SW AgCrash: leveraging motor vehicle crash data for injury surveillance and research in AFF.

Details

• Personal Author:
  Garcia T ; Jisung K ; Kum H-C ; Pant A ; Shipp E ; Subasish D ; Trueblood A ; Vasudeo S
• Description:
  Background: Workers in the Agriculture, Forestry, and Fishing sector (AFF) experience substantially higher motor vehicle related fatal injury rates compared to workers in other sectors. However, little is known about the magnitude of this problem and how to best address it specifically in the Southwest region (AR, LA, NM, OK, and TX). The objective of the Southwest Agricultural Crash Surveillance System (SW AgCrash) project is to inform the development of AFF injury surveillance systems while also identifying factors contributing to transportation-related injuries in AFF populations. In addition, crash narratives are free-text fields that are a rich source of data for identifying relevant crashes and illuminating causal factors. This resource has yet to be fully incorporated into surveillance and research in AFF or other sectors. Methods: This three-year project began in 2016 and involves constructing a regional crash surveillance system for the Southwest region. The system is comprised of a crash database largely populated by state crash reports that police officers collect on-scene. It also includes data cleaning and analysis protocols including descriptive and inferential statistics and mechanisms for dissemination of information including via an advisory panel. In addition, it involves developing simple and complex methods for mining the crash narratives such as a Microsoft Excel-based tool and advanced computer science techniques (e.g., Natural-language processing, support vector machines). Results: The database currently includes data on over 5 million crashes. Each state, at a minimum, contributed data for 2011-2015 with more recent years expected to be added as available. From 2012-2015, there were 2,084 crashes involving agricultural vehicles and 1,240 crashes involving logging vehicles. As expected based on population size and economic factors, the greatest number of crashes occurred in Louisiana and Texas. Analyses are currently underway to estimate crash rates for each state and identify contributing factors. Preliminary findings are available for logging-related crashes in Louisiana from 2010-2015 (n=368). In a multiple logistic regression model, behavioral factors associated (p<0.05) with an injury among logging drivers involved in a single motor-vehicle crash included: no or improper seatbelt use (OR=3.12), careless operation of a vehicle (OR=3.19), and other non-behavioral factors. Most states do not have a structured data field to indicate which vehicle is at-fault. Analyzing the crash narratives along with the structured data allowed for developing a method for assigning fault to a specific vehicle. Among the logging drivers involved in a crash with at least one other vehicle, approximately 46% were found to be atfault for the crash. Additional findings are forthcoming with respect to this analysis and additional benefits of mining the crash narratives. Discussion: Motor vehicle crash records are a credible and valuable source of data for monitoring injuries among agricultural and logging workers and producing data to support outreach and other interventions in the Southwest region. Structured data can provide information on injury frequency, severity, and contributing factors at the person, vehicle, crash, and environment levels. Narrative data can be mined to provide additional details and to assess the quality of the structured data. Both simple and complex data mining methods are useful to this end.
• Subjects:
  Accidents, Occupational Acids Occupational Health Preceptorship Safety Skin Diseases
• Keywords:
  Accidents Accident Statistics Logging Industry Motor Vehicles Surveillance Truck Drivers
• Publisher:
  National Institute for Occupational Safety and Health
• Document Type:
  Text
• Funding:
  Cooperative Agreement
• Genre:
  Abstract
• Place as Subject:
  OSHA Region 6 ; Texas
• CIO:
  National Institute for Occupational Safety and Health (NIOSH)
• Topic:
  Workplace Safety & Health
• Location:
  North America
• Pages in Document:
  143-144
• NIOSHTIC Number:
  nn:20070067
• Citation:
  National Occupational Injury Research Symposium 2018, (NOIRS 2018), October 16-18, 2018, Morgantown, West Virginia. Morgantown, WV: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, 2018 Oct; :143-144
• Federal Fiscal Year:
  2019
• NORA Priority Area:
  Agriculture, Forestry and Fishing
• Performing Organization:
  University of Texas Health Center at Tyler
• Peer Reviewed:
  False
• Start Date:
  20010930
• Source Full Name:
  National Occupational Injury Research Symposium 2018, (NOIRS 2018), October 16-18, 2018, Morgantown, West Virginia
• End Date:
  20270929
• Collection(s):
  National Institute for Occupational Safety and Health
• Main Document Checksum:
  urn:sha-512:c75c92aeb1c972e7f64f7240fa66aa99a7860c98493b5903a5450bed05be5381e32235484919a312bba3b641309bc8b6a3462c0865ea0e705e491d5f37f777c2
• Download URL:
  https://stacks.cdc.gov/view/cdc/181996/cdc_181996_DS1.pdf
• File Type:
  [PDF - 609.65 KB ]