[sohex]

The authors' own conclusion seems to directly contradict your overall argument here: "Implications for practice The totality of evidence now supports the benefits of statins for primary prevention. The individual patient data meta-analyses now provide strong evidence to support their use in people at low risk of cardiovascular disease. Further cost-effectiveness analyses are now needed to guide widening their use to these low risk groups."

And as haldujai mentioned this is explicitly regarding their use for primary prevention, not with regards to usage for secondary prevention which has strong supporting evidence.

[rgbgraph]

Generally, you read the papers for their methodology and their data — not for the author commentary; and then make up your own mind. One man’s “benefits of statins for primary prevention” is another’s “the benefits are too meager to be notable.”

Please provide me literature from a reputable publication (viz. the AHA, Cochrane, or the New England Journal of Medicine), that has not been funded by a pharmaceutical company — that demonstrates strong supporting evidence for the usage of statins in secondary prevention; wherein the experiment does not extrapolate from LDL values to determine mortality risk (I will concede defeat if you can find any paper that utilizes CAC scans and shows a reversal in atherosclerosis), and/or shows a greater than 2% absolute reduction all-cause or CVD-only mortality risk.

You will not find such a paper, because it does not exist. Most funding has gone towards primary prevention in young adults — while little more than weak associative studies have been published for secondary prevention (and countless others I no doubt have never seen the light of day).

[haldujai]

So every link I provided give you a risk of MACE. Reversal of atherosclerosis is not the outcome measure we care about lol. Certainly not lowering coronary calcium which is not possible. You’re literally making this up…

Statins work amazingly not just for LDL reduction but plaque stabilization.

As an aside a 2% ARR is huge, it means the number needed to treat is 50 to save a life. For something with next to no serious side effects, rhabdo/diabetes is dramatically overstated.

Pertinently, the number needed to treat for MACE is 39. That’s hugely significant.

[rgbgraph]

Then we are at an uncrossable philosophical chasm.

I don’t consider 2% ARR huge — especially when the risks of side-effects have been down-played. We can argue about this all we want, but it’s no longer a matter of fact, but of opinion and values.

[haldujai]

You seem to be misunderstanding how evidence works, it is not about “what you consider” and is entirely based on fact. We also don’t talk ARR in isolation when we decide on interventions, it’s NNT vs NNH and considering the specific risk being reduced and the specific harm.

I’ll use your 2% ARR for death although there are better numbers in different patient populations.

In other words: Statins will save 1 life for every 50 patients treated and prevent 1 in ~20-40 non-fatal cardiac events, a medically significant result period. The NNH is > 100, and the harm is a self-limiting myopathy (and a possible risk of accelerated diabetes-onset in observational studies, that is still outweighed by the reduction in all-cause mortality and MACE).

The evidence is unequivocal that the benefits far outweigh the harms.

Separately, you have a personal choice to take/not take any treatment, and you may personally feel treating 50 people to save 1 life is not worth it for you, because you subjectively feel the numbers don’t fit your personal risk/benefit model. This is where you are saying 2% ARR is insignificant to you but this says nothing about the evidence or rationale behind the treatment.

> risks of side-effects have been down-played.

Except every study looking at side-effects has shown they were overstated in the initial trial.

“The most severe complication of SI is discontinuation of effective cholesterol-lowering treatment in patients who, by virtue of their CVD risk and cholesterol level, might otherwise benefit.”

https://www.sciencedirect.com/science/article/abs/pii/S00219...

[rgbgraph]

This conversation is going in circles.

Linked study equivocating on the overstatement of statin side-effects has the first author as a pharmaceutical consultant (you do not see this as a problem, I do).

2% ARR is meager in relation to lifestyle changes that can account for 3x-15x the in ARR compared to statins (you do not see this as a problem, I do).

Those are the facts. How you interpret them is subjective.

All of the money and manpower thrown into statins, could have been thrown into smoking cessation programs or preventing onset of type 2 diabetes. This is an opinion.

Saving 1 life for every 50 patients, at the cost of untold resources, instead of saving 6-30 for every 50 is myopic. This is an opinion.

Medical significance is an opinion. The determination of significance is a subjective interpretation. This is not a "this idea concurs with my sentiments, so I will say it is so. Period." This is math. Statistical interpretation is an opinion. The difference in statin efficacy vs. lifestyle changes is a quantifiable fact. The difference is 3-15x. To take a statistical finding without incorporating it into the larger context is poor practice, bordering on deception (the former is a fact, the latter is an opinion).

Here are 50 people aged 50 y.o. from the general population:

OOOOO_OOOOO | OOOOO_OOOOO | OOOOO_OOOOO | OOOOO_OOOOO | OOOOO_OOOOO

Here is the same group of people after 25 years (14 have died due to CVD-related mortality):

XXXXX_XXXXX | XXXXO_OOOOO | OOOOO_OOOOO | OOOOO_OOOOO | OOOOO_OOOOO

Here is the same group of people after 25 years, but they were on statins:

XXXXX_XXXXX | XXXOO_OOOOO | OOOOO_OOOOO | OOOOO_OOOOO | OOOOO_OOOOO

Here is the same group of people after 25 years, but instead of statins, the smokers ceased smoking:

XXXXX_XXXXX | OOOOO_OOOOO | OOOOO_OOOOO | OOOOO_OOOOO | OOOOO_OOOOO

Here is the same group of people after 25 years, but instead of a treatment, all of them developed diabetes:

XXXXX_XXXXX | XXXXX_XXXXX | XXXXX_XXXXX | XXXOO_OOOOO | OOOOO_OOOOO

[haldujai]

> at the cost of untold resources

Not going to argue the methodology but it seems like a starting point is $22,000-37000/QALY[0-1].

> instead of

This is a false dichotomy, guidelines all start with lifestyle modifications.

[0]https://jamanetwork.com/journals/jamacardiology/fullarticle/...

[1]https://jamanetwork.com/journals/jama/article-abstract/23964...

[sohex]

> Generally, you read the papers for their methodology and their data — not for the author commentary; and then make up your own mind.

I'd say that generally it's advisable to take both into consideration given that in most cases the author of a paper likely has more domain expertise than you in that specific area. Not always, obviously, and not to the exclusion of an outside objective analysis of their data and results, but it's certainly more informative than referring someone to a page from a study with no additional context.

> that has not been funded by a pharmaceutical company

I get where you're coming from here, but it's kind of silly. And the question becomes where do you draw the line? Is a meta-analysis of a large group of studies each of which has been supported at least in part by funding from a pharmaceutical company guilty by association? That aside, the structure of research funding with regards to pharmaceuticals (at least in the US so far as I'm aware) makes the likelihood of conducting any long term, large scale study without receiving any funding from a pharmaceutical company vanishingly small. There have certainly been issues with studies funded and conducted by those companies, but that doesn't mean that all studies funded by them are instantly invalid. Nor does it mean that it's impossible to conduct a study that has received their funding without compromising its integrity. It is entirely possible to take sufficient measures to isolate those companies from the actual process and analysis of the research.

> wherein the experiment does not extrapolate from LDL values to determine mortality risk

As I made reference to earlier, I am not a subject domain expert here. That being said I did do a brief survey of the literature, reading fifteen papers published on either studies of statin efficacy or meta-analysis thereof. I may be misunderstanding what you're saying, but the studies I looked at assessed efficacy by looking at the actual number of cardiac events, strokes, et al suffered by those in the control and experimental groups. Their analysis was based on those numbers, not an extrapolation from LDL values.

> and/or shows a greater than 2% absolute reduction all-cause or CVD-only mortality risk.

Why that number? And why that number in two very different contexts? Regardless, statins have been show in numerous studies to be highly effective.

> while little more than weak associative studies have been published for secondary prevention

That's simply not true. There have been a number of large scale, long term studies on the efficacy of statins for secondary prevention and the preponderance of evidence is on the side of them being very effective.

Some of the studies I looked at:

Mega, J. L., Stitziel, N. O., Smith, J. G., Chasman, D. I., Caulfield, M. J., Devlin, J. J., … Sabatine, M. S. (2015). Genetic risk, coronary heart disease events, and the clinical benefit of statin therapy: an analysis of primary and secondary prevention trials. The Lancet, 385(9984), 2264–2271. doi:10.1016/s0140-6736(14)6173

- The primary focus of this study was looking at the efficacy of statins with relations to genetic risk profiles, but as a component of that we can see the overall efficacy of the statins across those risk profiles as well.

- "The relative risk reductions were 34% in low, 32% in intermediate, and 50% in high genetic risk score categories in the primary prevention trials, and 3% in low, 28% in intermediate, and 47% in high genetic risk score categories in the secondary prevention trials. When the data were combined, the gradient of relative risk reductions with statin therapy across low, intermediate, and high genetic risk score categories were 13%, 29%, and 48%, respectively (p value for trend=0·0277, figure 2)."

- "With a focus on the primary prevention trials, in JUPITER, the number needed to treat to prevent one coronary event in 10 years was 66 for those individuals with a low genetic risk score, 42 for those with an intermediate score, and 25 for those with a high score. In ASCOT, the number needed to treat to prevent one coronary heart disease event in 10 years was 57, 47, and 20, respectively, across the three genetic risk score categories."

MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20 536 high-risk individuals: a randomised placebocontrolled trial. (2002). The Lancet, 360(9326), 7–22. doi:10.1016/s0140-6736(02)09327-3

- "All-cause mortality was significantly reduced (1328 [12·9%] deaths among 10 269 allocated simvastatin versus 1507 [14·7%] among 10 267 allocated placebo; p=0·0003), due to a highly significant 18% (SE 5) proportional reduction in the coronary death rate (587 [5·7%] vs 707 [6·9%]; p=0·0005), a marginally significant reduction in other vascular deaths (194 [1·9%] vs 230 [2·2%]; p=0·07), and a non-significant reduction in non-vascular deaths (547 [5·3%] vs 570 [5·6%]; p=0·4). There were highly significant reductions of about one-quarter in the first event rate for non- fatal myocardial infarction or coronary death (898 [8·7%] vs 1212 [11·8%]; p<0·0001), for non-fatal or fatal stroke (444 [4·3%] vs 585 [5·7%]; p<0·0001), and for coronary or non- coronary revascularisation (939 [9·1%] vs 1205 [11·7%]; p<0·0001). For the first occurrence of any of these major vascular events, there was a definite 24% (SE 3; 95% CI 19–28) reduction in the event rate (2033 [19·8%] vs 2585 [25·2%] affected individuals; p<0·0001)."

Sever, P. S., Dahlöf, B., Poulter, N. R., Wedel, H., Beevers, G., Caulfield, M., … Östergren, J. (2003). Prevention of coronary and stroke events with atorvastatin in hypertensive patients who have average or lower-than-average cholesterol concentrations, in the Anglo-Scandinavian Cardiac Outcomes Trial—Lipid Lowering Arm (ASCOT-LLA): a multicentre randomised controlled trial. The Lancet, 361(9364), 1149–1158. doi:10.1016/s0140-6736(03)1294

- "The primary endpoint of non-fatal myocardial infarction, including silent myocardial infarction, and fatal CHD was significantly lower by 36% (hazard ratio 0·64 [95% CI 0·50–0·83], p=0·0005) in the atorvastatin group than in the placebo group (figure 2, table 3)."

- "There were also significant reductions in four of the seven secondary endpoints, some of which incorporated the primary endpoint: total cardiovascular events including revascularisation procedures (21%); total coronary events (29%); the primary endpoint excluding silent myocardial infarction (38%); and fatal and non-fatal stroke (27%, figures 3 and 4). All-cause mortality was non-significantly reduced by 13%, with non-significantly fewer cardiovascular deaths (figures 3 and 4) and no excess of deaths from cancer (81 assigned statin vs 87 assigned placebo) or from other non-cardiovascular causes (111 vs 130)."

It appears that the evidence in support of the use of statins is quite overwhelming.

[rgbgraph]

> I'd say that generally it's advisable to take both into consideration given that in most cases the author of a paper likely has more domain expertise than you in that specific area. Not always, obviously, and not to the exclusion of an outside objective analysis of their data and results, but it's certainly more informative than referring someone to a page from a study with no additional context.

I do not agree. I do not have the time to elaborate further.

————

> I get where you're coming from here, but it's kind of silly. And the question becomes where do you draw the line? Is a meta-analysis of a large group of studies each of which has been supported at least in part by funding from a pharmaceutical company guilty by association? That aside, the structure of research funding with regards to pharmaceuticals (at least in the US so far as I'm aware) makes the likelihood of conducting any long term, large scale study without receiving any funding from a pharmaceutical company vanishingly small. There have certainly been issues with studies funded and conducted by those companies, but that doesn't mean that all studies funded by them are instantly invalid. Nor does it mean that it's impossible to conduct a study that has received their funding without compromising its integrity. It is entirely possible to take sufficient measures to isolate those companies from the actual process and analysis of the research.

Again, I do not agree. These are matters of values, and no arguments can be made for what we innately value. I draw a nuanced line based on my values, that I have tried to express here; but making it finer and finer will serve no purpose but as fuel for disagreement — because it is wholly subjective.

Possibility is not actuality. Most researchers are not a Platonic ideal: perfectly noble and virtuous and vigilant. They are real people: lazy, prone to error, requiring money to survive, self-interest at the very forefront.

I will not call your viewpoint naive, but it’s something that can only be formed when one’s exposure to this field is limited to papers and doctor’s visits.

> Why that number? And why that number in two very different contexts? Regardless, statins have been show in numerous studies to be highly effective.

Because 2% ARR is the highest change I’ve seen in any statin experiment — in either context. I do not consider one out of every fifty people being saved by a statin significant, or my definition of “highly effective.”

> It appears that the evidence in support of the use of statins is quite overwhelming.

My patience for reiterating this point is gone: relative changes are not absolute changes.

A starting risk profile of 2.25%, reduced to 1.25%, will have been reduced an absolute 1%, but a relative 44%.

This is why you read the methodology, and not the authors’ interpretation of their own data.