Operative death rates higher at weekend, warn researchers

There is a higher risk of death for patients who have elective surgery later in the week and at the weekend, compared with those earlier in the week, a paper published today on bmj.com suggests.

Previous research has suggested a significantly higher risk of death if admitted as an emergency patient at the weekend compared with a weekday. Plus, other papers have described the "weekend effect".

Researchers offer two potential explanations for this: poorer quality of care at the weekend (which can be attributed to staffing levels and / or less senior / experienced staff) and patients who are admitted or operated on at the weekend being more severely ill.

In this, the first study to focus on day of elective surgery to report a 'weekday effect', researchers from Imperial College London investigate death rates for planned admissions by day of the week of procedure, hypothesising that if there is a quality of care issue at weekends, higher death rates would be seen.

National hospital administrative data were used, linked with death certificates. Information was used on patients' age, gender, source of admission and diagnoses. Mortality outcome was defined as any death occurring within 30 days of the index procedure.

The researchers focussed their study on five higher-risk major surgical procedure groups: excision of oesophagus and / or stomach; excision of colon and / or rectum; coronary artery bypass graft; repair of abdominal aortic aneurysm and excision of lung.

There were 4,133,346 elective inpatient surgical procedures with 27,582 deaths within 30 days of the date of procedure during 2008/2009 to 2010/2011. 4.5% of this surgery was performed at the weekend.

Weekend patients tended to have fewer diseases, fewer admissions, longer waiting time and lower-risk surgery.

The overall risk of 30-day death for patients undergoing elective surgery increased with each day of the week the procedure was performed (by an odds ratio factor of 1.09 per day from Monday). Compared with Monday, the risk of death was significantly higher if procedures were carried out on a Friday. There were also significant differences in the observed rates of death for each day of the week, compared with Monday, for all procedures.

The 30-day mortality rates (for the five selected major survival procedures) per 1000 admissions were: 35 for excision of oesophagus and/or stomach; 24 for excision of colon and/or rectum; 20 for coronary artery bypass graft; 34 for repair of abdominal aortic aneurysm and 20 for excision of lung. All procedures, apart from repair of abdominal aortic aneurysm, had statistically significant trends towards higher mortality at the end of the working week and weekends compared with Monday.

The findings suggest that the weekend effect might be more pronounced for patients with more diseases and for patients with three or more previous admissions, than for patients with none.