Background: Despite increasing emphasis on safety culture assessment, little is known about the factors that affect hospitals' participation in such studies. Factors affecting recruitment of 30 Department of Veterans Affairs (VA) hospitals into a study to evaluate perceptions of safety culture, or safety "climate," were examined.
Methods: To minimize selection bias, hospitals were recruited that represented the spectrum of safety performance on the basis of Patient Safety Indicator scores. Invitations and additional mailings, informational conference calls, and personal contact with hospitals were used to encourage participation. Investigators worked closely with hospitals' key stakeholders to obtain support and buy-in for the study. Relationships among safety performance, organizational culture, and other hospital characteristics with hospitals' participation and ease of recruitment were examined. Findings were compared with those of a companion study in the non-VA setting.
Results: Despite attempts to optimize recruitment, it was necessary to contact more than 90 hospitals to obtain a 30-hospital sample. Having a more entrepreneurial culture (associated with risk-taking, innovation, and quality improvement) was significantly related to shorter recruitment time in VA and non-VA settings. Safety performance was significantly related to participation in the VA (that is, "better-performing" hospitals were more likely to be recruited than "lower-performing" hospitals), but not in the non-VA study, where recruitment was based on size and region.
Discussion: Researchers should recruit representative samples of hospitals based on measures of safety performance. Hospital selection bias could lead to erroneous findings, ultimately impeding efforts to improve safety within organizations.