Few errors would have caused any serious harm to the patient
Nearly one in 10 hospital prescriptions contains a mistake, ranging from the minor to the potentially lethal, research has found.
But the study, commissioned by the General Medical Council, found very few errors would have caused serious harm.
It also found that, contrary to belief, novice doctors were no more responsible for mistakes than the more experienced.
To eliminate one area of confusion, the GMC is calling for a UK-wide standard prescription chart as exists in Wales.
The research team, led by Professor Tim Dornan of the University of Manchester, examined the issue amid rising fears inexperienced doctors were making prescription errors which could, at worst, result in a patient dying.
They examined 124,260 prescriptions across 19 hospitals - and found just under 9% contained errors.
Of these 11,077 errors, overwhelmingly intercepted and corrected before reaching the patient, about 2% contained potentially lethal instructions - such as failing to take account of a patient's allergies.
More than half involved errors in which a patient's medication was not prescribed on admission, during a rewrite of a prescription, or when the patient was sent home.
Another 40% were accounted for by prescriptions where the writing was illegible or the wording ambiguous.
Very few of these mistakes caused actual harm to a patient because on the whole they were stopped by senior doctors, nurses - and in particular pharmacists.
There were however concerns that so effective was this safety net, some doctors relied on it to pick up their mistake.
'Off the hook'
But the study did not find doctors fresh out of medical school were making the most mistakes - as has often been suggested.
Doctors in their first year of medical training in fact made slightly fewer mistakes than the average, although that rose slightly in their second year. However at 8.3% their rate was the same as registrars. Consultants made the fewest, with 5.9%.
While the curriculum at medical schools could always be improved, it was clearly not at the root of the problem, the team concluded - noting many factors - from fatigue to unfamiliarity with a prescription form - produced errors.
"The research shows the complexity of the circumstances in which errors occur and argues against education as a single quick-fix solution.
"Education can always be improved but it must be very practically oriented and include all phases of a doctor's career as well as the undergraduate stage," said Professor Dornan.
The chairman of the GMC, Professor Peter Rubin, said: "Prescribing decisions in a hospital setting often have to be made quickly, so it is important that a procedure is as simple as possible to minimise the chance of an error being made.
To avoid confusion as doctors move between hospitals with very different prescribing forms - including paper and electronic - the GMC wants to see a standardised system across the UK. Wales introduced this in 2004.
A Department of Health spokesman said it would continue to look into the benefits of electronic prescribing systems, "taking into account the evidence gained where standardisation of the paper chart has been successfully implemented."
Dr Hamish Meldrum, of the doctors' union, the BMA, said: "It would certainly help if there was greater uniformity in the prescription forms used in the NHS and the BMA would encourage prescribing procedures to be kept as simple as possible."
Joyce Robins, Co-Director of Patient Concern commented: "For patients, one of the most astonishing and frightening findings is that so many errors happen because doctors are confused by widely differing prescription forms, both paper and electronic.
'Wales has had a standard prescription form throughout the service for five years but in England we continue to rate patient safety as less important than the sensitivities of hospital managers, who are allowed to operate their own independent little fiefdoms."
Professor Simon Maxwell, of the British Pharmacological Society said: "Like everyone else, I am extremely concerned by this error rate but I am dismayed at the suggestion that improved education and training is not a central part of the solution.
"There is plenty of evidence from around the world to show that when appropriate education and training are delivered, prescribing improves."
He said no credible observer would say this was the sole answer but that it must play a role.