Volume-outcome relationship in neurotrauma care.

Clement RC, Carr BG, Kallan MJ, Wolff C, Reilly PM, Malhotra NR.

Source

Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania 19104, USA.

Abstract

OBJECT:

A positive correlation between outcomes and the volume of patients seen by a provider has been supported by numerous studies. Volume-outcome relationships (VORs) have been well documented in the setting of both neurosurgery and trauma care and have shaped regionalization policies to optimize patient outcomes. Several authors have also investigated the correlation between patient volume and cost of care, known as the volume-cost relationship (VCR), with mixed results. The purpose of the present study was to investigate VORs and VCRs in the treatment of common intracranial injuries by testing the hypotheses that outcomes suffer at small-volume centers and costs rise at large-volume centers.

METHODS:

The authors performed a cross-sectional cohort study of patients with neurological trauma using the 2006 Nationwide Inpatient Sample, the largest nationally representative all-payer data set. Patients were identified using ICD-9 codes for subdural, subarachnoid, and extradural hemorrhage following injury. Transfers were excluded from the study. In the primary analysis the association between a facility's neurotrauma patient volume and patient survival was tested. Secondary analyses focused on the relationships between patient volume and discharge status as well as between patient volume and cost. Analyses were performed using logistic regression.

RESULTS:

In-hospital mortality in the overall cohort was 9.9%. In-hospital mortality was 14.9% in the group with the smallest volume of patients, that is, fewer than 6 cases annually. At facilities treating 6-11, 12-23, 24-59, and 60+ patients annually, mortality was 8.0%, 8.3%, 9.5%, and 10.0%, respectively. For these groups there was a significantly reduced risk of in-hospital mortality as compared with the group with fewer than 6 annual patients; the adjusted ORs (and corresponding 95% CIs) were 0.45 (0.29-0.68), 0.56 (0.38-0.81), 0.63 (0.44-0.90), and 0.59 (0.41-0.87), respectively. For these same groups (once again using < 6 cases/year as the reference), there were no statistically significant differences in either estimated actual cost or duration of hospital stay.

CONCLUSIONS:

A VOR exists in the treatment of neurotrauma, and a meaningful threshold for significantly improved mortality is 6 cases per year. Emergency and interfacility transport policies based on this threshold might improve national outcomes. Cost of care does not differ significantly with patient volume.