Patients undergoing emergency surgery are more likely to survive if they are treated in hospitals with more doctors, consultants and nurses, a major study has found.
The new research - carried out over five years and involving almost 300,000 cases of emergency surgery at 156 NHS trusts - found that hospitals with the highest staffing levels had the lowest death rates.
This was despite the fact their patients were often sicker before surgery and were more likely to suffer complications following operations.
The study - described as "ground-breaking" by the researchers - suggests that the way complications were identified and managed by staff was key in determining whether patients lived.
The research also found that patients admitted to the hospitals on weekends had lower chances of survival than those admitted during the week.
The study will add to the row over staffing levels in NHS hospitals, particularly at weekends, after previous studies suggested higher death rates for people admitted on Saturdays and Sundays.
Previous research has also suggested that a lack of consultants on wards contributes to higher death rates.
The new study, published in the British Journal of Anaesthesia, looked at patients admitted to hospitals in England for emergency surgery for bowel problems, peptic ulcers, removal of the appendix, hernia repair and pancreatitis.
Their chance of dying within 30 days of admission was analysed. Some 14 NHS trusts were found to have significantly higher than expected death rates, while nine had lower than expected death rates.
Researchers found that those trusts with lower than expected death rates had more nurses, consultant surgeons and surgical junior doctors per bed than those with the higher death rates. They also had more critical care beds and operating theatres.
Compared to those with the higher death rates, trusts that performed the best had 44% more doctors of any grade or speciality per bed, 40% more surgical doctors per bed, 40% more consultant surgeons per bed, 38% more junior surgical doctors per patient admitted and 24% more nurses per bed.
They also had 65% more critical care beds per standard ward bed and 32% more operating theatres per bed.
Overall, in those trusts with the least nurses and doctors per bed, patients were 7% more likely to die, the researchers said.
Patients were also 11% more likely to die if they were admitted to hospital on a weekend.
Senior author Dr Peter Holt, reader in vascular surgery at the St George's Vascular Institute in London, said: "This is the first analysis on this scale for emergency general surgical admissions examining the associations between the numbers of senior doctors, junior doctors and nurses per hospital bed and patient survival rates.
"Although we have not demonstrated causation, we believe the findings should form an important part of the debate over disparities in staffing levels and resource provision between NHS trusts."
Report author Mike Grocott, professor of anaesthesia and critical care at the University of Southampton, said: "Our study has identified a striking association between staffing levels, clinical experience and patient outcomes. This raises important questions about how emergency surgery is delivered within hospitals and across the NHS as a whole."
In trusts that performed better than expected and higher numbers of staff, patients had a greater number of pre-existing medical conditions.
They were also 27% more likely to suffer complications during treatment.
But, despite these complications, these trusts had 60% fewer deaths at 30 days than trusts performing less well than expected.
Those trusts with the lowest death rates had significantly greater provision per bed of doctors (0.85 vs 0.59 doctors per bed), consultant doctors per bed (0.28 vs 0.20), nursing staff per bed (2.33 vs 1.88), critical care beds per bed (0.038 vs 0.023), and operating theatres per bed (0.029 vs 0.022) than high mortality trusts.
Commenting on the weekend death rates, Professor Ravi Mahajan, editor of the British Journal of Anaesthesia, said: "It seems a fair assumption that the number of senior staff and availability of resources at the weekend would be less than during the week."
Dr Mark Porter, chair of the British Medical Association (BMA), said: "The findings suggest that more investment in hospital staff, beds and resources yields better outcomes for patients.
"This is in line with the BMA's calls for greater investment in urgent and emergency care to ensure patients have access to high-quality care, seven days a week."
Clare Marx, president of the Royal College of Surgeons, said: "While many hospitals provide high-quality emergency surgery, it is wholly unacceptable that death rates vary so significantly for the sickest patients across the country.
"The published data is old and must be interpreted carefully.
"However, the Royal College of Surgeons has been clear there is still a current and urgent need to review how emergency care is provided in the NHS, including staffing arrangements. This has to become one of the top priorities for the NHS."
Janet Davies, chief executive of the Royal College of Nursing (RCN), said: "Poor staffing levels have a serious impact on patient care.
"Evidence clearly shows the positive effect that having the right number of registered nurses has on patient care, outcomes and experience.
"When staffing levels fall, there are no longer the right number of skilled professionals in the right settings, and when that happens patients and their families sadly feel the effects."
Professor Sir Bruce Keogh, national medical director for the NHS, said: "Patients accept the risks associated with their condition and treatment, but they should never have to accept an increased risk because of the way healthcare services are designed and delivered."
He added: "It is vital we have the right number of clinical staff for each individual ward in every hospital, and that these staff should have the right skills and expertise to meet the needs of our patients, 24 hours a day and seven days a week."