BACKGROUND: Specialized trauma intensive care unit (TICU) care impacts patient outcomes. Few studies describe where and how TICU care is delivered. We performed an assessment of TICU structure and function at a sample of US trauma center TICUs. METHODS: This was a multicenter study in which participants supplied information about their trauma centers, staff, clinical protocols, processes of care, and study TICU (the ICU admitting the majority of trauma patients). RESULTS: Forty-five Level I trauma centers trauma centers enrolled through the American Association for the Surgery of Trauma multi-institutional trials platform; 71.1% had less than 750 beds and 55.5% treated 1,000 to 2,999 trauma activations/year. The median number of hospital ICU beds was 109 [66-185]. 46.7% were "closed" ICUs, 20% were "open," and 82.2% had mandatory intensivist consultation. 42.2% ICUs were classified as trauma (>/=80% of patients were trauma), 46.7% surgical/trauma, and 11.1% medical-surgical. Trauma ICUs had a median 10 [7-12] intensivists. Intensivists were present 24 hours/day in 80% of TICUs. Centers reported a median of 8 (interquartile range [IQR], 6-10) full-time trauma surgeons, whose ICU duties comprised 25% (IQR, 20%-40%) of their clinical time and 20% (IQR, 20-33) of total work time. A median 16 (IQR, 12-23) ICU beds in use were staffed by 10 (IQR, 7-14) nurses. There was considerable variation in the number and type of protocols used and in diagnostic methods for ventilator-associated pneumonia. Daily patient care checklists were used by 80% of ICUs. While inclusion of families on rounds was performed in 91.1% of ICUs, patient- and family-centered support programs were less common. CONCLUSION: A study of structure and function of TICUs at a sample of Level I trauma centers revealed that presence of nontrauma patients was common, critical care is a significant component of trauma surgeons' professional practice, and significant variation exists in care delivery models and protocol use. Opportunities may exist to improve care through sharing of best practices. LEVEL OF EVIDENCE: Therapeutic/Care management, level IV.