Frontline Workers' Attitudes Toward Evidence-Based Practice: Shaped by Organizational Context?

Background and Purpose: To address the challenge of transporting evidence-based practices to everyday community settings, researchers are working to understand mechanisms that promote frontline workers' implementation of innovative and/or evidence-based practice. Such efforts have recently focused on organizational contexts and the effects of shared attitudes among frontline staff (Glisson et al., 2008). According to Rogers' (2003) theory of the diffusion of innovations, “collective innovation” within an organization reflects shared attitudes, norms, and decisions about practice. Certainly, in human services, practitioner attitudes are strongly associated with practice behaviors, including evidence–based practice (Nelson & Steele, 2007). But to what extent are attitudes shared within organizations? This study addresses individual and organizational influences on frontline practitioners' attitudes toward evidence-based practice. It addresses the question: To what extent do practitioner attitudes toward evidence-based practice reflect organizational contexts? Methods: The study involves a cross-sectional survey of frontline counselors who provide outpatient substance abuse treatment. The study involved a nested probability sample. First, organizations were randomly selected from those meeting selection criteria based on location, size and type. Of the 70 organizations selected, 57 participated, for an organization-level response rate of 81%. Surveys were distributed in person and by mail to 429 counselors in the organizations and 293 (68%) participated. Attitudes toward evidence-based practice were measured with the Evidence-Based Practice Attitudes Scale (EBPAS) (Aarons, 2004), a 15-item measure with good psychometric properties. Organizational contexts were measured with scales designed to assess specific dimensions of organizational climate and culture (UT-CMHSRC, 2002; TCU-SOF, 2005). Multi-level (HLM) regression models were conducted to appropriately address the nested sample while assessing the association of individual and organizational-level factors with practitioner attitudes (Raudenbush & Bryk, 1992). Results: An unconditional HLM model indicated that vast majority of variation in attitudes toward evidence-based practice was within rather than between organizations. Only 3% of the variation in the global EBPAS score was between organizations. Random intercept, fixed slope HLM regression models indicate that at the individual level, having a master's degree increased positive attitudes toward evidence based practice, whereas more years in an agency and being African American decreased positive attitudes toward evidence-based practice. At the organization level, as organizations were more oriented toward change, practitioners had more positive attitudes toward evidence based practice. Other organizational characteristics related to climate, such as role overload or emotional exhaustion, or to culture, such as organizational focus on outcomes, were not related to attitudes toward evidence-based practice. Conclusions and Implications: Research demonstrates an important link between practitioner attitudes and the implementation of evidence-based practice in community settings. This study finds, however, that practitioner attitudes vary substantially within organizations. Shared attitudes toward evidence-based practice are uncommon in this sample of community-based substance abuse treatment organizations. Findings are consistent with theory predicting that early in a diffusion process, organizations are likely to include a range of early and late adopters. The study findings suggest that to promote evidence-based practice, human service organizations might do more to promote shared positive attitudes toward evidence-based practice among frontline staff.