banner of Methodology for Determining Motorcycle Operator Crash Risk and Alcohol Impairment

3. Detailed Report of Discussion

Emergency Departments

This methodology would involve a go-team going to hospital emergency departments to interview injured motorcyclists to obtain BACs, demographic, and other data. An alternative would be to have staff of the emergency department conduct interviews and collect data. This methodology would be similar to the Contemporary Case Control study in that it is a way of providing in-depth data on crash victims and would require some way of obtaining comparison data. It would be similar to the Injury Data study in that it would be necessary to obtain cooperation of local hospitals, in this case to allow interviews and institute a policy of BAC measures for all injured motorcyclists. As with the Injury Data study it may be difficult to get data on riders with only minor injuries and fatally injured riders. OMB and IRB clearance would likely be issues. It would be necessary to consider all issues associated with whichever method is used to provide comparison data (e.g., Contemporary Case Control or Geo-General Comparison Data study roadside rider surveys). Some hospitals, such as some Level I Trauma Centers, are already collecting BAC data on many of their patients (primarily those who are admitted to the hospital). As with other studies, it would be beneficial to conduct this study in an area with a large riding population. It seems likely that the more frequently a given hospital has motorcyclist victims, the more likely they will be to remember that there is a study and to correctly perform the survey protocol.

A possible variation on this methodology would be to combine it with the Contemporary Case-Control method for obtaining crash data. This hybrid system is currently being used for research in Thailand (Kasatikul, 2001). The go-team monitors police frequencies and hospital admissions (typically via the 911 system). This provides leads to Property-Damage-Only (PDO) crashes via the police, while also providing injury cases not reported to the police (riders will often go straight to hospitals to avoid police involvement in order to avoid prosecution, insurance issues, etc.). In the case of the Thailand study, a go-team makes its headquarters at the hospital to pickup unreported cases as they enter the hospital. The go-team monitors police activity and, based on information provided by the police, decides whether to go to the crash site or to conduct the interview at the hospital. The go-team has an ambulance on-call so that they can easily get to the PDOs, should they choose to go to the crash site.


This study would include injury crashes, which would be more representative of all crashes, and would provide more cases from a given area/time, than studying fatalities only. The BACs would likely be obtained from a blood sample, which facilitates the study of contributions of other drugs to the crash.


This study would not include PDO crashes. BACs are more removed in time from BACs collected at the scene. This study also would require cooperation from the hospital and crash victims. If emergency department staff is used to collect data, there will be a higher potential for erroneous and missing data due to the fact that hospital staff have other, higher, priorities, and would not have as complete an understanding of the project’s research needs as researchers would have. Hospital staff will also have difficulty providing roadway data and other variables that are routinely included in a police-reported crash report. It may be difficult to determine the crash site from data collected at the hospital.  Lastly, limiting data collection to the area served by a single hospital may not provide sufficient numbers of cases.



Cost issues for this methodology would be similar to those of the Contemporary Case Control study. If a go-team is used to collect crash data at hospitals, the costs would be similar to a Contemporary Case-Control study. If hospital staff collects crash data, costs could be reduced but with the potential effect on data quality discussed above. In either case, the cost of collecting comparison data in the field will be high.