Kentucky Trauma Registry and CRASH Integration and Surveillance Quality Improvement Related to Commercial Vehicles

The May 2012 Kentucky Traffic Records Assessment report by the National Highway Traffic Safety Administration Technical Assessment Team recommended the development of additional datasets that link the CRASH and injury surveillance/Kentucky Trauma Registry (KTR) systems (at sections 1-C and 2-F). A FY 2014 grant under section 405 initiated a baseline linkage of the 2012 KY Trauma Registry data and the 2012 CRASH data. The linkage is in progress, and the same project has added new state-specific data fields to KTR data in order to 1) improve the integration between CRASH and the KTR; and 2) improve the quality of the data collected in the KTR related to the presence of drugs and alcohol in injured drivers. The new data will list the specific alcohol blood levels, specific medications administered as part of patient care, and toxicology screen results by drug classification. The KTR reporting hospitals started pilot data submission with the new data elements in the last quarter of 2013. The new project will perform probabilistic linkage of the 2014 CRASH and KTR data. The KTR data has presented challenges for linkage because many of the trauma patients are first stabilized in emergency departments or small hospitals that are not part of the trauma system, and then transferred to a trauma center that could be far from the place of injury. In order to improve the KTR –CRASH linkage, new elements were added to the KTR data: EMS run number, transport agency, transport origin, and referring facility. All these new fields should improve the linkage in 2014, but achieving better accuracy and completeness brings new challenges and makes the linkage more complicated. Information like this has never been collected and is not collected in hospital and ED discharge data so the 2014 linkage between KTR and CRASH data will represent the first attempt to overcome the related methodological challenges. In addition, our analysis of KTR data from previous years showed that certain required variables are not well populated. This is especially true of variables like occupation, industry, and work-related injury, although they are required elements for submission to the National Trauma Data Bank. The linkage with CRASH data will help to address the completeness and accuracy of coding in these variables, particularly regarding commercial vehicle drivers whose crashes are most likely work-related and whose occupation should be easily populated. Special attention will be given to the severity of the injury among passengers in commercial vehicles by using the KTR data on injury severity as a gold standard. The results from the assessment of trauma data completeness and coding will be used for the development of training material for trauma registrars. Another goal of the project is to continue our work with the second vendor of trauma registry software, Digital Innovation, for the inclusion of the new state-specific data elements in the software used by trauma centers that are their customers