Patients who have undergone a bowel operation could be five times more likely to need further emergency surgery depending on which hospital treats them, research suggests.
There are wide variations around England in the rates at which people need operating on again once they have undergone surgery, experts said.
The research, published in the British Medical Journal (BMJ), found unplanned reoperation rates were five times higher in some hospitals - prompting concerns about the performance of individual surgeons and trusts.
Experts at Imperial College London used statistics for 246,469 patients treated in 175 English hospital trusts who underwent bowel resection surgery for the first time between 2000 and 2008.
Bowel resection involves removing part of the bowel and joining the two ends together, and is used for conditions such as bowel cancer, bad cases of Crohn’s disease, blockages in the bowel and scarring.
Of the group, 15,986 (6.5%) patients needed further surgery within 28 days, including of the bowel, to control bleeding or for complications arising from their wound.
Most of these patients (83%) underwent emergency surgery while still in hospital while the remainder ended up being readmitted.
Analysis showed that patients who had gone in for emergency surgery in the first place had slightly higher reoperation rates than those whose surgery was planned (7% versus 6.2%).
Men and people with inflammatory bowel disease such as colitis and Crohn’s, or other pre-existing conditions, were also more likely to need emergency surgery.
For those whose original surgery was planned, there was a fivefold difference in highest and lowest reoperation rates (14.9% compared to 2.8%) among trusts performing more than 500 procedures during the study.
For those performing more than 2,500 procedures, there was a threefold difference in reoperation rates (11.5% compared to 3.7%) between the best and worst performing trusts.
The researchers said reoperation rates could be a “powerful means of checking quality of surgical care” when used alongside data on death rates.