Barriers of Research Utilization for Nurses

C L I N I C A L N U R S I N G IS S U E S Bridging the divide: a survey of nurses’ opinions regarding barriers to, and facilitators of, research utilization in the practice setting Alison Margaret Hutchinson BAppSc, MBioeth PhD Candidate, Victorian Centre for Nursing Practice Research, School of Nursing, University of Melbourne, Australia Linda Johnston BSc, PhD, Dip N Professor in Neonatal Nursing Research, Royal Children’s Hospital, Melbourne, and Associate Director, Victorian Centre for Nursing Practice Research, Melbourne, Australia Submitted for publication: 4 March 2003 Accepted for publication: 29 August 2003Correspondence: Alison M. Hutchinson School of Nursing University of Melbourne 1/723 Swanston St Carlton, VIC 3053 Australia Telephone: ? 61 3 8344 0800 E-mail: [email  protected] com H U T C H I N S O N A . M . & J O H N S T O N L . ( 2 0 0 4 ) Journal of Clinical Nursing 13, 304–315 Bridging the divide: a survey of nurses’ opinions regarding barriers to, and facilitators of, research utilization in the practice setting Background. Many researchers have explored the barriers to research uptake in order to overcome them and identify strategies to facilitate research utilization.However, the research–practice gap remains a persistent issue for the nursing profession. Aims and objectives. The aim of this study was to gain an understanding of perceived in? uences on nurses’ utilization of research, and explore what differences or commonalities exist between the ? ndings of this research and those of studies that have been conducted in various countries during the past 10 years. Design. Nurses were surveyed to elicit their opinions regarding barriers to, and facilitators of, research utilization.The instrument comprised a 29-item validated questionnaire, titled Barriers to Research Utilisation Scale (BARRIERS Scale), an eight-item scale of facilitators, provision for respondents to record additional barriers and /or facilitators and a series of demographic questions. Method. The questionnaire was administered in 2001 to all nurses (n ? 761) working at a major teaching hospital in Melbourne, Australia. A 45% response rate was achieved. Results. Greatest barriers to research utilization reported included time constraints, lack of awareness of available research literature, insuf? ient authority to change practice, inadequate skills in critical appraisal and lack of support for implementation of research ? ndings. Greatest facilitators to research utilization reported included availability of more time to review and implement research ? ndings, availability of more relevant research and colleague support. Conclusion. One of the most striking features of the ? ndings of the present study is that perceptions of Australian nurses are remarkably consistent with reported perceptions of nurses in the US, UK and Northern Ireland during the past decade. Relevance to clinical practice.If the use of res earch evidence in practice results in better outcomes for our patients, this behoves us, as a profession, to address issues surrounding support for implementation of research ? ndings, authority to 304 O 2004 Blackwell Publishing Ltd Clinical nursing issues Barriers to, and facilitators of, research utilization change practice, time constraints and ability to critically appraise research with conviction and a sense of urgency. Key words: barriers to research utilization, facilitators of research utilization, research dissemination, research implementation, research utilizationIntroduction and background For over 25 years research utilization has been discussed in the nursing literature with growing enthusiasm and amid increasing calls for the use of research ? ndings in practice. Additionally, the evidence-based practice movement, which emanated in the early 1990s (Evidence-Based Medicine Working Group, 1992) has highlighted the importance of incorporating research ? ndings into pra ctice. Furthermore, controversy surrounding the achievement of professional status has resulted in an increased awareness of the need for a research-based body of knowledge to underpin nursing practice.Gennaro et al. (2001, p. 314) contend: Using research in practice not only bene? ts patients but also strengthens nursing as a profession. If nursing is truly a profession, and not just a job or an occupation, nurses have to be able to continually evaluate the care they give and be accountable for providing the best possible care. Evaluating nursing care means that nurses also have to evaluate nursing research and determine if there is a better way to provide care. Twelve years prior, Walsh & Ford (1989) warned that the professional integrity of nursing was threatened by dependence upon experience-based practice.Similarly, Winter (1990, p. 138) cautioned that conduct of nursing practice in this manner is ‘the antithesis of professionalism, a barrier to independence, and a detrim ent to quality care. ’ Winter therefore, recommended that nurses ‘evaluate their status as research consumers, to identify problems in this area, and to develop means to better use research ? ndings’ (p. 138). Evidence-based practice, which should comprise the use of broad ranging sources of evidence, including the clinician’s expertise and patient preference (Sackett et al. , 1996), includes the use of research evidence as a subset (Estabrooks, 1999).Consistent with the classi? cation of knowledge utilization, three types of research use have been outlined (Stetler, 1994a,b; Berggren, 1996). The ? rst is described as ‘instrumental use’ and involves acting on research ? ndings in explicit, direct ways, for example application of research ? ndings in the development of a clinical pathway. The second is termed ‘conceptual use’ and involves using research ? ndings in less speci? c ways, for example changing thinking. The ? nal type o f research use, described as ‘symbolic use’, involves the use of research results to support a predetermined position.The nursing literature is replete with examples of limited use of research in practice and discussion surrounding perceived barriers to research utilization (Hunt, 1981; Gould, 1986; Closs & Cheater, 1994; Lacey, 1994). Despite this, the phenomenon of the research–practice gap, the gap between the conduct of research and use of that research in practice, remains an issue of major importance for the nursing profession. Many researchers have explored the barriers to research uptake in order to overcome them and identify strategies to facilitate research utilization (Kirchhoff, 1982; MacGuire, 1990; Funk et al. 1991a,b, 1995b; Closs & Cheater, 1994; Hicks, 1994, 1996; Lacey, 1994; Rizzuto et al. , 1994; Hunt, 1996; Walsh, 1997a,b). Hunt (1981) suggested that nurses fail to utilize research ? ndings because they do not know about them, do not understa nd them, do not believe them, do not know how to apply them, and are not allowed to use them. According to Hunt (1997), the barriers to research utilization and, therefore, to evidence-based practice fall into ? ve main categories: research, access to research, nurses, process of utilization and organization.Self-reported utilization of research is one method that has frequently been implemented to elicit the extent of research utilization. Responses to selected research ? ndings have been used to elicit and explore respondents’ awareness and use of respective ? ndings (Kete? an, 1975; Berggren, 1996). Numerous researchers have also undertaken to investigate, through self-reporting, the opinions of nurses’ in regard to barriers to research utilization in the practice setting. Funk et al. (1991b) explored research utilization in the US using a postal questionnaire titled the Barriers to Research Utilization Scale (BARRIERS Scale).Their purpose was to develop a tool to a ssess the perceptions of clinicians, administrators and academics in regard to barriers to research utilization in clinical practice. Rogers’ (1995) model of ‘diffusion of innovations’, a theoretical framework, which describes the process of communication, through certain channels within a social network, of an idea, practice or object over time, was used to develop a 29-item scale. The questionnaire was sent out to a random sample of 5000 members of the American Nurses’ Association with a resulting response rate of 40%. 305O 2004 Blackwell Publishing Ltd, Journal of Clinical Nursing, 13, 304–315 A. M. Hutchinson and L. Johnston On the data generated, Funk et al. (1991b) undertook an exploratory factor analysis, to elicit a four-factor solution which closely corresponded with Rogers’ (1995) ‘diffusion of innovations’ model. The factors translated into characteristics of the adopter comprising the nurse’s research values, s kills and awareness; the organization incorporating setting barriers and limitations; the innovation including qualities of the research; and communication including accessibility and presentation of the research.Items associated with the clinical setting, a characteristic of the organization, were perceived as the main barriers to research utilization. These included the views that nurses lack suf? cient authority to implement change; nurses have insuf? cient time to implement change; and there is a lack of cooperation from medical staff. Approximately 21% of the respondents in this study were classi? ed as administrators. Over three quarters of the items on the BARRIERS Scale were rated as great or moderate barriers by over half the administrators. The administrators identi? d factors relating to the nurse, the organizational setting and the presentation of research among the greatest barriers. Overall, they cited the organizational setting as the greatest barrier to research use. Approximately 46% of the respondents were classi? ed as clinicians (nurses working in the clinical setting). The clinicians overwhelmingly identi? ed factors associated with the organizational setting as being the greatest barriers to research utilization. They rated all eight factors associated with the setting in the top 10 barriers to research utilization.The clinicians rated perceived ‘lack of authority to change patient care procedures’, ‘insuf? cient time on the job to implement new ideas’ and being ‘unaware of the research’ as the top three barriers to research utilization. The BARRIERS Scale (Funk et al. , 1991b) has been used extensively since it was developed in 1991, as one method to explore the perceived in? uences on nurses’ utilization of research ? ndings in their practice. At least 17 studies that employed the BARRIERS Scale to elicit opinions of nurses regarding barriers to research utilization in practice have been rep orted in the nursing literature.Most studies reported the barriers in ranked order according to the percentage of respondents who rated items as moderate or great barriers. Insuf? cient time to read research and/or implement new ideas was rated in the top three barriers in 13 studies (Funk et al. , 1991a, 1995a; Carroll et al. , 1997; Dunn et al. , 1997; Lewis et al. , 1998; Nolan et al. , 1998; Rutledge et al. , 1998; Retsas & Nolan, 1999; Closs et al. , 2000; Parahoo, 2000; Retsas, 2000; Grif? ths et al. , 2001; Marsh et al. , 2001; Parahoo & McCaughan, 2001).A perceived lack of authority to change patient care procedures was reported in the top three barriers in eight studies (Funk et al. , 1991a; Walsh, 1997a; Nolan 306 et al. , 1998; Closs et al. , 2000; Parahoo, 2000; Retsas, 2000; Marsh et al. , 2001; Parahoo & McCaughan, 2001). In eight studies, the item ‘statistical analyses are not understandable’, was cited in the top three barriers (Funk et al. , 1995b; Dunn et al. , 1997; Walsh, 1997a,b; Rutledge et al. , 1998; Parahoo, 2000; Grif? ths et al. , 2001; Marsh et al. , 2001). ‘Inadequate facilities for implementation’ was cited in the top three barriers in ? e studies (Kajermo et al. , 1998; Nolan et al. , 1998; Retsas, 2000; Grif? ths et al. , 2001; Marsh et al. , 2001). Finally, the item ‘lack of awareness of research ? ndings’ was reported in the top three barriers in four studies (Funk et al. , 1991a, 1995a; Carroll et al. , 1997; Lewis et al. , 1998; Retsas & Nolan, 1999). It is acknowledged that these studies comprised varying populations of nurses, employed differing sampling methods, used sample sizes ranging from 58 to 1368 respondents and resultant response rates ranged from 27 to 76%.In some studies, minor rewording of a limited number of items in the tool had been undertaken. Furthermore, some studies included only 28 of 29 barrier items included in the original BARRIERS Scale. Factor analysis, a stat istical technique aimed at reducing the number of variables by grouping those that relate, to form relatively independent subgroups (Crichton, 2001; Tabachnick & Fidell, 2001), was undertaken in a limited number of these studies. In the UK, Dunn et al. (1997) tested the factor model proposed by Funk et al. (1991b), using con? rmatory factor analysis, a complex statistical technique used to test a heory or model (Tabachnick & Fidell, 2001). Attempts to load each item onto a single identi? ed factor were found to be unsuccessful and they concluded that the US model was inappropriate for their data. Closs & Bryar (2001) further explored the appropriateness of the BARRIERS Scale for use in the UK through exploratory factor analysis. The model identi? ed included the following four factors: bene? ts of research for practice, quality of research, accessibility of research, and resources for implementation. Finally, Marsh et al. (2001) tested, using con? matory factor analysis, a revised v ersion of the BARRIERS Scale. The revision comprised minor changes in wording such as substitution of the term ‘administrator’ with the term ‘manager’. A factor structure that was not possible to interpret resulted and they concluded that the model proposed by Funk et al. (1991b) was not supported and had limited subscale validity in the UK setting. In the light of these ? ndings and those of Dunn et al. (1997), Marsh et al. (2001) suggested that the factor model arising from the original BARRIERS Scale was not sustained in the international context.However, in Australia, Retsas & Nolan (1999) undertook an exploratory factor analysis resulting in a three-factor solution comprising: (i) nurses’ perceptions about the usefulness of research in O 2004 Blackwell Publishing Ltd, Journal of Clinical Nursing, 13, 304–315 Clinical nursing issues Barriers to, and facilitators of, research utilization clinical practice, (ii) generating change to practice based on research, and (iii) accessibility of research. Again, in Australia, a four-factor solution arose from another exploratory factor analysis undertaken by Retsas (2000).The resulting factors were conceptualized as: accessibility of research ? ndings, anticipated outcomes of using research, organizational support to use research, and support from others to use research. Given these ? ndings in the Australian context, an exploratory factor analysis was employed in the present study to explore what model would arise from data generated using the BARRIERS Scale. The aim of the present study was to gain an understanding of perceived in? uences on nurses’ utilization of research in a particular practice setting, and explore what differences or commonalities exist between the ? dings of this research and those of studies which have been conducted during the past 10 years in various countries around the world. This study was undertaken as part of a larger study designed to exp lore the phenomenon of research utilization by nurses in the clinical setting. The relative importance of barrier and facilitator items and the factor model arising from this data will in? uence development of future stages of this larger study. who then took responsibility for distribution. It cannot be guaranteed, however, that this process in fact resulted in all nurses receiving the questionnaire.The questionnaire included the 29-item BARRIERS Scale in addition to an eight-item facilitator scale and a series of demographic questions. The respondents were asked to return completed questionnaires in the self-addressed envelope supplied, by either placing them in the internal mail or placing them in the ‘return’ box supplied in their ward or department. Return of completed questionnaires implied consent to participate and all responses were anonymous. Setting The setting for this study was a 310-bed major teaching hospital offering specialist services in Melbourne, Aus tralia. SampleApproximately 960 nurses work in the organization. All Registered Nurses working during the 4-week distribution time frame were invited to complete the questionnaire. This self-selecting, convenience sample therefore, excluded nurses on leave at the time of the study. The study The research question addressed in this study was: What are nurses’ perceptions of the barriers to, and facilitators of, research utilization in the practice setting? Instrument The questionnaire comprised three sections. The ? rst section contained the 29 randomly ordered items from the Barriers to Research Utilization Scale (Funk et al. 1991b), which respondents were asked to rate, on a four-point Likert type scale, the extent to which they believed each item was a barrier to their use of research in practice. The options included 1 ? ‘to no extent’, 2 ? ‘to a little extent’, 4 ? ‘to a moderate extent’ and 5 ? ‘to a large extent’. A â €˜no opinion’ ? 3 option was also given. The respondents were then asked to nominate and rate (1 ? greatest barrier, 2 ? second greatest barrier, and 3 ? third greatest barrier) the items they considered to be the top three barriers.Further to this, the respondents were given the opportunity to list and rate, according to the above-mentioned Likert scale, any additional items they perceived to be barriers. The second section of the survey contained eight items (Table 4), which respondents were asked to rate according to the extent to which they considered them to be a facilitator of research utilization using the Likert scale described above. The respondents were also asked to nominate and rate, from 1 to 3, the items they considered to be the three greatest facilitators of research utilization.Again, the respondents were given the opportunity to list and rate, according to the 307 Method A survey design was chosen to elicit opinions of nurses. This method was selected bec ause the ‘BARRIERS Scale’, a validated questionnaire, based on the work of Funk et al. (1991b), and designed to elicit nurses’ views about barriers to, and facilitators of, research utilization in their practice, was found to have high reliability. Approval to use the tool was gained from the authors. Permission was also given to include questions crafted by the investigators to elicit nurses’ opinions about facilitators of research utilization.Approval to conduct the project was sought and granted by the hospital research ethics committee to ensure the rights and dignity of all respondents were protected. Nurses working during the 4-week survey distribution time frame (n ? 761) were invited to complete the self-administered questionnaire. It was intended that every nurse receive a personally addressed envelope containing the questionnaire and a self-addressed return envelope. To facilitate this, the envelopes were hand delivered to a nominated nurse on ea ch ward or department O 2004 Blackwell Publishing Ltd, Journal of Clinical Nursing, 13, 304–315A. M. Hutchinson and L. Johnston Likert scale, perceived facilitators not listed in the survey. Section 3 of the survey included a series of demographic questions. Validity Content validity, i. e. whether the questions in the tool accurately measure what is supposed to be measured (LoBiondo-Wood & Haber, 1998), of the instrument was supported by the literature on research utilization, the research utilization questionnaire developed by the Conduct and Utilization of Research in Nursing Project (Crane et al. , 1977), and data gathered from nurses. Input was also gained from experts in the ? ld of research utilization, nursing research, nursing practice and a psychometrician to establish face validity, i. e. whether the tool appears to measure the concept intended (LoBiondo-Wood & Haber, 1998), and content validity from an extensive list of potential items. Those items for which face and content validity were established were retained. Further to piloting of the instrument, two additional items were included and some minor rewording of other items resulted. The BARRIERS Scale has been found to have good reliability, with Cronbach’s alpha coef? ients of between 0. 65 and 0. 80 for the four factors, and item-total correlations from 0. 30 to 0. 53 (Funk et al. , 1991b). Cronbach’s alpha is a measure of internal consistency, which is related to the reliability of the instrument. A Cronbach’s alpha of †¡0. 7 is considered to be good. Internal consistency is the extent to which items in the scale measure the same concept (LoBiondo-Wood & Haber, 1998). Item total correlations refer to the relationship between the question or item and the total scale score (LoBiondo-Wood & Haber, 1998). Data analysisData analysis was performed using Statistical Package for the Social Sciences (version 10. 0; SPSS Inc. , Chicago, IL, USA) software. Frequency and descriptive statistics were employed to describe the demographic characteristics of respondents. Analysis of these data indicated that a wide cross section of nursing staff responded to the questionnaire. Factor analytic procedures were employed to reduce the 29 barrier items to factors. The ‘no opinion’ responses (coded to be in the centre of the scale) were included in the factor analytic procedure, on the basis of statistical advice.Suitability of the data for undertaking factor analysis is determined by testing for sampling adequacy and sphericity. The Kaiser–Meyer–Olkin Measure of Sampling Adequacy at 0. 83 was in excess of the recommended value of 0. 6 (Kaiser, 1974), indicating that the 308 correlations or factor loadings, which re? ect the strength of the relationship between barrier items, were high. The Bartlett test of sphericity at 2118. 3 was statistically signi? cant (P < 0. 001). On the basis of these results, factor analysis was considered appropriate.The factor analysis method employed consisted of principal component analysis (PCA), a method of reducing a number of variables (barrier items) to groupings to aid interpretation of the underlying relationships between the variables (Crichton, 2000) whilst capturing as much of the variance in the data as possible. PCA revealed eight components with an eigenvalue exceeding one, indicating that up to eight factors could be retained in the ? nal factor solution. Inspection of the scree plot, a plot of the variance encompassed by the factors, failed to provide a clear indication for the number of factors to include.Eight factors were considered too many to be meaningful, thus factor solutions from two to seven factors were explored. A solution comprising four factors was considered most meaningful. Examination of the factor loadings was then undertaken to determine which items belonged to each factor. Consistent with the procedure employed by Funk et al. (1991b), items were considered to have loaded if they had a factor loading of 0. 4 or more. Varimax rotation, a statistical method employed to simplify and aid interpretation of factors, was then applied.Whilst factor analysis assists in reducing the number of variables to groupings and aids in interpretation of the underlying structure of the data, it does not identify the relative importance of individual items. Thus, while one factor may account for the largest amount of variance in the factor solution it does not mean that the items within that factor are the greatest barriers to research utilization. In order to determine the relative signi? cance of each barrier item, the number of respondents who reported them as a moderate or great barrier was calculated and items were ranked accordingly.Additional barriers recorded by participants were grouped thematically. Similarly, to determine the relative signi? cance of each facilitator item, the number of respondents who reported them as a moderate or great facilitator was calculated and items were ranked accordingly. Additional facilitators recorded by participants were grouped thematically. Results Demographics A total of 317 nurses returned the questionnaires, representing a 45% response rate, assuming that all nurses did, in fact, receive a personally addressed envelope. The age range of respondents was 43 years (minimum ? 1 years, O 2004 Blackwell Publishing Ltd, Journal of Clinical Nursing, 13, 304–315 Clinical nursing issues Barriers to, and facilitators of, research utilization maximum ? 64 years) while the range in years since registration was 42 years. The demographic characteristics of the nurses (Table 1) were consistent with those of the State of Victoria’s nursing workforce (The Australian Institute of Health and Welfare, 1999). Factor analysis A four-factor solution was selected as the most appropriate model arising from PCA of the 29 barrier items. This accounted for 39. % of the total variance in re sponses to all barrier items. The factor groupings including the loading for each barrier item and the titles allocated to each factor are included in Table 2. According to the correlation coef? cient or factor loading measure of †¡0. 4, two items, ‘research reports/articles are not published fast enough’, and ‘the research has not been replicated’, failed to load on any of the four factors. Table 1 Nurse demographics (n ? 317) Variable Gender Male Female Missing Age (years) Experience Registered Nurse (years) Clinical experience (years) Years since most recent quali? ation Highest quali? cation Division 2 certi? cate for registration Division 1 hospital certi? cate for registration Tertiary diploma/degree for registration Specialist nursing certi? cate Graduate diploma Masters by coursework Masters by research Others (including education and management quali? cations) Missing Principle job function Clinical Administrative Research Education Others Mis sing Research experience Yes No Missing N (%) Mean (SD) 24 (7. 6) 291 (91. 8) 2 (0. 6) 33. 8 (9. 73) 12. 6 (9. 95) 11. 35 (8. 8) 4. 28 (6. 52) 14 (4. 4) 23 (7. 3) 104 (32. 8) 26 34 9 1 87 (8. 2) (10. 7) (2. 8) (0. 3) (27. ) Factor 1, comprising eight items with loadings of 0. 73 to 0. 43, includes items relating to characteristics of the organization that in? uence research-based change. Eight items loaded onto factor 2 with loadings of 0. 66 to 0. 40. These items are associated with qualities of research and potential outcomes associated with the implementation of research ? ndings. Factor 3 with seven items loading 0. 60 to 0. 41, relates to the nurse’s research skills, beliefs and role limitations. Factor four refers to communication and accessibility of research ? ndings onto which ? ve items loaded 0. 67 to 0. 42.The four factor groupings comprising setting, nurse, research and presentation, generated in the US study 10 years ago (Funk et al. , 1991b), were similar to gr oupings that arose from factor analysis in the present study (Table 2). Cronbach’s alphas were calculated for each factor generated. For factors 1–3 the alpha coef? cients were 0. 75, 0. 74 and 0. 70, respectively, demonstrating good reliability. The alpha coef? cient for factor 4 was lower at 0. 54. The total scale alpha was 0. 86, which indicates that the scale can be considered reliable with this sample. Item-total correlations ranged from 0. 1 to 0. 60. Although a low correlation between some items and the total score was evident, deleting any of these items would have resulted in a reduction in reliability of the scale. Relative importance of barrier and facilitator items The percentages of items perceived by nurses’ as great or moderate barriers are summarized in Table 3. The respondents were also given the opportunity to list and rate any additional perceived barriers not included in the questionnaire. About 27% (85) of respondents documented a total of 1 74 barriers. However, analysis revealed that only 11% (36) of respondents actually identi? d additional barriers. The remainder had reiterated or reworded barrier items already included in the tool. The additional barrier items listed by respondents were grouped into themes, which included funding, organizational commitment, research training, implementation strategy and professional responsibility. The percentages of items perceived by nurses’ as great or moderate facilitators are summarized in Table 4. The respondents were also given the opportunity to list and rate additional perceived facilitators. Eighteen per cent (57) of respondents took the opportunity to record a total of 90 facilitators. Of these, 7. % (24) actually identi? ed additional facilitators whereas the remainder had rephrased or repeated items already included in the tool. Consistent with the themes identi? ed for the additional barriers were funding, organizational commitment, active participation in rese arch 309 19 (6. 0) 252 28 6 10 15 6 (79. 5) (8. 8) (1. 9) (3. 2) (4. 7) (1. 9) 207 (65. 3) 105 (33. 1) 5 (1. 6) O 2004 Blackwell Publishing Ltd, Journal of Clinical Nursing, 13, 304–315 A. M. Hutchinson and L. Johnston Table 2 BARRIERS Scale factors and factor loadings US factor groupings Factor loadings Communalities Factor 1 Factor 2 Factor 3 Factor 4Barrier item Factor 1: Organizational in? uences on research-based change Physician will not cooperate with implementation Administration will not allow implementation The nurse does not feels she/he has enough authority to change patient care procedures The facilities are inadequate for implementation Other staff are not supportive of implementation The nurse feels results are not generalizable to own setting The nurse is unwilling to change/try new ideas Factor 2: Qualities of the research and potential outcomes of implementation The research has methodological inadequacies The literature reports con? cting results The conclu sions drawn from the research are not justi? ed The research is not relevant to the nurse’s practice The nurse is uncertain whether to believe the results of the research The research is not reported clearly and readably Statistical analyses are not understandable The nurse feels the bene? ts of changing practice will be minimal Factor 3: Nurses’ research skills, beliefs and role limitations The nurse sees little bene? for self The nurse does not feel capable of evaluating the quality of the research There is not a documented need to change practice The nurse does not see the value of research for practice The amount of research information is overwhelming The nurse is isolated from knowledgeable colleagues with whom to discuss the research There is insuf? cient time on the job to implement new ideas Factor 4: Communication and accessibility of research ? dings Research reports/articles are not readily available Implications for practice are not made clear The nurse is unaware of the research The relevant literature is not compiled in one place The nurse does not have time to read research Setting Setting Setting Setting Setting Setting Nurse 0. 55 0. 52 0. 42 0. 42 0. 34 0. 39 0. 36 0. 73 0. 71 0. 56 0. 54 0. 53 0. 49 0. 43 0. 09 0. 10 0. 06 0. 11 0. 17 0. 30 0. 01 A0. 02 A0. 01 0. 31 A0. 04 0. 19 0. 23 0. 41 0. 09 A0. 04 0. 05 0. 33 0. 02 0. 01 A0. 09 Research Research Research Presentation Research Presentation PresentationNurse 0. 46 0. 38 0. 44 0. 43 0. 46 0. 33 0. 33 0. 46 0. 17 0. 11 0. 11 0. 22 0. 27 0. 11 A0. 04 0. 36 0. 66 0. 59 0. 57 0. 55 0. 53 0. 49 0. 47 0. 40 0. 03 0. 12 0. 30 A0. 13 0. 32 0. 18 0. 03 0. 38 0. 00 0. 04 A0. 05 0. 25 0. 07 0. 19 0. 32 A0. 14 Nurse Nurse Nurse Nurse * Nurse Setting Presentation Presentation Nurse Presentation Setting 0. 57 0. 45 0. 35 0. 55 0. 29 0. 31 0. 38 0. 45 0. 47 0. 33 0. 25 0. 31 0. 23 A0. 04 A0. 04 0. 15 0. 05 0. 31 0. 28 0. 01 0. 06 A0. 04 0. 13 0. 22 0. 39 0. 26 0. 14 0. 47 A0. 01 0. 11 A0. 17 0. 00 0. 31 0. 09 0. 3 A0. 14 0. 60 0. 58 0. 57 0. 55 0. 51 0. 42 0. 41 0. 00 A0. 09 0. 16 0. 13 0. 26 0. 04 0. 21 0. 09 A0. 04 0. 15 0. 16 0. 31 0. 67 0. 60 0. 54 0. 45 0. 42 Two items, ‘research reports/articles are not published fast enough’ and ‘the research has not been replicated’, did not load at the 0. 4 level in this analysis. *The item, ‘the amount of research information is overwhelming’ failed to load on any factor in the Funk et al. model. process – experience, strategy to ensure project completion, implementation strategies, and professional attitude.Discussion The present study generated a four-factor solution with similarities to that produced in the US by Funk et al. (1991b) and in the UK by Closs & Bryar (2001). The ? rst factor comprises characteristics of the organization and re? ects health professional and other resource support for change 310 associated with the implementation of research ? ndings. More broadly , the theme ‘organizational commitment’ identi? ed following analysis of the additional perceived barriers listed by respondents, appears to be associated with this factor.Organizational commitment, many respondents felt, would facilitate mobilization of resources to promote change. Factor 2 relates to qualities of research and potential outcomes associated with the implementation of research ? ndings. This factor re? ects the nurse’s reservations about reliability and validity of research ? ndings and conclusions, O 2004 Blackwell Publishing Ltd, Journal of Clinical Nursing, 13, 304–315 Clinical nursing issues Table 3 BARRIERS Scale items in rank order Barriers to, and facilitators of, research utilization Barrier items The nurse does not have time to read research There is insuf? ient time on the job to implement new ideas The nurse is unaware of the research The nurse does not feel she/he has enough authority to change patient care procedures Statistica l analyses are not understandable The relevant literature is not compiled in one place Physicians will not cooperate with the implementation The nurse does not feel capable of evaluating the quality of the research The facilities are inadequate for implementation Other staff are not supportive of implementation Research reports/articles are not readily available The nurse feels results are not generalizable to own setting The amount of research information is overwhelming Implications for practice are not made clear The research is not reported clearly and readably The research has not been replicated The nurse is isolated from knowledgeable colleagues with whom to discuss the research Administration will not allow implementation The research is not relevant to the nurse’s practice The literature reports con? icting results The nurse feels the bene? s of changing practice will be minimal The nurse is uncertain whether to believe the results of the research Research reports/ar ticles are not published fast enough The nurse is unwilling to change/try new ideas The research has methodological inadequacies The nurse sees little bene? t for self There is not a documented need to change practice The nurse does not see the value of research for practice The conclusions drawn from the research are not justi? ed Reporting item as moderate or great barrier (%) 78. 3 73. 8 66. 2 64. 7 64. 1 58. 7 56. 1 55. 8 52 52 50. 8 50. 8 45. 7 45. 5 43. 3 41. 3 41 35 34. 4 34 31. 9 30. 9 30. 6 29. 4 25. 5 23. 3 22. 1 17 13. 8 Item mean score (SD) 4. 06 3. 9 3. 64 3. 51 3. 56 3. 51 3. 41 3. 3 3. 23 3. 16 3. 19 3. 09 3. 07 3. 0 3. 01 3. 16 2. 76 2. 88 2. 67 2. 87 2. 52 2. 58 2. 81 2. 34 2. 85 2. 25 2. 27 1. 9 2. (1. 21) (1. 3) (1. 4) (1. 39) (1. 32) (1. 26) (1. 33) (1. 39) (1. 3) (1. 29) (1. 35) (1. 26) (1. 35) (1. 22) (1. 25) (1. 14) (1. 49) (1. 18) (1. 28) (1. 11) (1. 3) (1. 29) (1. 21) (1. 34) (1. 0) (1. 26) (1. 24) (1. 21) (1. 02) Responding ‘no opinion’ or non- response (%) 0. 9 1. 6 1. 6 0. 9 3. 8 13 7. 6 3. 5 8. 8 6. 3 6. 3 3. 5 6. 9 5 8. 2 26. 1 3. 8 19. 6 4. 4 18. 9 3. 5 4. 7 25. 2 2. 2 32. 5 3. 5 8. 5 1. 6 21 Table 4 Facilitator items in rank order Reporting item as moderate or great facilitator (%) 89. 6 89. 5 84. 8 82. 3 82. 0 81. 4 81. 3 78. 2 Number (%) responding ‘no opinion’ or non-response 8 (2. 5) 6 9 6 10 (1. 8) (2. 8) (1. 8) (3. 2)Facilitator item Increasing the time available for reviewing and implementing research ? ndings Conducting more clinically focused and relevant research Providing colleague support network/mechanisms Advanced education to increase your research knowledge base Enhancing managerial support and encouragement of research implementation Improving availability and accessibility of research reports Improving the understandability of research reports Employing nurses with research skills to serve as role models Item mean score (SD) 4. 52 (0. 93) 4. 39 4. 21 4. 11 4. 15 (0. 94) (1. 02) (1. 13) (1. 08) 4. 12 (1. 11) 4. 16 (1. 1) 4. 04 (1. 22) 5 (1. 5) 8 (2. 5) 9 (2. 9)O 2004 Blackwell Publishing Ltd, Journal of Clinical Nursing, 13, 304–315 311 A. M. Hutchinson and L. Johnston in addition to bene? ts of use of ? ndings in practice. Factor 3 focuses on characteristics of the nurse. In particular, this factor is associated with the nurse’s beliefs about the value of research and their research skills, in addition to the limitations of their role. The fourth factor is concerned with characteristics of communication. The focus of this factor centres on access to research ? ndings and understanding of the implications of ? ndings. The issues encompassed within this factor re? ect organizational barriers to access, and research presentation barriers.These factors are congruent with the concepts characterized in Rogers’ (1995) model of ‘diffusion of innovations’, including characteristics of the adopter, organization, innovation and communication , on which the BARRIERS Scale was developed. Two barrier items, ‘research reports/articles are not published fast enough’ and ‘the research has not been replicated’, failed to load suf? ciently onto a factor and were subsequently discarded. Exclusion of these items from the model re? ects their minimal signi? cance in relation to the underlying dimensions of the factors. That these items were ranked 23 and 16, respectively, is not surprising because they become less relevant when there is a perceived lack of time to read research and implement change as re? cted in the top two nominated barriers to research utilization. It is also important to note that over one quarter of respondents selected the ‘no opinion’ option or failed to respond to both of these items, which further suggests their lack of importance to respondents. The majority of respondents in this study rated approximately 40% of the barriers items as moderate or great barriers. Thi s is compared with the majority of nurse clinicians in the US (Funk et al. , 1991a) and nurses in the UK (Dunn et al. , 1997), who rated about 65% of the barrier items as moderate or great barriers. Overall, this group of Australian nurses perceived there to be fewer barriers to esearch utilization than their colleagues in the UK or US, with a mean score of 43. 7% of respondents rating all the barriers as moderate or great. In the UK (Walsh, 1997a) and the US (Funk et al. , 1991a) mean scores of 59. 8 and 55. 7%, respectively, re? ect the proportion of respondents who rated all barriers as moderate or great. Possible in? uences such as time, population, nursing education programmes should be acknowledged when considering these comparisons. Content analysis of the data comprising additional perceived barriers elicited ? ve new themes respondents associated with barriers to research utilization. Revision of the instrument to re? ect the themes identi? d and changes that have occurred over the past 10 years may be warranted to achieve a more valid scale for the setting in which it was used in this study. The addition of items consistent with changes in the availability of technological resources, information availability and use, and education may enhance the content validity of the scale. The ranking of perceived barriers in practice resulting from this study showed considerable consistency with rankings reported in other studies, as previously discussed. The top three barriers reported in 12 other studies fell within the top 10 barriers identi? ed in this study. Furthermore, two of the top three barriers in an additional two studies fell within the top 10 barriers identi? ed in the present study. The barrier item ‘there is insuf? ient time on the job to implement new ideas’ was reported within the top three barriers in 13 studies, including this and another Australian study (Retsas, 2000). When Spearman’s rank order correlation coef? cients were generated to compare the rank ordering of perceived barriers, a strong positive correlation between this and several other studies was evident (Table 5). Whilst acknowledging differences in nursing populations, sample size, sampling methods, response rates, and minor variations in item wording and number, this suggests a large degree of consistency regarding Study Funk et al. (1991a) Funk et al. (1995a) Dunn et al. (1997) Rutledge et al. (1998) Lewis et al. (1998) Kajermo et al. (1998) Retsas & Nolan (1999) Parahoo (2000) Retsas (2000) Closs et al. 2000) Parahoo & McCaughan (2001) Grif? ths et al. (2001) Location USA USA UK USA USA Sweden Australia Northern Ireland Australia UK Northern Ireland UK r 0. 866 0. 779 0. 835 0. 816 0. 879 0. 719 0. 884 0. 837 0. 801 0. 762 0. 799 0. 912 P 0. 000 0. 000 0. 000 0. 000 0. 000 0. 000 0. 000 0. 000 0. 000 0. 000 0. 000 0. 000 Coef? cient of determination (%) 75 61 70 66 77 52 78 70 64 58 64 83 Table 5 Barrier rank order correlations 312 O 2004 Blackwell Publishing Ltd, Journal of Clinical Nursing, 13, 304–315 Clinical nursing issues Barriers to, and facilitators of, research utilization nurses’ perceptions of the relative importance of the barrier items. Marsh et al. 2001) however, caution against international comparisons with the original US data because changes in nursing education and roles, technology, funding and collaboration with other disciplines since then, may invalidate such comparisons. Nonetheless, despite these changes, the ? ndings of the present study have consistencies with not only the US data of 1991 but also more recent studies in the US, UK, Sweden, Northern Ireland and Australia (Table 5). Thus, notwithstanding the increasing momentum of the evidence-based practice movement in recent years, the pursuit of professional status by the nursing profession, the move of nursing education to the tertiary sector, increased access to systematic reviews and research databases, the research – practice gap persists.In the light of the plethora of research and theoretical literature on the research–practice gap and issues surrounding research utilization, it is of concern that nurses’ perceptions of the barriers to research utilization appear to remain consistent. In particular, issues surrounding support for implementation of research ? ndings, authority to change practice, time constraints and ability critically to appraise research continue to be perceived by nurses as the greatest barriers to research utilization. This raises important questions. Firstly, do such perceptions re? ect the reality of contemporary nursing? Or rather, do they represent unchallenged, traditionally held and ? rmly entrenched beliefs, which are founded on an understanding of nursing in a socio-historic context that is no longer relevant? If such perceptions do, in fact, re? ct the reality of current day nursing practice, despite the changes and progress that have been ma de in health care and nursing over the last decade, it behoves us, as a profession, to address the issues related to time, authority, support and skills in critical appraisal with conviction and a sense of urgency. Contextual issues including the socio-political environment, organizational culture and interprofessional relations need to be taken into serious consideration when exploring and formulating potential strategies to overcome these barriers. The hospital in which this study was conducted has since undertaken to explore and develop strategies to address and overcome barriers to, and reinforce and strengthen facilitators of research utilization highlighted in the ? ndings. ther studies using the BARRIERS Scale, may re? ect a response bias. That is, nurses with a positive attitude to research may have been more likely to complete the questionnaire. Internal consistency, the extent to which items in the scale measure the same concept (LoBiondo-Wood & Haber, 1998), of the tool w as reasonable, although not as high as that reported by Funk et al. (1991b). For seven items, more than 10% of the respondents nominated ‘no opinion’ or failed to respond. Furthermore, this study was conducted in one organization; the ? ndings are therefore context speci? c, which makes it dif? cult to generalize to other settings. However, there is consistency over ime and between countries in regard to nurses’ perceptions of the barriers to research utilization. Conclusion In order to gain an understanding of perceived in? uences on nurses’ utilization of research in a particular practice setting, nurses were surveyed to elicit their opinions regarding barriers to, and facilitators of, research utilization. Many of the perceived barriers to research utilization reported by this group of Australian nurses are consistent with reported perceptions of nurses in the US, UK and Northern Ireland during the past decade. Time was the most important barrier percei ved by nurses in this study, which is re? ected by responses to the items, ‘the nurse does not have time to read research’ and ‘there is insuf? ient time on the job to implement new ideas’, resulting in them being ranked as the top two barriers to research utilization. Consistent with this ? nding was the ranking of facilitator item ‘increasing the time available for reviewing and implementing research ? ndings’ as the most important facilitator to research utilization. The employment of qualitative research methods, such as observation and interview, will contribute further to our knowledge about barriers to, and facilitators of, research utilization by nurses by allowing deeper exploration of experiences, perception and issues faced by nurses in the utilization of research in their practice.Fundamental questions about whether nurses’ perceptions actually re? ect the reality of the current context of nursing need to be further investiga ted. Future research should also examine issues surrounding the use of time by nurses. Questions exploring how much additional time nurses require in order to read the relevant literature and how nurses can be given more time to implement new ideas, need to be addressed. Issues related to nurses’ perception of their authority to change patient care procedures, the support and cooperation afforded by doctors and others, the facilities and availability of resources, and their skills in critical appraisal, also require further 313 LimitationsReporting bias associated with the self-report method raises questions about the extent to which the responses accurately represent nurses’ perceptions of the barriers to research utilization. The low response rate achieved in this study, although consistent with response rates reported in several O 2004 Blackwell Publishing Ltd, Journal of Clinical Nursing, 13, 304–315 A. M. Hutchinson and L. Johnston exploration. Investigatio n of the information-seeking behaviour of nurses, the means by which they gain and synthesize new research knowledge and the way in which they apply that knowledge to their decision making, will further contribute to our understanding of the research–practice gap phenomenon.Measurement of the actual extent of research utilization by nurses in the practice setting presents a major challenge for researchers in this ? eld. Acknowledgements The authors thank Sandra Funk for her permission to use the BARRIERS Scale for the purpose of this study. We wish to acknowledge and thank the nurses who completed the questionnaire. The authors also wish to acknowledge the statistical assistance provided by Ms Anne Solterbeck, Statistical Consulting Centre, Department of Mathematics and Statistics, The University of Melbourne. Contributions Study design: LJ, AMH; data analysis: AMH; manuscript preparation: AMH, LJ; literature review: AMH. References Berggren A. 1996) Swedish midwives’ awareness of, attitudes to and use of selected research findings. Journal of Advanced Nursing 23, 462–470. Carroll D. L. , Greenwood R. , Lynch K. , Sullivan J. K. , Ready C. H. & Fitzmaurice J. B. (1997) Barriers and facilitators to the utilization of nursing research. Clinical Nurse Specialist 11, 207–212. Closs S. J. & Bryar R. M. (2001) The barriers scale: does it ‘fit’ the current NHS research culture? NT Research 6, 853–865. Closs S. J. & Cheater F. M. (1994) Utilization of nursing research: culture, interest and support. Journal of Advanced Nursing 19, 762–773. Closs S. J. , Baum G. , Bryar R. M. , Griffiths J. & Knight S. (2000) Barriers to research implementation in two Yorkshire hospitals.Clinical Effectiveness in Nursing 4, 3–10. Crane J. , Pelz D. C. & Horsley J. A. (1977) Conduct and Utilization of Research in Nursing Project. School of Nursing, University of Michigan, Ann Arbor, MI. Crichton N. (2000) Information point: p rincipal component analysis. Journal of Clinical Nursing 9, 815. Crichton N. (2001) Information point: factor analysis. Journal of Clinical Nursing 10, 550–562. Dunn V. , Crichton N. , Roe B. , Seers K. & Williams K. (1997) Using research for practice: a UK experience of the barriers scale. Journal of Advanced Nursing 26, 1203–1210. Estabrooks C. A. (1999) Will evidence-based nursing practice make practice perfect?Canadian Journal of Nursing Research 30, 273–294. Evidence-Based Medicine Working Group (1992) A new approach to teaching the practice of medicine. Journal of the American Medical Association 268, 2420–2425. Funk S. G. , Champagne M. T. , Wiese R. A. & Tornquist E. M. (1991a) Barriers to using research findings in practice: the clinician’s perspective. Applied Nursing Research 4, 90–95. Funk S. G. , Champagne M. T. , Wiese R. A. & Tornquist E. M. (1991b) Barriers: the barriers to research utilization scale. Applied Nursing Research 4, 39–45. Funk S. G. , Champagne M. T. , Tornquist E. M. & Wiese R. (1995a) Administrator’s views on barriers to research utilization.Applied Nursing Research 8, 44–49. Funk S. G. , Tornquist E. M. & Champagne M. T. (1995b) Barriers and facilitators of research utilization. Nursing Clinics of North America 30, 395–407. Gennaro S. , Hodnett E. & Kearney M. (2001) Making evidencebased practice a reality in your institution: evaluating the evidence and using the evidence to change clinical practice. MCN, the American Journal of Maternal/Child Nursing 26, 236–244. Gould D. (1986) Pressure sore prevention and treatment: an example of nurses’ failure to implement research findings. Journal of Advanced Nursing 11, 389–394. Griffiths J. M. , Bryar R. M. , Closs S. J. , Cooke J. , Hostick T. , Kelly S. Marshall K. (2001) Barriers to research implementation by community nurses. British Journal of Community Nursing 6, 501–510. Hicks C. ( 1994) Bridging the gap between research and practice: an assessment of the value of a study day in developing research reading skills in midwives. Midwifery 10, 18–25. Hicks C. (1996) A study of nurses’ attitudes towards research: a factor analytic approach. Journal of Advanced Nursing 23, 373–379. Hunt J. (1981) Indicators for nursing practice: the use of research findings. Journal of Advanced Nursing 6, 189–194. Hunt J. (1996) Barriers to research utilization. Journal of Advanced Nursing 23, 423–425. Hunt J. 1997) Towards evidence based practice. Nursing Management 4, 14–17. Kaiser H. (1974) An index of factorial simplicity. Psychometrika 39, 31–36. Kajermo K. N. , Nordstrom G. , Krusebrant A. & Bjovell H. (1998) Barriers to and facilitators of research utilization, as perceived by a group of registered nurses in Sweden. Journal of Advanced Nursing 27, 798–807. Ketefian S. (1975) Application of selected nursing research findi ngs into nursing practice: a pilot study. Nursing Research 24, 89–92. Kirchhoff K. T. (1982) A diffusion survey of coronary precautions. Nursing Research 31, 196–201. Lacey A. (1994) Research utilization in nursing practice: a pilot study.Journal of Advanced Nursing 19, 987–997. Lewis S. L. , Prowant B. F. , Cooper C. L. & Bonner P. N. (1998) Nephrology nurses’ perceptions of barriers and facilitators to using research in practice. ANNA Journal 25, 397–405. LoBiondo-Wood G. & Haber J. (1998) Nursing Research. Methods, Critical Appraisal and Utilization. Mosby, St Louis, MO. MacGuire J. M. (1990) Putting nursing research findings into practice: research utilization as an aspect of the management for change. Journal of Advanced Nursing 15, 614–620. 314 O 2004 Blackwell Publishing Ltd, Journal of Clinical Nursing, 13, 304–315 Clinical nursing issues Marsh G. W. , Nolan M. & Hopkins S. 2001) Testing the revised barriers to research utiliz ation for use in the UK. Clinical Effectiveness in Nursing 5, 66–72. Nolan M. , Morgan L. , Curran M. , Clayton J. , Gerrish K. & Parker K. (1998) Evidence-based care: can we overcome the barriers? British Journal of Nursing 7, 1273–1278. Parahoo K. (2000) Barriers to, and facilitators of, research utilization among nurses in Northern Ireland. Journal of Advanced Nursing 31, 89–98. Parahoo K. & McCaughan E. M. (2001) Research utilization among medical and surgical nurses: a comparison of their self reports and perceptions of barriers and facilitators. Journal of Nursing Management 9, 21–30. Retsas A. 2000) Barriers to using research evidence in nursing practice. Journal of Advanced Nursing 31, 599–606. Retsas A. & Nolan M. (1999) Barriers to nurses’ use of research: an Australian hospital study. International Journal of Nursing Studies 36, 335–343. Rizzuto C. , Bostrum J. , Suter W. N. & Chenitz W. C. (1994) Predictors of nursesâ€⠄¢ involvement in research activities. Western Journal of Nursing Research 16, 193–204. Rogers E. M. (1995) Diffusion of Innovations. The Free Press, New York. Rutledge D. N. , Ropka M. , Greene P. E. , Nail L. & Mooney K. H. (1998) Barriers to research utilization for oncology staff nurses and nurse managers/clinical nurse specialists. Oncology Nursing Forum 25, 497–506.Barriers to, and facilitators of, research utilization Sackett D. L. , Rosenberg W. M. C. , Gray J. A. M. , Haynes R. B. & Richardson W. S. (1996) Evidence based medicine: what it is and what it isn’t. British Medical Journal 312, 71–72. Stetler C. B. (1994a) Problems and issues of research utilization. In Nursing Issues in the 1990’s (Strickland O. L. & Fishman D. L. eds). Delmar, New York, pp. 459–470. Stetler C. B. (1994b) Refinement of the Stetler/Marram model for application of research findings to practice. Nursing Outlook 42, 15–25. Tabachnick B. G. & Fidell L. S. (2001) Using Multivariate Statistics. Allyn & Bacon, Needham Heights, MA.The Australian Institute of Health and Welfare (1999) National Health Labour Force Series. Number 20 – Nursing Labour Force 1999. The Australian Institute of Health and Welfare, Canberra. Walsh M. (1997a) How nurses perceive barriers to research implementation. Nursing Standard 11, 34–39. Walsh M. (1997b) Perceptions of barriers to implementing research. Nursing Standard 11, 34–37. Walsh M. & Ford P. (1989) Rituals in nursing: ‘we always do it this way’. Nursing Times 85, 26–35. Winter J. C. (1990) Brief. Relationship between sources of knowledge and use of research findings. The Journal of Continuing Education in Nursing 21, 138–140. O 2004 Blackwell Publishing Ltd, Journal of Clinical Nursing, 13, 304–315 315