Evidence-based practice (EBP) has evolved over the past few decades as an integral part of medical education and clinical practice. Multiple large-scale, double-blind, randomized, placebo-controlled studies have emerged in the last 10 years in child and adolescent psychiatry. In medicine, meta-analysis of these randomized, controlled studies has been increasingly more accepted as a means of summarizing the literature and guiding treatment (¹). With evolving methodology and research steering clinical practice, understanding how to gather, assess, and apply the literature to patient care is important for health care providers.

Evidence-based practice has been defined as the more general use of research to guide assessment and treatment of patients (²). Once evidence is drawn from research, clinical expertise and patient preference are combined to guide clinical decision making (³). To effectively draw evidence from the research, one must be able to define a clinical problem, create a question and search terms related to that question, and critically evaluate the literature (¹).

Studies have demonstrated resistance to using and learning about EBP. Some barriers include personal factors such as attitudes, values, and skills to analyze the evidence, as well as external factors such as lack of convenient access to resources and time to do so while tending to the demands of patient care (²). Many clinicians fear that using EBP will result in loss of autonomy for medical decision making, resulting in "cookbook" medicine (²). One study of medical school faculty and trainees found that the majority supported use of evidence-based medicine techniques but lacked adequate training or competence in their use (⁴). A recent survey of mental health practitioners revealed that significant predictors of EBP use were practitioner training, perceived openness toward EBP in the respective clinical setting, and practitioners’ attitudes toward treatment research (⁵). Another recent study of behavioral professionals identified the most frequently cited barriers to learning about EBP as negative attitudes about EBP and lack of training (⁶). There have been no studies to date about the feasibility of teaching EBP to an inpatient psychiatry staff.

This study was designed to determine if introducing EBP to inpatient child and adolescent psychiatry clinical staff would result in a gain in knowledge about EBP as well as a positive attitude toward EBP.

The population surveyed consisted of inpatient clinical staff, including nurses, social workers, teaching staff, and recreational and milieu therapists. The sample size and attendance per session ranged from seven to 10; however, data were not collected on the individual consistency of staff attendance.

The three-part lecture series was based on the Duke University Child and Adolescent Psychiatry Model (⁷). The first session, "Introduction to Evidence Based Practice (EBP)," focused mainly on the history of EBP, definitions, and understanding the hierarchy of literature. Session 2, "Applying the Evidence," focused on how to create an answerable clinical question using PECO (a pneumonic for framing foreground questions: What is the Population? What is the Exposure? What is the Control or Comparison condition? What is the desired Outcome?) and what resources one would use to gather relevant information (⁷). Session 3 was a presentation of case vignettes and a discussion of utilizing EBP to approach clinical questions. Session 3 was not evaluated for knowledge gain, because it was meant to demonstrate EBP principles. Institutional review board approval was obtained for this study.

Staff members were given anonymous pre- and postcourse surveys to assess prior education regarding EBP, attitudes, and quality of the course. Twelve statements regarding attitudes were created from the Evidence-Based Practice Attitude Scale (EBPAS) (⁸). Reliability of the EBPAS scale has been reported as good (Cronbach alpha =0.77) (⁸). These 12 statements were presented on both the pre- and postcourse surveys to determine if a change in attitudes would occur. Participants also reported the quality of their prior EBP education on a 5-point scale (1=poor, 5=very good). Data were collected, averaged, and tabulated from these responses.

On the postcourse survey, participants were asked to respond again to the 12 attitude statements and rate the lectures, teaching, and likelihood of using EBP in the future. The course and lectures were rated on a 5-point scale, (1=poor, 5=very good; 1=not useful, 5=very useful, respectively). Participants commented on the amount of material and the clarity of presentation by multiple-choice responses. Collected scaled ratings were averaged, and statement responses were tabulated.

The 12 attitude statements that appeared on both the pre- and postcourse surveys were rated on a 5-point scale (1=never agree, 3=sometimes agree, 5=always agree). Data sets were analyzed using the student’s dependent t test, with variable sample size. Ten participants completed the precourse survey, and seven completed the postcourse survey. Means, standard deviations, confidence intervals, and t test scores were obtained for each sample. The probability of the result was then calculated, assuming the null hypothesis.

A test to determine knowledge gained was given before and after Sessions 1 and 2. Each pre- and post-lecture test was the same for the respective lecture and contained four questions on the lecture key points and content.

Participants reported only one experience before college, five experiences during college, three experiences during graduate school, and two experiences with previous lectures at Children’s Memorial Hospital, Chicago. Three participants reported that they had never learned about EBP, and there was one report of "other." Participants’ average rating of their previous experiences was 2.81 (1=poor, 5=very good).

The test means for participants’ likelihood to use EBP in the future were 3.3 precourse and 4.67 postcourse. T score comparison of these means was −2.93, with p=0.01. The difference in the participant’s willingness to use EBP before and after the course was significant. These responses were compared, and the results are shown in Table 1 and Table 2.

For Session 1, participants scored 85.7% on the precourse test and 87.5% on the postcourse test. The Session 2 precourse test mean was 60.7%, and the postcourse test mean was 96.4%.

Participants’ average responses on the course itself were as follows (1=poor, 5=very good): course materials (handouts), 5; lecture content, 4.86; teaching, 4.86; and location, 4.71. Participant’s average responses for the usefulness of lectures were as follows (1=not useful; 5=very useful): Session 1, 4.57; Session 2, 4.83; and Session 3, 4.71. The participants unanimously (i.e., 7/7 participants) responded that the amount of material was "just right" and "clear and concise."

The hypothesis of this study was that the staff would gain a positive attitude regarding EBP and gain knowledge from this intervention. Staff performed better on the postcourse tests for knowledge gained, with a greater percentage gain from Session 2. After the course, the staff went from a neutral likelihood to a more significant likelihood of using EBP. Attitude statements reflected mild change with regard to being more likely to use well-researched treatments for patients and believing that research-based treatments are clinically useful.

Most participants had previous experience learning EBP in college, and those participants reported that the quality of these experiences was relatively poor. This could be a factor in explaining the more significant knowledge gain from Session 2.

Regarding attitude change, the staff presented initially with relatively positive attitudes toward EBP. Significant, but slight, changes were demonstrated from more neutral to mildly positive attitudes on two statements on using well-researched treatments and believing that research was relevant clinically. One of the main biases in this study is the lecturer’s familiarity with the clinical staff and inpatient milieu. The lecturer knew the patient population most commonly seen by the staff, as well as common clinical conundrums. Given this familiarity, the lecturer requested ideas from staff, including nurses and milieu therapists, on what information may be more useful. This may have impacted the way staff reported their feelings. After the course, staff members reported that they would be more likely to use EBP in their clinical work, that they liked the lectures, and that they believed the lectures were very clear and concise. Again, familiarity with the lecturer may be an important bias in this reporting.

Limitations in this study include the small sample size. Selecting an inpatient staff as participants is a bias as well. The inpatient staff members regularly see children with severe mental illness where medications and intensive behavioral treatments are given in a teaching hospital with emphasis and positive regard for EBP.

The gains in EBP knowledge by staff and their persisting positive attitudes were important outcomes in this study. Methods to support maintaining this outcome over time include having a "refresher" lecture and providing handouts of the lecture material to all staff. The "refresher" lecture has yet to be scheduled, but it would be interesting to see if the attitudes and knowledge would change at that time.

The overall contribution of this study is the finding that teaching EBP to inpatient child psychiatry staff is feasible, because there have been no previous studies in this area. More studies are needed to determine if teaching EBP would be feasible in other treatment settings (e.g., community hospitals, outpatient facilities) with outside lecturers. As the child and adolescent psychiatric literature continues to grow, knowing how to understand and utilize EBP will be of paramount importance for all mental health providers.

Special thanks to Dr. Mina Dulcan for her support and guidance with this project as well as Dr. Thomas Cummins and Dr. Mary Lou Gutierrez. At the time of submission, Dr. Kurth declared no competing interests.