The gold standard for testing a drug is to trial it on hundreds or thousands of people — then compare the results to an equal number given a placebo.
Such randomised controlled trials find out how a drug works overall, but there will be some people for whom the drug works well and others for whom it didn’t work at all.
But another type of trial, that tests the effects of a drug on just one patient, has hit the headlines recently.
Called an N-of-1, the aim is to give an individual result about whether a drug works for them, which is the best dose, or if there are side-effects.
The ‘N’ stands for the number of people in the trial and the ‘1’ means there is just one.
Up to eight million Britons take statins, but a fifth stop, or refuse to start, over worries about side-effects such as aches and fatigue
‘N-of-1 are the ultimate form of personalised care,’ says Nicky Britten, an emeritus professor of applied healthcare research at Exeter University.
‘If you’re prescribed a drug that has been through a randomised controlled trial, all your doctor can tell you is that on average, say, 70 per cent of people get better when taking it. They can’t tell you if it will work for you.’
Another reason why conventional trial results will not apply to everyone is that older people and women have traditionally been excluded from studies — over concerns that women’s hormones could affect the results; and that existing health conditions in older people, or other medication they are taking, could skew the findings. N-of-1 trials address this by focusing on just one patient.
John Chapple, 77, a retired engineer from West London, is among a handful to benefit from an N-of-1 trial. He had been prescribed statins after heart surgery in 2010, but stopped taking them five months later due to side-effects.
‘My muscles ached all over,’ says John. ‘It was like constantly having the flu. My consultant suggested I try two other statins, but that didn’t make a difference.’
Up to eight million Britons take statins, but a fifth stop, or refuse to start, over worries about side-effects such as aches and fatigue.
In 2016, John had a heart attack. When he told his consultant why he’d stopped taking statins, he was put on an N-of-1 trial.
He was given 12 pill bottles by researchers at Imperial College London. Four contained the statin atorvastatin, four contained identical-looking sugar pills and four were empty, to allow tablet-free months.
Called an N-of-1, the aim is to give an individual result about whether a drug works for them, which is the best dose, or if there are side-effects
He used one bottle each month for a year; and logged side-effects on an app (including tablet-free months, which helped show if his aches were due to statins or not).
After a year, John was shown his results and found, to his surprise, he had the same aches and pains whether taking a statin or a placebo. ‘That convinced me,’ he says, and now takes atorvastatin daily without side-effects.
So why did John feel unwell when first taking statins? The researchers, who analysed John’s results alongside those of another 59 patients on similar N-of-1 trials, said almost all the ‘side-effects’ were caused by the patients’ belief the drug would make them sick, reported the New England Journal of Medicine in November. Half of the patients went back on statins when shown their results.
An analysis of N-of-1 trials published in The BMJ in February led to two-thirds of participants saying they’d take statins again.
N-of-1 trials can produce good results but with fewer people, says Dr Elizabeth Williamson, an associate professor of medical statistics at the London School of Hygiene & Tropical Medicine.
This is because when you’re comparing two groups in a trial, one on a drug and the other a placebo, there will be differences between participants, from their age and weight, to how they perceive pain. To smooth out these differences, hundreds or thousands of people are studied.
When N-of-1 results are pooled, in contrast, you’re comparing how individuals fared. This means there are far fewer differences to account for and, you may get an answer from studying a few dozen people.
However, as N-of-1 trials involve switching from a drug to a placebo several times, they’re only suitable for testing medicines with a rapid but short-lasting effect on chronic conditions, such as statins and painkillers.
Other obstacles include the price of placebos, which can cost a hundred times more than the drug.
Currently N-of-1 trials are only being used in research. Liam Smeeth, a professor of clinical epidemiology at the London School of Hygiene & Tropical Medicine, is unsure if GPs will have time to set up N-of-1 trials for individuals.
But, he says, centrally organised research projects, like the two statin studies, would be more feasible, and ‘could easily answer some really big questions about what treatments are best’.