An Application of CODES Data Linkages for Crashworthiness Computations

Multiple datasets (crash, emergency department, inpatient discharge, and death certificate) were linked in order to (a) follow crash victims through the healthcare system, and (b) examine types and severities of injuries sustained by vehicle occupants in head-on, rear-end, angle, and rollover crashes occurring on Massachusetts roadways in 2002 and 2003. The crash-involved vehicles were classified by type (passenger car or light truck/van) and age (pre-1995 or 2002-newer). Moreover, the injury types, hospital charges, and length of stay of the crash-involved occupants were considered. Finally, logistic regression models were developed to compare the injury odds ratios for each of the four types of crashes. Some of the findings included the following:

• New vehicles are more likely to be involved in property damage only crashes while old vehicles are more likely to be involved in all levels of injury crashes (fatal, incapacitating, non-incapacitating, and possible injury).
• The highest emergency department charges are associated with rollover and head-on crashes.
• For head-on crashes, the emergency department charges were significantly lower for occupants of new vehicles than occupants of old vehicles.
• In angle crashes, occupants of new vehicles were less likely to suffer spine and back injuries or sprains and strains. Occupants of new vehicles were more likely to suffer extremity injuries, fractures, and superficial contusions.
• In angle crashes, LTV occupants were more likely to suffer injuries to the spine and back area or sprains and strains than passenger car occupants; LTV occupants were less likely to suffer superficial contusions than passenger car occupants.