Abstract Details

To compare time from admission to tracheostomy (TTT) for patients with Severe Acute Brain Injury (SABI) in US hospitals in regions with high vs. low density of long-term acute care hospital (LTACs) beds.

We identified patients with SABI, defined as acute stroke, anoxic injury following cardiac arrest, or traumatic brain injury and prolonged mechanical ventilation, from the Nationwide Inpatient Sample (NIS), 2006-2011. Among those who underwent tracheostomy, we evaluated TTT in days. We measured LTAC bed density in hospital referral regions by linking the 2011 Healthcare Cost Report Information System and US Census Bureau to hospitals identifying ZIP code in the NIS. We defined high density LTAC regions as >1.5 beds per 100,000 (approximately top 20^th percentile).  We used multivariable regression to evaluate the relationship between LTAC beds per 100,000 population and TTT.

Overall, 21,580 patients with SABI received tracheostomy in 1,761 hospitals in 193 hospital referral regions with 205 LTACs and 13,340 LTAC beds. Eighty-four (44%) referral regions had zero LTAC beds. High density LTAC regions (range 1.55-32.37 beds per 100,000) had mean TTT of 12.9 vs. 14.1 days in other regions (p<0.0001), and were more likely to perform early tracheostomy (within 7 days; 27% vs. 19%; p<0.0001).  After controlling for patient and hospital factors, each additional LTAC bed per 100,000 population was associated with a 12% decrease in TTT (?-coefficient -0.12, 95% CI [-0.22, -0.02]) and high density LTAC regions had 44% greater odds of early tracheostomy (adjusted OR 1.44, 95% CI 1.30, 1.59).

Regions with high LTAC bed availability are associated with earlier time to tracheostomy for SABI.