Abstract
Demand for hospital resources may increase over time; we have examined all emergency admissions (51,136 episodes) from 2005 to 2013 for underlying trends and whether resource utilization and clinical risk are correlated. We used logistic regression of the resource indicator against 30-day in-hospital mortality and adjusted this risk estimate for other outcome predictors. Generally, resource indicators predicted an increased risk of a 30-day in-hospital death. For CT Brain the Odds Ratio (OR) was 1.37 (95% CI: 1.27, 1.50), CT Abdomen 3.48 (95% CI: 3.02, 4.02) and CT Chest, Thorax, Abdomen and Pelvis 2.50 (95% CI: 2.10, 2.97). Services allied to medicine including Physiotherapy 2.57 (95% CI: 2.35, 2.81), Dietetics 2.53 (95% CI: 2.27, 2.82), Speech and Language 5.29 (95% CI: 4.57, 6.05), Occupational Therapy 2.65 (95% CI: 2.38, 2.94) and Social Work 1.65 (95% CI: 1.48, 1.83) all predicted an increased risk. The in-hospital 30-day mortality increased with resource utilization, from 4.7% (none) to 27.0% (five resources). In acute medical illness, the use of radiological investigations and allied professionals increased over time. Resource utilization was calibrated from case complexity/30-day in-hospital mortality suggesting that complexity determined the need for and validated the use of these resources.