they have serous side effects?
The overwhelming outcome of all of these studies, no matter which combination used, even conducted with biases to give the best possible results, is that to possibly save a single life in a high-risk group one has to treat nearly 50 people in a low-risk group. In the population, between 100 and 1,000 people must be treated, with the possibility of dangerous side effects, to potentially save one life. Perhaps we might say that every life is worth that. Indeed, every life is valuable! A nutritional and lifestyle approach can save many lives. Unfortunately the use of statins comes with a high economic price, especially considering that nutrition and lifestyle changes can bring about much greater real benefits and even save money.
[i] Bartolucci, A.A., S. Bae, et al. (2009). A Bayesian meta-analysis approach to address the effectiveness of statins in preventing death after an initial myocardial infarction. 18th World IMACS/MODSIM Congress. Cairns, Australia.
[ii] Thavendiranathan, P., A. Bagai, et al. (2006). "Primary prevention of cardiovascular diseases with statin therapy: A meta-analysis of randomized controlled trials." Archives of Internal Medicine 166: 2307-2313.
[iii] Vrecer, M., S. Turk, et al. (2003). "Use of statins in primary and secondary prevention of coronary heart disease and ischemic stroke. Meta-analysis of randomized trials." International Journal of Clinical Pharmacology and Therapeutics 41(12): 567-577.
[iv] Hayward, R.A., T.P. Hofer, et al. (2006). "Narrative review: Lack of evidence for recommended low-density lipoprotein treatment targets: A solvable problem." Annals of Internal Medicine 145(7): 520-530.
[v] University of British Columbia (2003). "Do statins have a role in primary prevention? A review by the Therapeutics Initiative of the Department of Pharmacology & Therapeutics of the University of British Columbia." Therapeutics Letter (48).
[vi] Smith, A. and J. Warner (2009). "Psychiatrists’ appreciation of statistical v. clinical significance: a quick test." Psychiatric Bulletin 33: 293-295.
[vii] Fidan, D., B. Unal, et al. (2007). "Economic analysis of treatments reducing coronary heart disease mortality in England and Wales, 2000–2010." QJM 100: 277-289.
[viii] Franco, O.H., A. Peeters, et al. (2005). "Cost effectiveness of statins in coronary heart disease." Journal of Epidemiology and Community Health 59: 927-933.
[ix] Franco, O.H., E.W. Steyerberg, et al. (2006). "Effectiveness calculation in economic analysis: the case of statins for cardiovascular disease prevention." Journal of Epidemiology & Community Health 60: 839-845.
[x] Capewell, S. (2008). "Will screening individuals at high risk of cardiovascular events deliver large benefits? No." British Medical Journal 337: a1395.
[xi] Nuovo, J., J. Melnikow, et al. (2002). "Reporting number needed to treat and absolute risk reduction in randomized controlled trials." Journal of American Medical Association 287: 2813-2814.
[xii] Medical Research Council Working Party (1985). "MRC trial of treatment of mild hypertension: principal results." British Medical Journal 291: 97-104.
[xiii] Miall, W.E. and G. Greenberg (1987). Mild Hypertension: Is There Pressure to Treat? An account of the MRC trial. New York, Cambridge University Press
[xiv] Shepherd, J., S.M. Cobbe, et al. (1996). "Prevention of coronary heart disease with Pravastatin in men with hypercholesterolemia." New England Journal of Medicine 333: 1301-1307
[xv] Heart Protection Study Collaborative Group (2002). "MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20,536 high-risk individuals: A randomised placebo-controlled trial." Lancet 360: 7-22.
[xvi] Ravnskov, U. (2002). "Statins as the new Aspirin. Conclusions from the heart protection study were premature." British Medical Journal 324: 789.
[xviii] Cholesterol Treatment Trialists' Collaborators, C. Baigent, et al. (2005). "Efficacy and safety of cholesterol lowering treatment: Prospective meta-analysis of data from 90,056 participants in 14 randomised trials of statins." Lancet 366: 1267-1278.
[xix] Ridker, P.M., E. Danielson, et al. (2008). "Rosuvastatin to prevent vascular events in men and women with elevated C-reactive protein." New England Journal of Medicine 359(21): 2195-2207
[xx] Ray, K.K., S.R.K. Seshasai, et al. (2010). "Statins and all-cause mortality in high-risk primary prevention: A meta-analysis of 11 randomized controlled trials involving 65 229 participants." Archives of Internal Medicine 170(12): 1024-1031.
[xxi] de Lorgeril, M., P. Salen, et al. (2010). "Cholesterol lowering, cardiovascular diseases, and the Rosuvastatin-JUPITER controversy: A critical reappraisal." Archives of Internal Medicine 170(12): 1032-1036.
[xxii] Brugts, J.J., T. Yetgin, et al. (2009). "The benefits of statins in people without established cardiovascular disease but with cardiovascular risk factors: meta-analysis of randomised controlled trials." British Medical Journal 338: b2376
[xxiii] Franco, O.H., A. Peeters, et al. (2005). "Cost effectiveness of statins in coronary heart disease." Journal of Epidemiology and Community Health 59: 927-933.
[xxi] Fidan, D., B. Unal, et al. (2007). "Economic analysis of treatments reducing coronary heart disease mortality in England and Wales, 2000–2010." QJM 100: 277-289.