MORTALITY FOLLOWING DELAY TO SURGERY IN HIP FRACTURES IN THE PROVINCE OF ONTARIO.

Abstract

The purpose of this study was to evaluate mortality following delay to surgery in hip fractures in the province of Ontario. All patients undergoing a surgical procedure for a hip fracture between 1993 and 1999 were identified using administrative databases. For every day that surgery was delayed, the adjusted odds of in-hospital mortality increased by a factor of 1.12 times (95%CI), with similar results at three months and one year. A significant relationship exists between delay to surgery and mortality in elderly hip fracture patients. Every effort should be made to avoid non-medical delays in providing operative treatment for hip fractures.

A significant relationship exists between delay to surgery and mortality in elderly hip fracture patients. Every effort should be made to avoid non-medical delays in providing operative treatment for patients with fractured hips.

This finding will have far reaching implications for the allocation of health resources in the future.

All patients undergoing a surgical procedure for a hip fracture between 1993 and 1999 in the Province of Ontario were identified using administrative databases and the provincial mortality database. Multivariable logistic regression models were used to adjust for age, gender, medical comorbidity, type of hip fracture, and teaching status of the treating hospital.

For every day that surgery was delayed, the adjusted odds of in-hospital mortality increased by a factor of 1.12 times (95% CI). The adjusted odds of in-hospital mortality increased as the delay to surgery lengthened from a factor of 1.2 [95% CI] with a one day delay to a factor of 1.5 [95% CI] for a delay over two days as compared with patients operated within twenty-four hours. Similar relationships were observed at three months and one year. Even when considering only healthy patients, < 70 years old with no comorbid conditions, the relationship between mortality and surgical delay remained significant (p < 0.0001), suggesting that surgical delay was unlikely to be caused by patient factors alone.