The Problem of Undercounting Seriously Injured Bicyclists in Anchorage
In order to effectively improve bicycle safety in Anchorage, the Alaska Injury Prevention Center seeks strong data, which, in turn, allows it to prioritize funding for safety interventions, determine necessary engineering improvements, define performance measure and evaluate success.
Alaska’s Strategic Highway Safety Plan only used police crash data to measure the magnitude of the bicycle safety problem on the roads. However in many cases, cyclists who are seriously injured on roads do not call police, even if a car is involved. The result is a significant undercounting of the problem.
METHODS:
The AIPC manually matched Police Crash reports of serious bicycle injuries with Alaska Trauma Registry (ATR) bike injuries cases, which occurred on roads from 2001-2010. Bicyclists injured in places other than roadways were not considered; to remove non-highway safety related cases.
The definition of a case in the ATR was a bicycle injury, which occurred on a roadway in the Municipality of Anchorage. The assumption was made that an injury resulting in a minimum of 24 hours in a hospital warrants a finding of being a serious injury. For Police Crash (Crash) reports, a case was defined as involving an injured cyclist in the Municipality of Anchorage. The definition of a Crash case was expanded to include all bicycle crashes except those determined to result in “property damage only.” Variables used to match cases were: Gender, Age, Date of injury (with allowance of two days for delay in seeking medical), Time of day of injury, Review of ATR narrative to assist with inconsistencies
RESULTS:
Out of 1429 bike Crash cases, and 499 ATR roadway bike injuries, there were 113 matched cases. AIPC discovered that police crash reports significantly underreport bicycle crashes. Seventy-eight percent of hospitalizations from roadway bicycles crashes were not reported in police reports. And out of the 113 matched cases, only 2 were reported in police reports as major injuries.
CONCLUSIONS:
This is not an issue of one database being better than the other. Both databases have valuable information. Police crash reports provide timely and accurate information on the use of safety equipment, the location of the crash, a person’s behavior preceding a crash, and environmental conditions. Trauma registry is well equipped to define the severity of the injury, the medical outcome and the costs. Together, these two data sources provide a more complete story of the bicycle injury problem.
By using both data sources AIPC is now able to calculate costs and severity of bicycle injuries from trauma data and helmet use, roadway and behavior factors from crash data. For example, we can use geo data to determine the most dangerous areas for cyclists, advocate for improved engineering and provide costs of injuries to support our policy initiatives. Using good solid data will assist with choosing programs to implement and performance measures to track to calculate success.
Both of these data sources are now measures in Alaska’s State Highway Safety Plan (SHSP). The SHSP identifies a State’s key safety needs and allocates funding based on where it would have the most impact in saving lives and preventing injuries, and performance measures to determine successes.