Statins: finding safety in numbers
Statins are the most commonly prescribed medicines in the UK and are truly life-saving. They’ve been shown to have a significant impact in reducing the risk of heart attacks, some types of strokes and other cardiovascular problems.
Over the past three decades, scientists here at the Nuffield Department of Population Health (NDPH) in Oxford have painstakingly carried out detailed clinical trials and data analysis to produce as accurate a picture of the risks and benefits of statin treatment as possible. This impressive body of work highlights the power of the NDPH’s approach to answering big questions about public health: big numbers.
What are statins and how do they work?
Statins are an important family of cholesterol-lowering drugs. Cholesterol is a fatty molecule that’s needed by all our cells to keep them healthy, and is also involved in making vitamin D and some hormones. Although we hear a lot about cholesterol in fatty foods, most of the cholesterol found in the body is actually made directly in the liver.
Research has shown that people with higher levels of certain types of cholesterol-containing molecules in the bloodstream (referred to as LDL or so-called ‘bad’ cholesterol) are more likely to develop hardening and narrowing of the arteries, known as atherosclerosis. They’re also at a higher risk of cardiovascular disease, including coronary heart disease, angina, strokes and heart attacks. Animal research and genetic studies in humans have also shown a strong link between LDL cholesterol and the risk of atherosclerosis.
It has been known since the 1960s that high cholesterol in the bloodstream was a risk factor for heart disease, so it makes scientific sense that lowering cholesterol should reduce that risk. However, early clinical trials testing various cholesterol-lowering techniques – including drugs and changes to diet – couldn’t achieve big changes in cholesterol levels. Because of the relatively weak effects of the approaches that were available at the time (combined with the fact that most of the trials were quite small and short), it looked as if lowering cholesterol wasn’t very effective at cutting the risk of heart attacks and other cardiovascular problems.
To try and figure out what was going on, in the 1980s NDPH scientists Richard Peto and Rory Collins gathered together all the data from trials testing cholesterol-lowering interventions, and crunched all the numbers together (a statistical technique known as a meta-analysis). They noticed that reducing cholesterol could cut the risk of cardiovascular problems, and the longer and lower it was reduced the more effective it was. So it was clear that cutting cholesterol could make a difference, but the approaches available at the time to do it simply weren’t good enough.
Then came statins. First developed in the 1970s from chemicals found in fungi, statins work by blocking the activity of a substance in the liver called HMG-CoA reductase. This is involved in one of the first steps towards making cholesterol, so blocking its function cuts off cholesterol production at the source. There are now several different statin drugs on the market and most are now off-patent, making them cheap and widely-available.
How do we know that statins work and are safe?
The effectiveness and safety of statins have been tested in large-scale trials involving many thousands of patients over many years, several of which have been run by scientists at the NDPH. Importantly, many of these studies are randomised, placebo-controlled trials – the ‘gold standard’ for clinical research – which are probably the best way we have of finding out how well a treatment works and the chances of developing any side effects.
A placebo-controlled trial means that patients are either given the genuine treatment or a dummy pill (placebo), although they don’t know which is which. This means that any effects – whether good or bad – can be accurately put down to the treatment. Randomisation means that patients are randomly assigned to one of these groups, so researchers can’t inadvertently or deliberately give certain types of people either the drug or the placebo.
To take a fictional example, if the same number of people randomly assigned to receive a placebo as a real drug complain of headaches, that just says that headaches are common and doesn’t mean that the treatment causes them. But if twice as many people get earache if they’re taking the drug compared to the placebo, say, then we can infer that earaches are probably a side effect of the treatment.
Another important aspect to consider is the size of a trial. Simply put, bigger is better. The larger the trial, the greater the likelihood that anything researchers observe (whether it’s the expected effect of the treatment or any side effects) is likely to be genuine, rather than a random ‘blip’ or chance event. Bigger numbers also mean more accurate statistics and estimation of benefits and risk. And combining results of large trials together with a meta-analysis - effectively creating a ‘supertrial’ – makes the picture even clearer.
Furthermore, researchers can increase the diversity of types of patients by combining trials together. For example, some trials may not include many women or elderly patients, but adding all the results together gives big enough patient groups to provide statistically meaningful results and provide recommendations covering wider sections of the population.
It all adds up
In 1994, NDPH researchers set up the Cholesterol Treatment Trialists’ (CTT) Collaboration. Over the years they have brought together and analysed data from several large-scale trials, building a bigger, better analysis of the impact of statin treatment. They have also revealed a clearer picture of the side effects of statins, discussed in more detail below, and proved that there is no increased risk of cancer from taking the drugs.
Overall, these large-scale randomised trials show that statins reduce the risk of heart attacks, strokes caused by blocked blood vessels (ischaemic strokes) and certain other cardiovascular problems. For example, a commonly prescribed statin (atorvastatin, which costs around £2 a month in the UK) will approximately halve someone’s LDL-cholesterol levels. And the benefits don’t stop there. As well as treating people with high cholesterol levels, statins can also help individuals who are at increased risk of cardiovascular disease for other reasons such as having already suffered a heart attack, or having diabetes.
Achieving this in 10,000 patients over five years would prevent these kinds of events happening in around 1,000 of people with pre-existing cardiovascular problems (known as secondary prevention). For people who are at increased risk of a stroke or heart attack but haven’t yet had one (so-called primary prevention), the benefits are smaller but still there, with up to 500 cardiovascular events prevented per 10,000 people treated.
That’s not to say that the other 9,500 people on statins will get no benefit from the drug at all. There is some evidence to suggest that statins have a long-term impact on atherosclerosis (hardening of the arteries) for a larger number of people. And taking the drugs for ten years prevents more heart attacks and strokes than five years of medication.
Clinical trials have also revealed things that don’t help. For example, sometimes (particularly in the US) a cheap and widely-available drug called niacin (also known as vitamin B3) is used in combination with or instead of statin therapy. It’s a treatment approach that has been used for more than 30 years. However, a large trial led by NDPH researchers showed that this combination wasn’t any more effective than statins alone, and also increased the risk of side effects such as facial flushing, bleeding, type two diabetes and life-threatening infections.
Balancing benefits and risks
Overall, large scale placebo-controlled clinical trials of statins, adding up to more than 100,000 people treated, have found very few side effects. And those that do arise seem to be rare. One side effect that has been consistently found is myopathy - a type of severe muscle pain. Statin use also appears to cause a small increase in the chances of developing type two diabetes. More rarely, the drugs have been linked to a particular kind of stroke known as haemorrhagic (bleeding) stroke – this is much less common than ischaemic stroke, which is caused by artery blockage and can be prevented by statins.
To put this in perspective, given 10,000 patients treated for 5 years with a standard 40 milligram daily dose of atorvastatin, doctors could expect to see around 50 to 100 people with muscle pain or weakness as well as five cases of myopathy, one of which could progress to more serious muscle breakdown (rhabdomyolysis) if they don’t stop taking the drug. They can also expect around 50 to 100 new cases of diabetes and about five haemorrhagic strokes.
Some researchers have suggested other side effects arising from statins, including an increased cancer risk, memory problems, cataracts, kidney problems, liver disease, aggression, suicidal behaviour, erectile dysfunction and more. None of these have consistently shown up in large clinical trials or combined meta-analyses, yet some people still have doubts and concerns.
Together with co-investigator Professor Colin Baigent, Professor Collins and the CTT Collaboration are now gathering as much data as possible about all these ‘adverse events’ in the large-scale trials. They’re working out whether any of them are genuine side effects of statin use (i.e. are more common in people taking statins compared to those on a dummy placebo) or not. Furthermore, the CTT team is using standardised coding, so everything is open and transparent. It will take time to gather everything together and carry out the necessary statistical analysis, but the results of this work should help to clarify the issue. And, of course, all this information about side effects needs to be balanced against the proven benefits of statins in reducing the risk of life-threatening cardiovascular events.
What do we still need to find out?
Large-scale clinical trials involving many thousands of participants and meta-analyses by NDPH scientists and others around the world have shown that statins are broadly safe and can save lives, whether for people who have already had a heart attack or other cardiovascular problems, or those who haven’t but are thought to be at increased risk due to high cholesterol levels. However, we still need to find ways of identifying people who are more likely to suffer from the known side effects of statins. For example, NDPH researchers and others are investigating how differences in individual genetic makeup can affect the chances of developing side effects such as diabetes.
It’s also important to acknowledge that statins aren’t the single solution to heart health. Stopping smoking, eating a healthy diet, keeping a healthy weight and staying active are all ways that people can reduce the risk of cardiovascular disease and other conditions such as diabetes or cancer. Statins are another important and proven tool on top of these interventions, with measurable and wide-spread effectiveness that has been shown through large-scale, randomised clinical trials, careful statistical analysis and long-term detailed follow-up of patients.
Given how important these drugs are, it’s important that information for the public is clear and accurate, providing as much as possible about the risks and benefits of statins and enabling people to make a properly informed decision about whether to take them. In a major analysis published in the Lancet journal in 2016, NDPH director Professor Sir Rory Collins and his colleagues have laid out the scientific evidence surrounding the effectiveness and safety of statin use. They also argue that misrepresentation of the risks and inaccurate information in the media could mean that people who would benefit from these drugs may decide not to take them.
This is a tough challenge for researchers, doctors, providers of patient information and journalists: even the biggest randomised controlled trials can’t provide a definite ‘yes or no’ answer – as with so much in life, we are dealing here with probabilities and uncertainties – but they can provide good measurements about the relative risks and benefits of a given treatment, and it is essential that the public is properly and transparently informed about them. It’s just as important not to over-hype side effects as it is to boast excessively of a drug’s effectiveness.
The biggest potential harm from misleading media coverage is that people who could benefit from the protective effects of statins decide either to stop taking them or not take them at all, risking the long-term debilitating effects of stroke or the devastating impact of a heart attack. Statins may not be right or necessary for everyone, but everybody is entitled to a clear, factual explanation of the scientific evidence surrounding them.
- Interpretation of the evidence for the efficacy and safety of statin therapy. Collins et al (2016) The Lancet, Volume 388: 2532–2561
- Information about statins from NHS Choices
- Information about statins from the British Heart Foundation