Introduction

During the last two decades the paradigm of evidence-based medicine (EBM) has been introduced and then fostered by societal and patients’ demand for professional and resource accountability in health care [1]. Nowadays, executive boards stress the use of EBM to prevent unsafe practices that lack empirical support to reduce unacceptable individual variance and ultimately to increase efficiency and health care quality [2].

Reality, however, proves that not all health care professionals use EBM in their daily practice. In general medicine, roughly half of all medical treatments are evidence-based and only about one-fourth of all surgical treatments were found to be based on evidence [3, 4]. This is related to a lack of surgical evidence, the fact that surgical questions cannot always be answered by a randomized clinical trial, and restrictive inclusion criteria that hamper application of trial recommendations to the average surgical patient [5]. Clinicians also report a lack of knowledge and basic skills—particularly about searching the literature and critical appraisal [6]—as major barriers to practicing EBM as well as a lack of personal time [7].

It often requires a joint venture between surgeons and surgical nurses to make evidence-based surgery (EBS) work. For example, evidence is accumulating that fast-track surgery is useful for enhancimg patient recovery [8]. This result can only be achieved when surgeons and nurses collaborate. However, nurses also were found to face difficulties with EBS in that they believe research reports are too “academic” and do not offer the desired level of clinical direction [9].

Improvements in evidence-based behavior can only be realized, and measured, if awareness of EBS and a positive attitude toward EBS are secured first [10]. Moreover, it has been suggested that strategies to promote change in clinical practice are more likely to be successful if based on an analysis of barriers and facilitators specific to the context [11]. To our knowledge, awareness of EBS, attitude, and barriers to the utilization of research findings have not yet been investigated simultaneously among physicians and nurses in the surgical realm.

Therefore, the aim of this study was to determine the attitudes and awareness among surgeons and surgical nurses as to the principles of EBS and the barriers experienced when practicing EBS to be able to define future tailor-made interventions to improve evidence-based behavior in surgery.

Methods

Context

This study was carried out at the Department of Surgery of the Academic Medical Center (AMC), a university teaching hospital in Amsterdam, The Netherlands. It is part of a larger, ongoing inventory regarding attitudes, awareness, and perceived barriers among doctors and nurses of several specialties within the hospital. At the Department of Surgery, learning EBM is encouraged through yearly EBS courses and by producing critically appraised topics (CATs) based on actual routine clinical questions [12]. Although the surgical nurses are not yet involved in these local EBS initiatives, they are offered some courses to learn EBS skills.

To assess the extent to which the EBS principles are known and applied, a questionnaire and a quiz were developed. All surgeons, surgical trainees, and qualified nurses on all surgical wards were invited to complete the questionnaire.

Questionnaire

The questionnaire combined the BARRIERS scale and the McColl questionnaire [7, 13]. The BARRIERS scale investigates health care professionals’ perceptions of barriers to the utilization of research findings in clinical practice. The scale has been validated and used in various settings worldwide [14–16]. The McColl et al. questionnaire addresses attitude, awareness, and actual use of EBM. It has also been applied ubiquitously [7, 17, 18]. In this study, “attitude” was defined as the mind-set of the responders as to the principles of EBM; “awareness” is defined as familiarity with the meaning of certain EBM terms.

Other validated instruments, such as the Berlin questionnaire and the Fresno test, objectively measure EBM skills [19, 20]. These measures were considered inappropriate in this study as their questionnaires aim to evaluate the effectiveness of an EBS course, whereas in our situation only optional EBS courses had been available.

Because the BARRIERS scale and the McColl et al. questionnaire are in English, they were translated into Dutch by the process of forward-backward translation [21]. Subsequently, a reliable Cronbach’s α value of 0.95 was recalculated for the part containing the BARRIERS scale.

EBS quiz

Available CATs were used to measure if surgeons and trainees were aware of the surgical evidence communicated and how much of the evidence was in agreement with present practice. Two staff surgeons (D.A.L. and the head of the department) were consulted to select 20 CATs, representative for the four surgical subspecialties (general, gastroenterologic, trauma, and vascular surgery) and reflecting knowledge any member of the Department of Surgery should have. The EBS quiz consisted of multiple-choice questions formulated for each of the 20 topics. An example is shown in Table 1, considering evidence that polyethylene glycol is not indicated to prepare the bowel before colon surgery [22] The other questions and corresponding anwers can be found in the Appendix).

Table 1 Example of a question in the quiz
Full size table

During one of the surgical research meetings, the EBM quiz was conducted by showing each of the 20 questions for 30 seconds to allow the attendees to fill in their answer on a reply form. Correct answers were defined by the conclusion of the CAT and the actual policy of the surgical department based on the local guidelines and protocols available.

Data analysis

Means and standard deviations (SD) were calculated after checking for normal distribution. Student’s t-test was used to compare the means of the attitude scores toward EBS between certain subgroups (surgeons and nurses, males and females, staff and nonstaff personnel). Correlation coefficients (according to Pearson’s or Spearman’s, as determined by the presence or absence of a normal distribution, respectively) and scatter graphs were made to find possible correlations between age or working experience in health care and attitudes toward EBS.

Nonparametric Kruskal–Wallis and Mann–Whitney tests were used for subgroup analysis. Statistical analysis was performed using SPSS 15.0 (SPSS, Chicago, IL, USA).

Results

We achieved response rates of 67% (29/43) for the surgeons and trainees and 60% (73/122) for the nurses. Personal characteristics of the respondents were representative for the doctors and nurses working on the surgical wards (Table 2).

Table 2 Demographic data for respondents to the questionnaire
Full size table

McColl questionnaire

Current attitudes of doctors and nurses toward EBS are shown in Table 3. Doctors in particular were extremely positive. They never disagreed with the statement: “practicing EBS improves patient care” (mean score 78, SD 14). Actual surgical practice was considered to be 45% (SD 16%) evidence-based. Nurses’ attitude toward the current promotion of EBS was welcoming, although there was wide variation in response (mean score 52, SD 21). Senior nurses appeared to have a significantly more positive attitude toward the principle of EBS than registered nurses (P = 0.01) and considered research findings in daily practice significantly more useful (P = 0.005). When comparing surgeons to nurses, surgeons were found to have a more positive attitude toward EBS (P = 0.00), judged their colleagues to be more positive toward EBS (P = 0.00), and thought research findings were more useful in daily practice. No correlations between age, sex, or work experience in health care and attitude toward EBS were found.

Table 3 Current attitudes toward experience-based medicine
Full size table

Figure 1 shows the extent to which surgeons said they understand the most cited terms in articles about EBM that are likely to be relevant for surgical practice. Approximately 90% of the surgeons showed (some) understanding of six of eight terms. Particularly “absolute risk increase” and “confidence interval” were known by all surgeons. We added “fixed event rate” and “random benefit ratio”—nonexisting dummy terms—as indicators of possible socially desirable answering [23]. Overall, the dummy terms scored a lower understanding than the other terms, as can be seen in Fig. 1 (approximately 40% compared to 90% of the existing terms). Hence, overestimation of competence was less likely in our study, which makes the scores about reported knowledge more reliable.

Fig. 1
figure 1

Surgeons’ and nurses’ knowledge of evidence-based surgery (EBS) terms. The last two terms are meaningless “dummy terms.” (1) I understand the meaning and could explain to others. (2) I partially understand the meaning. (3) I do not understand the meaning but would like to. (4) I do not understand the meaning; it would be of no use to understand. (5) Missing value. * N.N.H. number needed to harm; ** N.N.T. number needed to treat

Full size image

Ten common terms used in articles about EBS in the nurses’ surgical practice were presented to the nurses. They were less aware of EBS terms than surgeons (about 40%). “Absolute treatment increase” and “dosage chance” were inserted as dummy terms. Their understanding of these terms was similar to that of the existing terms (approximately 40%), suggesting socially desirable answering to this part of the questionnaire.

BARRIERS scale

A five-point scale was used to rate the extent to which 29 statements were considered a barrier to using research in practice. By dichotomizing the answers, a list of major barriers for surgeons and nurses was produced (Table 4).

Table 4 Barriers to research utilization
Full size table

To surgeons, major barriers were the conflicting results (79%, 23/29) and methodologic inadequacies (73%, 21/29) of research reports, whereas nurses mentioned the unawareness of research reports in the first place (67%, 49/73). A lack of time (to read research reports and to implement new ideas) and unclear implications for practice were rated as major barriers by both surgeons and nurses.

To facilitate the utilization of data from medical research, both surgeons and nurses stressed constant involvement and reminding each other to practice EBS. Nurses claimed to need time to learn the basic skills of EBS. Surgeons thought articles should be selected and discussed based on their relevance to surgical practice. Subsequently, both groups asked for clear, practical documents (e.g., a protocol or guideline) that would be easily obtainable through a periodical newsletter or an online database.

EBS quiz

The quiz was completed by 86% (36/42) of the attendees at the meeting (Table 5). Their mean age was 33 years (SD 9.4 years).

Table 5 Respondents’ answers to the quiz
Full size table

Surgeons (P = 0.001) and surgical trainees (P < 0.001) showed significantly more awareness of evidence than interns. This was expected because the latter group had just recently joined the department and had received no previous EBM training.

Of the 20 questions, 8 referred to convincing evidence that was actually applied in surgical practice according to local guidelines and protocols. Surgeons (n = 11) answered affirmatively to these questions correctly in 50% (44/88) and trainees (n = 12) in 41% (39/96), respectively. Thus, surgeons and trainees used nearly half of the convincing evidence as presented in the quiz, despite moderate to good awareness of the evidence itself.

Discussion

In this study, we have shown that surgeons at a university hospital in The Netherlands appear to have a positive attitude toward the paradigm of EBS and are familiar with EBS terminology but that conflicting results and methodologic shortcomings of research reports appear to be major barriers to practicing EBS. These barriers have also been reported in nonsurgical settings [24–26]. To implement EBM successfully in daily practice, further teaching of the EBM principles to nurses and stimulating and reminding surgical colleagues to apply these principles is required.

Although not specified by the surgeons stating this barrier, the “major methodologic shortcomings” generally are well known flaws in the design or conduct of the studies that jeopardize study validity and thus the strength of the evidence. Among the shortcomings are nonrandomized trials (most surgical studies are case series), no allocation concealment or blinding used (patients and surgeons are usually aware of the treatment given), insufficient follow-up, no intention-to-treat principle applied, and small sample size [27].

Surprisingly, surgeons and trainees indicated that they use only about half of the convincing evidence as presented during the quiz, even though it was decided earlier in the group of surgeons to include this evidence in local guidelines or to change local surgical policy. Apparently, the decision whether to apply evidence is often difficult (e.g., because of other available evidence showing a favorable outcome on one endpoint but a harmful effect on another relevant clinical endpoint).

Nurses had a moderately positive attitude toward the principles of EBS but were rather unfamiliar with EBS sources and terminology. They indicated that unawareness of research reports was the major barrier to using research findings in surgical nursing practice.

To facilitate the practice of EBS, both surgeons and nurses suggested constant involvement of EBS in daily practice, interactive education, and the availability of a digital expert system to support practice, which is also described in the literature [11, 28].

A limitation of this study design is that self-perception of attitude, awareness, and barriers toward EBS were assessed. One would rather determine these parameters during “real-time” surgical practice and ultimately measure if patient outcomes improve owing to the application of surgical evidence by surgeons and nurses. Although the number of instruments to evaluate EBS is growing [29], valid assessment instruments to evaluate EBS behavior are still lacking.

Second, the absolute number of participants in this study was limited and derived from one university hospital only. However, the response rate of surgeons and nurses within our department was satisfactory. Our findings may be indicative of the situation in other (academic) centers because similar implementation barriers and positive attitudes regarding EBM have been reported [7, 14, 26]. Despite these similarities, further research is needed in other specialties and other medical centers to see if general recommendations can be made on interventions to promote EBM.

Implementation of EBS needs further improvement. Several tailor-made interventions have become clear from our findings and experience in creating a professional environment that incorporates EBS. First, continual confrontation with available evidence through frequent critical appraisal meetings or grand rounds is necessary [30]. Second, as our nurses were unfamiliar with EBS and received less postgraduate EBS training than surgeons, it is expected that they will benefit from EBS training focusing on basic skills, integrated in everyday surgical nursing practice [31, 32]. Third, surgeons may enhance their efficiency by using more aggregate sources of evidence [33], such as the National Guideline Clearinghouse or BMJ Clinical Evidence. Finally, collaboration is advocated among doctors and nurses with the same zest for EBS.