State Data Reporting Systems Division
E N D
Presentation Transcript
State Data Reporting Systems Division State Data Program • State Data System (SDS) • Model Minimum Uniform Crash Criteria (MMUCC) • Crash Outcome Data Evaluation System (CODES) Fatality Analysis Reporting System (FARS) 29th International Traffic Records Forum Denver, Colorado July 13, 2003
NHTSA’s State Data System Expansion Effort
SDS • A collection of crash data received directly from the 21 participating states. • The data consists of all state-reported crashes. Census data received annually. • The primary use of state data is for specialized internal studies.
SDS • State data are converted into a standard SAS format—the data structure is similar to FARS. • State variable attributes are retained. No recoding or standardization of attributes is done in SDS. • State annual Traffic Facts are used to ensure that data is processed correctly.
Data Confidentiality • Data files are sanitized: • All personal identifiers are removed during creation of the SAS data files. • VINs are truncated to 12 characters to protect vehicle owners. • General information may be retained, for example: • State of Vehicle Registration • Driver Zip Code
Internet Access • Internet access to state raw data files is strictly prohibited. • Internet access to NHTSA’s SAS data files is strictly prohibited. • If state chooses to allow its crash statistics to be published, statistical summary information may be posted on the Internet.
How SDS Data are Used • Rollover Analysis: • Firestone/Ford Explorer rollover analysis. • Implementation of rollover ratings in NHTSA’s New Car Assessment Program. • Supports NHTSA’s Office of Defects Investigation. • Useful for evaluating state voluntary implementation of MMUCC. • Useful for linkage to injury outcome data in CODES.
How SDS Data are Used • Useful for evaluating the effectiveness of vehicle safety equipment and safety campaigns: • Safety belts. • Anti-lock braking systems. • High mounted stop lamps and daytime running lights. • Alcohol awareness campaign. • Moving young children to the back seat.
Crash Data Report Crash Data Report: 1990-1999 available for downloading. Includes descriptive stats in tables and charts. http://www-nrd.nhtsa.dot.gov/departments/nrd-30/ncsa/sds.html
California Florida Georgia Illinois Indiana Kansas Maryland Michigan Missouri New Mexico North Carolina Ohio Pennsylvania Texas Utah Virginia Washington SDS States Before Expansion Effort
Why Expand SDS? • Dr. Jeff Runge has made data collection one of NHTSA’s top priorities. • Additional state data strengthens NHTSA’s ability to provide accurate assessments, leading to better public policy and improved traffic safety.
Why Expand SDS? • State data are diverse—Although they share a common purpose, each state’s PAR consists of a unique set of data elements and attributes. • The success of specialized studies depends on pertinent information being available on the state PARs. • Expanding SDS can potentially increase the sample size for these studies.
Main Expansion Approaches • Regional Office Solicitations—Regional Staff contact state officials directly regarding SDS. • Traffic Records Coordinating Committee (TRCC) meetings—SDS representative makes direct pitch with assistance from Regional Staff in attendance.
Both Approaches Successful Direct Solicitation by RO Staff • South Carolina (1/03), thanks to Region IV’s Erick Moran. • Wisconsin (5/03), thanks to Region V’s Bob Pollack. TRCC Meetings • Connecticut (3/03), thanks to Region I’s Mario Damiata. • Delaware (7/03), thanks to Region III’s Rod Chu.
Probable New Members • Kentucky—Based on work by Region IV’s Erick Moran. NHTSA is expecting an MOU from KY. • West Virginia—Positive response to TRCC briefing on 6/18, with valuable assistance by Bill Naff.
Other SDS Invitations Sent • Arkansas • Massachusetts • Mississippi • Montana • New Jersey • Tennessee
States Declining Membership • New York—Concerns over losing data sales revenue, data publication pre-emption, data misinterpretation, and lack of resources to reconcile discrepancies between NHTSA and state totals. • Rhode Island—Concern over data misinterpretation and incomplete nature of RI data files. • Vermont—"No compelling reason to join".
Two Levels of SDS Membership • State permits dissemination of state-specific summary information via publications and other methods. • State prohibits dissemination of state-specific summary information. For research purposes, the data is aggregated, making it impossible to determine the state’s contribution to the analysis pool.
US DOT Data Access Policy • US DOT researchers outside of NHTSA do not have access to SDS data files unless written permission is obtained from state. • During the approval process, state is advised of the specific intended use of state data by the requesting US DOT modality.
Public Data Access Policy • Public access to SDS data files is prohibited unless the researcher obtains written permission from state. • During the approval process, state is advised of the specific intended use of state data by the researcher.
Where Do We Go From Here? • Are the current SDS marketing materials sufficient for state solicitations? • Should we continue SDS presentations at TRCC meetings? • What additional assistance is needed by Regional staff to help with Dr. Runge’s data initiative?
MODEL MINIMUM UNIFORM CRASH CRITERIA • MMUCC 2nd Edition (2003), published June 2003. • Standardizes state crash data to generate the comprehensive information needed to improve highway safety within each state and nationally. • Facilitates inter- and intra state comparisons.
MMUCC FORMAT • Data element name, definition, set of attribute values. • Based on existing standards (ANSI D16.1, ANSI D20.1, FARS, NASS and FMCSA). • Data elements collected at scene, derived, or obtained through linkage. • Voluntary implementation. No MMUCC police!!
REPORTING THRESHOLD • Death, personal injury or property damage of $1,000 or more. • All involved persons (injured and non-injured). • Consistent and uniformly implemented statewide.
BENEFITS OF MMUCC • Facilitates inter- and intra state comparisons. • Standardizes population-based data to identify national trends and issues.
NEXT STEPS • MMUCC incorporated into TraCS software • Next revision in 2007; publication in 2008 • Assistance to be available: • web-based training • marketing materials • roll call video for law enforcement • surveys of state implementation • best practices guides, etc.
CRASH OUTCOME DATA EVALUATION SYSTEM • 29 CODES States funded. • Fifty percent or more of the states in nine of the 10 NHTSA Regions. • 41% of CODES states also part of SDS. • Goal is to add all CODES states.
PURPOSE OF CODES • Expands crash data so that all components of highway safety can be evaluated in terms of death, injury, injury severity and total inpatient charges.
CODES MODEL Links injury outcome to: • specific person, vehicle and event characteristics for all persons, injured or uninjured, involved in crashes statewide.
CRASH DATA LINKED TO: • Injury Data • EMS, ED, Inpatient, Trauma Registry • Death Certificate, Medical Examiner • Insurance Claims (health or vehicle) • Other Traffic Records • Roadway, Traffic Citation, Adjudication • Driver License • Vehicle Registration • Training: motorcycle, impaired driver
CODES DATA NETWORK • 21 (72%) of CODES states participate in the CODES Data Network • Purpose: • Facilitate use of linked data by NHTSA analysts • Assist states to institutionalize CODES
SIGNIFICANCE OF CODES FOR STATES • Only source of population-based state-specific highway safety-related outcome data that: • Indicates effectiveness of countermeasures in terms of death, injury, injury severity and costs. • Justifies the priorities in the state highway safety strategic plan. • Provides a permanent data base to monitor trends over time.
