Author's response to reviews Title:Evidence based practice beliefs and implementation among nurses: A cross-sectional study Authors: Kjersti Stokke (KST@ous-hf.no) Nina R Olsen (Nina.Rydland.Olsen@hib.no) Birgitte Espehaug (birgitte.espehaug@hib.no) Monica W Nortvedt (Monica.Wammen.Nortvedt@hib.no) Version:3Date:1 November 2013 Author's response to reviews: see over
Author s response to reviewers reports Title: Attitudes and behaviour towards evidence-based practice among nurses: A cross- sectional study Authors: Kjersti Stokke (KST@ous-hf.no) Nina Rydland Olsen (nina.rydland.olsen@hib.no) Birgitte Espehaug (Birgitte.Espehaug@hib.no) Monica Wammen Nortvedt (monica.wammen.nortvedt@hib.no) Version 1: submitted 16.september 2013
Kjersti Stokke Department of Oncology The Cancer-, Surgery- and Transplantation Clinic Oslo University Hospital Oslo, Norway 31. October 2103 To: Journal Editorial Office BioMed Central Subject: Attitudes and behaviour towards evidence-based practice among nurses: A cross- sectional study Reviewed Subject (November 2013): Evidence based practice beliefs and implementation among nurses: A cross-sectional study Resubmission manuscript: 1319113676964997 Dear Journal Editorial Office, BioMed Central, dear reviewers, Thank you for the opportunity to submit a revised version of our manuscript. We appreciate constructive criticism and helpful feedback from the reviewers Rebecca P Winsett, Joyce Wilkinson and Ann Catrine Eldh. All reviewers comments are addressed in detail below, in the order of the reviewers remarks and comments. Sincerely, Kjersti Stokke
Reviewer: Rebecca P Winsett Major compulsory 1. What was the reason you chose to use multiple linear regression? You may wish to describe the purpose of using regression in your data analysis section. For example, multiple linear regression was used to determine the contribution of background variables, knowledge, and attitudes towards EBP on implementation behavior. (You used linear regression as you first established that there was correlation among variables and your scatterplot was somewhat linear in shape). Reply: Multiple linear regression analyses were used to determine the contribution of background variables on EBP beliefs and implementation, respectively, while controlling for the other variables. With the exception of the variable indicating seniority, the background variables were all categorical and in the analyses each category was compared against a chosen reference category. Regarding the variable seniority, we did check for a possible non-linear relationship with the outcome variables but this was not found. Correction is performed in the manuscript. 2. Figure 2. The x and the y axis are unlabeled. Does the figure have a purpose? It does not match the text where you refer to it. Where was r coefficient.59 obtained? Reply: Figure 2 was included in order to visually show the correlation between beliefs toward EBP and implementation of EBP (r = 0.59). We agree that Figure 2 perhaps was unnecessary to include and have removed it. 3. Please describe in more detail your results of the multiple linear regression. If you are exploring what contributes the most to implementation behaviour, describe the beta values, confidence intervals. Table 5 should include your dependent variable and your explanatory (independent variables) with corresponding correlation coefficients standardized beta scores, p values. Reply: More details have now been added to the results part on beta values and corresponding 95 % confidence intervals. Table 5 has also been changed accordingly and now includes estimated beta values, standardized beta values, confidence intervals and p- values. Correlation coefficients have not been provided as, with one exception, the independent variables were all categorical with a limited set of values.
Minor essential 1. The last sentence in the Result section of the abstract is unclear. Where was the r coefficient.59 obtained? 2. Would it be clearer to state that there was a positive moderate correlation in attitudes and knowledge of EBP (r=.38, p<.0001)? This pertains to both the abstract Result section and the Result/Link between attitudes and behaviour in EBP section in the text. Reply: 1. & 2. : The last sentence in the Result section is now changed to display the there was a positive correlation (r) between the EBP Belief Scale and the EBP Implementation Scale (r = 0.59, p < 0.001). Further, the text has been corrected to state more clearly that statistically significant, but moderate positive correlations were found between all the four subscales of the EBP Beliefs Scale (attitudes related to: 1) knowledge, 2) resources, 3) the value of EBP and 4) difficulty and time) and the EBP Implementation Scale, with the strongest correlation observed for attitudes related to knowledge of EBP (r=0.38, p<0.001). The results section of the abstract has also been changed accordingly. 3. Scores by nurse level were introduced in the Discussion section (2nd sentence first paragraph and paragraph 4). There was no data presented earlier by nurse level, so discussing it here is not appropriate. If you feel that scores by occupational position is important add this to your results section. Reply: We agree with the referee, and have chosen to remove this information from the Discussion section (2nd sentence first paragraph and paragraph 4). Discretionary revisions 1. Delete spaces between numbers and % symbol throughout manuscript. Reply: Corrections are performed. 2. What do you mean by the statement correlated most? (result section, link between attitudes and behaviour in EBP, 2 nd paragraph, first sentence) Reply: The statement is moderated from correlated most to a positive moderate correlation. We have now chosen to describe the correlation this way: There was a statistical significant positive, but moderate correlation between all the four subscales of the EBP
Beliefs Scale (beliefs related to: 1) knowledge, 2) resources, 3) the value of EBP and 4) difficulty and time) and the EBP Implementation Scale, with the highest correlation observed for beliefs related to knowledge P (r=0.38, p<.0001). 3. Use of? mark is not needed. Reply: Question mark is removed. 4. Quotation mark in wrong direction. Methodological issues, 3 rd paragraph, 4 th sentence. Reply: Correction is performed. 5. Do you mean relationship instead of relation in the Conclusion section, 2nd sentence? Reply: Yes, and the correction is performed. 6. Typo in reference #40. Reply: Correction is performed.
Reviewer: Joyce Wilkinson 1. The paper seems to assume the stance that EBP implementation is an event rather than a process that is ongoing and it may benefit the paper to acknowledge this more fully. 2. The paper seems to suggest that EBP implementation is a straightforward event which is largely a linear process rather than the complex process that it is now know to be. Reply: 1 & 2: We agree that it can appear as we mean that EBP implementation is a happening rather than a process. We have revised most of the background section and clarified that EBP is a process that is far from straightforward and does not follow a prescribed, logical and linear path, but is both challenging and complex. The complexity of EBP implementation is also discussed further in the discussion section as implementing EBP requires a whole system change implicating individuals, teams and the organization. 3. The limitations of the study could be more fully acknowledged in that surveys rely on memory (in this instance of practice in the last 8 weeks before the survey) and this may have impacted on results. In addition, surveys of this nature are often prone to socially desirable responses and this may have been a specific factor in this study as nurses may not have wanted to admit that they were not using EBP. Reply: We have followed this recommendation and have added in the Methodological issues that : There is a risk of receiving socially desirable responses where answers reflect an anticipated social norm, and retrospective self-reports about EBP beliefs and behaviour have been criticized for being biased. 4. The discussion assumes that having access to knowledge (EBP resources) is the most significant factor in the use of evidence in practice, however this fails to acknowledge that there is also a need for knowledge of implementation processes, of wider contextual factors and the need for access to financial and human resources for implementation. Reply: This is clarified by expanding the theme beyond individual factors, at the end of the discussion: Effective change management plays a fundamental role facilitating an organizational environment that encourages EBP implementation. The lead management plays a essential role in the technical and facilitative leadership, the organization s policies, procedures, values, established habits, routines, financial and human resources and supervision of the clinical and non-clinical processes involved in EBP implementation.
5. On page 13 there is reference made to the UKCC, this was superceded by the Nursing & Midwifery Council some years ago. Reply: The findings and the reference from this study is now removed. 6. On page 14 of the paper the final sentence of the discussion section refers to the best way to organise implementation of EBP in health services which seems to get to the heart of the problem. It might be helpful to use this point to highlight the particular contribution of this study to this point. Reply: We have tried to highlight this by raising this point by supplementing at the end of the discussion section that: In this study there was a significant difference between the nurses who took part in evidencebased working groups and those who did not. Findings from a survey support the idea that mentorship in EBP facilitates the implementation of evidence-based care. Polit and Beck stressed the importance of education, administrative support, resources, and developing collaborations with potential mentors who can provide guidance and direction in the search for and appraisal of evidence. Health political visions and goals require health staff to have competence in EBP and work in an evidence-based way, but implementing EBP among nurses in clinical practice is challenging. Although the use of questionnaires replied by individual nurses implies that we measure practice of EBP at the individual level, as nurses actually perform the components of EBP, we do not believe that practicing EBP is a purely individual responsibility. As nurses often say that they lack the autonomy to change practice, implementing EBP requires a whole system change implicating individuals, teams and the organization. Effective change management plays a fundamental role facilitating an organizational environment that encourages EBP implementation. The lead management plays a essential role in the technical and facilitative leadership, the organization s policies, procedures, values, established habits, routines, financial and human resources and supervision of the clinical and non-clinical processes involved in EBP implementation. Changes has also been made in the Conclusion where we have included the contribution of this study to how to the best way to organise implementation of EBP in health services, and where we also point out that there is still a need to find more about how best to organise the implementation of EBP in the health services effectively.
To reviewer: Ann Catrine Eldh 1. The study is relatively small and with a rather low response rate. In a parallel study (Dalheim et al) you reached a more acceptable response rate. Could you elaborate somewhat on the response rate in this paper, in particular in the methodological discussion on what actions you applied or could have applied to improve or better understand the sample, i.e. if you consider reminders and/or additional interviews could have improved your data set and thus the understanding of the conditions prior to implementing EBP in the hospital? Reply: We have now elaborated the actions we applied in the Methodological in the Method, in the beginning of Procedure: Each hospital unit assigned a contact person for the project. The contact person and the senior charge nurse at each unit were informed about the aim of the study, that the study was anonymised and that participation was voluntary. The first author disseminated information about the survey to all nurses at the hospital a week before the data collection started. The contact persons daily reminded the nurses about the survey. In addition, they were responsible for handing out and collecting the questionnaires, including the information letters, consent forms and reply envelopes. And we added in the Methodological issues, second sentence that: Possibly a higher response rate could have been achieved with more reminders. 2. With regards to EBP; In the Background, you argue for the concept of evidencebased practice to be central to high quality patient care. While EBP is one of many concepts contributing to best practice, it is by far exclusive. Rather, EBP relates to concepts such as knowledge translation, or knowledge utilisation, and research utilisation (RU). In order to set your study in a larger perspective, I suggest the paper should incorporate these EBP-related concepts. This further relates to a tendency to mismatch use of the concepts: while you set off with EBP, you later (particularly in the Discussion) write about RU. Please assure that you not only set the picture in the background section but use the concepts consistently, as they are not interchangeable. Reply: We agree that EBP is one of many concepts contributing to best practice and that EBP relates to concepts such as knowledge translation (KT), or
knowledge utilization (KU), and research utilisation (RU). However, as our aim is directly related to EBP as such, and the same applies for the survey instruments, we do not expand on these KT or RU. The EBP Beliefs Scale and the EBP Implementation Scale developed by Melnyk and Fineout Overholt measures the strength of beliefs in EBP and the frequency of implementing EBP. We recognize that we have used the terms EBP and RU interchangeably and appreciate this comment. We now use the term EBP consequently throughout this paper. To our knowledge EBP represents a broader concept that goes beyond just the rigorous scientific research steps. When clinicians use the EBP approach, they involve more than using research findings in practice; it involves several welldefined activities and steps as we clearify in the first paragraph in the introduction. 3. In addition, the background section is concluded by stating that is most important for nurses to have a positive attitude for the implementation of EBP. What do you mean by implementation of EBP if you define EBP as the integration of best research evidence with clinical experience and patient values? Please clarify, and as suggested above, state and sty with your main concept throughout the paper. Reply: Please see the changes in the first paragraph in the introduction: EBP involves the following steps: asking clinical questions, searching for and collecting the most relevant best evidence, critically appraising the evidence, integrating the evidence with one s clinical expertise, patient preferences and values, and evaluating outcomes of the practice decision or change made on the evidence (Melnyk & Fineout-Overholt 2011). EBP implementation involves use of the EBP steps and strategies that promote integration of best available evidence with practitioner expertise and other recourses (Melnyk, Fineout Overholt & May 2008). 4. Further on EBP, there are a number of powerful statements that lack references and/or clarification; in the second paragraph, you state that this challenges nurses to examine their nursing practice in a new way. Please clarify this sentence, as it is not clear whether it is the EBP, the definition or the integration that challenges nurses, or what is your basis for stating this. Reply: These statements were removed as we changed the introduction to meet other recommendations.
5. Please elaborate on what is nurses to you. From an international perspective, nurses include a number of positions such as licensed practical nurses, registered nurse etc. I suggest you clarify that in your study, you focused on registered nurses only. Also, it seems unclear why you included the RNs that were not involved in patient care, if you wish to relate to what you start off with in the paper, i.e., that it s about improving practice in direct health care. Reply: Under Methods, Participants, we do account what a nurse is in Norway and why nurses that were not involved in patient care was included: the Norwegian nursing education is a 3-years bachelor program (180 ECTS). Further in the article we refer to them as registered nurses (RN). Nurses have the opportunity to conduct specialization after having gained some clinical experience for some years. Education which provides specialist expertise in nursing is done by universities and university colleges and leads in some cases to a master s degree. The program takes from 1 to 2 years fulltime (referred as specialist nurses). We also included the nurses that were not always directly involved in patient care, such as senior charge nurses and professional development nurses at the hospital units, as support from nursing administrators and leaders, is seen as a key element to promote EBP. 6. I suggest you update the references, since there are a number of later studies which could inform your study, and place your findings of attitudes and behaviours in a larger/international perspective. For example, there are supposedly relevant studies performed in Sweden (Boström et al in BMC Health Services Research 2013) Canada (Doran et al in Implementation Science 2012) and the US (Nickerson & Thurkettle in Journal of Nursing Education 2013). Reply: In the revised introduction, we have updated the references and added among others: Yoo & Oh 2012; Thorsteinsson 2013, Melnyk et al. 2010; Squires et al. 2011, Boström et al 2103. 7. In the Discussion, avoid lengthy paragraphs on what others have found (such as the second paragraph). Rather, this section or the Conclusion could benefit from you further exploring what your study findings add to the scientific arena of EBP and/or translation och knowledge into practice. Reply: In the Discussion, the second paragraph is shortened. We wish though to retain the paragraph, as it clarifies our result compared with other studies that have used the same scales. Changes have been made in the Conclusion where
we have included the contribution of this study. Please see the reply made to Joyce Wilkinson, point 6. 8. The tables and figures are duplicated most likely a mistake or possibly by technical error. In the first set of tables, there are track changes, indicating that those are redundant. Reply: We believe this could be a technical error as it is not visible to us. In addition: We have changed the heading from Attitudes and behaviour towards evidence-based practice among nurses: A cross- sectional study to the Reviewed Subject (November 2013): Evidence based practice beliefs and implementation among nurses: A cross-sectional study. The reason is that we are now of through the manuscript consistently have kept us to the terms that the EBP Implementation Scales and the EBP Belief Scale make use.