Ben Goldacre 'Bad Pharma' - Cochrane 2011 report 'Statins for the primary prevention of cardiovascular disease'

reposted from:

crabsallover highlights, key points, comments / links.

Ben Goldacres' 'Bad Pharma' mentions Cochrane as a independent systematic reviewer of trial results.

I blogged about The Lancets review of the Cochrane systematic study 'Statins for the primary prevention of cardiovascular disease' in January 2011 but until now I've not looked at the original Cochrane review.

The Cochrane website gives the 2011 Abstract, full Article and previous versions with free pdf downloads.

Abstract

Background

Reducing high blood cholesterol, a risk factor for cardiovascular disease (CVD) events in people with and without a past history of coronary heart disease (CHD) is an important goal of pharmacotherapy. Statins are the first-choice agents. Previous reviews of the effects of statins have highlighted their benefits in people with coronary artery disease. The case for primary prevention, however, is less clear.

Objectives

To assess the effects, both harms and benefits, of statins in people with no history of CVD.

Search methods

To avoid duplication of effort, we checked reference lists of previous systematic reviews. We searched the Cochrane Central Register of Controlled Trials (Issue 1, 2007), MEDLINE (2001 to March 2007) and EMBASE (2003 to March 2007). There were no language restrictions.

Selection criteria

Randomised controlled trials of statins with minimum duration of one year and follow-up of six months, in adults with no restrictions on their total low density lipoprotein (LDL) or high density lipoprotein (HDL) cholesterol levels, and where 10% or less had a history of CVD, were included.

Data collection and analysis

Two authors independently selected studies for inclusion and extracted data. Outcomes included all cause mortality, fatal and non-fatal CHD, CVD and stroke events, combined endpoints (fatal and non-fatal CHD, CVD and stroke events), change in blood total cholesterol concentration, revascularisation, adverse events, quality of life and costs. Relative risk (RR) was calculated for dichotomous data, and for continuous data pooled weighted mean differences (with 95% confidence intervals) were calculated.

Main results

Fourteen randomised control trials (16 trial arms; 34,272 participants) were included. Eleven trials recruited patients with specific conditions (raised lipids, diabetes, hypertension, microalbuminuria). All-cause mortality was reduced by statins (RR 0.84, 95% CI 0.73 to 0.96) as was combined fatal and non-fatal CVD endpoints (RR 0.70, 95% CI 0.61 to 0.79). Benefits were also seen in the reduction of revascularisation rates (RR 0.66, 95% CI 0.53 to 0.83). Total cholesterol and LDL cholesterol were reduced in all trials but there was evidence of heterogeneity of effects. There was no clear evidence of any significant harm caused by statin prescription or of effects on patient quality of life.

Authors' conclusions

Reductions in all-cause mortality, major vascular events and revascularisations were found with no excess of cancers or muscle pain among people without evidence of cardiovascular disease treated with statins.  Other potential adverse events were not reported and some trials included people with cardiovascular disease. Only limited evidence showed that primary prevention with statins may be cost effective and improve patient quality of life. Caution should be taken in prescribing statins for primary prevention among people at low cardiovascular risk.

Plain language summary

Statins for the primary prevention of cardiovascular disease

Cardiovascular disease (CVD) is ranked as the number one cause of mortality and is a major cause of morbidity world wide. Reducing high blood cholesterol which is a risk factor for CVD events is an important goal of medical treatment. Statins are the first-choice agents. Since the early statin trials were reported, several reviews of the effects of statins have been published highlighting their benefits particularly in people with a past history of CVD. However for people without a past history of CVD (primary prevention), the evidence is less clear. The aim of this systematic review is to assess the effects, both in terms of benefits and harms of statins for the primary prevention of CVD. We searched the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE and EMBASE until 2007. We found 14 randomised control trials with 16 trial arms (34,272 patients) dating from 1994 to 2006. All were randomised control trials comparing statins with usual care or placebo. Duration of treatment was minimum one year and with follow up of a minimum of six months. All cause mortality. coronary heart disease and stroke events were reduced with the use of statins as was the need for revascularisations. Statin treatment reduced blood cholesterol. Taking statins did not increase the risk of adverse effects such as cancer. and few trials reported on costs or quality of life. This current systematic review highlights the shortcomings in the published trials and we recommend that caution should be taken in prescribing statins for primary prevention among people at low cardiovascular risk.

Full Report

Very detailed. A Key paragraph is:-

On the basis of our systematic review and these recent meta-analyses, it is clear that 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. Furthermore, the tendency of trial protocols to remove patients suffering with co-morbidities limits their generalisability to typical patient populations in whom decisions to prescribe statins have to made.

Further analysis of this report and the CTT 2012 report is required to be made by crabsallover.

Statins—should we adjust the risk:benefit ratio?

reposted from: http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(11)60111-6/fulltext?rss=yes

crabsallover highlights, key points, comments / links.

The Lancet

The prevailing opinion that statins are an elixir for long life was challenged with the recent publication of the Cochrane review, Statins for the primary prevention of cardiovascular disease. After analysing 16 trial arms with 34 272 participants, the authors found no evidence of harm, and mortality, composite cardiovascular endpoints, and revascularisations were reduced. But they concluded: “caution should be taken in prescribing statins for primary prevention among people at low cardiovascular risk.”

The conclusion seems at odds with the findings. However, the authors found the evidence of insufficient quality to allow them to conclude differently; many trials included patients with a previous cardiovascular event, and the authors state that poor reporting of adverse events and selective reporting of outcomes contributed to their failure to draw a positive conclusion. They state that the evidence is “impossible to disentangle without individual patient data”.

The authors' conclusion is also at odds with the recent Cholesterol Treatment Trialists' (CTT) Collaboration meta-analysis, published in The Lancet. Individual patient data from around 70 000 patients with no previous cardiovascular disease were analysed, and showed that statins significantly reduced the relative risk of a cardiovascular event by 0·75 per 1 mmol fall in LDL cholesterol.

Unfortunately, the media are quick to forget, and have reported the Cochrane's headline-grabbing details with scant regard for the preceding evidence. This approach will have left many patients at best bewildered. The media have also largely ignored the Cochrane authors' conclusion that the results support guidelines from the UK's National Institute of Health and Clinical Excellence, which recommends statins are considered for primary prevention in patients with an annual incidence of cardiovascular events of over 2%.

So what should general practitioners (GPs), who are faced with increasing numbers of low-risk patients worried about mildly raised cholesterol, do? The Cochrane review has muddied the water, but the available evidence shows that statins are safe, and evidence from the CTT Collaboration also shows that reductions in cholesterol per se can produce benefits. So, the simple answer is that GPs should do what they always have done—clearly explain the risks and benefits to patients so that the patient is able to choose the strategy that is best for them.

Statin benefit for low risk people 'questionable'

crabsallover highlights, key points, comments / links.

Listen to PM 19/1/11 (25'30" - 30' 30") on BBC iPlayer

Prof. Shah Ibrahim (study author) said

• half the trials didn't give data on adverse effects
• all trials but one where paid for by industry

Prof. Nicholas Wald, Wolfson Institute of Preventive Medicine

• data of trials show benefits of taking statins
• 20-30% reduction in heart attacks and strokes
• 16% reduction in (all-cause) total mortality

Risk Ratio 0.84 = 16% reduction in mortality due to statins

• benefits are way ahead of any risks
• not sensible to probe trials for risks
• why not?
• severe muscle problems occur 1 per 100,000 per year
• the trials are not designed to show these risks - why not?
• 20-30 year olds should not take statins
• risk of heart attack or stroke doubles every 7 years
• 1 in 2 will have a heart attack or stroke
• 1 in 3 will die from a heart attack or stroke
• taking statins together with blood pressure lowering drugs can reduce risk of heart attacks and strokes if you are in your 50s or older
• ignor bias from fact that studies are paid by industry - they have been undertaken by reputable researchers

reposted from: http://www.nhs.uk/news/2011/01January/Pages/cholesterol-lowering-statin-drugs-examined.aspx

“Up to three million people are taking statins needlessly,” says The Daily Telegraph. It reports that a comprehensive study suggests statins are “ineffective in many cases and could be doing more harm than good”.

The news story is based on a review of trials of statins in people who had not (yet) suffered a cardiovascular event such as a heart attack or stroke. There was some evidence that statins reduced the risk of dying from any cause, and the risk of any cardiovascular outcome. However, the trials and review have several limitations, including some indications that adverse events within the trials were not recorded.

It is important to point out that the benefit of statins in people with cardiovascular disease, who have already suffered a heart attack or stroke, or who are considered to be at high risk of an event, is not in question here.

This review supports the need for careful consideration of the overall cardiovascular risk of the individual when deciding whether to prescribe a statin. In higher-risk populations, the benefits of a drug often clearly outweigh the risks. However, when lower-risk populations are considered, this balance can often tip the other way. The results here do not support the widespread use of statins in people at low risk of cardiovascular events.

Where did the story come from?

The news reports follow a Cochrane systematic review conducted by researchers from the London School of Hygiene and Tropical Medicine and the University of Bristol.

The main conclusion of this review is that there is a lack of quality evidence to support the use of statins in people with low cardiovascular risk. This has been generally reflected in the articles by The Daily Telegraph, the Daily Mirror and the Daily Express. However, the Daily Mail’s headline (“Statins 'may cause loss of memory and depression'”) is incorrect. The researchers’ main concern is that there is not enough reporting of adverse events, not that there is evidence for any particular harm.

What kind of research was this?

This Cochrane systematic review and meta-analysis investigated whether statins lower blood cholesterol, thereby reducing cardiovascular risk in people without a history of coronary heart disease (known as “primary prevention”). There is already clear evidence of their benefit in people who have already suffered a heart attack or stroke (known as “secondary prevention”).

A high quality systematic review that searches the medical literature to identify all relevant randomised controlled trials of a particular intervention is the most reliable method of assessing the evidence of its safety and effectiveness. Systematic reviews do have some inherent limitations, in that they rely on individual studies with varied quality, methods, outcomes and follow-up.

What did the research involve?

The researchers searched medical databases for all randomised controlled trials of at least 12 months of statin treatment compared with placebo or usual care, with at least another six months follow-up. To be eligible, trials had to have focused mainly on primary prevention, with less than 10% of the participants having a history of cardiovascular disease. Trials that also investigated other drug treatments were allowed if both the experimental group and control groups took them. The main outcomes that the researchers were interested in were:

• death from any cause
• fatal or non-fatal cardiovascular events
• fatal or non-fatal heart attack or stroke

Secondary outcomes of interest were changes in blood cholesterol, need for revascularisation procedures, adverse effects and quality of life effects. The individual trials were assessed for quality and risk of bias and the trial results were combined, taking into account the variability between study populations, interventions and follow-up (heterogeneity).

What were the basic results?

Fourteen randomised controlled trials met the inclusion criteria. These were in a total 34,272 people who had been followed for between one and five years, equating to 113,000 patient-years of follow-up. The average age of the participants was 57 and 66% were male. The trials were dated 1994-2006 and conducted mostly in Europe, the USA and Japan.

Eleven trials recruited patients with specific conditions that put them at higher cardiovascular risk. In eight of the trials this was raised blood lipids (fat levels), but others included populations with diabetes or hypertension. All trials tested the effectiveness of a statin compared with placebo, of which the most common statin used was pravastatin 10-40mg per day (the drug used in nine trials). Five trials also included advice, counselling or information on lifestyle behaviour, such as smoking cessation, diet and exercise.

Overall, eight trials reported data on death from any cause. A total of 2.8% of the overall study population in these eight trials died during follow-up. Risk of death from any cause was reduced by about 17% (relative risk 0.83, 95% CI 0.73 to 0.95) by statins.

Risk Ratio 0.84 = 16% reduction in mortality due to statins

Three large trials demonstrated that statins reduced the risk of any fatal or non-fatal cardiovascular event (relative risk 0.70, 95% CI 0.61 to 0.79).

Regarding secondary outcomes, there was evidence that statins reduced the need for revascularisation interventions (RR 0.66, 95% CI 0.53 to 0.83). Cholesterol levels reduced in all trials, but the studies were too different to allow the results to be combined for this outcome, mostly because of different statins and doses across the studies.

There was no evidence of any significant harm caused by statins, with no difference in rate in the statin and placebo groups, though the trials had all variously reported adverse effects (ranging from cancer to muscle pain). There was no reliable data to allow assessment of the effect on individuals’ quality of life.

How did the researchers interpret the results?

The researchers conclude that statins reduced death from any cause, any cardiovascular event and the need for revascularisation. There was also no evidence that adverse events increased with statins. However, they caution that there is also evidence of “selective reporting of outcomes, failure to report adverse events and inclusion of people with cardiovascular disease”. More details of these are given in the conclusion.

The researchers say that when these things are considered together, there is only “limited evidence” that primary cardiovascular prevention with statins is cost-effective and improves quality of life. They advise caution in the prescription of statins for people who are at low cardiovascular risk.

Conclusion

This review examined the use of a statin for primary prevention in people who have not yet suffered a cardiovascular event, and whose risk of having one varied considerably. It is important to point out that the benefits of statins in people with established cardiovascular disease and who have already suffered a heart attack or stroke, or who are considered to be at high risk of suffering a cardiovascular event, is not in question.

In high-risk populations, the benefits of a drug that prevents disease often clearly outweigh the risks (such as side effects on health and quality of life). However, in lower-risk populations, this balance often begins to tip the other way and the size of the drug’s benefits compared to its harms can become negligible. This is particularly the case with drugs such as statins, where a relatively large proportion of the population are at low cardiovascular risk, some with raised cholesterol but having no other risk factors. There are also feasibility and cost issues to consider when giving a drug to such a potentially large population.

Although there was some evidence of a reduction in death from any cause, any cardiovascular outcome, and need for revascularisation with no increased risk of adverse events, the researchers acknowledge several limitations to this review and the trials within it. These included:

• The small number of individual cardiovascular events (e.g. stroke or heart attack) that actually occurred during the trials. When separate disease outcomes are quite rare, researchers in the individual trials often compensate for this by instead recording the occurrence of any one of a defined set of outcomes (e.g. all strokes, all heart attacks, all cases of peripheral arterial disease that occurred within the trial, combined all together in one endpoint). The trial then has better ‘power’ to calculate the risk of this ‘composite endpoint’ in the intervention group compared with the control group, than it would have for examining the risk of a single outcome, such as a heart attack. Some of the trials did not even report the numbers that experienced outcomes individually and only reported the numbers with the composite outcome. Therefore it is difficult for the reviewers to gain an accurate picture of whether or not taking a statin has any effect on the person’s risk of suffering a heart attack, or suffering a stroke, as individual disease outcomes.    
• Some of the trials included people with previous cardiovascular events (i.e. it was not a pure primary prevention population). The researchers only included new studies that had less than 10% secondary prevention populations, but they also included data from previous systematic reviews in their current review, some of which may not have been so stringent in which studies they analysed.
• It is possible that some of the trials suffered from selective reporting of outcomes, particularly adverse events. The researchers point out that eight of the trials did not report adverse effects at all.
• Two of the large trials were stopped early due to an observation of benefit in the statin arm. This may lead to an overestimation of the treatment effect.
• As the researchers indicate, all but one trial had received funding from the pharmaceutical industry, which may potentially allow for biased reporting.
• The trials included predominantly white, middle-aged populations. Their results may not be applicable to people outside of these groups.

Overall, this review supports the need for careful consideration of the overall cardiovascular risk profile of the individual when deciding whether to prescribe a statin. As the authors of this review conclude, the evidence does not support the widespread use of statins in people at low risk of a cardiovascular event (expected annual risk of death from any cause below 1%, or expected annual risk of any cardiovascular event below 2%).

The findings of the review also highlight the need for further quality trials when statins are used in primary prevention populations, which give full reporting of outcomes.

Links to the headlines

Statins 'may cause loss of memory and depression'. Daily Mail, January 19 2011

Millions taking statins 'needlessly'. The Daily Telegraph, January 19 2011

'Statins overused'. Daily Mirror, January 19 2011

Why the 'worried well' should not take statins. Daily Express, January 19 2011

Links to the science

Taylor F, Ward K, Moore THM et al. Statins for the primary prevention of cardiovascular disease. Cochrane Library of Reviews

• Standard Trial
• Full Trial