Factors Associated With Nurses’ Knowledge of and Perceived Value in Evidence-Based Practices

We developed a cross-sectional, electronic survey and administered it to intensive care nurses in 12 adult ICUs within 6 hospitals in a single, integrated health care system. This survey was associated with a larger study examining ICU team function and the delivery of evidence-based practices. All nurses employed in these ICUs were eligible to participate. We obtained a list of the names and email addresses for these individuals from the health care system’s Office for Quality Improvement and Patient Safety.

We developed the survey instrument in the spring of 2017. Various theories guided us in this process: theories about the psychosocial underpinnings of organizational performance; the theory of planned behavior, which posits a central role for individual attitudes, subjective norms, and perceived behavioral control in the intention to act; and the Consolidated Framework for Implementation Research, which posits that the characteristics of individual clinicians interact with local contexts to influence the implementation of best practices.^17–19  In addition, we applied both existing knowledge of evidence-based practices that improve outcomes in patients receiving mechanical ventilation and data from prior studies of nurse characteristics that influence patient outcomes.^3,20–22  We designed the survey to measure 4 domains: nurse demographics, perceived knowledge of evidence-based practices used in the ICU, perceptions about the value in evidence-based practices in the ICU, and individual psychosocial beliefs that are potentially associated with perceptions of evidence-based practices.^23–26 

Demographic questions evaluated education level (associate’s, bachelor’s, master’s, or doctoral degree, dichotomized as BSN or higher vs not), specialty certification status (CCRN certification or not), age, and race. Questions pertaining to evidence-based practices encompassed perceived knowledge of and perceived value in 3 evidence-based practices used when caring for patients receiving mechanical ventilation: daily interruption of sedation, spontaneous breathing trials, and ventilation at a low tidal volume (a lung-protective ventilation strategy).^3,5,21  Individual psychosocial beliefs included 3 composite measures: professional identity, defined as a feeling of self-worth derived from one’s profession; self-efficacy, defined as the belief in one’s ability to succeed in a task; and role clarity, defined as the notion that one knows what one needs to do and what is expected.^3,5,20–26  Given that, on the basis of findings from prior studies, nurses who have strong professional identity and feel they have self-efficacy and role clarity may be more capable of providing evidence-based care, we hypothesized that these psychosocial beliefs would mediate the relations between nurse characteristics and perceived knowledge of and perceived value in evidence-based practices.^26,27 

The survey contained 44 items, with a combination of items answered by using a 5-point Likert-type scale (ranging from 1 [strongly disagree] to 5 [strongly agree]) and demographic questions. It took approximately 10 minutes to complete. We pilot tested the survey in a single ICU to determine readability and content validity; we implemented changes on the basis of participant feedback. The final survey is available in Supplement 1; psychometric testing results are available in Supplement 2. We assessed composite item reliability using Pearson correlation for self-efficacy (a 2-item construct) and the Cronbach α for professional identity and role clarity (constructs with more than 2 items); we found acceptable item reliability and within-item correlation.

We administered the survey via Qualtrics, an online survey platform hosted by the University of Pittsburgh, from June through September 2017. All eligible participants received an email explaining the study goals and providing a link to the survey. We sent up to 4 email reminders to nonrespondents in order to increase participation. Each nurse could complete the survey only once. Participants who completed the survey received a $5 gift card as an incentive. Institutional review board approval was obtained from the University of Pittsburgh Human Subjects Protection Office before we began collecting data.

We analyzed nurses’ demographic characteristics by using descriptive statistics. All other items were measured on a 5-point Likert-type scale. Data are reported as the mean (SD) and 95% confidence interval. We used the Student t test to examine the bivariate relationships between nurses’ education level or certification status and the 3 individual psychosocial beliefs. We used sequential multiple regression to examine independent relationships among the independent variables (nurses’ education level or certification status) and the dependent variables (perceived knowledge of and perceived value in specific evidence-based practices). The unit of analysis was the individual nurse. We used Huber-White confidence intervals to adjust the standard errors for clustering by ICU. We first fit separate models in which perceived knowledge of or perceived value in an evidence-based practice was the dependent variable, education level or certification was the primary independent variable, and age (a proxy for years of experience) was a covariate. Next, we fit the same models but added the 3 individual psychosocial beliefs as covariates. This approach allowed us to estimate the independent associations of nurse education or certification with our outcomes of interest while also testing whether these associations were mediated by individual psychosocial beliefs about nurses’ place on the ICU team.

To enhance the interpretability of the models’ results, we fit all models using logistic regression. Given the response distributions, we dichotomized the measures of nurses’ perceived knowledge of or perceived value in specific evidence-based practices into “high level of knowledge” or “highly valued practice” (a score ≥ 4 on a 5-point scale) or “less high level of knowledge” or “less highly valued practice” (a score < 4 on a 5-point scale). Exponentiated coefficients in the models could then be interpreted as odds ratios for high perceived knowledge of and high perceived value in the practices. The threshold for statistical signifi-cance was set at α = 0.05. We performed all statistical analyses in Stata software version 15.0 (StataCorp).

A total of 268 nurses working in adult ICUs responded to the survey (Table 1). Respondents were primarily white and female, and most worked in a teaching hospital. Although the majority of the nurses had a BSN or higher degree, only about a quarter of respondents had CCRN specialty certification. Among nurses with a BSN or higher degree, 34.4% (n = 62) also had specialty certification, whereas only 12.5% (n = 9) of those without their BSN had CCCN certification (P = .001).

Respondents reported a high level of perceived knowledge of the 3 evidence-based practices used in caring for patients receiving mechanical ventilation (Table 2). We found no significant differences in nurses’ perceived knowledge of or perceived value in the 3 evidence-based practices related to nurses’ education level. Nurses with specialty certification reported significantly more perceived knowledge of spontaneous breathing trials and lung-protective ventilation and reported greater perceived value in all 3 evidence-based practices than did nurses without CCRN certification.

Scores of the 3 individual psychosocial beliefs thought to be associated with the selected evidence-based practices are presented in Table 3. Overall, nurses reported strong professional identity, self-efficacy, and role clarity, all with mean scores greater than 4.0; for professional identity and self-efficacy, scores ranged from 2.5 to 5.0, and for role clarity, scores ranged from 1.7 to 5.0. We found no significant differences in reports of the individual psychosocial beliefs based on nurses’ level of education. With regard to specialty certification, however, we did find significant differences, with higher levels of self-efficacy (P = .001) and role clarity (P = .05) among nurses who reported having specialty certification (Table 3).

Analysis of unadjusted effects indicated no significant associations between nurses’ education level or certification status and perceived knowledge of evidence-based practices (Table 4). In our full model, we did not find significant associations between education level and perceived knowledge of the evidence-based practices or between certification status and perceived knowledge of the practices (Table 4). The 3 individual psychosocial beliefs were inconsistently associated with nurses’ perceived knowledge of or perceived value in evidence-based practices. In the full model, among the 3 individual psychosocial beliefs, professional identity was most often associated with nurses’ perceived knowledge of evidence-based practices (Table 4).

Analysis of unadjusted effects indicated no significant associations between nurses’ education level and perceived value in evidence-based practices. We did, however, find that certification status was associated with the perceived value in 2 of the 3 evidence-based practices for patients receiving mechanical ventilation: daily interruption of sedation and lung-protective ventilation (Table 4). These associations persisted in the full model (Table 4). In addition, a nurses’ professional identity was significantly associated with perceived value in spontaneous breathing trials among nurses with their CCRN, but overall, we found no significant differences in the perceived value in this practice between nurses with and nurses without CCRN certification (Table 4).

In a cross-sectional study of ICU nurses, we examined the associations between nurses’ education level and certification status, and their perceived knowledge of and perceived value in specific evidence-based practices used to care for patients receiving mechanical ventilation. We explored whether 3 individual psychosocial beliefs (professional identity, self-efficacy, and role clarity) were important mediators of the relationships between nurse education or certification and perceived knowledge of and value in evidence-based practices. We found that ICU nurses’ specialty certification status was strongly associated with perceived knowledge of and perceived value in the use of these evidence-based practices, whereas nurses’ educational level was not. The nurses’ place on the ICU team, assessed via 3 psychosocial beliefs, had an inconsistent mediating effect on nurses’ reported perceived knowledge of and perceived value in evidence-based practices.

These findings have important implications for nurses when delivering care in the ICU and adopting evidence-based practices in caring for patients receiving mechanical ventilation. Adoption of evidence-based practice is a strategy proven to improve outcomes.^5,16,28  Many evidence-based practices remain underused, partly because of gaps between providers’ attitudes toward practices and the delivery of care at the bedside.²  The strong association we found between specialty certification and nurses’ perceived knowledge of and value in specific evidence-based practices is consistent with current knowledge regarding the value of specialty certification^8,9,15  and supports the value of nurses with specialty certification, especially among institutions that aim to increase the adoption of evidence-based practices.

Contrary to our hypothesis, we did not find a significant relationship between nurses’ education level and perceived knowledge of or perceived value in evidence-based practices. Nurses in this sample were highly educated, which could be a factor contributing to the null findings. In addition, characteristics of individual psychosocial beliefs, based on 3 measures, had only a small association with perceived knowledge of and perceived value in these practices. Prior studies have suggested that these factors are associated with improved outcomes for patients.^6,20,22,29  However, given the limited associations with perceived knowledge of and perceived value in evidence-based care among ICU nurses found in this study, future studies are needed in order to elucidate better the clinical significance of these findings and to determine the mechanisms that result in improved outcomes for patients. Future studies of processes by which to increase the value nurses give to evidence-based practices may speed the implementation of these practices by institutions.

As an important next step, researchers should examine how ICU nurses’ knowledge of and value in evidence-based practices translate to the delivery of evidence-based care—not just their perceptions of those practices, as measured here. Prior studies demonstrate that knowledge and perceived value alone are not enough reason to adopt evidence-based practices, and thus future studies should evaluate similarities and differences among nurses’ practices.^2,16  Efforts to identify barriers to the use of evidence-based care in practice will allow for targeted interventions to address such barriers, to increase the adoption of condition-specific evidence-based practices such as those used with patients who require mechanical ventilation, and to improve the quality of care delivered in the ICU.

This study has several limitations. First, it was conducted within a single health care system, and thus the results may not generalize to ICU nurses working at other institutions or in other regions of the country. Our measure of self-efficacy asked about a nurse’s general ability to perform tasks and may not be sensitive enough to detect true relationships between nurses’ characteristics and their ability to implement specific evidence-based practices. These data are self-reported, and the respondents’ perceived knowledge of and perceived value in evidence-based practices may not reflect the attitudes of all nurses working in ICUs, especially given the high number of nurses with a BSN degree in this sample. We did not gather information regarding nurses’ years of experience or about certifications other than CCRN. These factors could be associated with nurses’ individual psychosocial beliefs and could relate to their perceptions of evidence-based practices. In addition, because we administered this survey through the institution that employed the nurses, a risk of acquiescence existed; in an effort to reduce this effect, however, we administered the survey through a web-based platform, and responses were anonymous. Finally, the responses from nurses with different education levels and certification statuses showed minimal variation, which limits our ability to assess relationships among outcomes.

This study increases our knowledge of the role of education, certification, and professional identity among ICU nurses in their implementation of evidence-based practices for patients requiring mechanical ventilation. Specialty certification was associated with greater professional identity and higher perceptions of knowledge of and value in evidence-based practices used in the ICU, whereas education level was not. Supporting specialty certification among nurses is a plausible way to assist with the adoption of evidence-based practices as a means to improve ICU quality and should be evaluated further.

This work was performed at the University of Pittsburgh Medical Center and the University of Pittsburgh.