Cincinnati, OH: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, Contract 200-2000-08018, 2006 Jun; :1-508
The purpose of this project was to evaluate options for the collection of occupational health and safety surveillance data in the healthcare industry used for planning the National Exposures at Work Survey (NEWS). NEWS will collect descriptive data on health and safety activities in the workplace and will assess workers' potential exposures to chemical and physical agents and biologic, ergonomic, and safety hazards. This feasibility assessment involved five major project activities which are summarized in this report. Working in close cooperation with NIOSH technical managers on this contract, Battelle has: 1. Conducted a comprehensive review of the literature pertaining to the occupational safety and health concerns and exposures of healthcare workers; 2. Solicited and synthesized information from healthcare stakeholders in order to identify health and safety priorities and the potential uses of national hazard surveillance data; 3. Developed two questionnaires - one targeted to facility health and safety managers and the other to employees, both designed to elicit the types of data deemed critical by NIOSH and stakeholders; 4. Extensively refined the questionnaires through a series of focus groups and cognitive interviews with healthcare managers and employees, obtaining input on clarity, instructions, and content; and, 5. Conducted pilot tests to evaluate different survey modes (i.e., paper and pencil versus web-based) and methods (i.e., targeting, recruitment, and distribution) for conducting the surveys in participating hospitals. In order to design effective survey instruments, it was important to gain an understanding of how national surveillance data would be used. While many different potential uses for survey data were provided by stakeholders, the three most frequently cited uses for the NEWS data were to: 1) provide national norms for healthcare facilities to compare themselves against (e.g., benchmarking); 2) describe and assess the use and effectiveness of various engineering or administrative controls (e.g., best practices); and 3) assess the health and safety culture of the organization. Based on NIOSH surveillance goals and objectives and on input received from industry stakeholders, we designed and extensively refined Management and Employee Questionnaires through a series of focus groups, cognitive interviews, and post-survey validation interviews with employees and managers. Both questionnaires are designed to take a broad approach to surveillance by eliciting descriptive information related to workplace and employee characteristics, health and safety policies and practices, and information on a diverse range of specific hazards, rather than an in-depth study of any particular hazard. The Employee Questionnaire is modular in format, consisting of a core module and ten hazard-specific modules. The Core Module is envisioned to be completed by all employees and is focused on perceptions of health and safety culture, as well as broad-based issues including ergonomics, workplace violence and blood-borne pathogen exposure; while the Hazard Modules focus on selected chemical agents to gather information that would permit a qualitative assessment of workers' potential exposure to the respective hazards. Each module is designed to elicit information related to the location, frequency and duration of use, work practices, and exposure controls. The Management Questionnaire elicits facility-wide information regarding health and safety resources, programs, policies, and practices. We pilot tested the survey tools and methods in two federal hospitals to evaluate the quality of the information obtained by the Employee and Management Questionnaires as well as to assess the survey process itself. We evaluated two modes of administration and different methods of recruiting employees and distributing the Employee Questionnaire. We established sampling and distribution procedures designed to attain the highest possible response rates, especially among employees potentially exposed to those hazards addressed by the Hazard Modules. Specific consideration was given to issues (such as limited resources and overall survey burden) which will apply to a future national survey. The lessons learned from the pilot tests indicate that: 1. The burden of conducting an establishment-based survey was greater and more complicated than expected. Obtaining the cooperation and approval to conduct an employee survey in healthcare facilities was complicated and time-consuming, especially due to different requirements of facility Institutional Review Boards (IRBs). 2. Survey participation should be endorsed by top management or other "key informants." Participation rates in pilot tests were significantly affected by perceived and explicit support for the survey provided from "key informants" (including the management, supervisors, and peers at an establishment) who help to establish the legitimacy and value of the survey. 3. Targeting of employees with specific exposures to receive customized questionnaires is not feasible. Targeting employees to receive "customized" hazard-specific questionnaires based on a priori assumptions of exposure (using job titles and/or management guidance) was frequently inaccurate and inefficient and imposed additional burden on the institutions involved. 4. Passive recruitment did not increase participation by non-sampled employees. While not explicitly evaluated in these pilot tests, the limited use of informational posters in one pilot test was ineffective for increasing participation among employees not included in the survey sample. 5. Highest response rates were obtained when the paper version was provided at time of invitation. Employees who initially received a recruitment letter and paper questionnaire responded at higher rates than those who initially received only a recruitment letter with instructions on how to obtain a paper questionnaire or how to access the web version of the survey. 6. Response rates were lower, but acceptable, even when the paper questionnaire was not initially provided. Employees who received only instructions for obtaining a paper questionnaire or for completing a web-based questionnaire strongly preferred the web version. 7. Employees generally responded at similar rates across job categories. 8. Requiring participants to request any additional Hazard Modules decreased completeness of responses. Participants who completed a web version completed all applicable modules since these were presented seamlessly. The burden for obtaining and completing all applicable questionnaires should be minimized to increase both participation and the completeness of responses. 9. An employee survey effectively describes the variability of health and safety conditions at an establishment. Cumulatively, responses to the Employee Questionnaire provided a more comprehensive description of the scope and extent of occupational exposure issues than provided by the Management Questionnaire. 10. Self-reported data collected from employees was accurate. Validation of survey data suggested that employees were generally able to provide accurate responses to specific issues concerning health and safety exposures, including the identification of health hazards, presence of engineering controls, and their use of PPE/C. Based on the results from this limited feasibility evaluation, we offer the following recommendations for planning and implementing the NEWS in the healthcare sector: 1. A national-scale population-based Employee Survey of the healthcare sector should be conducted through professional associations and union organizations to help maximize the efficient collection of information from a large and diverse pool of respondents. 2. A strong commitment of support of the survey should be obtained from management and communicated through key informants within participating organizations. 3. Incentives to prospective participants of the NEWS should be considered to help maximize participation rates. 4. The method(s) for distributing survey correspondence should be selected based on accessibility by NIOSH to information required to contact employees (such as name and address). 5. Survey correspondence to employees should be distributed via email, whenever possible, to minimize costs and maximize efficiency. 6. The current Employee Questionnaire that was developed during this feasibility evaluation should be modified for use in a population-based survey. 7. A web-based questionnaire should be the primary mode for the Employee Survey to simplify logistics, to minimize implementation and data management costs, and to maximize data quality (including completeness of responses). 8. A paper questionnaire should also be available, upon request, for the Employee Survey since some employees will not have access to computers and/or prefer the paper mode. 9. Management Survey should be conducted of those establishments identified from the Employee Survey. 10. Site visits should be conducted at a sample of establishments in the Management Survey to validate data collected from both the Employee and Management Surveys. 11. Results from both the Employee and Management Surveys should be posted on a NEWS website.