Project Results

Public Policy Goal:

To develop and test a workers survey to determine the feasibility of creating a nationwide database for Workers' Compensation medical care outcomes.

Project Objectives:

• Develop a set of core questions and variables of value to multiple groups interested in Workers' Compensation health policy
• Determine a feasible methodology to measure as many of the core variables as possible
• Create a draft survey tool for the measurement of these variables
• Pilot the proposed methodology and draft survey tool
• Outline a feasible plan for the long-term development and implementation of a national Workers' Compensation database

Project Description:

This project represented a collaboration of 5 universities, 7 state WC agencies, the National Institute of Occupational Safety and Health (NIOSH), the International Association of Industrial Accident Boards and Commissions (IAIABC), and the Picker Institute. The project is testing an approach for creating a national WC medical care database by:

1) Querying key stakeholders to define the essential data elements
2) Developing a worker self-report survey
3) Exploring the feasibility of linking survey and administrative data
4) Implementing and maintaining a public-use database

Key Findings and Accomplishments:

• The greatest challenge to this project was (and still is) to obtain consensus on which policy questions most need to be asked and thus which variables are of greatest value to health policy.
• A number of specific predictors of health and economic outcomes can be proposed as a starting point, and with continued piloting of survey drafts the items and questions they address will evolve.
• The largest number of most salient predictive variables of cost and health outcomes can only be obtained at an individual level from the injured worker. A survey of injured workers is therefore a requisite part of any proposed implementation plan for a national Workers' Compensation database and a survey of injured workers obtained through state claims databases would be the most feasible option for the development of a national database.
• Given the limited resources of state agencies and skill limitations of agency staff, the preferred methodology for conducting interviews would be for state agencies to provide contact information to a survey organization which, in turn would conduct the interviews. By recommending the use of a professional survey organization at a state university, it is hoped that in most jurisdications, obstacles regarding confidentiality of data would be overcome. For those states that cannot provide personally identifying information about injured workers to anyone outside the agency, state agency employees would have to be trained to conduct the interviews. A formal WINS training of interviewers should be developed so that the interviewers will be consistent in their implementation of the survey.
• The first survey draft was designed to balance the need for detail with the need for brevity, and the need for existing (validated) measures with tailored questions of interest to the authors. Health outcomes included the SF-12 (encompassing the Physical Components and Mental Components scales of health, as well as Vitality and Fatigue), and Work Role Functioning. Economic outcomes included Current Work Status, Pre/Post-Injury Income Change, Economic Impact, and Days Lost from Work. Predicting variables included gender, age, race, Quality of Provider Interaction, Quality of Evaluating Physician Interaction, Claims Manager Fairness, Workers' Compensation Literacy, Worksite Retribution, Employment Insecurity, Seeking of Legal Help, Unionization Status, Employer Size, Nature of Injury, Part of Body Injured, Years of Education, and Type of Occupation.
• To make interstate comparisons it will be important to make sure each state consistently interviews persons injured between three and five months prior to that interview date.
• Since the interviews will be fairly lengthy, several strategies had to be developed to maximize response rate and interview completion rate. Defining calling strategies and scripts was very important.

Tools Developed:

• Worker Injury National Survey (WINS)
• How to Administer the "WINS" Survey:   If a high quality, comprehensive database is to be built through the completion of workers' surveys coupled with administrative data, strict protocols must be followed. In a part of this project, such protocols were developed and tested. Details on how a state would go about conducting the "WINS" survey are provided via link to the "WINS" website.

Website:
The WINS Website (under construction until March 2003) is currently housed on the University of Texas server,  http://www.sph.uth..edu/csph/wins/default.htm. It is also reachable by use of the alias,  http://www.winsonline.net.

Reports and Publications:

Journal Publications:

• Amick BC, Bazzani LG, Vridine D, Roberts K, Boden L, Markiewicz B, Zaidman, B, Pransky G. "Performance of State Workers' Compensation Systems from the Perspective of the Injured Worker: Scale Reliability and Validity for the Worker Injury National Survey (WINS) Project." [future link]
• Amick BC, Bazzani LG, Vridine D, Roberts K, Boden L, Markiewicz B, Zaidman B, Pransky G. "Comparison of Florida and Minnesota Claimant Profiles." [future link]

White Papers:

• The WINS Translations Process
• Sampling Tradeoffs
• State Interest in a National WC Survey and Database
• Administrative Data Availability
• Privacy Concerns
• The Importance of Using Valid and Reliable Scales
• Sample Size for Valid Economic Analyses
• Prototype Report for WINS Data Presentation
• Recommended Plan and Cost Estimates

Presentations:

• WCRG Conference, November 17, 2001.
• IAIABC Convention, October 2001.