Table 2.
Study sites | One shock trauma, One ED in referral centres | Two EDs in university system | EDs in level I, II, III trauma hospitals | EDs in all acute care hospitals |
State of study site | Maryland | Kentucky | Colorado | Massachusetts |
Setting: type of hospital | Large academic hospitals* | Acute care hospitals | Acute care hospitals in 11 urban counties† | All acute care hospitals in the state |
Number of hospital EDs in the study | 2 | 2 | 29 | 73 |
Percentage of state population covered by hospitals (catchment area) | 15 | 38 | 83 | 100 |
Dates of the ED visits in the billing dataset | January 2016- December 2018 |
January 2016- June 2018 |
January 2017-December 2017 | October 2015-September 2016 |
Number of diagnosis fields in the ED billing dataset | 30 | 25 | 30 | 34 |
Type of medical record reviewer | 18 clinical researchers | 2 trauma nurses | 1 professional MR coder | 3 professional MR coders |
Access to electronic medical record | Yes | Yes | ED report and EMS transport‡ | ED report and EMS transport‡ |
*In Maryland, one of the two sites was a neurotrauma referral centre, not an ED, and its trauma registry was used for sampling.
†The Colorado team selected the 29 acute care hospitals that had a trauma designation of level I, II or III, and located in Colorado’s 11 most populous counties. These 29 hospitals had 77% of all ED visits for intracranial injury and 81% of all ED visits for skull fractures (without an intracranial injury) in the state during the study time period.
‡Additional medical documents requested in Colorado: face sheet, radiology reports, toxicology reports. Massachusetts: face sheet, radiology reports, lab work/lab notes, triage notes, history and physical.
ED, emergency department; EMS, emergency medical services; ICD-10-CM, International Classification of Diseases, Tenth Revision, Clinical Modification; MR, medical record.