VR -- Thanks for the info on the testing of new screening techniques.  I'm going to look up "tomosynthesis"!

This new guideline BLOWS MY MIND!

I guess it's cheaper to let "sick" people die!

Way to go Canada! Let's make it even more difficult to get diagnosed! Later stage cancers need more treatment, thus pharmaceutical companies will benefit $ from our suffering!

BASTARDS!

My doctor didn't even know how to do a breast exam! He said he couldn't feel my lump!! It was 8cm!!!!

And breast self exams are free! Why would they not recommend that?!

I agree, it's nothing to leave to chance.  A cluster of microcalfications was discovered on my first mammogram.  It turned out to be DCIS...but high grade with comedo necrosis (very fast growing) and I triple negative invasive tumour was lurking nearby. The invasive tumour was discovered when I had my lumpectomy. I was so lucky mine was caught. I hope others in a similar situation will also be so lucky with these new guidelines.

I apologize that the charts cannot be reproduced.  But everyone can use these resources to find the primary articles and then examine the charts in the primary articles. 

NY Times medical reporter Gina Kolata wrote yet ANOTHER story this week about the topic of whether or not to call certain cells "cancer" or "weird cells."  The title of the article is "Cancer" or "Weird Cells"  Which Sounds Deadlier.  Back in 2009 she wrote a series about cancer entitled "Forty Years' War" about the progress and LACK of progress on the War on Cancer that President Nixon began.  Well worth reading.

Bottom line...please read the above post TWICE and then share it.  It really gives you an understanding of the SCREENING controversy and is an excellent primer.  I think it answers many of the thoughtful questions that many women are asking on this thread.

Littlemelons..We need to find a way to save the lives of younger patients with more aggressive cancers.  And that begins with allocating funds to research.  I just think if we continue to advocate for screening methods that clearly do NOT improve mortality, then we are doing a disservice to our brethren.  We have to support the researchers who are toiling in labs looking for treatments.  This controversy with screening needs to be settled once and for all so we can move on and find modalities of screening THAT save lives and we need to support research that finds a cure.  I just keep thinking of all of the monies and energy wasted on Autism research derived from Dr. Wakefield.  We need to wake up too and move on...lives depend on it!

Read Dr. Welch's book.  He eloquently explains the controversy of screening.  It's unfortunate that when other doctors try to explain the controversy in a minute or two, the impression left is that the individual is flippant.  That is unfortunate.  That's why it is imperative to read the literature.  Regarding the folks who wrote the letter to the Lancet, I would not be surprised if a number of them are radiologists.  The folks who are presenting the ant-screening ideas are researchers who specialize in statistics.  They are merely explaining what CAN be explained using statistics.  I understand that statistics can be tailored to explain whatever one wants to explain.  But this controversy regarding screening has been gathering steam for over a decade and the evidence supporting the anti-screening evidence continues to build.  Why else in the last few weeks have we been told that most men can defer prostate screening? 

When we have to change our ideas, we have to step outside of our comfort zone and that isn't easy.  Nor is it easy to accept that even a single life will be lost if we let go of an idea that we've come to accept as a given.  We've been told for years and years how screening leads to early diagnosis and early diagnosis leads to a cure.  Is there a place for screening? Yes.  But we must not be vigilant in our belief that screening is the holy grail.  Instead, we need to accept that the prevailing wisdom might just not be true and we have to face the harsh reality that scientific evidence, while flawed, is all we have. We must focus our attention on dismissing our previously held beliefs and instead covet the evidence, even if it means letting go on what we thought was "The Truth."  And yes, lives will be lost due to lack of screening.  But, hopefully, one day soon, many, many more lives will be saved because this controversy will lead to screening modalities that DO save lives.  Those modalities that are found to work may be less expensive and then more money can be secured to find a cure for the most aggressive cancers. 

Don't upset yourself after watching a small segment on television.  Instead, spend time reading the British Medical Journal and the Lancet.  Read Dr. Welch's book.  Also read John Abramson, MD's book Overdosed America and then decide what you believe in.  And by all means, if you haven't studied Statistics, then by all means study it!  I look back on my education and believe the most important and most difficult class I ever took was in Statistics and I am most grateful now that I took it.  It has shaped my adult life in countless ways.

Understanding uncertainty: Breast screening, a statistical
controversy

by David
Spiegelhalter

Issue
53




Submitted by plusadmin on December 1, 2009
in

• 53
• cancer
• confidence
  interval
• medical statistics
• medicine and
  health
• public understanding of mathematics
• risk analysis
• statistics
• understanding
  uncertainty

December 2009

Back to
the What's the best
medicine package
Back to the Do you know
what's good for you package

One in nine women will get breast cancer in her lifetime, and it seems
sensible to try and find and treat those cancers as early as possible by taking
regular X-rays of the breasts - a process known as screening with mammography.
But heated arguments are going on in the USA about how often women should be
screened, and in the UK about what women should be told when invited for
screening. Both controversies rest on what one believes are the benefits and
harms of screening. For example, it's been claimed that if 2,000 women are
screened for ten years, only one will avoid dying from breast cancer, but ten
will unnecessarily become cancer patients, since their tumour would never have
caused a problem if it had remained undetected.

Disentangling these arguments involves subtle issues about statistics,
mathematical modelling, scientific uncertainty and risk communication. This is
very topical: newspapers and the medical journals repeatedly discuss the issue,
and Sense about Science, a UK
charity trying to improve public understanding of science, have recently
released a guide to
screening.

The idea of screening seems simple: if a cancer can be detected when it is
small, then it can be treated appropriately and that should reduce the chance of
the woman eventually dying of breast cancer. But women may be harmed by
screening in a variety of ways: first, a woman may be made anxious and suffer
unnecessary further investigation after an X-ray that suggests a cancer but
which turns out to be a false alarm; second, she may genuinely have a form of
cancer but one which would never have caused a problem if it had not been found
and unnecessarily treated; third, she is exposed to radiation from the
mammography that, paradoxically, may give her cancer (however this is a very
small risk given the current low X-ray dose).

So first of all the health service has to decide how often to offer screening
(which is the current concern in the US), and second each individual woman has
to decide whether to accept the offer (the current concern in the UK). Each
decision becomes one of weighing up the potential benefits and harms of
screening, and this means doing research to help us put numbers on what those
benefits and harms might be.

Benefits and harms: evidence from clinical trials

The best type of research study is a randomised controlled trial (RCT), in
which a large group of women are randomly allocated to be offered screening or
not, and then followed up for years to see what happens to them. Note, however,
that some of the women offered screening will not accept it, while many of the
women not offered screening will have it anyway. So we are comparing a group of
women offered the screening to a group not offered it, rather than a group of
women who have been screened to a group who haven't. Therefore, comparing the
two groups allows us to estimate the benefits of a health service offering a
screening programme, not the benefits for an individual woman going for
screening.

The US
researchers, for example, wanted to see whether screening benefited women
under 50, and found eight relevant trials with the results shown in Table 1.

 Without screening With screening  TrialNumber of women (A)Deaths (B)Deaths/1000 women (C=1000B/A) Number of women (D)Deaths (E)Deaths/1000 women (F=1000E/D) Death rate difference (C-F)11069562512.3 538841051.9 0.42252161084.3 252141054.2 0.1313740826.0 13740644.7 1.3414271594.1 11724342.9 1.258021131.6 14303342.4 -0.8612279665.4 13568533.9 1.575031163.2 9582222.3 0.9810459302.9 10285313.0 -0.1 TotalTotalOverall death rate TotalTotalOverall death rate   1959736253.19 1523004482.94  

Table 1: Evidence from eight
randomised trials of screening women aged 39-49, showing the number of women
randomised to be offered screening or not, and the subsequent number of them who
died of breast cancer.

Table 1 shows the number of women who died per 1,000, and the difference in
the death rates in each trial - for example, in the first trial the death rate
was 2.3 per 1,000 in the group without screening, and 1.9 per 1000 in the
screened group, with a difference of 0.4.

There has been a lot of research into methods for summarising such a table.
It would be tempting to just add up the columns to create the total number of
women and deaths and then calculate the overall death rates and their
difference. This suggests an overall saving of 3.19-2.94 = 0.25 of a life per
1,000 women offered screening.

However this is not an accepted procedure. Instead we should "respect the
randomisation" - in other words calculate the effect of the screening in each
study, and combine those measures of effect. For example, we might take the
average of the death rate differences (the average of the values in the
right-most column) to produce an estimate of 0.57 lives saved per 1000 women
offered screening, representing a reduction in risk of 0.57/3.19 or 18%.

In fact, the authors' methods were somewhat more complex and they arrived at
an estimate of a 15% risk reduction. This estimate of the true underlying risk
is imprecise in spite of the large numbers of women recruited into the studies:
the authors also provided a rather wide 95% confidence interval, which
ranges from a 4% risk reduction to a 25% risk reduction. This means that they
are 95% confident that the true risk reduction lies within this range. They
estimated similar benefits of a 14% reduction for women aged 50-59, and a
greater risk reduction of 32% for women aged 60-69.

Benefits and harms: evidence from mathematical models

However, this analysis was not the main evidence used to come up with the US
recommendations for screening. Randomised trials can only give part of the
picture, and detailed mathematical modelling is necessary using data from a
variety of sources. The US
Preventive Services Task Force (USPSTF) published its recommendations
last month (November 2009), based on work by six different teams to estimate the
effect of a range of different screening strategies. A "preferred" strategy was
chosen, based on the average of the models, which only started screening at 50,
and then screened every two years.

This final strategy was estimated to have the following benefits and harms
when applied to 1,000 women:

Benefits Breast cancer deaths avoided7.5Extra years of life121Costs and harms Mammographies11109False-positive diagnoses requiring further investigation940Unnecessary biopsies (tissue samples taken with needle)66Over-diagnoses (cancers detected and treated that wouldnot have caused any harm if left alone)N/A

Table 2: Summary of benefit and
harms for 1000 women under the screening strategy suggested by US Preventive
Services Task Force, screened each two years between 50 and
74.

The Task Force found that screening more often and starting at a younger age
saved more lives, but at a large increase in false positives (erroneous
diagnoses of cancer when the woman was healthy) and unnecessary biopsies.

Screening can detect cancers that would have been harmless, leading to
unnecessary treatment.

These recommendations are more intensive than those in the UK, which
recommends screening only every three years, but were extremely controversial in
the US, as previous guidelines had recommended screening every year from age 40.
There have been dramatic newspaper headlines and claims that this reduction in
screening was a denial of health care under the Obama administration, and the American Cancer Society
has come out against the recommendations. It is noticeable, however, that the
proposed strategy is supported by the major breast cancer charity - the National Breast Cancer Coalition.
There is also no explicit attempt to cost the interventions - in the UK the cost
of all the medical treatment would have been included in the analysis.

Potential harm: over-diagnosis

Most remarkable is the lack of any estimate of over-diagnosis - the
diagnosis of cancers that would not have caused harm if left alone. In fact, the
authors estimated over-diagnosis in each of the six models, but did not publish
the results. They felt that the "absolute estimates were unreliable" due to
limitations in knowledge about the types of tumour that are detected in these
screenings.

Other authors do try and estimate over-diagnosis. For example, a July editorial in
the British Medical Journal provides the following assessment:

BenefitsBreast cancer deaths avoided1Costs and harmsFalse-positive diagnoses requiring further investigation100-500Over-diagnoses (cancers detected and treated that wouldnot have caused any harm if left alone)2-10

Table 3: Summary of benefit and
harms for 1000 50-year old women screened each two years for 10 years, as
estimated by Welch (2009).

Another paper
in the British Medical Journal recently argued that women are not given full
information when considering screening, and a new leaflet has been written which
contains the following estimates of what would happen to 1,000 average women
going for screening.

BenefitsBreast cancer deaths avoided0.5Costs and harmsFalse-positive diagnoses requiring further investigation100Over-diagnoses (cancers detected and treated that wouldnot have caused any harm if left alone)5

Table 4: Summary of benefit and
harms for 1000 women screened each two years for ten years contained in a
proposed information leaflet for women deciding whether to go for breast cancer
screening.

The latter table suggests that 2,000 women need to be screened to save one
life over ten years, while ten of those 2,000 will have been unnecessarily
treated as cancer patients. These figures, however, have been strongly disputed.
Other researchers
point to evidence that six lives will be saved in 1,000 women screened over
20 years (similar to Table 2), while other studies "suggest that
over-diagnosis in mammography screening is a minor phenomenon".

There is also another problem with including the results from Table 4 in an
information leaflet for women. The table claims to estimate what will happen to
1,000 different women who are offered screening, some of whom will accept and
some not. But this seems inappropriate for an information leaflet: when a woman
is asked whether she wants to attend screening, the relevant information is the
benefit if she actually gets screened, not just offered it.

This is a real statistical issue - numerical evidence is being used to decide
important policies that can affect the lives of all women, yet the evidence from
different studies does not always agree and there is substantial scientific
uncertainty. This is partly because trials (whose results also feed into the
mathematical models) are done on different populations and in different ways,
and also because the number of women benefiting from mass screening is small, so
the benefit is hard to measure. Larger clinical trials would help, but it would
now be very difficult, and probably unethical, to have a control group that was
not offered screening. So it looks like careful statistical analysis and
mathematical modelling is likely to be vital.

Communicating the evidence

Of course, the final decision on whether to be screened depends on an
individual woman's feelings about the trade-off between the possibility of
benefit and the risks of harm. People find these decisions difficult, and tend
to be influenced by the opinions and behaviour of people around them whom they
trust. But women should still have an idea of the magnitudes of the numbers
involved in order to compare options.

It is often, however, difficult to make comparisons between the options. Not
just because outcomes of studies don't always agree, but also because different
ways of expressing benefits and harms are used. Even the recommendation by the
USPSTF presents the evidence in an unclear way, focusing on the percentage
reduction in risk, rather than the actual chance of a women benefiting from
screening - for example, a 50% reduction in risk might seem like a huge
improvement, but if the risk was very small to start with, then the chance for
an individual woman to benefit from screening may remain small. Fortunately, the
presentations of results are increasingly expressed in a common way - what is
expected to happen to 1,000 typical women - a unified approach which is
improving communication.

It may seem extraordinary that such disagreement and controversy exists in
such an important and high-profile issue. As mentioned before, this is partly
because the actual numbers of women who benefit from a mass screening programme
are fairly small, and so measuring benefit is difficult. But the controversy
about information leaflets may also reflect a caution in being too up-front to
patients about both the possibly small size of the benefit, and the fact that
there is so much uncertainty.

The current buzz-phrase is shared decision-making, which means that
patients should be making fully-informed choices in cooperation with their
doctor. This inevitably needs communication of the magnitudes of possible
benefits and harms, and so emphasises the role of robust and clear statistical
reasoning. These are difficult and delicate issues, but where numerical and
mathematical insights can be very valuable.

I don't understand the hesitation to promote breast self-examination?? I found my own lump quite by accident while showering....absolutely no family history.....my cancer was grade 3, very aggressive.  I consider myself lucky.

My teenage son recently had a physical and the doctor showed him how to exam himself for any testicular lumps and gave him a big lecture on the importance of self examination......explain the difference...?

Olivia...Statistics tries to prove logic. Many times logic will be refuted because the statistics based on evidence based medicine will prove otherwise. Regarding the data that jumps out regarding mortality, when clinical trials are formulated they must have measurable endpoints. In this situation the endpoint is mortality.

Voraciousreader,

I went ABD (all but disseration) for PhD in statistics many years ago.  I have quibble with the statement that you made in your answer to Olivia. It hit a couple of hot buttons from my time as  TA working with students and aiding researchers with experimental design.

Statistics doesn't prove logic -- it can show that a relationship is not likely to be caused by chance.  

Data relationships can jump out in statisical studies, but  correlation is not causation.  A recent posting in the research area illustrated this.  It found that daughters of BC survivors were diagnosed earlier than their mothers.   Does this mean that the disease presents earlier in the daughters or is it an artifact that the daughters are screened for the disease earlier than their mothers?  Or maybe a combination?

The large task force studies are meta-studies that combine the results of a large number of studies.  

To do this the studies have to be using the same measures.   The studies out there tend to treat BC as uniform disease and use mortality as the measure.

With any approach there are issues that come up. 

As most of us here are aware breast cancer has many variations and there are  vast differences in aggressiveness.  The most common type is  ER+/PR+ HER- (luminal A).    Most of us who are diagnosed with this are told that the cancer was there for 10 years before it was detectable.  It is largely seen in menopausal women.  If this were the only type of breast cancer, delaying mammograms to 50 might have a bit more support.

The other issue is using mortality as the sole measure of effectiveness of screening.  It's very important, but discounting the value of discovering  the cancer at more treatable, less lethal stage is troubling to many people. 

Studies using mortality as the measure, must by definition have a long duration.  There are confounding factors in interpreting this since both the sensiitivy of screening and the efficacy of treatment are improving at an accelerated rate.

There is no question that screening process  has problems, but it's currently the best we have to offer.    About three-quarters of us who are diagnosed have no risk factors other than being female and growing older. Too many are diagnosed initially  with advance stages

The University of Pennsylvania study you pointed to and some goals of ASCO of how to  better target the screening process. are examples of what needs to happen.

There needs to be more work done  in researching cures, prevention and better screening methods.

The question remains what are considered  "acceptable losses" for limiting access to our current imperfect methods while better ones are being found.

Finding  the right  balance is  tricky business. 

Megadotz...for folks who are unfamiliar with statistics and scientific methodology and the use of endpoints, there needs to be some simplification.  I wish I could explain Statistics to everyone...but then why would there be a class and why, why, why is THAT class so difficult to understand? 

All of you raise important points.  Yes. Choosing endpoints is very contentious and difficult.  Yes, if the endpoint is mortality, it might take a long time to determine if a treatment works.  Yes, we have surrogate endpoints that try to extrapolate important data.   Yes, most of the studies are based on meta-analysis that have their own drawbacks.  Yes, "acceptable losses" is tricky business when you're in the business of saving lives. Yes, we've been told FOREVER that early detection leads to less aggressive treatment and longer life...PERHAPS. And then again, maybe not. And THAT is the conundrum. I think the NCI/University of Pennsylvania announcement is probably the best news that we've heard in a long time. And at $7.5 million, I think it is a BARGAIN if it resolves the screening controversy ONCE AND FOR ALL.

 Beesie...Here's the statin controversy in a nutshell.  Kinda like the screening mammography controversy.....

Cochrane review stirs controversy over statins in primary prevention

                                        January 20, 2011                                                 Sue Hughes

inShare0

• 76Comments
• Read later
  • 
  • 
  • 
• Print
• Font sizeaAA
• Cite

London, UK - A new Cochrane review has provoked controversy by concluding that there is not enough evidence to recommend the widespread use of statins in the primary prevention of heart disease [1].

The authors of the new Cochrane meta-analysis, led by Dr Fiona Taylor (London School of Hygiene and Tropical Medicine, UK), issued a press release questioning the benefit of statins in primary prevention and suggesting that the previous data showing benefit may have been biased by industry-funded studies. This has led to headlines in many UK newspapers saying that the drugs are being overused and that millions of people are needlessly exposing themselves to potential side effects.

This has angered researchers who have conducted other large statin meta-analyses, who say the drugs are beneficial, even in the lowest-risk individuals, and their risk of side effects is negligible. They maintain that the Cochrane reviewers have misrepresented the data, which they say could have serious negative consequences for many patients currently taking these agents.

The Cochrane authors reviewed data from 14 trials involving 34 272 patients. Outcomes in patients given statins were compared with outcomes in patients given placebos or usual care. Although results suggested that deaths were reduced on statins, the researchers say the effect is not large enough to justify the cost/effort and risk of adverse effects.

Senior author Dr Shah Ebrahim (South Asia Network for Chronic Disease, New Delhi, India) told heartwire that their review differed from others done in primary prevention in that it looked at just those at low risk, limiting the studies included to just those with populations where <10% had a previous history of CVD.

It is probably a real effect but it means a lot of people have to be treated to gain this small benefit.

Ebrahim commented to heartwire: "If you look at the hard end points of all deaths and coronary deaths, the effects are consistent with both benefit and with the play of chance. But importantly, the absolute benefits are really rather small-1000 people have to be treated for one year to prevent one death. It is probably a real effect, but it means a lot of people have to be treated to gain this small benefit. As we don't know the harms, it seems wrong-minded to me to treat everyone with a statin. In these circumstances, lifestyle changes and stopping smoking would be far preferable."

I object to the conclusions they have drawn from their review.

But Dr Colin Baigent (Clinical Trials Service Unit, Oxford, UK) commented to heartwire: "I object to the conclusions they have drawn from their review. They say there is not good evidence of benefit, but their own data show significant reductions in deaths and cardiac events." And Baigent further objects to the Cochrane authors' suggestion that harms are not known with statins. "They didn't show any increase in adverse events in their review, but they then say the benefit is not worth the risk. That doesn't make sense."

Cochrane results

The Cochrane review showed that in the eight trials that reported on total mortality, none of the individual trials showed strong evidence of a reduction in total mortality, but when the data were pooled, a relative risk reduction of 17% was observed with statin treatment. On combined fatal and nonfatal CHD events, nine trials reported on this end point, with four trials showing evidence of a reduction in this combined outcome, which was maintained in the pooled analysis, with a 28% relative reduction. Seven trials reported on fatal and nonfatal stroke, and on pooled analysis, statin treatment was associated with a 22% relative reduction.

Cochrane review: Risk ratio of major events with statins in lower-risk primary-prevention patients

Outcome	Risk ratio (95% CI)
Total mortality	0.83 (0.73-0.95)
Fatal and nonfatal CHD events	0.72 (0.65-0.79)
Fatal and nonfatal stroke	0.78 (0.65-0.94)

No excess in combined adverse events, cancers, or specific biochemical markers were found.

The authors conclude: "This current systematic review highlights the shortcomings in the published trials of statins for primary prevention. Selective reporting and inclusion of people with cardiovascular disease in many of the trials . . . in previous reviews of [statins'] role in primary prevention make the evidence impossible to disentangle without individual patient data."

They say that in people at high risk of cardiovascular events (>20% 10-year risk), "it is likely that the benefits of statins are greater than potential short-term harms, although long-term effects (over decades) remain unknown." They conclude: "Any decision to use statins for primary prevention should be made cautiously and in the light of an assessment of the patient's overall cardiovascular risk profile. Widespread use of statins in people at low risk of cardiovascular events-below a 1% annual all-cause mortality risk or an annual CVD event rate of below 2% observed in the control groups in the trials considered here-is not supported by the existing evidence."

Latest Oxford meta-analysis not included

The Cochrane review did not include the recent meta-analysis from the Oxford group, published late last year, which showed a clear reduction in events with statin therapy in primary-prevention patients. Baigent noted that this meta-analysis was more reliable than the Cochrane review, as the Oxford researchers used individual patient data from all the trials. "Our 2010 meta-analysis in primary prevention is substantially more complete than the Cochrane review and provides direct and overwhelmingly statistically convincing evidence of a clear reduction in events in all patient groups, right down to those at the lowest risk."

On the possible hazards of taking these drugs, Baigent says: "Statin therapy is very safe. The most serious hazard, rhabdomyolysis, is very rare, and most often seen at high doses. There is a possibility that reducing LDL cholesterol might increase the risk of hemorrhagic stroke, but even in primary prevention these hazards would be much smaller than the benefits, and there is no reliable evidence for other hazards mentioned by the Cochrane authors, such as depression and cognitive impairment."

It all comes down to economics

Baigent says the only argument against using statins in low-risk people is economic. "The absolute benefits of statin therapy become very small when used among people at low absolute risk, so it is important that the costs of such treatment are considered when weighing how widely statins should be used. That is a government decision."

In the UK, the National Institute for Clinical Excellence [NICE] currently recommends that statins not be used for people with a CHD risk below 20% over 10 years. Ebrahim says the Cochrane conclusions are in line with this.

But Baigent argues that the benefits of statins are clear at levels far below this threshold. "Whether or not it is economic to use them in the lowest-risk individuals is not for me to say, but generic statins are now very cheap, and there is clear evidence of benefit and safety based on substantial numbers of individuals studied in large-scale trials. So, when all the relevant randomized evidence is considered, there does not seem to me to be any justification at all for the Cochrane authors' claim that the evidence is unclear on this issue."

Educational programs also of little benefit In a separate Cochrane review [2], the same group looked at the use of "healthy heart programs" that use counseling and educational methods to encourage people to reduce their risks for developing heart disease. These risk factors include high cholesterol, excessive salt intake, high blood pressure, excess weight, a high-fat diet, smoking, diabetes, and a sedentary lifestyle. They reviewed 55 trials that aimed to reduce more than one risk factor in people without evidence of cardiovascular disease. Results showed that after a median duration of 12 months of follow-up, multiple risk-factor intervention was associated with small reductions in risk factors, including blood pressure, cholesterol, and smoking, but had little or no impact on the risk of coronary heart disease mortality or morbidity. They conclude: "The methods of attempting behavior change in the general population are limited and do not appear to be effective. Different approaches to behavior change are needed and should be tested empirically before being widely promoted, particularly in developing countries where cardiovascular disease rates are rising." In an accompanying editorial [3], Dr Carl Heneghan (University of Oxford, UK) suggests an alternative approach for policy is to focus on populationwide prevention. He reports that "legislating for smoke-free public spaces, redesigning public spaces to improve exercise, or reducing daily dietary salt intake prove generally effective and can be cost-saving interventions. Given the scale of the worldwide CVD problem, large-scale commissioned studies of multiple risk-factor interventions are urgently required."

• inShare0
  
  

« Previous heartwire article
Plaques linked to future events have signature features; clinical implications unclear: PROSPECT
Jan 19, 2011 17:00 ESTNext heartwire article »
Less may be more in post-heart-transplant immunosuppression trial: TICTAC
Jan 20, 2011 11:30 ESTSources

1. Taylor F, Ward K, Moore THM, et al. Statins for the primary prevention of cardiovascular disease. Cochrane Database Syst Rev 2011; 1 (CD004816).
2. Ebrahim S, Taylor F, Ward K et al. Multiple risk factor interventions for primary prevention of coronary heart disease. Cochrane Database Syst Rev 2011; 1 (CD001561).
3. Heneghan C. Considerable uncertainty remains in the evidence for primary prevention of cardiovascular disease [editorial]. Cochrane Libr 2011 (January 19, 2011). Available here.

Related links

• Coronary calcium weeds out JUPITER-like patients who will benefit from statins, beats CRP for risk prediction
  [Lipid/Metabolic > Lipid/Metabolic; Nov 19, 2010]
  
• ASCOT analysis fuels debate over JUPITER-based CRP indication for statins
  [Lipid/Metabolic > Lipid/Metabolic; Nov 18, 2010]
  
• Lower is better for LDL even at very low levels: New meta-analysis
  [Lipid/Metabolic > Lipid/Metabolic; Nov 08, 2010]
  
• High-dose statin therapy reduces recurrent cardiovascular events
  [Lipid/Metabolic > Lipid/Metabolic; Dec 14, 2009]
  
• JUPITER: Low LDL and low CRP best for reducing events in primary prevention
  [Lipid/Metabolic > Lipid/Metabolic; Mar 29, 2009]
  
• JUPITER hits New Orleans: Landmark study shows statins benefit healthy individuals with high CRP levels
  [Clinical cardiology > Clinical cardiology; Nov 09, 2008]
  
• SEARCH me: Null trial "morphed" into positive meta-analysis
  [Lipid/Metabolic > Lipid/Metabolic; Nov 09, 2008]
  
• ENHANCE published, presented, discussed, and debated: Experts mull over what the findings mean
  [Lipid/Metabolic > Lipid/Metabolic; Mar 30, 2008]
  
• A "sea change" in the prevention of CVD: Experts discuss the PROVE-IT trial
  [Lipid/Metabolic > Lipid/Metabolic; Mar 08, 2004]
  

Your comments <td valign="top" bac

	Cochrane review stirs controversy over statins in primary prevention
	# 1 of 76	January 20, 2011 12:59 (EST)
	RICHARD KONES  MD	Statins have a major and important role in primary prevention...                             Given the immense threat of a growing cardiovascular risk burden in our population, the prudent but practical physician must treat his/her patients now, using the best evidence available, even if it is not ideal, rather than chose inaction. This study adds another meta-analysis to those available concerning the effectiveness of statins in primary prevention. The imperfections in these trials are well known. In the August, 2010 issue of the Archives of Internal Medicine, Ray et al concluded that statins did not reduce mortality in primary prevention patients. In this Cochrane review, methodological problems are cited in such studies, even though benefits in hard end points were realized. Having read the first review, one is left to doubt whether statins are worthwhile, thus favoring primordial prevention. However, in a second related Cochrane review, the effects of counseling, education, and "programs" upon lifestyle behavior were also termed "dubious." The seasoned cardiologist will not be surprised at either conclusion. What is the reasonable, committed physician to do in the interim, wait years to begin treatment while atherosclerosis disseminates and progresses?  Should we not bother to talk about lifestyle change or risk factors to patients while we also withhold statins, since both are "unclear "interventions? ACCF/AHA and NCEP guidelines, supported by major professional organizations, advise the opposite, are evidence-based, and do not offend common sense. The epidemiological and pathological evidence suggesting that the incubation period for CHD spans decades is strong and accepted. Tracking of risk factors in the Bogalusa Heart study, combined with additional data comparing worsening CIMT and CAC scores between youth and adulthood in proportion to the risk burden, clearly connect the dots. For several reasons, there will not be a study in the near future that definitively demonstrates statin treatment in a low-risk primary prevention population is an air-tight method of reducing hard end-points, including mortality, at a cost acceptable to everyone. Some models indicate it may do so under given conditions, but RCTs, no. Rather than chose inaction, which would be disastrous in years to come, we should continue both methods-primordial prevention, and intensive evaluation followed by aggressive treatment in high-risk patients. The techniques used for each and cutoffs need continuous reevaluation and improvement, to be sure. The best evidence, while admittedly relying on logical inference, indicates statins should be used, since there are no equals. The safety of statins was not an issue in the first review, and in the second, the behavior change observed was simply insufficient. Assuming that behavior change in humans is impossible has serious social implications. Acting on that assumption is irresponsible. Richard Kones MD
	# 2 of 76	January 20, 2011 01:00 (EST)
	RICHARD KONES  MD	Statins have a major and important role in primary prevention...                             Substantial support now exists for earlier and more aggressive therapy of risk factors, rather than waiting because of "lack of evidence", "clinical inertia", or indecision due to lack of familiarity with current data or fear of potential (and exaggerated) side effects. In fact, failure to treat early enough, and/or with sufficiently potent statins contributes to the lack of progress in lowering consequences of uncontrolled CV risk factors, rather than the opposite, as suggested in these reviews.  Similar data exist with respect to the management of other risk factors, such as hypertension, diabetes, inflammation. Failure to address all risk factors completely leaves "residual risk" unchallenged. Failing to include the Baigent study, and inadequately weighing results of JUPITER and other supportive evidence, has resulted in a conclusion which defies logic and reality. Perhaps economic considerations played a part in this process, but even then, other data confirm that statin-based and primordial prevention are in fact cost-effective. Therapeutic nihilism based upon both these reviews is unwarranted and potentially dangerous. Waiting for "clarification" involves far greater peril than beginning statins when needed and intensifying primordial prevention efforts on a national scale. The evidence for primordial prevention is summarized at  The evidence for early and aggressive reduction of risk factors is outlined at   and  Richard Kones MD
	# 3 of 76	January 20, 2011 01:44 (EST)
	Don Kelsey	All-cause mortality                             What has struck me in several statin studies is that, even though CHD and other events were modestly reduced, all-cause mortality HR still ended up very close to 1.0.  I first noticed this in the comparison study of 10mg to 80mg Lipitor for the subgroup 65 and older.  And in the article above "none of the individual trials showed strong evidence of a reduction in total mortality".  Even a recent study of fenofibrate reported in HeartWire, the total mortality was 1.0 even though the CHD events were reduced by, I think, 13%.  So, I wonder if artificially lowering cholesterol is really just shifting the total biochemistry in such a way as to aggravate, and perhaps accelerate, other diseases? And the notion that rhabdomyolysis is the only "serious" side effect of statins is ludicrous. The literature is ripe with estimates that somewhere between 5 and 15% experience myalgia and myopathy, which can be serious enough to cause quality of life issues and discontinuance of medication. The drug companies (and many researchers) have underplayed or even ignored the side effects of statins and, eventually, this will come back to haunt them.
	# 4 of 76	January 20, 2011 02:07 (EST)
	Jon Whitney	Increased risk of diabetes.                             What about the Jupiter study that showed a 25% increase in adult-onset diabetes in the group on statin medication?
	# 5 of 76	January 20, 2011 02:35 (EST)
	Jared Johnson	Behavioral  Counseling is counter to current science                             The most glaring is the high-fat diet, as this has never been shown to even associate with CHD definitively. In fact, many recent studies on low-carb (high-fat) diets show greater improvement in cardiovascular risk factors than low-fat, calorie-restricted diets. It is my belief that correct behavioral counseling will have a much greater impact on not only CHD and CVD risk, but also overall mortality. Put in terms of absolute risk reduction, statin effectiveness is ridiculously lacking. Even if the 1 in 1000 statistic were off by an order of magnitude I don't believe the cost to be justified. There can be little doubt that the bulk of the benefit for statin use as primary prevention is to drug companies and the doctors whose practices are supported by this relatively useless 'medicine.'
	# 6 of 76	January 20, 2011 02:41 (EST)
	david nott	supposed statin benefits                             Several cadriologists have assured me that statins (in this case Lipitor) have positive effect on the walls of heart arteries and, separately,  the covering carapace on fat deposits adhering to them. Why don't the manufacturers say so in their fact sheets and websites? They talk only of reducing bodyfat. Who is right?
	# 7 of 76	January 20, 2011 03:03 (EST)
	Allen Leier	Ultimately NO Drugs is the best alternative                             We are continuously trying to come up with a pill to fix all ills and save people from themselves.  Is it not time we hammered home that life style is most indiviuals biggest problem.  If they are not going to help themselves by quiting smoking, over eating and not getting any exercise then they should be left alone to suffer the consequences of their own lack of interest in their health.  All we are doing is making the drug companies rich and the medicare system over burdend with drug costs which are ten to twenty times as expensive as they should be.
	# 8 of 76	January 20, 2011 03:11 (EST)
	Ibrahim Mubarak	Therapeutics Letter #48                             Therapeutics Letter Issue 77 / Mar - Apr 2010 in continuation with there first review  few years ago   come with the same conclusion of the Cochrane group in their review both worth reading. from well-respected non profited journal
	# 9 of 76	January 20, 2011 03:39 (EST)
	William Feeman, Jr	Primary Prevention                                  The position of Taylor, et al, is that many people are taking statins who don't need to be.  This position, typically British, is not new, and indeed is just another part of the "the Cholesterol Wars."  What is really needed is a means of identifying the population at risk of atherothrombotic disease (ATD).  I have no quarrel with Taylor, as long as she is not mis-quoted and as long as her suggestions are not overly broad.  At any given time, not everyone should be on a statin, which is why I would not put statins in the drinking water.      To identify the population at risk of ATD, may I suggest the online version of Advanced Studies in Medicine, 2005; 5(1): 20ff.  This letter also shows how to predict when the ATD event will begin.      Just so no one makes a mistake, I have fought on the "pro" side of the Cholesterol Wars for decades.  Heart attacks and strokes are uncommon to rare in my practice.
	# 10 of 76	January 20, 2011 04:58 (EST)
	James J. King	Primary or Secondary?                             Is treating patients with statins who have a high Calcium Score of > 40  AJ called Primary or Secondary statin therapy?
	# 11 of 76	January 20, 2011 05:15 (EST)
	Colin Rose	Statins for high calcium scores in primary prevention                             There are at least three good trials showing that treating patients with high calcium scores with statins has no effect either on the progression of the score or in clinical events.
	# 12 of 76	January 20, 2011 05:33 (EST)

megadotz: re your comment on daughters of bc women being diagnosed younger than their mothers were. This is why my GP began ordering mammograms for me when I was in my early 30s, because my mother developed bc at 38. If my mother carried a genetic mutation that caused her bc and if I inherited it (we didn't have genetic testing then), I would develop bc earlier than 38. I don't think these new guidelines affect daughters at high risk from having mammograms, US or MRIs. The case could be made from my result that I had 25 years of unnecessary mammograms, but studies don't consider such factors as peace of mind. On the other side of the fence, consider that because 80% of women diagnosed with bc have no family history whatsoever of the disease. No family history should mean no mammogram until you are 50? Life is a crap shoot. You may not be able to statistically quantify the wild cards in life or peace of mind in a medical study, but they sure matter for a person whose cancer is found years after it could have been dealt with.

I have found doctors to generally be scientists and conservative first and considerate of the human factors in treatment second. They base treatments on actual test results, first-hand experience and standard care as set by medical communities/governing bodies. Medical advice changes over time as more is known about various diseases. My doctor thought I'd stay on HRT for decades given how young I was when I went through menopause. Knowing more now about the connection between HRT and bc, I am glad I took myself off as soon as I could. I am concerned about too much testing and over-treating, but surely there must be a balance somewhere between effective prevention/early treatment and pulling right back from detection tests for women under 50. 

 Here's a little information about how endpoints are chosen in clinical trials.  While it doesn't address meta-analysis, you can see the relevance of "endpoints."  According to this article, in cancer research, the "gold standard" endpoint is survival....

Whenever clinical study results are publicly announced, the endpoints are the
focus; more specifically, the focus is on whether the clinical trial met or
failed its primary endpoint. Clinical trial endpoints are important because they
are proof that the drug does what its manufacturer claims it does for the
patient. Even though clinical trial endpoints are measurements at an individual
patient level, the results are extrapolated to entire populations of patients
based on clinical similarities to patients in the clinical trials. The Food and
Drug Administration bases approval decisions on clinical trial
endpoints.

The most important requirement for a clinical trial endpoint
is clinical relevance, or what the effect means to the patient. Appropriate
measurement or detection of a clinical endpoint is secondary to the appropriate
selection of a clinically relevant endpoint; measuring an endpoint well means
nothing to the patient if the endpoint has poor clinical relevance. In addition
to clinical relevance, trial endpoints should reflect the research question, the
drug's mechanism (or mechanisms) of action and patient well-being. In phase I
clinical studies, safety is the primary endpoint; in phase II clinical studies,
a measure of efficacy is the primary endpoint. Quality of life (QOL) is becoming
an increasingly important endpoint in chronic or terminal conditions, including
cancer.

Clinical trials should have only one primary endpoint -- hence
the "primary" designation. The primary endpoint should be the most meaningful
result in a clinical trial. An example of a primary endpoint in cancer clinical
trials is survival. Survival has long been considered the "gold standard"
endpoint for cancer therapies, based on the importance of survival to cancer
patients. Other results that contribute to the interpretation of the clinical
trial are then considered "secondary endpoints." A clinical trial can have, and
often does have, many secondary endpoints. Still, when a clinical trial attempts
to capture too many endpoints, complications can arise that impede the critical
goal of answering a primary question. If multiple endpoints yield conflicting
results, the study cannot be interpreted meaningfully. Multiple endpoints also
increase the chance of a Type 1 error (when a difference is shown when there is
no actual difference).

Composite
endpoints

Composite endpoints are used in studies of diseases that
occur rarely or in studies that may require a sample size so large that the
studies cannot be completed in a reasonable amount of time. A combination of
measurements can be examined and assigned as "an" endpoint. Components of the
composite endpoint must be related to a common pathophysiology and be closely
related in the degree of severity. For example, a composite endpoint in
cardiovascular disease can include all-cause mortality, fatal coronary heart
disease and non-fatal myocardial infarction; all of these components are
considered closely related in the degree of severity, such that an occurrence of
one or more of these component endpoints is as serious as death (mortality).
Composite endpoints can sometimes yield conflicting results that complicate
interpretation of the study. For example, if one treatment group has a higher
event rate in one component but another treatment group shows a higher event
rate in another component, the interpretation of the study becomes less
clear-cut than comparing singular endpoints between treatment
groups.

Quality of life endpoints may be considered "missed
opportunities" in clinical trials, but QOL endpoints are not often used because
defining QOL parameters can be a challenge. In addition, assessment by patients
is subjective (one can debate that the patients' opinions are ultimately what
matter). Quality of life may capture perceptions that are not always
study-related, and can affect the results of a study (for example, a
life-changing event can affect a patient on many physiological and psychological
levels). Still, QOL is especially relevant in therapies that can be toxic, such
as cancer therapies. I remember reading a story from the husband of a cancer
patient: His wife fell into a coma as a result of the toxicities of a treatment,
but did not die until many months later. This patient was considered a
statistical "success story" based on survival, yet this example begs the
question of whether survival with poor QOL is better than death without
suffering.

Surrogate endpoints are increasingly used in lieu of the
"true" endpoints of disease eradication or survival, and the role of surrogate
endpoints in clinical trials is a controversial subject. Disease eradication and
survival can take years to prove. Surrogate endpoints in clinical trials lead
not only to shorter clinical trial timelines, but also to less cost. Surrogate
endpoints also reduce follow-up lapses that can be common in long-term studies:
Patients may miss visits and not follow up, which leads to missing data and
suboptimal study analysis. Examples of surrogate endpoints are blood pressure
effects or prostate-specific antigen levels. The use of surrogate endpoints in
clinical trials requires researchers to identify the appropriate surrogate
markers, which is not an easy task. Surrogate endpoints are useful if they can
be easily measured and are highly correlative with direct endpoints of clinical
relevance. Surrogate endpoints are usually closer to the biology of the disease
and may not always correlate with clinical outcomes.

Oliva,

I'm sorry if what I said came across as dismissive, it certainly wasn't intended that way. It was that some studies claim one conclusion, when there are several that may be.  A better example might be a study that looked for a genetic reason for why redheads had a lower  rate  skin cancer, when a plausible alternative might be that with sun sensitive skin, they tend to look for shade sooner.

What I was trying to say about the breast cancer study, not very effectively, was that with current methods that perhaps your mother might have been detected earlier.   I happen to think it's very good idea to screen early  for any one whose risk factor has been identified.  We need better ways to identify that risk.

New information and new techniques are coming forward all the time.   I'm concerned that recent testing guidelines, we are in danger abandoning the good in search of the perfect  and  putting those who would  identified to receive needed treatment at risk.

I think we're actually on the same page for this.  I just stated things poorly.

All the best,

Meg 

  

  From the Journal of The National Institute of Cancer:

  

Screening Trials Are Even More Difficult Than We Thought They WereIn this issue of the Journal, Black et al. (1) present an important analysis of methodologic pitfalls associated with randomized studies of screening interventions. The                  authors compare disease-specific and all-cause mortality from the 12 published randomized trials of cancer screening for which                  these end points were available. In seven of the 12 studies, major inconsistencies were detected in the direction or magnitude                  of these two outcomes. Black et al. propose that the use of disease-specific mortality as the primary end point renders screening                  trials subject to at least two forms of serious bias:               Sticky-diagnosis bias-where deaths from other causes in the screened group are wrongly attributed to the target cancer or deaths in the control                        group are wrongly attributed to other causes. This bias favors the control group.Slippery-linkage bias-where deaths due to diagnostic or therapeutic interventions that are stimulated by the screening test are not included in                        disease-specific mortality. This bias favors the screened group.Because both of these sources of bias arise from determination of cause of death, they do not affect all-cause mortality.                  This provides a strong argument for the use of all-cause mortality as the primary endpoint in screening trials.               The comparison of all-cause mortality and disease-specific mortality in the article by Black et al. (1) suggests that the net effect of these biases has favored screening and that slippery linkage is more important than sticky                  diagnosis. Slippery linkage is also inherently more important because it recognizes that screening has the capacity to cause death as well as to prevent it. In any screening study (or in the population to which its results are applied), the risk of the screened population dying                  of the target disease is low. Even if screening is "effective" in detecting disease in a phase at which it is still curable,                  a large number of subjects must be evaluated, and a substantial number of those must be treated to save one life from cancer.                  Although modern diagnostic tests and even cancer operations appear to be remarkably safe, rare fatal complications do occur,                  and more subtle effects to hasten death through cardiovascular or other causes may be missed completely. Thus, there may be                  a fine balance between benefit and harm from screening.               Disease-specific mortality rather than all-cause mortality has been the accepted end point of screening studies because fewer                  patients are required to provide adequate power. It has been assumed that disease-specific mortality is a good surrogate end                  point for all-cause mortality, but the current study raises serious doubts about the validity of this assumption. Whether                  measured directly or not, a decrease in all-cause mortality should be the ultimate aim of screening programs. A death from                  a nonmalignant cause is just as important as a cancer-related death.               What is the feasibility of all-cause mortality as a primary end point? Seven of the 12 studies assessed in the article by                  Black et al. (1) were of screening mammography for breast cancer. Because screening mammography is widely available in Western countries, performing a randomized study without substantial contamination in the observed arm may now be impossible. Therefore, careful attention needs to be paid to the available data. Some caution should be exercised in interpreting all-cause mortality in studies that were not prospectively designed to address this end point. Thus, as acknowledged by the authors, some of the inconsistencies observed may be the result of chance alone. Pooling of the available data in a meta-analysis may overcome this problem. However, in the controversial article published by Gotzsche and Olsen (2), only two of eight published randomized trials of screening mammography were thought to be of sufficient quality to include                  in a meta-analysis. Neither of these studies demonstrated a benefit for screening mammography, leading the authors to conclude                  that there is no evidence that breast cancer screening reduces mortality. This article has sparked intense debate both in                  the lay press and in scientific literature. One of the criticisms was the authors' emphasis on all-cause mortality as an end                  point (3), but the article by Black et al. (1) adds substantial credence to that approach.               The methodologic lessons provided by the analysis of Black et al. (1) should be applied to future trials. One recommendation of the authors is that future screening studies concentrate on high-risk                  groups. This recommendation has the advantage of requiring a smaller number of patients to detect statistically significant                  differences between study groups. Unfortunately, studies of high-risk patients may not be applicable to the general population                  because of differences in patient characteristics, the biology of disease, and the level of risk. For example, women at high                  risk of breast cancer, such as known carriers of BRCA1 and BRCA2 gene mutations, are at risk of developing the disease from                  a young age. The reduced sensitivity of mammography in premenopausal women is a well-documented problem (4,5), and the behavior of these cancers may be different. Therefore, even a well-conducted randomized study of screening for breast                  cancer in such high-risk women with all-cause mortality as an end point will not answer the question of whether screening                  is beneficial to the general population. Screening programs should be implemented only in the populations in which they have                  shown benefit in a clinical trial.               The concept of slippery-linkage bias applies equally to prevention trials in which mortality related to unexpected complications                  of the intervention may go unrecognized if all-cause mortality is not reported. For example, the National Surgical Adjuvant                  Breast and Bowel Project (NSABP) P-1 study has led to the widespread introduction of the use of tamoxifen for prevention of                  breast cancer in the United States, based on a demonstrated reduction in the incidence of breast cancer (6). This event is an even more proximal surrogate end point than disease-specific mortality, which was not different between                  the two arms of the NSABP study. Proponents argue that a decrease in incidence will translate into a difference in disease-specific survival, which in turn will translate into a difference in all-cause survival. However, tamoxifen can cause harm (a low incidence of endometrial cancer                  and thromboembolic disease) as well as benefit, and the above reasoning seems to be a tenuous justification for its use in                  well women. Both the Study of Tamoxifen and Raloxifene for the Prevention of Breast Cancer and the Randomized Study of Selenium                  and Vitamin E for the Prevention of Prostate Cancer have incidence rates of the respective cancers as their primary endpoints.                  Even if these large and expensive trials demonstrate a decrease in cancer incidence, this finding is not necessarily proof                  of lives saved.               The biases demonstrated in the article by Black et al. (1) may make it almost impossible to demonstrate (or rule out) the value of screening for prostate cancer. The potential for                  misdiagnosis of the cause of death and mortality as a direct or indirect result of procedures resulting from screening will                  be particularly high in a screened population of elderly men. Because of competing causes of death, the influence of cancer                  screening on all-cause mortality will be difficult to detect, whereas cause-specific mortality will be subject to these types                  of bias. The International Prostate Screening Trials Evaluation Group is a collaboration of multiple centers conducting randomized                  trials of prostate cancer screening. It is hoped that the pooled sample will be about 200 000 men, with a follow-up period                  of at least 10 years. Even with this large sample and long follow-up, the primary end point is mortality from prostate cancer.               Population-based screening trials that are designed with improvement in all-cause mortality as the primary end point will                  require very large numbers of patients, lengthy follow-up, and great expense. However, we cannot justify implementation of                  screening programs that are costly to the individual and to the community if we are uncertain of their true benefit.