Does size matter? The relationship between volume and outcome in the care of major trauma

To investigate the relationship between case volume and outcome for major trauma patients.

Prospective follow-up study of all major trauma patients (with injury severity score >15) arriving alive, with no invariably fatal injury, at 14 English emergency departments between 1990 and 1993. Using the stratified W statistic, an age and severity adjusted measure of outcome, the relationship between volume of cases and outcome was initially examined using the Spearman correlation coefficient. Multiple regression analysis was used to explore further the relationship, after adjustment for hospital-level characteristics.

The smallest department saw five major trauma cases each year, the largest saw 96. The results of the initial correlation analyses indicated that there was little evidence that outcome improved with increasing volumes for all major trauma (rho(s) = 0.12, 95% confidence interval [CI]: -0.36 to 0.55) nor for the cases presenting out-of-hours (rho(s) = 0.30, 95% CI: -0.19 to 0.67). However, there was evidence that patients with multiple injury (rho(s) = 0.65, 95% CI: 0.27 to 0.86) and those with severe head injuries (rho(s) = 0.52, 95% CI: 0.08 to 0.79) did better in high volume departments. This pattern, of a positive relationship for more complex cases was also in evidence from the results of the multiple regression analyses and, in particular, for patients with multiple injuries, was stable over time.

While there was little evidence that all patients with major trauma do better in higher volume departments, there was evidence that patients with complex needs, such as the multiple injured or those with head injuries, had better outcomes.