Framework of policy recommendations for implementation of evidence-based practice: a systematic scoping review
- 1Department of Quality Assurance & Process Innovation, Academic Medical Center, Amsterdam, The Netherlands
- 2Department of Surgery, Academic Medical Center, Amsterdam, The Netherlands
- 3Department of Clinical Epidemiology & Biostatistics, McMaster University, Hamilton, Canada
- 4Amsterdam School of Health Professions, University of Amsterdam, Amsterdam, The Netherlands
- Correspondence to Dr Dirk T Ubbink;
- Received 26 July 2012
- Accepted 21 December 2012
- Published 24 January 2013
Objectives Evidence-based practice (EBP) may help improve healthcare quality. However, not all healthcare professionals and managers use EBP in their daily practice. We systematically reviewed the literature to summarise self-reported appreciation of EBP and organisational infrastructure solutions proposed to promote EBP.
Design Systematic review. Two investigators independently performed the systematic reviewing process.
Information sources MEDLINE, EMBASE and Cochrane Library were searched for publications between 2000 and 2011.
Eligibility criteria for included studies Reviews and surveys of EBP attitude, knowledge, awareness, skills, barriers and facilitators among managers, doctors and nurses in clinical settings.
Results We found 31 surveys of fairly good quality. General attitude towards EBP was welcoming. Respondents perceived several barriers, but also many facilitators for EBP implementation. Solutions were proposed at various organisational levels, including (inter)national associations and hospital management promoting EBP, pregraduate and postgraduate education, as well as individual support by EBP mentors on the wards to move EBP from the classroom to the bedside.
Conclusions More than 20 years after its introduction, the EBP paradigm has been embraced by healthcare professionals as an important means to improve quality of patient care, but its implementation is still deficient. Policy exerted at microlevel , middlelevel and macrolevel, and supported by professional, educational and managerial role models, may further facilitate EBP.
Systematic review of the literature to summarise self-reported appreciation of evidence-based practice (EBP) and organisational infrastructure solutions proposed to promote EBP.
More than 20 years after its introduction, the EBP paradigm has been embraced by healthcare professionals as an important means to improve quality of patient care, but its implementation is still deficient.
Policy exerted at microlevel, middlelevel and macrolevel, and supported by professional, educational and managerial role models, may further facilitate EBP.
Strengths and limitations of this study
Worldwide overview of EBP appreciation and implementation strategies useful for all centres striving at a better EBP implementation.
Self-reporting may have led to an overestimation of the results.
The success of implementation strategies is still unclear.
Evidence-based practice (EBP) provides a structure for the bedside use of research and consideration of patient values and preferences to optimise clinical decision-making and to improve patient care.1 ,2 EBP could potentially be used to improve quality of healthcare.3 ,4 In 2001, the Institute of Medicine's Quality Chasm series suggested EBP as one of the five core competencies for professional healthcare curricula.5 More recently, the growing societal demand for quality, safety, equality and accountability of healthcare and credentialing programmes as exerted by the Joint Commission International and Magnet hospitals have further promoted EBP.6 ,7 To date, hospital executive boards, insurance companies and consumers recognise that EBP may help prevent unsafe or inefficient practices, as part of a strategy to achieve quality improvement in healthcare.8
Thus far, however, educational efforts have failed to achieve EBP at the bedside or in daily clinical problem-solving. While there is an ongoing debate on how to measure quality of care in general, attitude, awareness, knowledge or behaviour are relevant to understand application of EBP. Various questionnaires have been developed and used to appreciate these aspects (eg, McColl and Funk).9 ,10 This information suggested the implementation of EBP by doctors is hampered by a perceived lack of time, knowledge or EBP resources,9 ,11 whereas in the nursing realm EBP awareness, the body of knowledge and research utilisation, as well as managerial support are still developing.12 ,13 Based on these findings, many different recommendations for improvement have been proposed. Hence, it is timely to synthesise these recommendations for more structural organisational initiatives that may help overcome barriers and facilitate the uptake of EBP.
Therefore, the purpose of this study was to collect surveys of healthcare professionals’ views on EBP in terms of self-reported attitude, knowledge, awareness, skills, barriers and behaviour regarding EBP among clinical doctors, nurses and managers, and to summarise proposed recommendations as derived from these views to improve the use of EBP. We subsequently used the findings of this review to propose a framework for implementation of EBP, tailor-made for different managerial levels and suitable to structurally facilitate and sustain evidence-based behaviour in clinical healthcare organisations.
Literature search and study selection
Two of the authors (DTU, HV) searched the MEDLINE (using PubMed), EMBASE (using Ovid) and Cochrane databases from 2000 through 2011 for surveys or reviews of EBP attitude, knowledge, awareness, barriers and facilitators among nurses, physicians and managers in any clinical setting, that is, hospitals or other healthcare institutions, rather than general practice settings, on which a review has recently been published.14 Reference lists of the included studies and reviews were checked for additional eligible papers.
In brief, our search strategy was (evidence-based[ti] practice OR evidence-based medicine OR EBM OR EBP) and (questionnaire* OR survey OR inventory) and ((barriers OR McColl) AND (knowledge OR attitude* OR aware* OR behavio*) and (hospital* OR clinic* OR medical cent*)). No language restrictions were applied. Papers in foreign languages, if any, would be translated if possible.
We excluded studies in an undergraduate educational setting, studies with a purely qualitative design, studies not including clinical doctors or nurses, and those focusing on a specific disorder, guideline, model or technique. We focused on surveys rather than the latter studies, because merely following (particularly expert-based) guidelines or focusing on a specific disorder or technique does not necessarily indicate the general application of the five steps of EBP. Studies before 2000 were also excluded because in these years the EBP paradigm was in an early phase with a limited dispersion among healthcare professionals. Study selection and quality assessment was performed by two investigators independently.
Judgment of the quality of the surveys was based on the number of centres and respondents involved, response rates and robustness of the questionnaires used (through pilot testing, prior validation or internal consistency based on a Cronbach's α).
Data items and synthesis of results
By means of a structured form, two researchers independently extracted data on study characteristics (including country of origin, publication year, type and number of respondents and type of clinics included), questionnaires used and EBP characteristics studied, in particular EBP attitude, knowledge, skills and awareness, and perceived barriers and facilitating factors for EBP implementation. We extracted in a qualitative manner the reported recommendations, if any, on how to overcome these barriers or how to exploit facilitators. These were grouped into solutions to be executed at various organisational levels. After one investigator had entered the data in the database, these data were checked for accuracy by a second.
Meta-analysis was not planned because of the expected large range in geographical locations, caregivers investigated and questionnaires used. To summarise the results of the studies reporting on EBP attitudes and knowledge, we calculated the medians and report the ranges of the scores given for each item, for doctors and nurses separately. A possible association between response rate, year of publication and attitude towards EBP was calculated using Spearman's correlation coefficient. Statistical analysis was performed using PASW Statistics V.18.0 (IBM Inc, Armonk, New York, USA).
Our search yielded 286 potentially relevant studies. We also found two recent reviews of studies on barriers towards EBP,15 ,16 from which other relevant studies were derived. Some more recent studies not included in these reviews were also found by hand-searching the references of included studies. Four surveys among medical postgraduates were excluded because these publications were in Chinese. In total, 31 studies that included 10 798 respondents from 17 countries proved eligible (table 1). Studies represented nearly all continents, one-third (11/31) were European and a quarter (8/31) were from North America (figure 1). In four of the studies, EBP questions were administered in the context of an educational meeting. Seventeen studies focused specifically on doctors, 11 on nurses. Three of the 31 studies enrolled both doctors and nurses.24 ,30 ,43 Wherever possible, results from doctors and nurses are presented separately.
All studies applied postal or electronic questionnaires. To assess EBP attitude, knowledge, skills and awareness, most studies used the questionnaires developed by McColl, Upton or Estabrooks.9 ,47 ,48 To assess EBP barriers and facilitators, most investigators used the Funk questionnaire.10 Half of the studies investigated both EBP attitude and barriers.
The studies enrolled from 1910 to 115624 respondents (median 273), consisting of doctors (residents, specialists) and nurses (ward and staff nurses, nurse managers and educators) from various clinical specialties. Seven of the 31 studies were conducted in a single centre. Response rates varied from 9% in nationwide surveys to 100% in questionnaires during trainings, with a median of 72%. Twenty-four of the 31 studies (77%) used robust questionnaires. So, overall quality of the included studies was good (table 2). Most studies addressed EBP attitude, skills and barriers (table 1).
Fifteen of the 18 studies addressing EBP attitude used a (sometimes modified) McColl questionnaire. Based on these 15 studies, both doctors and nurses strongly felt that EBP improves patient care and is important for their profession (table 3). Their overall attitude towards EBP was welcoming and appreciated the use of research evidence in daily clinical practice. However, they considered only half of their clinical practice to be evidence-based, although what they meant by this was, in most cases, not specified and unclear. These findings were consistent among the various countries. We did not find significant correlations between either response rate (−0.112; p=0.703) or year of publication (−0.286; p=0.321) and attitude towards EBP.
EBP knowledge and skills
The majority (median 64%) of doctors and nurses reported they considered their EBP knowledge was insufficient. Similarly, a median of 70% of the respondents regarded their skills as insufficient, even in the most recent studies, and desired (more) EBP training. The percentage of doctors who had had EBP training ranged from 13% (Indian surgical trainees) to 80% (Iranian internal medicine doctors). The most appropriate way, respondents thought to move towards EBP, was through evidence-based guidelines (median 68%), evidence summaries (median 39%) or critical appraisal skills (median 36%).
PubMed accessibility was high (at least 88%, except for India, 58% and Jordan, 70%), either at home or at work. However, clinical decision-making was based on consulting textbooks and colleagues rather than by searching electronic databases.
Figure 2 depicts the knowledge of common EBP terms among doctors. Not all studies used the same EBP terms, but in general, half of the doctors had at least some knowledge about 83% (20/24) of the presented EBP terms. Three of the four terms they were unfamiliar with were meaningless dummy terms. Hence, the results of this part of the questionnaire seemed not biased by socially desired answering.
Only one study examined the nurses’ knowledge of EBP terms (figure 3).43 Half of the nurses had at least some knowledge of 4 (40%) of the 10 terms presented. The dummy terms appeared more familiar than terms like ‘bias’, ‘power calculation’ and ‘number needed to treat’, suggesting some socially desired answering.
Awareness of common sources of evidence
Eight studies addressed this issue (table 1). About a quarter of the responding doctors used the Cochrane Library (median 25%), whereas 39% of them were unaware of this database. The journal Evidence-Based Medicine was used by 14%, but unknown in 34% of the doctors. Guidelines from the National Guideline Clearinghouse were used by 8% and unknown in 48%, the ACP Journal Club used by 3% but unknown in 68% and the TRIP database was used by 15% and unknown in 71%. Two studies showed this awareness was even less among nurses.24 ,43
EBP barriers and facilitators
Responses regarding the 29 barriers presented in Funk's questionnaire were usually dichotomised, that is, items scored as ‘barrier’ or ‘large barrier’ were counted as barriers. To give an overview of the barriers to EBP most frequently mentioned by doctors and nurses, we merged our data with the barriers found among nurses in the systematic review by Kajermo et al.15 These barriers are summarised in table 4. Worldwide, EBP barriers were strikingly convergent, except the language barrier for non-English speaking countries and the limited access to electronic databases in some countries.
The major facilitating initiatives as desired by doctors and nurses were mostly collected through open questions. These facilitators include continuing EBP-teaching efforts in pregraduate and postgraduate curricula, constant involvement by colleagues in daily practice, staff and management support to learn and apply EBP in daily clinical practice, structural promotion and facilitation of EBP activities by the management and experts, and clear and easily accessible sources of evidence, protocols and guidelines.
Recommendations reported to implement EBP
All studies gave recommendations to overcome or address the identified barriers (table 5). From macrolevel, middlelevel and microlevel perspectives, that is, at (inter)national, hospital and ward levels, various solutions were proposed, ranging from advocating EBP by national regulatory bodies to specific interventions at ward level, including availability of computers and internet.
A qualitative evaluation of the recommendations shows they mainly focused on education for both pregraduates and postgraduates. The following aspects were considered important: how and with whom to build EBP curricula, tiered education based on needs assessments, learning by interaction and transfer of the education from the classroom to the bedside.
Regarding preconditions to strategically implement EBP, authors put emphasis on the role of the management in terms of facilitating prerequisites as well as creating a positive culture towards EBP. They also suggested that solutions to the problems encountered when implementing EBP should start with an analysis of the organisation to identify problems at both local and organisational levels to tailor the interventions.
Our systematic review shows that, worldwide, many professionals in clinical healthcare welcome EBP, although the awareness of, education in and actual bedside application of EBP leaves room for improvement. Based on the reasons given for the limited uptake of EBP, a structural implementation of EBP in clinical healthcare organisations will require a culture change at various organisational levels, that is, patient care, education and management. The framework of policy recommendations, as presented here, encompasses the wide range of possible entries to implement in a multifocal manner and sustain EBP. Because recommendations were found for virtually all levels of management, a general policy seems indicated to address and govern these EBP implementation issues. Some recommendations might also be useful as indicators to monitor the usage of EBP in daily clinical practice. Furthermore, this review could stimulate the testing of some of our recommendations through appropriately designed studies.
Although the majority of healthcare professionals appear quite EBP-minded and the uptake of EBP is progressing,49 important barriers are still obstructing the full implementation of EBP in daily clinical practice. These findings occur consistently among the various medical specialists and nurses alike, and in many specific settings and specialties throughout the world. However, Brown et al found in a multiple regression analysis that perceived barriers to research use predicted only a fraction of practice, attitude and knowledge/skills associated with EBP.23 Apparently, the most frequently reported barriers are not necessarily the main reason for a poor implementation of EBP. Rather, a change in mind set seems indicated among the various healthcare professionals who perceive these barriers. Additional barriers to EBP implementation may lie at the organisational level.4 Hence, an integrative approach, involving all professionals and supported by initiatives from various organisational levels, may be a more fitting solution.
An integrative approach to overcome perceived barriers to EBP has also been suggested by other authors,50 who reasoned that the best implementation strategy should be a multifocal, comprehensive programme involving all professionals and should be tailored to their desires and perceived barriers. A systematic review of 235 studies on (multifaceted) guideline implementation strategies presented imperfect evidence to support decisions about which guideline dissemination and implementation strategies are likely to be efficient under different circumstances.51 Opinion leaders and role models appear to have a key function.52 A recent systematic review, comprising seven observational studies, described the relation between EBP implementation and leadership among nurses.53 The evidence suggested that initiatives on the level of leadership, organisation and culture are pivotal for the process of implementing EBP in nursing. However, available evidence for the effectiveness of organisational infrastructures in promoting evidence-based nursing is scarce.4 In the medical realm, such evidence is also limited.28 ,54–56
Other frameworks or multidimensional programmes have been proposed to improve research utillisation,13 or to stimulate the use of EBP by nurses,57 or on specific wards.58 Others have promoted a dedicated research agenda,59 integrated EBP education56 ,60 or the implementation of EBP in specific medical specialties.16 ,61 Clinically integrated rather than stand-alone EBP teaching initiatives have been shown to improve EBP behaviour and may therefore help implement EBP in clinical practice.62 These initiatives per se seem defective because none of these aspects can be omitted to arrive at a truly evidence-based healthcare: if EBP education falls short, managers do not facilitate EBP activities, doctors do not apply EBP in their daily practice or nurses are lagging behind in EBP knowledge, optimum evidence-based healthcare eventually will not (fully) reach the patients who deserve it. This has been one of the reasons why a European teaching project has started to incorporate evidence-based medicine in clinical practice.63
Although not all studies found were performed in teaching hospitals, the majority may have been performed in centres that already had the aim, or were in the process of implementing EBP. Many other centres are likely to be lagging further behind. However, higher response rates were not associated with more positive attitudes towards EBP. Given the settings and types of respondents in the studies included here, the inferences of our review appear primarily valid for clinical doctors and nurses from various specialties in centres that aim at implementing EBM.
Second, the questionnaires used were self-reported and response rates varied considerably. For both reasons, our results may overestimate enthusiasm, knowledge and uptake of EBP. On the other hand, the framework of implementation recommendations we derived from these studies may be useful for all centres striving at a better EBP implementation.
Third, in our review, we searched for surveys of EBP attitude, knowledge, awareness, barriers and facilitators rather than studies specifically focusing on testing alternatives to improve implementation of EBP. Such studies, however, are rare.4 ,28 ,53 The implementation factors these studies mentioned also became clear from our review, while the success of these implementation strategies is still unclear. One of the reasons for this is the absence of a valid means of assessing actual EBP behaviour during daily practice.62 ,64–66
Finally, we realise EBP is an essential but not the sole factor to improve quality of care. Even if clinicians are aware of available evidence, the right thing to do does not always happen. Continuous quality improvement strategies also involve active implementation of available evidence and existing guidelines. Nevertheless, a critical evidence-based attitude towards current practice remains the first step towards quality improvement.
Our review of all available surveys on the barriers for, and promotion of, EBP activities as perceived by clinical doctors and nurses suggests that EBP implementation needs a multilevel approach, involving interventions in the policy-making, managerial, educational and practical areas. We offer a summary of the suggested interventions at these different levels. These may be used not only to implement, but also to monitor the usage of EBP in daily clinical practice. This requires a joint effort and cultural change within the whole healthcare organisation, but is likely to result in a better quality of care.
Contributors DTU and HV conceived the idea of the study and were responsible for its design. They also performed data acquisition and analysis. DU drafted the paper, HV revised it critically. GHG contributed substantially to the draft of the article, provided input to the data analysis and the interpretation of the results, and revised the manuscript critically. All authors gave approval for the final version.
Funding This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement There are no additional data available .
This is an open-access article distributed under the terms of the Creative Commons Attribution Non-commercial License, which permits use, distribution, and reproduction in any medium, provided the original work is properly cited, the use is non commercial and is otherwise in compliance with the license. See: http://creativecommons.org/licenses/by-nc/3.0/ and http://creativecommons.org/licenses/by-nc/3.0/legalcode