Nursing cutbacks are directly linked to higher patient death rates in hospitals, a major study has found.
Every extra patient added to a nurse's workload increases the risk of death within a month of surgery by 7%, according to data from 300 European hospitals in nine countries.
The situation is made worse by employing poorly qualified nurses, the research showed.
When nurses had university degrees, this went a long way towards making up for reduced staffing levels.
A 10% increase in the proportion of nurses holding a bachelor degree was associated with 7% lower surgical death rates.
The findings, published in The Lancet journal, included figures for 30 English hospitals showing that on average every one of their nurses looked after around nine patients.
In some other countries the patient-to-nurse ratio was significantly smaller. Norway had a ratio of 5.2 to one, the Irish Republic 6.9, the Netherlands seven and Finland and Sweden 7.6.
Spain appeared to have the most overworked staff, with an average 12.7 patients per nurse. But in Spain every nurse had a bachelor degree, compared with only 28% in England at the time the data were collected in 2009-10.
US expert Professor Linda Aiken, from the University of Pennsylvania School of Nursing, who led the research, said: "Our findings emphasise the risk to patients that could emerge in response to nurse staffing cuts under recent austerity measures, and suggest that an increased emphasis on bachelor's education for nurses could reduce hospital deaths."
The study analysed information on more than 420,000 patients admitted to hospitals in Belgium, England, Finland, the Irish Republic, the Netherlands, Norway, Spain, Sweden, and Switzerland.
Researchers compared nurse workload and education and patient outcomes, taking into account a number of factors that could influence the result.
These included the age and sex of patients, types of surgical procedure, chronic conditions, and the kind of technology available in a hospital.
The overall percentage of surgical patients who died within 30 days of admission was low, ranging between 1% and 1.5% per country. In England, the average rate was 1.4%.
However, in every country death rates varied significantly between individual hospitals, with some losing up to 7% of surgical patients after 30 days.
The highest risk of death after surgery was found in hospitals where nurses with lower levels of education cared for the most patients, said the researchers.
A comparison was made between hospitals where every nurse cared for an average of six patients and the proportion of degree-educated staff was at least 60%, and those where nurses looked after eight patients each and just 30% had bachelor degrees.
The hospitals with lighter workloads and more qualified nurses were expected to have 30% lower surgical death rates.
"Our data suggest that a safe level of hospital nursing staff might help to reduce surgical mortality, and challenge the widely held view that nurses' experience is more important than their education," Dr Aiken added."
The European findings closely mirrored similar trends seen in the US, said the researchers.
Peter Carter, chief executive of the Royal College of Nursing (RCN), said: "It is worrying to see that researchers found the mean ratio of patients to nurses in England is above eight, as we know that this can compromise patient safety.
"The RCN has also expressed concern at the skills mix in UK hospitals as trusts get rid of more senior nurses to save money, meaning there is far less experience on many wards, and the full extent of this will be revealed in our upcoming Frontline First report."
The findings supported the UK's recent decision to make nursing an all-graduate profession, Dr Carter added.
"Compassion and caring will always be at the heart of nursing practice and nurse training," he said.
"However, the demands of modern health care mean that this must be backed up by a higher level of education so that patients, who often have complex needs, receive skilled and compassionate care..
"Modern medicine means that a nurse's role is far more technical and requires complex decision making which demands a degree level education as well as the practical experience which currently makes up at least half of a nursing degree."
In November, the Government's response to the inquiry into Stafford Hospital stopped short of introducing a minimum staff-patient ratio on wards or enshrining this in law.
Organisations including the RCN have called for a minimum of one nurse for every eight patients.
Health Secretary Jeremy Hunt said the Government would not introduce a legal minimum because staff requirements were a "different number for different wards".
He said forcing hospitals to publish monthly data on ward staffing would be a "huge step" forward.
Figures from the Health and Social Care Information Centre (HSCIC) in January showed a rise in the number of nurses working in hospitals under the acute, general and elderly category, with most of the increase attributable to first level (more junior) nurses.
There has been a drop in the number of other types of nurses, including community nurses, those for people with mental health problems and learning disabilities, community services and those working in schools.
A poll of 526 nurses for Nursing Times magazine earlier this month found 57% believed their wards were "dangerously understaffed".
A year after Robert Francis QC published his report into failings at Stafford Hospital, where hundreds of people died needlessly contributed to by neglect, 39% of nurses said staffing levels had worsened where they work over the last 12 months.
Some 37% said they have stayed the same while only 22% reported an improvement.