Statins For Healthy People – Tweaking the Numbers

I assume it is every pharmaceutical company’s dream to market a drug that people take for a lifetime to cut the risk of certain diseases or simply to delay the process of aging or prolong life. Indeed statins, drugs that are often used to lower blood cholesterol, have commonly been marketed as such wonder drugs. Although primarily used among people with cardiovascular disease (CVD), from a business perspective it would surely be much more interesting if their use became widespread as a preventive measure among the majority of healthy individuals.

Statins in healthy People - Tweaking the Numbers

Although scientific evidence supports the use of statin drugs for certain groups of individuals, advertisement and other marketing procedures appear to have facilitated their use among people who probably derive no benefit from their use. Unfortunately, the medical profession has played along in this process, not surprisingly because many doctors believe that these drugs have transformed cardiovascular medicine.

Generally, a physician will receive more critique from his/her colleagues for not treating a patient who may derive benefit from statin therapy than for treating a patient who will likely not benefit. In many countries around the world, about 25-30 percent of adults are taking statin drugs. Although many specialists believe there are a number of high-risk individuals who are not receiving therapy, many low-risk individuals who derive very little benefit are also being treated.

Statins are potent inhibitors of cholesterol biosynthesis and lower the concentration of LDL cholesterol. Most specialists believe that the lowering of LDL-cholesterol matters most when it comes to treating heart disease. However, the overall benefits observed with statins appear to be greater than what might be expected from changes in lipid levels alone, suggesting effects beyond cholesterol lowering. The non-cholesterol lowering effects of statins are often termed their “pleiotropic effect.” Among the pleiotropic effects of statins is an anti-inflammatory effect. It is not known whether the clinical benefits of treatment with statins are due to the reduction of LDL-cholesterol alone, to inflammation inhibition, or to a combination of both processes.

Clinical trials of statin therapy have addressed patients with cardiovascular disease as well as healthy individuals with increased risk for heart disease. The evidence of statins for secondary prevention, such as after a heart attack, is much stronger. Statins significantly reduce the risk of a second heart attack and they lower mortality. The higher the risk of disease the greater the potential benefit for any preventive measure, and the easier it is to measure the benefit in clinical trials.

A meta-analysis published 2012 indicates that statins reduce the risk of vascular events among healthy individuals with a relatively low cardiovascular risk. The authors concluded that “this benefit greatly exceeds any known hazards of statin therapy.” However, another recent meta-analysis shows that statins increase the risk of diabetes and the frequency of abnormal liver function tests. The increased risk of diabetes is about 9%, and it is estimated 250 patients need to be treated with a statin for one case of diabetes to be diagnosed.

I often wonder how to best inform individual patients about the role of statins in cardiovascular prevention. Almost every day I am faced with the decision on whether to treat or not to treat. I also receive a number of e-mails and comments through my website from people who are having a hard time deciding whether they should take statins or not. Unfortunately, I’m only able to answer a fraction of those and usually only in general terms.

Statins for Healthy People

I’m not going to discuss statin treatment for patients who have been diagnosed with cardiovascular disease because the evidence is strongly in favor of treatment. My concern is for the healthy individual who may be prescribed statins for years, risking side effects, and possibly deriving a very small or possibly no benefit.

Deciding whether to prescribe statins in primary prevention is a typical example of when “shared decision making” or “patient centered care” should be employed. Shared decision making is a collaborative process that allows patients and their providers to make health care decisions together, taking into account the best scientific evidence available, as well as the patient’s values and preferences.

Shared decision making honors both the provider’s expert knowledge and the patient’s right to be fully informed of all care options and the potential harms and benefits.

So, to inform individuals on the effects of statin therapy, the physician has to start teaching them about clinical trials.  I’m not going to spend to much time on that here, but let me reflect shortly on the issue of “absolute” and “relative ” risk reduction.

Let’s say that a trial on statin therapy among high-risk men randomized to a statin drug or placebo for six years showed that 6 percent in the statin group had a cardiovascular event (heart attack or death of cardiovascular causes), but 9 percent in the placebo group. This is actually what The West of Scotland Coronary Prevention Study (The WOSCOPS trial) showed. The risk of an event is 3 percent less over six years if you take a statin drug compared to if you don’t. This is “absolute” risk reduction. However, 6 percent is 33 percent lower than 9 percent. Thus, “relative” risk reduction is 33 percent. Understandably, this is usually what the pharmaceutical companies highlight in their marketing efforts.

Randomized clinical trials are not perfect. You can tweak the numbers one way or the other, depending on what picture you want to paint. For example, if I were a patient I would like to ask my doctor: “What is the likelihood of escaping an event over six years if I don’t take a statin drug?” According to the above study, the answer is 91 percent. On the other hand, if I take a statin drug for six years the likelihood of escaping an event is 94 percent. Not a big difference, is it? The absolute difference is the same as before, 3 percent.

However, if you are a fan of “relative” risk like the pharmaceutical companies, the likelihood of escaping an event if you take a statin is increased only by 3.3 percent. These are indeed very small numbers keeping in mind that the relative risk of having a statin induced diabetes is 9 percent. So, obviously, it all depends on how the physician presents the available data to the patient.

Is There a Difference Between Men and Women?

The question whether women benefit from statins in primary prevention depends on how the trials are interpreted. In her book, The Truth About Statins, Dr. Barbara Roberts criticizes the use of the term “need for revascularization” as a cardiovascular event in clinical trials.

Most of the statin trials count hard end-points such as nonfatal heart attack and cardiac death. However, many use composite end points that are softer, often including “need for revascularization” which basically is the number of times the patient was referred to bypass surgery or stent treatment. This “need” is usually determined by the person’s physician, not the physician carrying out the study, and there is wide variability in the alacrity with which physicians decide whether or not to refer a patient for revascularization. Dr. Roberts points out that if this end point is omitted from the major trials on statins in primary prevention which have included women, and only hard end points used, there is no significant effect of statins compared with placebo among women.

The Number Needed to Treat (NNT) is often used to describe the efficacy of certain treatments. It is the number of patients you need to treat to prevent one additional bad outcome (death, stroke, etc.). For example, if a drug has an NNT of 5, it means you have to treat five people with the drug to prevent one additional bad outcome.

In their meta-analysis published in JACC (Journal of The American College of Cardiology) 2012, William J Kostis and coworkers concluded that statins are effective among both men and women, in primary as well as secondary prevention. In an accompanying editorial, Dr. Lori Mosca discusses some of the problems associated with addressing the effects of statins in primary prevention by meta-analysis. Firstly, within each primary prevention population, there may be a substantial difference in risk. Secondly, the authors did not have enough data to critically evaluate adverse side effects. Thirdly, concerns have been raised about the long term safety of statins for primary prevention, both in men and women, due to a potential increased risk of incident diabetes. Fourthly, women without CVD have a lower risk for mortality and CVD than men without CVD. Therefore the absolute benefit of statins will typically be less for women than men, suggesting it might be appropriate that women receive statins less frequently than men in the setting of primary prevention. 

In October 2012 William J Kostis responded with a letter in JACC providing interesting results from an additional meta-analysis examining absolute risk reduction, where they looked at the primary end-point of the respective studies. For women, the number needed to treat (NNT) over a 4-year period was 148 for primary prevention. In men, the corresponding NNT over a 4-year period was 43.

Thus, looking at the numbers with regards to absolute risk reduction, the difference between men and women is quite striking. In primary prevention, the NNT for women almost four times higher than for men. This confirms the suggestion that it “might be appropriate that women receive statins less frequently than men in the setting of primary prevention.” Furthermore, a recent observation indicated that the risk of diabetes associated with statin therapy might be higher among women than men.

The Bottom Line

Giving unnecessary and potentially harmful treatment to people is bad medical practice. It goes against one of the oldest and most important general rules of medicine, “first do no harm.” Statins are used by about 25 percent of Americans 45 years and older. Between six and seven million people in the UK take statin drugs every day. The widespread use of these drugs is partly due to effective marketing by pharmaceutical companies.

The largest part of those who take statins is healthy individuals. For any therapy with such widespread use, the evidence of efficacy has to be substantial, and the risk of serious side effects has to be close to zero. This is not the case for statin use among healthy people.

However, it is important to keep in mind that among healthy individuals taking statin drugs, there are high-risk individuals who may benefit more than others. The decision on when to treat and not to treat can indeed be quite complex. It is a typical situation where shared decision making should be employed, allowing the patient to have a say in the final decision.

21 thoughts on “Statins For Healthy People – Tweaking the Numbers”

  1. Good post, I was thinking of something along these lines recently, perhaps you can shed some light.

    When a doctor is thinking of prescribing a statin for a healthy patient, I assume they would first consider looking at lifestyle options over drugs, this seems pretty logical. That said, is there an average..or expected value by which one can lower their LDL without resorting to statins?

    I’ve read that it can be hard for some to lower their own LDL due to genetic conditions or refusal to stick to a certain eating lifestyle so I imagine these are the “healthy” candidates who are prescribed statins?

    My own experience was that a couple of years ago my LDL was 104, a year later is was up to 161…now I was now eating very well at all (loads of processed foods, tons of sat and trans fat) and trigs of 199…..and 10 mths later via a much better eating plan, my LDL is at 107 and trigs at 78. So I infer that I’m a “responder” to dietary changes (which weren’t very onerous).

    I just see that 25-30% number for “healthy” people and I get a sense that beyond those who are genetically predisposed, there seems to be a large number of people who don’t need to be taking these drugs.

    Just my two cents.

    • Bill. I agree with you on the lifestyle issue. It´s extremely important. The PREDIMED study showed a 30% reduction of cardiovascular events on a Mediterranean diet among individuals at risk. That´s similar to the effects of statins in primary prevention, but of course without any side effects.

      It varies a lot how much people can lower their LDL-C by diet. LDL-C levels are very dependent on the liver´s own production and clearance of LDL cholesterol. Statins may be very important drugs for or individuals with familial hypercholsterolemia (FH).

      Of course, other lipid parameters are important as well as other risk factors in general. LDL-C is not the only thing that matters when it comes to risk assessment.

  2. Apart from diabetes, have you formed an opinion on any other possible side effects statins may have? For instance on the brain, say as discussed by Dr. Graveline?

    • Ágústa. I have mentioned a few other possible side effects in previous posts. Of course the muscle side effects are most common. Also, last year the FDA informed consumers and health care professionals that cognitive (brain-related) impairment, such as memory loss, forgetfulness and confusion, has been reported by some statin users.

  3. ” For any therapy with such a widespread use, the evidence of efficacy has to be substantial, and the risk of serous side effects has to be close to zero.”

    What are the serious side effects of aspirin? What are the serious side effects of various dietary supplements? Can you name one drug that has a risk of serious side effects close to zero?

    • Charles. There are serious side effects of aspirin. That´s why it shouldn´t be recommended for the general public. Neither should statins, beta blockers or ACE inhibitors for that matter. Fish oils (EPA and DHA) are an example of dietary supplements that probably don´t have any serious side effects. You can however debate about their efficacy when it comes to preventing heart disease. In my opinion statins should be used much more selectively than they are, due to the risk of side effects.

      • “Fish oils (EPA and DHA) are an example of dietary supplements that probably don´t have any serious side effects.”

        In general, not that kind that would be comparable to medication, I suppose. However, when you consider the side effects (e.g. excessive use can increase LDL levels among those suffering from hypertriglyceridemia; increased bleeding etc.) TOGETHER with the fact that their efficacy in primary prevention is close to non-existent, then …

        And of course even diets can have adverse effects.

  4. Hello Doctor. I came upon this article while conducting research on statins over the last few days. Just last week, I was diagnosed (with ultrasound) with the beginnings of left carotid artery plaque (not intermediate or advanced), and my Cardiologist wants me to start 5mg Crestor (I filled the prescription, but have NOT taken any pills yet) for the first time ever. No blockages in my heart. I also take Norvasc, Toprol XL, baby aspirin, fish oil, COQ10, and a GNC daily men’s multivitamin, and my BP is normal (with the drugs). I have heard many bad things about statin side effects (like through Dr. Mercola’s website and newsletters), and many articles do point out that one should not take them unless absolutely necessary. I am very afraid of taking them! I’ve been obese the last few years (231 lbs in August this year), but am now on a serious lifestyle and diet change. I’ve lost 16 lbs since August, and my goal is to weigh 180 pounds by next year sometime. I’m 52, male, and about 5′ 8″ tall, and in otherwise good health, and now exercise every day. If the primary goal of statins are to reduce cholesterol, and my cholesterol numbers are good (and improving with diet and exercise), is it really necessary to take the statin? What would you do? I just want to be around for a long time, if possible, for my beautiful 8 year old daughter! I’m very scared right now. Thank you.

    • George. It is a difficult task to give advice through a blog post. I assume your Cardiologist believes you have early signs of atherosclerosis considering the carotid artery plaque. It appears that all your risk factors are well controlled. You´re on medication for high blood pressure. Fish oil is probably a good idea. I understand your lipid numbers are good as well. It is possible that statins might reduce the number of cardiovascular events slightly if we had a large group of individual like you and compared it with placebo. However, we also know that there are side effects although they are very seldom serious. If you take good care of your self by healthy eating and regular exercise I don´t believe that adding statins would make any miracles. Discuss it with your doctor, it´s all about “shared decision making”. Good luck 🙂

      • Thank you Doctor! I am, for now, refraining from taking the 5mg Crestor, and am awaiting my Cardiologist’s response (I wrote and delivered him a letter outlining my concerns today, and hope he responds soon). I have also done some online research and came across an apparently good book (with high ratings) titled ‘Prevent And Reverse Heart Disease’ by Dr. Caldwell Esselstyn from The Cleveland Clinic. From what I understand, he strongly advocates a strict vegan diet, and claims to have obtained scientific results and many positive outcomes. I ordered the book today, and am looking forward to reading it, and even following this radically different, strict diet if it will help me. My only confusion would be, does carotid artery disease/plaque fall under the category of ‘heart disease’ or something else? I would think it does.

  5. Do you believe that substantial evidence has accumulated in support of the hypothesis that elevated cholesterol levels increase the risk of developing Alzheimer’s disease?

    If yes would not statins be of use?

    • Charles. I think there is some discrepancy between the experimental studies and the clinical data on the role of cholesterol in Alzhemer´s disease.
      Amyloid-β peptide (Aβ) is a primary component of neuritic plaques in this disease.
      The majority of in vivo and in vitro studies show an interaction between cholesterol and Aβ. Much of the clinical data however, do not support involvement of elevated cholesterol levels as a causative factor in Alzhemer´s disease.
      Finally, the efficacy of statins whether in prevention or treatment of Alzheimer´s disease is not proven.

  6. Association of Alzheimer disease pathology with abnormal lipid metabolism : The Hisayama Study
    T. Matsuzaki, K. Sasaki, J. Hata, et al.
    Neurology 2011;77;1068
    DOI 10.1212/WNL.0b013e31822e145d

    Objective: The relationship between lipid profiles and Alzheimer disease (AD) pathology at the
    population level is unclear. We searched for evidence of AD-related pathologic risk of abnormal
    lipid metabolism.

    Methods: This study included brain specimens from a series of 147 autopsies performed between
    1998 and 2003 of residents in Hisayama town, Japan (76 men and 71 women), who underwent
    clinical examinations in 1988. Lipid profiles, such as total cholesterol (TC), triglycerides, and
    high-density lipoprotein cholesterol (HDLC), were measured in 1988. Low-density lipoprotein
    cholesterol (LDLC) was calculated using the Friedewald formula. Neuritic plaques (NPs) were assessed
    according to the Consortium to Establish a Registry for Alzheimer’s Disease guidelines
    (CERAD) and neurofibrillary tangles (NFTs) were assessed according to Braak stage. Associations
    between each lipid profile and AD pathology were examined by analysis of covariance and logistic
    regression analyses.

    Results: Adjusted means of TC, LDLC, TC/HDLC, LDLC/HDLC, and non-HDLC (defined as TC–
    HDLC) were significantly higher in subjects with NPs, even in sparse to moderate stages
    (CERAD _ 1 or 2), compared to subjects without NPs in multivariate models including APOE _4
    carrier and other confounding factors. The subjects in the highest quartiles of these lipid profiles
    had significantly higher risks of NPs compared to subjects in the lower respective quartiles, which
    may suggest a threshold effect. Conversely, there was no relationship between any lipid profile
    and NFTs.

    Conclusion: The results of this study suggest that dyslipidemia increases the risk of plaque-type
    pathology. Neurology® 2011;77:1068–1075

    • Interestingly many different lipid parameters seem to be associated with NP´s in this study. Of course this does not prove any causal relationship. However, this may be quite confusing when it comes to the efficacy of different dietary approaches. Although LDL-C may be most effectively lowered by reducing the consumption of saturated fats and eating more PUFA´s and whole grains, TC/HDL ratio is best lowered by eating a low carb high fat diet, including saturated fats.

      We know that there is an association between the metabolic syndrome and the risk of Alzheimer´s disease. Most patients with the metabolic syndrome have abnormal lipid metabolism. There is still a limited understanding however on the role of lipid abnormalities as a causative factor for Alzheimer´s disease in these individuals. Other factors, such as high blood pressure and abnormal glucose metabolism may play an important role as well.

  7. “TC/HDL ratio is best lowered by eating a low carb high fat diet, including saturated fats”

    Dr Thomas Dayspring says
    “Why does the Dean Ornish extreme low fat diet so effectively reduce all cholesterol levels? Well the initial substrate from which cholesterol is synthesized is acylCoA (acetoCoA, acetylacetyl CoA) which is derived from fatty acid breakdown (oxidation). So eliminating fat from the diet will drastically reduce endogenous cholesterol synthesis and all cellular cholesterol levels will lessen. As cellular cholesterol synthesis reduces, less is effluxed via ABC family transporters into HDL particles: HDL-C will lessen. Also in people significantly restricting fat intake, the liver will have less cholesterol (less chylomicron delivery of fat, less production, less being brought back to the liver in HDLs: the results is when the liver makes VLDLs and IDLs, they carry a lot less cholesterol (less VLDL-C, less IDL-C and this will ultimately result in less LDL-C. Of course Ornish showed that by drastically reducing TC levels (as well as LDL-C) via fat restriction angiographic improvement occurs in persons with CHD. It mattered little that because of reduced cellular cholesterol, HDLs were no longer being fully lipidated (thus reducing HDL-C).”

    SO – does it matter if HDL-C is lower due to diet as long as the overall TC/HDL ratio is still >3.5 (optimal)?

    Also – isn’t the apoB/apoA-I ratio a better predictor than the TC/HDL ratio?

    • That depends on the types of individuals/risk groups we’re talking about. There are different types of dyslipidemia and for some lower HDL levels matter more than to others.Same with apoB/apoA-I ratio.

    The apoB/apoA-I ratio is better than the cholesterol ratios to estimate the balance between plasma proatherogenic and antiatherogenic lipoproteins and to predict coronary risk.

    High apoB and a high apoB/apoA-I ratio were strongly related to increased coronary risk, while high apoA-I was inversely related to risk. The apoB/apoA-I ratio was superior to any of the cholesterol ratios in predicting risk. This advantage was most pronounced in subjects with LDL-C levels <3.6 mmol/l. Addition of lipids, lipoproteins or any cholesterol ratio to apoB/apoA-I in risk models did not further improve the strong predictive value of apoB/apoA-I.

  9. Esselstyn has an hour long talk on youtube that gives a good overview – plus he has more than a few other, shorter ones.

    He refers to one of his cases that seems somewhat difficult to accept: one secondary patient, who had been getting some plaque regression, starts instead to progress. The reason, he says, is that the patient had been consuming commercial foods with a mere 1/2 g of fat here and there (which doesn’t have to be listed on the nutrition labels). That supposedly was enough, totaling at maybe <10 g fat per day, to cause his plaque to progress. Well, possibly if it was trans fat and that person was also especially susceptible to plaque buildup? Seems an extreme case, though.

    On the other hand, Dr, Greger (a vegan M.D. with a prolific web presence) has a video talk from several years ago where he points to the studies that show vegans doing *worse* than other diets on some health outcomes. As you can see from the title, he has his view on why that was so:
    "40 Year Vegan Dies of a Heart Attack! Why? The Omega-3 and B12 myth"

    Here's a paper: "Mortality in vegetarians and nonvegetarians: detailed findings from a collaborative analysis of 5 prospective studies"
    which says, "The death rate ratio for ischemic heart disease… was 0.76… There was significant heterogeneity between studies… There were no significant differences between vegetarians and nonvegetarians in mortality from cerebrovascular disease, stomach cancer, colorectal cancer, lung cancer, breast cancer, prostate cancer, or all other causes combined"

    Then again, the DIRECT-Carotid study by Iris Shai showed the same reduction in CIMT on 3 different diets: low fat, low carb and Mediterranean.So maybe it's really mostly about the excess calories. Then again, the study concludes that the reductions might only have been from reduced BP and the resulting lessening of smooth muscle in the artery wall.

    But then a Cochrane review from 2008 showed, IIRC, no reduction in CVD events via lowering of BP through meds.

  10. Hi, George. The ‘V’ in CVD (Cardio Vascular Disease) includes the carotids. But a person might have a great CIMT scan and still have bad plaques in their coronary arteries – or vice versa. I believe the phrase is that CIMT is “not a good proxy” for coronary heart disease.

    The differences in blood pressure might make for the difference? Though carotids (and the brain itself) are included in ‘central pressure’, both of which can be quite different from the brachial pressure taken with an arm cuff.

  11. Speaking of purported inflammation from linoleic acid, I’ve just read through a very well-ordered 2012 slideshow by a certain “pronutritionist ” at:
    which was recommended here by Mie back in March. It seems rather persuasive that generally:
    1) consuming LA doesn’t really raise arachidonic acid (AA), including in platelets
    2) consuming O3 doesn’t really provide much anti-inflammatory activity, except somewhat in cardio or renal diseases

    However, actually consuming too much AA might still be bad.

  12. We found consistent results in subgroup analyses examining the use of statins for primary and secondary prevention. Relative to pravastatin, treatment with atorvastatin (adjusted hazard ratio 1.20, 95% confidence interval 1.10 to 1.30), rosuvastatin (1.12, 1.02 to 1.23), or simvastatin (1.12, 1.02 to 1.23) was associated with a significantly increased risk of new onset diabetes in the primary prevention cohort, while no increased risk was observed for patients treated with lovastatin (0.98, 0.79 to 1.22) or fluvastatin (1.01, 0.82 to 1.23). Similar findings were observed among the secondary prevention users (table 3 ⇑ ).


Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.