Evidence-based practice barriers and facilitators from a continuous quality improvement perspective: an integrative review

Journal of Nursing Management, 2011, 19, 109 120 Evidence-based practice barriers and facilitators from a continuous quality improvement perspective: an integrative review NANM.SOLOMONS MS 1,2 and JUDITH A. SPROSS PhD, RN, FAAN 3 1 Doctoral Student, Arizona State University, Phoenix, AZ, 2 Data Analyst, MaineHealth Center for Quality and Safety, Portland, ME, and 3 Professor, University of Southern Maine School of Nursing and Health Professions, Portland, ME, USA Correspondence Nan M. Solomons MaineHealth Center for Quality and Safety 400 Woodford St Portland ME 04103 USA E-mail: nan.solomons@gmail.com SOLOMONS N.M. & SPROSS J.A. (2011) Journal of Nursing Management 19, 109 120 Evidence-based practice barriers and facilitators from a continuous quality improvement perspective: an integrative review Aims The purpose of the present study is to examine the barriers and facilitators to evidence-based practice (EBP) using ShortellÕs framework for continuous quality improvement (CQI). Background EBP is typically undertaken to improve practice. Although there have been many studies focused on the barriers and facilitators to adopting EBP, these have not been tied explicitly to CQI frameworks. Methods CINAHL, Academic Search Premier, Medline, Psych Info, ABI/Inform and LISTA databases were searched using the keywords: nurses, information literacy, access to information, sources of knowledge, decision making, research utilization, information seeking behaviour and nursing practice, evidence-based practice. ShortellÕs framework was used to organize the barriers and facilitators. Results Across the articles, the most common barriers were lack of time and lack of autonomy to change practice which falls within the strategic and cultural dimensions in ShortellÕs framework. Conclusions Barriers and facilitators to EBP adoption occur at the individual and institutional levels. Solutions to the barriers need to be directed to the dimension where the barrier occurs, while recognizing that multidimensional approaches are essential to the success of overcoming these barriers. Implications for nursing management The findings of the present study can help nurses identify barriers and implement strategies to promote EBP as part of CQI. Keywords: continuous quality improvement, evidence-based practice, integrative review, nursing Accepted for publication: 9 June 2010 Introduction There are many reasons for using EBP: to improve or update clinical practice (Institute of Medicine 2001), to meet organizational credentialing requirements (ANCC ND) and to improve the quality of care and outcomes for patients (McBride 2008, CMS ND). Evidence-based practice (EBP) is defined as Ôthe integration of best research evidence with clinical expertise, and patient valuesõ (Sackett et al. 2001, p. 1). Using EBP requires a DOI: 10.1111/j.1365-2834.2010.01144.x ª 2010 The Authors. Journal compilation ª 2010 Blackwell Publishing Ltd 109

N. M. Solomons and J. A. Spross systematic search and evaluation of research; clinical expertise; and consideration of the patientõs preferences and values (Melnyk & Fineout-Overholt 2005). EBP also requires personal and organizational commitment to change practice. Clinical research, evidence-based practice (EBP) research, and quality improvement (QI) research are distinct yet related areas of inquiry. QI activities often provide the local context for EBP efforts whereas clinical research provides empirical evidence for EBP (Newhouse 2007). QI studies may be used to confirm that evidence-based practices are being implemented. It seems important to integrate findings from these three areas in order to understand and address barriers and facilitators to adopting the best evidence and improve care delivery and patient outcomes. The link between EBP and continuous quality improvement (CQI) was recognized as early as 1998 in a classic article by Shortell, a health services researcher who continues to write extensively about healthcare improvement and reform. Shortell et al. (1998, p. 605) defined four foci of CQI. CQI focuses on: determining and meeting the needs of patients or customers; a holistic approach to quality improvement, based on identification of underlying causes of poor performance; fact-based management and scientific methodology, which make it culturally compatible with the values of health professionals; and empowering its practitioners to improve quality on a daily basis. CQI is different than other quality approaches because it incorporates system-level root cause analysis to understand and correct the underlying work processes so as to add value, not point out individualsõ mistakes (Shortell et al. 1998). The selected studies describe the barriers and facilitators to EBP adoption. ShortellÕs four dimensions of quality improvement provide a means of analysing and organizing findings about EBP barriers and facilitators across studies that have used disparate theories and instruments. The barriers to implementing EBP and research findings have been under study for a significant period of time, yet the remedies are few and far between. If these barriers are to be overcome, according to Shortell, all of the dimensions that contribute to the barrier must be addressed. This is a weakness in many of the studies that were examined in this analysis. CQI is a critical contextual organizational factor in adopting EBP. The purpose of the present integrative review is to examine individual and institutional barriers and facilitators to EBP using ShortellÕs CQI framework. Context To analyse and organize the literature, the authors used ShortellÕs CQI framework and the Pravikoff et al. national (US), comprehensive descriptive study of EBP from 2005. ShortellÕs CQI framework is comprised of four interrelated dimensions: strategic, cultural, technical and structural. The strategic dimension includes those activities and processes that are most important to the organization and provide the greatest opportunity for improvement (Shortell et al. 1998, p. 606), such as vision, budget priorities and long-term strategy. The cultural dimension represents the organizationõs Ôbeliefs, values, norms and behaviorsõ (Shortell et al. 1998, p. 605). The technical dimension encompasses training and information infrastructure (Shortell et al. 1998). The structural dimension refers to the ways that knowledge is acquired and dispersed throughout the organization (Shortell et al. 1998). Although there were other studies of barriers to EBP prior to 2005, Pravikoff et al. (2005) was the only national study found in the time period considered in this review. Using stratified random sampling, Pravikoff et al. (2005) surveyed 3000 nurses in the US to assess their overall readiness for widespread implementation of EBP. Eighty-six per cent of the sample was white and 41% received their most recent nursing education prior to 1985 (Pravikoff et al. 2005). Excluding time, the top three individual barriers identified were not seeing the value of research for practice; unfamiliarity with bibliographic databases such as CINAHL and PubMed; and difficulty accessing research materials (Pravikoff et al. 2005). The research also identified institutional barriers that prevented EBP adoption. The top three institutional barriers identified were: more important priorities than EBP, difficulty with recruitment and retention of nurses and insufficient monies to subscribe to information sources (Pravikoff et al. 2005). This work has been used in both the design and interpretation of subsequent research on EBP. Cadmus et al. (2008) replicated this study with 32 acute care hospitals in New Jersey. Beke-Harrigan et al. (2008) modified the Pravikoff study for their research on EBP at the Ohio Aultman Health Foundation. Pravikoff et al.õs (2005) results provide a basis for analysing individual and institutional barriers, facilitators and interventions for EBP implementation, also analysed using ShortellÕs CQI framework. 110 ª 2010 The Authors. Journal compilation ª 2010 Blackwell Publishing Ltd, Journal of Nursing Management, 19, 109 120

EBP barriers and facilitators from a CQI perspective Methods CINAHL, Academic Search Premier, Medline, Psych Info, ABI/Inform and LISTA databases were searched using the keywords: nurses, information literacy, access to information, sources of knowledge, decision making, research utilization, information seeking behavior, nursing practice and EBP. References from selected articles and bibliographic database referral articles were also reviewed. A total of 941 non-unique articles were retrieved. A preliminary search of the Cochrane database, using the term evidence-based practice nursing, yielded 65 articles in which the majority addressed specific clinical topics and were not relevant to this analysis. Inclusion criteria for this review were: a peerreviewed publication between 2004 and 2009 conducted in the United States or Canada, and samples that included practicing nurses. Exclusion criteria were editorial and opinion pieces and research conducted outside of the United States and Canada. The 5-year time span was chosen because of the expansion of internet access during that time, availability of online resources and a greater number of publications on EBP. Concerns about varied cultural influences and financing mechanisms and their impact on interpretation of EBP and CQI led to a decision to exclude studies from Europe and the Asia-Pacific. Canadian studies were included because the Canadian healthcare system has been under discussion as a model for reforming the healthcare system in the United States (Lasser et al. 2006). CanadaÕs proximity to the United States, participation and access to each otherõs EBP resources such as the Registered Nursing Association of Ontario, and evidence of collaboration among EBP and CQI authors in both countries were also factors that were relevant for including Canadian studies. After these criteria were applied, 23 articles were eligible for review. A summary of these studies including setting, sample size, study design and findings can be found in Table 1. Results Overall study comparisons This section describes the group of 23 studies based on author discipline, theoretical frameworks, setting, respondents and informants, study design and barriers and facilitators measures. Broadly, the descriptive studies tended to answer the question ÔWhat are the barriers to implementing EBP in the workplace?õ Studies that described solutions to barriers of EBP adoption measured their success as increased EBP utilization among participants. Author discipline One usually thinks of EBP as a topic solely within the domain of health professionals; however, five out of 23 studies were authored by librarians or information science professionals (Byrnes et al. 2004, Andrews et al. 2005, Dee & Stanley 2005, McKnight 2006, Beke- Harrigan et al. 2008). Information and library science are different disciplines from nursing. Excluding Beke- Harrigan et al. (2008), these studies focused on information access and usage patterns among nurses and other providers. Finding and accessing the research is one part of EBP. Andrews et al. (2005) assessed information seeking behaviour such as access to information, use of technology for information and usage patterns. Byrnes et al. (2004) reported on the effectiveness of PubMed and internet search training on a group of providers in rural, upstate New York. Dee and Stanley (2005) compared information access and information use between graduate nursing students and clinical nurses working in not-for-profit facilities without a medical library. Theoretical frameworks Seven studies were guided by RogersÕ Diffusion of Innovations Theory (Fink et al. 2005, Kenny 2005, Karkos & Peters 2006, Atkinson et al. 2008, Brown et al. 2009, Gale & Schaffer 2009, Schoonover 2009). Melnyk et al. (2004) used the Transtheoretical Model of Organizational Change and Control Theory. The remaining 15 studies did not describe a theoretical framework. Settings Five out of the 23 studies were Canadian (Estabrooks et al. 2005, Kosteniuk et al. 2006, Profetto-McGrath et al. 2007, 2009, Borycki et al. 2009). Of these, one study compared information seeking behaviour of nursing students in two simulated environments. In one environment all of the information was on paper, in the other environment some information was on paper and some information was computerized (Borycki et al. 2009). One report was part of a larger study of nurses in four hospitals (Estabrooks et al. 2005), while Profetto-McGrath et al. (2007) studied seven Clinical Nurse Specialists (CNSs). The 2009 study by Profetto- McGrath was based on results from the earlier research. Kosteniuk et al. (2006) examined information usage among rural and remote nurses. The remaining 18 ª 2010 The Authors. Journal compilation ª 2010 Blackwell Publishing Ltd, Journal of Nursing Management, 19, 109 120 111

N. M. Solomons and J. A. Spross Table 1 EBP Barriers and Interventions literature First author, Year Setting Sample Design & Instruments (Reliability) Key findings Andrews, 2005 Members primary care practice-based research network in KY Atkinson, 2008 Community hospital Illinois Beke-Harrigan, 2008 US OH Aultman Health Foundation (health system N.E. OH. Mid size tertiary care facility Borycki, 2009 Lab simulated hospital environment (nursing station) Canadian Brown, 2009 Academic medical center in California N = 59 (16 RNs, 41 MDs) Cross sectional Instrument: ID (NR) N = 249 (RN) Descriptive quantitative Instrument: B (NR) N = 443 (RN primarily clinical bedside) Descriptive correlational design. Instrument: P-based (NR) N = 35 (novice RN) Within group lab experimental study block randomized N = 458 (RN convenience sample) Descriptive cross sectional Instrument: B (NR) EBPQ (0.87) Burns, 2009 196 bed rural hospital N = 25 (RN) Implementation plan for EBP. Instrument: ID (NR) Quasi-experimental. Byrnes, 2004 10 primary care practices upstate New York Cadmus, 2008 32 acute care hospitals in New Jersey Dee, 2005 3 not for profit healthcare facilities without libraries Estabrooks, 2005 7 units in 4 hospitals in Alberta and Ontario Canada Fink, 2005 Large university affiliated Magnet hospital N =70 (clinician) N = 3411 (acute care RN) N = 50 (25 graduate level nursing students; 25 clinical RNs) Pre-survey, post-survey, 3 month post-survey Instrument: ID (NR) Descriptive exploratory Instrument: P-based (NR) content reliability validated Descriptive exploratory Instrument: ID (NR); interviews and observation N = 230 (RN) Descriptive. Part of a bigger study Instrument: ID (Reported elsewhere) N = 215 (RNs) Pre-survey; N = 239 (RNs)Post-survey Gale, 2009 Level 1 Trauma Center N = 92 (staff nurses & nurse managers) Karkos, 2006 Community Magnet hospital N = 275 (222 staff nurses; 53 other) Quasi experimental cross-sectional pre- post-survey design Instrument: B (0.91) RFQ (0.89) Descriptive Instrument: ID (NR) Descriptive quantitative BARRIERS instru ment. Reliability not reported Diffusion of Innovations Theory Information accessed for patient care; little use of online resources Top 2 barriers not enough authority to change practice and not enough time to implement new ideas New findings: Gen Y seek more work-life balance and Lack of formal research infrastructure Top two barriers: lack of understanding of electronic databases and difficulty accessing research materials; 70% did not use library; 59% had no instruction in electronic databases. 42%; rarely need research in their job Novice nurses seek less information in a hybrid environment; may be due to information overload and result on dependency of memory. Systems implementation should be aware of information overload for novice nurses Top 2 barriers lack of time and lack of autonomy; education and mentorship needed to change practice; change in culture important Importance of: CNO champion, staff involvement, access to research expertise, continuing education opportunities for EBP. Increase use of MEDLINE Increase internet searching; perceived importance of EBM to patient care increased 35% Library used infrequently; clinical nurses limited engagement in research. Top 2 institutional barriers: allocation and acquisition of resources Continued preference for colleagues and print sources for information; nursing students more likely to use all means; both groups lack database searching skills Preference for knowledge received through co-workers, experience and patients over journal or textbook; results similar to 2 previous studies; lack of standardization across instruments Interventions at many levels; improvement in perception of barriers and organizational culture; journal club participation linked to increase in research utilization EBP barriers: insufficient time, lack of staff, lack of equipment and supplies; EBP facilitators: interest in practice change, avoid risk of negative consequence to patient, personally valuing the evidence Top 3 barriers: not enough time to read research; insufficient authority to change patient care procedures; insufficient time on the job to implement new ideas; Magnet status may make setting a facilitator instead of a barrier 112 ª 2010 The Authors. Journal compilation ª 2010 Blackwell Publishing Ltd, Journal of Nursing Management, 19, 109 120

EBP barriers and facilitators from a CQI perspective Table 1 (Continued) First author, Year Setting Sample Design & Instruments (Reliability) Key findings Kenny, 2005 US Army medical treatment facilities. 3 hospitals north Atlantic regional medical command Koehn, 2008 Large urban medical center Kosteniuk, 2006 Rural and remote not hospital Canada N = 290 (130 civilian 160 military RNs) N = 422 (328 staff nurses, 53 unit managers/advisors, 41 other) N = 3933 (rural & remote RNs) Descriptive organizational Instrument: Research Utilization by Estabrooks (0.77 0.91) MyllesÕ Organizational Climate Survey (0.72 0.89) Descriptive cross sectional Instrument: CE-EBP (0.91) Descriptive cross sectional Instrument: Satisfaction With Home Community Scale (Henderson Betkus & MacLeod 2004) (NR); Work Satisfaction Scale (Stamps 1997) (NR). Used evidence in practice <50% of time although research was deemed to be important to practice; 38.6% did not know if there was a dedicated champion 2 most frequently cited barriers were time and knowledge; education levels a factor in attitudes about EBP. More likely to use colleagues, inservice, newsletter as top 3 sources of information; access to evidence is difficult for this population McKnight, 2006 Critical care hospital southern US Melnyk, 2004 Various settings N = 160 Convenience sample from conferences & workshops OÕLynn, 2009 Rural northern US SD, MT, OR N = 6 Observation, in context interviews Behavior centered on patient, seeking information from people, patient record, other systems. Reading journals on the job was considered ethically wrong by some nurses N = 200 (rural nurses) Pravikoff, 2005 US N = 760 Stratified sample Profetto-McGrath, 2007 Profetto-McGrath, 2009 Schoonover, 2009 Community hospital Washington Descriptive Instrument: ID (NR) Descriptive cross-sectional Instrument: ID (0.643 0.863) Descriptive exploratory Instrument originally used by Tanner and Pierce (NR) Western Canada N = 7 (CNS) Descriptive exploratory Interview Western Canada N = 94 (CNS) Descriptive cross sectional Instrument: ID (NR) via phone N = 79 Descriptive Instrument: B (0.30 0.53) Knowledge of EBP was lower than beliefs; use of evidence in practice correlates to having a mentor and using EBP databases Greater internet access than expected; preference for using coworkers for information; defecit in research interpretation among sample Barriers are both individual and institutional; top two individual barriers are lack of value of research in practice, understanding electronic databases; top two institutional barriers are goals with higher priority, problems recruiting and retaining nursing staff CNS same barriers as staff nurses; evidence is used from many sources including research, experience and coworkers Evidence from many sources; people and experience most prevalent sources of information used among CNS; journal clubs were the least accessed source of evidence Organizational characteristics greatest barriers to research utilization; top two barriers are lack of authority and no time to read research. B, Barriers; CE-EBP, Clinical Effectiveness Evidence-based Practice Questionnaire; ID, Investigator Designed; NR, Not Reported; P, Pravikoff; RFQ, Research Factor Questionnaire. ª 2010 The Authors. Journal compilation ª 2010 Blackwell Publishing Ltd, Journal of Nursing Management, 19, 109 120 113

N. M. Solomons and J. A. Spross studies were done in the United States. Only the report by Pravikoff et al. (2005) was a national study. Ten studies were done in hospital settings of which three were community hospitals (Karkos & Peters 2006, McKnight 2006, Atkinson et al. 2008), one was a tertiary care facility (Beke-Harrigan et al. 2008), two were academic medical centres (Fink et al. 2005, Brown et al. 2009), one compared nurses in three health care facilities without medical libraries (Dee & Stanley 2005), one was a trauma centre (Gale & Schaffer 2009), one was a rural agency (Burns et al. 2009) and one was an urban medical centre (Koehn & Lehman 2008). Of these ten, four hospitals had obtained Magnet status (Fink et al. 2005, Karkos & Peters 2006, Atkinson et al. 2008, Beke-Harrigan et al. 2008). Two hospitals were in the process of applying for Magnet status (Brown et al. 2009, Gale & Schaffer 2009). Of the 32 New Jersey hospitals in the Cadmus et al. (2008), study 11 hospitals had obtained Magnet status. The remaining six studies occurred in other settings. OÕLynn et al. (2009) studied the sources of information used by rural nurses in the northern Midwest and Cadmus et al. (2008) examined barriers to EBP among nurses at 32 acute care hospitals in New Jersey. Melnyk et al. (2004) used a convenience sample of nurses attending EBP workshops in the Eastern United States. Two studies examined information seeking behaviour among interdisciplinary clinical teams in primary care practices (Byrnes et al. 2004, Andrews et al. 2005). Kenny (2005) studied research utilization among nurses at three army medical treatment centres of differing sizes in the North Atlantic Regional Medical Command. Respondents/informants Most of the studies used a sample of registered nurses. However, Andrews et al. (2005) and Byrnes et al. (2004) used an interdisciplinary sample of clinicians that included nurses. Profetto-McGrath et al. (2007, 2009) specifically studied Clinical Nurse Specialists (CNS). Dee and Stanley (2005) compared graduate student nurses with staff nurses. Designs and measures Two studies were qualitative (McKnight 2006, Profetto-McGrath et al. 2007) and one was mixed methods (Dee & Stanley 2005). One study used an experimental design (Borycki et al. 2009). Two studies were quasiexperimental (Byrnes et al. 2004, Fink et al. 2005). One study was a case study that reported on an EBP implementation plan (Burns et al. 2009). The remaining 16 studies used descriptive, quantitative designs (Melnyk et al. 2004, Estabrooks et al. 2005, Kenny 2005, Pravikoff et al. 2005, Andrews et al. 2006, Karkos & Peters 2006, Kosteniuk et al. 2005, Atkinson et al. 2008, Beke-Harrigan et al. 2008, Cadmus et al. 2008, Koehn & Lehman 2008, Brown et al. 2009, Gale & Schaffer 2009, OÕLynn et al. 2009, Profetto-McGrath et al. 2009, Schoonover 2009). Instruments varied across studies, with the most common being investigator designed. Pravikoff et al. (2005) modified the instruments used by Tanner (2000) and Pierce (2000). Investigators in nine of the 23 studies developed their own instruments or modified other instruments (Byrnes et al. 2004, Melnyk et al. 2004, Andrews et al. 2005, Dee & Stanley 2005, Estabrooks et al. 2005, Burns et al. 2009, Gale & Schaffer 2009, OÕLynn et al. 2009, Profetto-McGrath et al. 2009). The BARRIERS to Research Utilization scale (Funk et al. 1991) was used in five studies (Fink et al. 2005, Karkos & Peters 2006, Atkinson et al. 2008, Brown et al. 2009, Schoonover 2009). The Clinical Effectiveness and Evidence-based Practice Questionnaire (EBPQ) (Upton & Upton 2006) was used by Koehn and Lehman (2008) and Brown et al. (2009). The investigator-designed instrument used by Burns et al. (2009) was influenced by Upton and Upton and Estabrooks. Beke-Harrigan et al. (2008) and Cadmus et al. (2008) modified the Pravikoff instrument originally designed by Tanner and Pierce. Fink et al. (2005) used an investigator-designed instrument. Kenny (2005) adapted the Research Utilization Survey (Estabrooks 1997) and MylleÕs Organizational Climate Survey (1998). Kosteniuk et al. (2006) used the Satisfaction with Home Community Scale (Henderson et al. 2004) and the Work Satisfaction Scale (Stamps 1997). McKnight (2006) was a participantobserver and also interviewed the subjects using open-ended questions. Dee and Stanley (2005) used interviews and observation in addition to surveys. Borycki et al. (2009) observed and interviewed student nurses. The instrument(s) and reported reliability for each study are reported in Table 1. Estabrooks et al. (2005) commented that few studies, methodological weaknesses and lack of replication make it difficult to draw conclusions and generalize findings of EBP. Analysis of barriers, facilitators and interventions by ShortellÕs CQI dimensions Stephen ShortellÕs dimensions of CQI (1998) were used as a framework to analyse EBP barriers, facilitators and 114 ª 2010 The Authors. Journal compilation ª 2010 Blackwell Publishing Ltd, Journal of Nursing Management, 19, 109 120

EBP barriers and facilitators from a CQI perspective interventions. This comprehensive framework, which includes individual and institutional factors, provides insight into the underlying sources of barriers. His framework addresses the strategic, cultural, technical and structural dimensions of an organization. Shortell et al. (1998) maintained that all four dimensions must be addressed if CQI is to be lasting and effective. Based on this review, factors affecting the success of EBP adoption seemed to occur at the individual and institutional levels therefore it is important to understand these factor as well as the CQI dimensions in evaluating EBP adoption. The dimensions affected by the barriers and interventions for each study are described in Tables 2 and 3. The authors recognize that there may be additional information not report in the articles that would result in a different analysis of the barriers. Strategic When the strategic dimension is not emphasized, then important organizational issues are not addressed (Shortell et al. 1998). Organizations where goals are not aligned with mission and priorities or where quality improvement is not important are examples of inattention to the strategic dimension. Barriers to EBP within the strategic dimension included time constraints; leaders and managers having other goals that are higher priority than EBP; difficulty in recruiting and retaining staff; lack of resources; and heavy workload. ÔLack of timeõ was listed as a barrier in 11 studies. Four of the 11 studies further qualified time to include no time to read the research (Fink et al. 2005, Karkos & Peters 2006, Atkinson et al. 2008, Gale & Schaffer 2009). Other time constraints included a demanding workload characterized by high acuity and many responsibilities and that performing EBP takes too long. Atkinson et al. (2008) reported lack of infrastructure for research-related activities. Lack of support for changing practice by the administration was noted in three studies (Fink et al. 2005, Brown et al. 2009, Schoonover 2009). At baseline in the study by Fink et al. (2005) 33 per cent of study participants reported lack of administrative support for changing practice. This value dropped slightly, to 28%, in the post-test. Difficulty in recruiting and retaining staff was mentioned in two studies that surveyed nurses across a number of hospitals (Pravikoff et al. 2005, Cadmus et al. 2008). Four studies described interventions to address some of the barriers defined in the strategic dimension. Fink et al. (2005) was the only study that provided pre- and post-test intervention findings. The interventions included integrating EBP philosophy and skills into job descriptions and clinical ladders for promotion. Both Fink et al. (2005) and Brown et al. (2009) reported on the importance of including a nursing presence on hospital-wide committees as a factor that supports EBP (Fink et al. 2005, Brown et al. 2009). Incorporating EBP into new employee orientation was discussed in two studies (Fink et al. 2005, Burns et al. 2009). These interventions, presence on committees and incorporating EBP philosophy into new employee orientation, can also be considered part of the cultural dimension. In some reports, nurses were given time during the workday to read and develop practice change activities (Kenny 2005, Brown et al. 2009). Burns et al. (2009) described an environment where the Chief Nursing Officer (CNO), the executive leader of nursing within the hospital, was very supportive and committed to EBP Table 2 EBP barriers by dimension First author, Year Strategic Cultural Technical Structural Andrews, 2005 x x Atkinson, 2008 x x x x Beke-Harrigan, 2008 x x Borycki, 2008 Brown, 2009 x x x x Burns, 2009 Byrnes, 2004 Cadmus, 2008 x x Dee, 2005 x x x Estabrooks, 2005 x Fink, 2005 x x x x Gale, 2009 x x Karkos, 2006 x x x Kenny, 2005 Koehn, 2008 x x x x Kosteniuk, 2006 x McKnight, 2006 x x Melnyk, 2004 x x x x OÕLynn, 2009 Pravikoff, 2005 x x x Profetto-McGrath, 2007 x x x Profetto-McGrath, 2009 x x Schoonover, 2009 x x Table 3 EBP interventions by dimension First author, Year Strategic Cultural Technical Structural Beke-Harrigan, 2008 x x Brown, 2009 x x x x Burns, 2009 x x x Byrnes, 2004 x Dee, 2005 x Fink, 2005 x x x x Karkos, 2006 x Kenny, 2005 x ª 2010 The Authors. Journal compilation ª 2010 Blackwell Publishing Ltd, Journal of Nursing Management, 19, 109 120 115

N. M. Solomons and J. A. Spross and included the definition of EBP in all communications. This CNO of the community hospital made resources available for EBP training, mentored members of the EBP council to take on leadership roles and published a monthly EBP newsletter to staff that described research activities and the importance of EBP, as well as solicited ideas for quality improvement projects (Burns et al. 2009). Interventions to promote EBP addressed many barriers described in the literature. A CNO who was committed to EBP was evidence of leadership commitment. Nursing presence on hospital-wide committees helped to promote EBP within the institution. These two interventions can be considered both strategic and cultural interventions that enhance nursesõ authority and autonomy to implement EBP. Two studies reported allocating time to research and implementing practice change, targeting time constraints (Kenny 2005, Brown et al. 2009). In addition, training in EBP and promotion based on demonstrating EBP competencies are technical and strategic interventions. Cultural dimension According to Shortell et al. (1998), when there is no attention to the cultural dimension, improvement results are not acknowledged by the organization, success is not rewarded and improvement behaviours do not become embedded in practice. Twelve studies described barriers to EBP that can be considered cultural barriers. The most frequently reported barrier (seven studies) was resistance to changing practice from co-workers and managers (Melnyk et al. 2004, Dee & Stanley 2005, Fink et al. 2005, Karkos & Peters 2006, Profetto- McGrath et al. 2007, Atkinson et al. 2008, Koehn & Lehman 2008). Three of the seven studies also noted a lack of authority to change practice, which could also be considered strategic (Fink et al. 2005, Karkos & Peters 2006, Atkinson et al. 2008). Pravikoff et al. (2005) reported that institutional leaders perceived that nurses were not interested or ready to adopt EBP and it would not be possible to actually adopt EBP. Lack of authority to change practice was described in five studies (Fink et al. 2005, Karkos & Peters 2006, Brown et al. 2009, OÕLynn et al. 2009, Schoonover 2009). Within the study by Fink et al. (2005), 40% of the pretest participants reported a lack of authority to change practice which dropped to 29% after the intervention. Karkos and Peters (2006) reported lack of authority to change practice was rated second within the top ten barriers to EBP. Schoonover (2009) reported lack of authority to change practice as the top barrier among 29 factors. OÕLynn et al. (2009) reported 50.3% agreement that it was difficult to influence change in the workplace. A unique finding, desire for work-life balance, was reported by Atkinson et al. (2008). Four studies described a lack of respect for research (Melnyk et al. 2004, Fink et al. 2005, Pravikoff et al. 2005, Beke-Harrigan et al. 2008). Subjects in the Fink et al. study commented that it (EBP) ÔdoesnÕt apply to what I doõ, is Ônot related to bedside careõ and Ônurses are not trained to think deductivelyõ (2005, p. 126). Interventions to address the barriers in the cultural dimension were described in three studies (Fink et al. 2005, Beke-Harrigan et al. 2008, Brown et al. 2009). EBP champions were appointed to cultivate staff interest and ownership in the research and attend an annual research symposium (Fink et al. 2005). Beke-Harrigan et al. (2008) reported that health science library staff strengthened their relationships with nursing staff. In the study by Brown et al. (2009), the nursing staff was rewarded for critical thinking and developed a culture of respect across disciplines. Of the three aforementioned studies only Fink et al. (2005) described their strategy to improve research utilization and reported measurable outcomes from the interventions. The remaining two studies described the interventions without discussing outcomes (Beke-Harrigan et al. 2008, Brown et al. 2009). Technical dimension When the technical dimension is not addressed then there are many aborted attempts at CQI. Employees may not have enough training to participate effectively in multidisciplinary teams or the information systems are not powerful enough to support their efforts. A lack of initial and ongoing training and poor information systems are known barriers to effective support of the technical dimension of CQI (Shortell et al. 1998). Fourteen studies reported barriers in the technical dimension. The most frequent barrier, reported in eight studies, was difficulty accessing resource materials (Melnyk et al. 2004, Fink et al. 2005, Karkos & Peters 2006, Kosteniuk et al. 2006, Beke-Harrigan et al. 2008, Cadmus et al. 2008, Gale & Schaffer 2009). Six studies noted that nurses did not feel confident in their ability to evaluate the quality of the research (Fink et al. 2005, Atkinson et al. 2008, Beke-Harrigan et al. 2008, Cadmus et al. 2008, Gale & Schaffer 2009, Schoonover 2009). Four studies reported lack of informationseeking skills as a barrier (Andrews et al. 2005, Dee & Stanley 2005, Profetto-McGrath et al. 2007, Gale & Schaffer 2009). Lack of understanding of online 116 ª 2010 The Authors. Journal compilation ª 2010 Blackwell Publishing Ltd, Journal of Nursing Management, 19, 109 120

EBP barriers and facilitators from a CQI perspective research databases such as CINAHL and MEDLINE was a barrier in five studies (Melnyk et al. 2004, Dee & Stanley 2005, Beke-Harrigan et al. 2008, Cadmus et al. 2008, Gale & Schaffer 2009). Using Google or Yahoo! for a literature search rather than the scientific research databases was identified as a barrier in two studies (Kosteniuk et al. 2006, Gale & Schaffer 2009). Difficulty understanding the statistics found in the journals was a barrier in four studies (Melnyk et al. 2004, Fink et al. 2005, Atkinson et al. 2008, Gale & Schaffer 2009). Three studies stated that subjects found research to be overwhelming (Atkinson et al. 2008, Gale & Schaffer 2009, Schoonover 2009). Beke-Harrigan et al. (2008) reported that the hospital blocked access to online bibliographic databases and other online resources. Difficulty navigating online systems was a barrier in the McKnight (2006) study. Nurses in the Dee and Stanley (2005) study were not confident in their library skills at any library be it the health science library or the local public library. Estabrooks et al. (2005) found that sources of evidence most often used by nurses were: patient information, individual clinical experience and interactions with others; journals and the internet were the least used sources of information. Five studies resolved many of these barriers through training (Byrnes et al. 2004, Dee & Stanley 2005, Beke- Harrigan et al. 2008, Brown et al. 2009, Burns et al. 2009). Three studies implemented hands-on training sessions for bibliographic database searches (Byrnes et al. 2004, Dee & Stanley 2005, Burns et al. 2009). In the Beke-Harrigan et al. (2008) the hospital formed a subcommittee to promote the use of EBP among nurses. In addition, classes were offered on how to access and interpret research (Beke-Harrigan et al. 2008). The hospital studied by Brown et al. (2009) developed a curriculum to address educational deficiencies related to EBP. The classes were developed based on the results of low and high scores on topics in the BARRIERS survey. Burns et al. (2009) reported training 25 of the hospitalõs nursing staff on the five steps of EBP: format a specific clinical question, collect the best evidence, critically appraise the evidence, integrate evidence, clinical expertise and patient preferences in the decision and evaluate the outcomes. A group of staff nurses, the Professional Practice Council, was responsible for identifying solutions for patient care delivery problems (Burns et al. 2009). Monthly meetings included some time devoted to learning and understanding research (Burns et al. 2009). In the Magnet hospital experience reported by Fink et al. (2005) the technical barriers were addressed by developing a manual for nurses to learn more about how research is used in practice. Among the 14 studies described in this section the main themes were lack of online access to the research databases and understanding how to use the databases. The solutions were primarily aimed at the individual rather than the institution. Classes were offered at least once, but the studies do not say whether the training efforts were sustained. Also, these interventions did not describe how the lack of time issue was addressed. Structural dimension When the structural dimension is not considered, new knowledge is not recorded and distributed among personnel at all levels of the organization. Barriers in the structural dimension include not taking advantage of system-wide committees that would provide an allsystem means of guiding the organization and communicating quality information or lack of a library (Shortell et al. 1998). Lack of awareness of research was the most frequently mentioned barrier within the structural dimension (Dee & Stanley 2005, Fink et al. 2005, Atkinson et al. 2008, Koehn & Lehman 2008, Schoonover 2009). Other barriers were: information was not compiled in one place (Atkinson et al. 2008, Schoonover 2009); there were too many journals (Melnyk et al. 2004); and difficulties with the information format (Andrews et al. 2005). Four studies provided interventions to address these barriers (Fink et al. 2005,. Karkos & Peters 2006, Brown et al. 2009, Burns et al. 2009). Journal clubs were established in three studies (Fink et al. 2005, Karkos & Peters 2006, Burns et al. 2009). Quarterly research workshops and yearly grand rounds were used by Fink et al. (2005). Two studies converted research findings into a format that was easy to comprehend and disseminated through email and online forums (Karkos & Peters 2006, Brown et al. 2009). A bulletin board became a venue to display current EBP (Fink et al. 2005, Burns et al. 2009). Nurse scientists were consulted for EBP and research projects (Fink et al. 2005, Karkos & Peters 2006). Fink et al. (2005) reported improvements in perceptions regarding authority to change practice, awareness of research, less isolation from knowledgeable colleagues and hospital administration became more tolerant of changes in practice. The interventions that were described in these studies made the research more accessible to the nurses either through formal discussion in journal clubs or distilled into an email or an online posting. Research would diffuse through the organization as a result of these ª 2010 The Authors. Journal compilation ª 2010 Blackwell Publishing Ltd, Journal of Nursing Management, 19, 109 120 117

N. M. Solomons and J. A. Spross activities and in some cases the outcome was less resistance to change in practice. Discussion Using ShortellÕs CQI framework to analyse EBP barriers demonstrates that barriers to EBP can occur at both individual and institutional levels and within the four dimensions of an organization. In the same vein, interest in EBP may occur at the individual or institutional level. The individual may practice EBP as a means to provide better care or as a result of interest in a particular condition. The institution may be interested in EBP as a way to maintain Magnet accreditation, attract more payers or to be eligible for incentives. Problems occur when individuals or institutions do not know about, or see the value in, EBP. It is likely that effective EBP that improves practice depends on strategies at both levels and one that addresses all of the four dimensions of CQI. Five studies out of 23 reported barriers across all four dimensions of the institution (Melnyk et al. 2004, Fink et al. 2005, Atkinson et al. 2008, Koehn & Lehman 2008, Brown et al. 2009). The common thread among these studies was a theory or measure that addressed the dimensions of the institution. Three studies (Fink et al. 2005, Atkinson et al. 2008, Brown et al. 2009) used the BARRIERS instrument which is based on RogersÕ Diffusion of Innovations Theory. This theory is important to EBP adoption because it describes how diffusion is dependent upon the characteristics of innovation, ease of adopting the innovation, characteristics of the individual and institution, and whether the innovation is aligned with the values and beliefs of the individual and organization. Melnyk et al. (2004) used the Transtheoretical Model of Change and the Control Theory. The Clinical Effectiveness and Evidence-based Practice questionnaire (Upton & Upton 2006), used by Koehn and Lehman (2008), did not have an underlying theory; however it was influenced by two studies that addressed institutional factors (Davies et al. 2000, Hughes et al. 2002). Although there was no common instrument used across the studies, common barriers were identified such as lack of time, inability to access research, difficulties comprehending the statistics and research language and inability to change practice. Common interventions were trainings on bibliographic database searching and understanding research reports. Journal clubs were also mentioned. Such interventions may affect both individuals and institutions but reports did not address how the improvement was measured. With regard to interventions to promote EBP, it is important to note that addressing one barrier, for example a cultural barrier such as lack of autonomy, may be addressed by appointing nurses to decision-making committees, a strategic and cultural intervention. Limitations of this review Different instruments and different theories made it difficult to compare study results. Many of the studies selected for this analysis did not use an instrument with established psychometrics and not all studies reported on their validity. There were no longitudinal studies that measured the persistence of training outcomes, for example, whether training in literature search and research comprehension continued to have any effect on practice or patient outcomes. This review only examined studies from the United States and Canada. Issues of patients refusing EBP-based care was not mentioned in any of the literature. This is important as patientõs values are part of the definition of EBP. Conclusions and recommendations Based on this review several gaps in our understanding of EBP barriers, facilitators and interventions can be identified. These descriptive studies did not link EBP to CQI process improvement and their relationship to CQI, or to patient and institutional outcomes. There is a need for more research to understand how well the interventions address EBP barriers in the short term and whether effects persist. What individual and institutional factors support and sustain changes in EBP? How do changes in EBP affect CQI processes and patient and institutional outcomes? The literature describes several models for implementing EBP, however, the authors found no studies comparing the success or failure of these models by environment/setting. The most frequently used instrument among the selected studies was the BARRIERS instrument which was created to address research utilization using RogersÕ Diffusion of Innovations Theory. However, much of the literature was atheoretical. More studies guided by theory and using instruments that are theoretically based and psychometrically sound are needed in order to strengthen our understanding of EBP, CQI, and barriers and facilitators. Lack of time to read and seek out literature was a common theme throughout these studies. Managers and institutions should look at the availability of EBP and determine the best ways to implement those practices on site. Sites such as TRIP (http://www.tripdat 118 ª 2010 The Authors. Journal compilation ª 2010 Blackwell Publishing Ltd, Journal of Nursing Management, 19, 109 120

EBP barriers and facilitators from a CQI perspective abase.com) and the National Guideline Clearinghouse (http://www.guideline.gov) are two sites that provide aggregated information from the literature. Further review of the Cochrane database to understand barriers and facilitators to adoption of evidence-based diseasespecific practices may also provide additional ideas. One technical intervention may be to ÔpushÕ the information out to nurses through e-mail and WEB 2.0 tools such as wikis and Twitter in a Ôjust the factsõ format. The literature shows that nurses most often obtain information, evidence-based or not, from each other. In order to ensure that this information is evidence-based, the research has to get into the hands and heads of the disseminators so that it is diffused throughout the organization. Research has to move from journals to practice more expeditiously than at the present rate. The information has to be in a format that is accessible to all, no matter how long ago the nurse was educated. It is important to start at the strategic dimension, making EBP and CQI an explicit institutional priority and individual job expectation. However, since this will be insufficient according to Shortell et al. individual and institutional barriers in all dimensions must be addressed. References ANCC Available at: http://www.nursecredentialing.org/magnet/ NewMagnetModel.aspx, accessed 25 May 2010. Andrews J.E., Pearce K.A., Ireson C. & Love M.M. (2005) Information-seeking behaviors of practitioners in a primary care practice-based research network (PBRN). Journal of the Medical Library Association 93, 206 212. Atkinson M., Turkel M. & Cashy J. (2008) Overcoming barriers to research in a Magnet community hospital. Journal of Nursing Care Quality 23 (4), 362 368. Beke-Harrigan H., Hess R. & Weinland J.A. (2008) A survey of registered nursesõ readiness for evidence-based practice: a multidisciplinary project. Journal of Hospital Librarianship 8 (4), 440 448. Borycki E.M., Lemieux-Charles L., Nagle L. & Eysenbach G. (2009) Evaluating the Impact of Hybrid Electronic-paper Environments Upon Novice Nurse Information Seeking. Methods of Information in Medicine 48 (2), 137 143. Brown C.E., Wickline M.A., Ecoff L. & Glaser D. (2009) Nursing practice, knowledge, attitudes and perceived barriers to evidence-based practice at an academic medical center. Journal of Advanced Nursing 65 (2), 371 381. Burns H.K., Dudjak L. & Greenhouse P.K. (2009) Building an evidence-based practice infrastructure and culture: a model for rural and community hospitals. Journal of Nursing Administration 39 (7 8), 321 325. Byrnes J.A., Kulick T.A., Schwartz D.G., Byrnes J.A., Kulick T.A. & Schwartz D.G. (2004) Information-seeking behavior changes in community-based teaching practices. Journal of the Medical Library Association 92 (3), 334 340. Cadmus E., Van Wynen E.A., Chamberlain B. et al. (2008) NursesÕ skill level and access to evidence-based practice. Journal of Nursing Administration 38 (11), 494 503. CMS Available at: http://www.cms.hhs.gov/acuteinpatientpps/ downloads/cms-1533-fc.pdf, p. 368, accessed 12 June 2009. Committee on Quality of Health Care in America, Institute of Medicine (2001) Crossing the Quality Chasm: a New Health System for the 21st Century. National Academy Press, Washington, DC. Davies H., Nutley S.M. & Mannion R. (2000) Organisational culture and quality of health care. Quality in Healthcare 9 (2), 111 119. Dee C. & Stanley E.E. (2005) Information-seeking behavior of nursing students and clinical nurses: implications for health sciences librarians. Journal of the Medical Library Association 93 (2), 213 222. Estabrooks C.A. (1997) Nursing Research Round-up. Research Utilization in Nursing: An Alberta Context. AARN Newsletter 53 (3), 16. Estabrooks C.A., Chong H., Brigidear K. & Profetto-McGrath J. (2005) Profiling Canadian nursesõ preferred knowledge sources for clinical practice. Canadian Journal of Nursing Research 37 (2), 118 140. Fink R., Thompson C.J. & Bonnes D.B. (2005) Overcoming barriers and promoting the use of research in practice. Journal of Nursing Administration 35 (3), 121 129. Funk S., Champagne M., Wiese R. & Tornquist E. (1991) Barriers: the Barriers to Research Utilization Scale. Appl Nurs Res 4, 39 45. Gale B.V. & Schaffer M.A. (2009) Organizational readiness for evidence-based practice. Journal of Nursing Administration 39 (2), 91 97. Henderson Betkus M. & MacLeod M. (2004) Retaining public health nurses in rural British Columbia. Canadian Journal of Public Health 95, 54 58. Hughes J., Humphrey C., Rogers S. & Greenhalgh T. (2002) Evidence into Action: changing Practice in Primary Care. Occasional Paper 84. Royal College of General Practitioners, London. Karkos B. & Peters K. (2006) A Magnet community hospital: fewer barriers to nursing research utilization. Journal of Nursing Administration 36 (7-8), 377 382. Kenny D.J. (2005) NursesÕ use of research in practice at three US Army hospitals. Nursing Leadership (Toronto, Ont) 18 (3), 45 67. Koehn M.L. & Lehman K. (2008) NursesÕ perceptions of evidence-based nursing practice. Journal of Advanced Nursing 62 (2), 209 215. Kosteniuk J.G., DÕArcy C., Stewart N. & Smith B. (2006) Central and peripheral information source use among rural and remote Registered Nurses. Journal of Advanced Nursing 55, 100 114. Lasser K., Himmelstein D.U. & Woolhandler S. (2006) Access to care, health status, and health disparities in the United States and Canada: results of a cross-national population-based survey. American Journal of Public Health 96 (7), 1300 1307. doi: 10.2105/AJPH.2004.059402. McBride S. (2008) Using administrative data to answer state policy questions. AHRQ Workshop. Available at: http:// archive.ahrq.gov/qual/kt/workshop1208/mcbride3.htm, accessed 6 December 2009. ª 2010 The Authors. Journal compilation ª 2010 Blackwell Publishing Ltd, Journal of Nursing Management, 19, 109 120 119