Evidence-based practice (EBP) promotes safe, effective, and appropriate patient care. Supported by evidence-based guidelines, EBP includes compilations of the best available evidence on a given topic, readily accessible from the National Guideline Clearinghouse Web site (http://www.guideline.gov). Despite availability, the inconsistent use of guidelines means that 30% to 40% of patients do not receive required care.1 Evidence-based pain management (EBPM) guidelines, one of the first compiled and published by the Agency for Healthcare Research and Quality, chronicled evidence available for well over 30 years.2 Yet, despite strong evidence ratings, only 27% of hospital chief executive officers reported using EBPM guidelines in their facilities.3 As a result of inconsistent guideline application, patients suffer from postoperative surgical pain. Only until behaviors change, as recommended by the guidelines, will the impact of EBPM to improve patient outcomes be recognized.
Effective EBPM guideline implementation relies on administrative enforcement and individual practitioner adoption. The evolving nature of pain management requires organizations to continually integrate new programs, new equipment, and new standards into policies and procedures. Programs to successfully incorporate EBPM into practice are multifaceted, aimed at both the organizational and individual level.4 Implementing EBPM interventions requires an assessment of environmental readiness with an accurate portrayal of care providers. Successful interventions occur when providers change their behavior and assimilate changes into practice. The wide gap between pain management evidence and practice in the acute care setting continues to challenge administrators and clinical specialists accountable for guideline implementation.
The Theory of Planned Behavior (TPB)5 integrates individual and normative aspects of predicting behavior and offers a framework to analyze factors involved when implementing EBPM. The TPB explains that human behavior is intentional. The intention is a direct result of personal factors such clinical expertise, normative factors such the expectations of the practice environment, and control factors, such as the degree of control to act independently.5 The TPB provides insight and may assist clinical nurse specialists (CNSs) in targeting programs to implement pain management evidence in the acute care setting. The purpose of this study was to examine the contribution of individual nurses' perceptions of their practice environments and their clinical expertise in the use of EBPM.
BACKGROUND
Evidence-based pain management, one of the first Agency for Healthcare Research and Quality guidelines introduced to the practice community, occurred in 1992.6 The guidelines are a compilation of the best available evidence, with some references dating back to the early 1970s. Lauded as a helpful resource, implementation of these guidelines was voluntary. The American Pain Society offered a set of guidelines in 1995 specifically geared toward organizations. The American Pain Society cited the need to "move beyond traditional education and advocacy to focus on increasing pain's visibility in the clinical environment."7(p1874) In 1999, the Joint Commission approved standards for acute care pain management to be implemented in 2001.8
The Joint Commission's standards set forth patients' rights to pain relief and mandated the implementation of organizational systems for patient and staff education, quality monitoring, and consistent pain assessment and documentation processes.9 The Joint Commission mandate definitively shifted the paradigm of EBPM from individual to organizational accountability. The Joint Commission standards resulted in quality improvement monitoring that demonstrated a critical lack of individual adherence.10 Complicating the implementation of EBPM was the complexity of poor pain management practices. Berry and Dahl8 highlighted poor documentation, poor practitioner knowledge, and lack of organizational priorities as contributing factors.
Perception of the Practice Environment
Organizations vary in the implementation of EBP, with some organizations more flexible and open to change than others. According to Estabrooks and colleagues,11 organizational qualities amenable to EBP include a responsive administration, presence of collaborative relationships, and a large hospital size. However, without direct financial incentives, EBPM succumbs to other administrative priorities and becomes problematic only when patient pain satisfaction declines or fails to meet expected standards of care. Lack of physician cooperation has long been identified as a barrier to EBPM guideline implementation.12 Lack of time and resources13 are also considered barriers. The process of EBPM involves complex decision making, adequate nurse-patient communication, planning, and evaluation in the context of a busy, hectic environment.
Given the organizational barriers to EBPM implementation, it seems plausible that Magnet environments with supportive leadership, staffing adequacy, positive collaborative relationships, nurse control over practice, and accountability would embody the organizational characteristics necessary for EBPM implementation. Recent studies on Magnet facilities have provided an extensive body of research connecting the environment of care to positive patient outcomes such as morbidity, mortality, failure to rescue, and patient satisfaction.14 Studies have also suggested a connection between unit culture and pain management practice,15,1610 yet it is unknown whether the Magnet environment specifically supports pain management. The relative importance of organizational characteristics in contributing to EBPM has not been studied.
Clinical Expertise
Individual adoption of EBP guidelines requires the ability to integrate new paradigms of thought. Practicing EBPM includes acknowledging the subjectivity of a pain experience, anticipating and preventing pain by titrating complex medication regimens, and using reassessment to maximize the pain management plan. The concept of expertise as a reflective understanding molded through practice17 seems essential for EBPM. Expertise is the ability to make complex decisions in ambiguous or uncertain situations and relies heavily on the skill of involvement or the ability to connect with patients and colleagues.17 Yet, gaps in pain management expertise are well documented. Erroneous individual myths and beliefs about pain management and a critical lack of knowledge perpetuate inaccuracies and bias in pain management practice.13 Inadequate communication among healthcare providers and patients yields discrepancies between patient and nurse.18
Early research showed that adopters of evidence relied on outside sources such as professional journals for scientific information.19 Today's nurses use knowledge for practice gained from social interactions, experience, and previous learning.17,20 Only 60% of nurses responding to a national survey believed they needed new knowledge for practice weekly.21 Less than 25% of nurses frequently or always used journal articles for information.21 Knowledge used for pain management may reflect the same knowledge nurses learned in their basic programs. In fact, younger, less experienced nurses have scored better on pain management knowledge and attitude tests.22 However, studies have shown that nurses' overall knowledge of pain management best practices is lacking, particularly in pharmacology where test results rarely exceed 65%.22,23 Knowledge obtained through practice and experience is also problematic. Practical knowledge can be devoid of reflection and progress into practice inertia,24 practice based exclusively on past experiences, potentially incongruous with current evidence.
Evidence-Based Pain Management
Evidence-based pain management guidelines require comprehensive initial and frequent assessments with reliable and valid instruments, assessment-based interventions, reassessment, and further intervention as required.9 Compliance with EBPM is evaluated through a review of nursing documentation.9 A new rating scale, the Samuels' Pain Management Documentation Rating Scale (SPMDRS),25 advances previous instruments measuring documentation compliance. The SPMDRS provides a mechanism to score the quality of the nurse's pain management documentation (PMD) to identify compliance with EBPM. Measurement of PMD compliance allows monitoring of interventions designed to promote EBPM using higher-level statistical analyses.25
Analyzing the contribution of the practice environment and nurses' clinical expertise may offer insight into the factors influencing the implementation of EBPM. The research question guiding the study is: What are the effects of nurses' practice environment perception and clinical expertise on the implementation of EBPM?
METHODS
The effects of practice environment perception and expertise were evaluated using a convenience sample of surgical nurses from 3 units of 2 metropolitan teaching hospitals. One hospital was a designated Magnet facility. Staffing for the three 24- to 27-bed surgical units ranged from 46 to 85 full-time equivalents. After obtaining approval from the institutional review boards of both hospitals, nurses were approached during staff meetings of the 3 units. Nurses were informed that after completing a questionnaire, 4 of their patient's records would be reviewed for PMD. One hundred three nurses agreed to participate, with 92 submitting completed questionnaires.
The patient records of the 92 nurses who practiced on the general surgical units were reviewed. Seven of the completed nurses' questionnaires were omitted because of an inability to identify 4 of their record entries. The record of any patient who required epidural opioid or patient-controlled analgesia for pain control was not included until the epidural or patient-controlled analgesia was discontinued. To detect a medium effect size of.03 at the.05 level of significance at a power of.80, at least 82 nurse participants were required. The final number of participants for this study was 85.
Instruments
PERCEPTION OF THE PRACTICE ENVIRONMENT
The Practice Environment Scale of the Nursing Work Index (PES-NWI) is 1 of 15 nurse-sensitive performance measures.26 Research has indicated that PES-NWI is a valid and reliable composite measure of nurses' perception of the practice environment.27 The major contribution of the PES-NWI had been on the organizational level of measurement. The PES-NWI is composed of 5 equally weighted subscales: nurse participation in hospital affairs; nursing foundations of quality of care; nurse manager ability, leadership, and support; staffing and resource adequacy; and collegial nurse-physician relationships. Respondents select 1 of 4 categories from agree to strongly disagree to indicate their agreement with the 31 items. A high score on the PES-NWI indicates a positive practice environment. The ability of the PEW-NWI to discriminate between Magnet and non-Magnet hospitals supports content and criterion validity.28 Lake27 reported Cronbach [alpha]'s at the individual practitioner level of.82, and intraclass correlation coefficient (ICC) (1,k) of 0.96 for an aggregated or organizational score.27
CLINICAL NURSING EXPERTISE
The Clinical Nursing Expertise (CNE) instrument, constructed by Lake,29 based on Benner's novice to expert framework,17 was used to measure clinical expertise. The CNE instrument lists 34 nursing activities common to general medical surgical nurses and asks the respondent to indicate their perceived level of expertise on each item using a 5-point scale (1 = novice, 5 = expert). The clinical expertise score is calculated as a mean of the items. Four questions ask respondents how often they are asked to precept or are consulted on difficult cases to validate the expertise level. In field testing with a purposive sample of 84 nurses, Cronbach [alpha] was.97.29 Construct validity was evaluated by Lake29; comparing respondents and their managers, correlations ranged from 0.69 to 0.81. In this study, unit-based CNSs, responsible for evaluating nurses' performance, rated the expertise of 10% of the randomly selected participants. A high correlation (r = 0.75) between nurses' CNE and CNS ratings supported the expertise self-report.
EVIDENCE-BASED PAIN MANAGEMENT
Evidence-based pain management quality was evaluated with the SPMDRS.25 A quality score (1 = excellent to 7 = very poor) to 55 PMD patterns of assessment, intervention, reassessment, and further intervention is attached to the pain documentation found in the patient care record. Four entries from different patients are rated, with a mean score for the 4 entries determining the nurse's PMD quality score. Content validity, criterion validity, intrarater and interrater reliability have been previously supported.25
Procedure
After providing written informed consent, subjects completed the PES-NWI, CNE instrument, and a demographic questionnaire. Four general surgical patient records, representing a typical patient assignment, were obtained and reviewed for each participant. The record review procedure consisted of identifying the pain assessment, intervention, reassessment, and further intervention found in the progress notes, flow sheets, or medication administration records. Results of the record review were sorted by the findings on assessment documentation to identify the PMD pattern. The identified pattern was matched to the SPMDRS pattern ranks and assigned a score. Two raters matched patterns, obtaining an inter rater reliability of 0.88. After discussion, the inter rater reliability rose to 0.99.
Data analysis included Pearson correlation to determine the relationship among variables and multiple regression statistics to identify predictors of EBPM. Of the 85 CNE instrument questionnaires submitted, 2 participants reported more than 15 years of experience yet rated themselves as novices. These CNE instrument scores were omitted from further analysis. Normality assessment revealed a significant finding in CNE instrument score (Kolmogorov-Smirnov = 0.145; df = 83; P < .000), and a squared transformation was attempted to adjust the normality. However, the transformation was deemed to affect the interpretability of the CNE instrument, and therefore, transformation was not included in the analysis.30 The independent variables were entered into the regression simultaneously. Standardized residuals met the assumption of normality, and no colinearity was identified.
RESULTS
Nurse Sample
The general surgical nurses who consented to a record review ranged in age from 23 to 63 years (mean, 36.9 years [SD, 10.7 years]) and reported 8 months to 42 years of nursing experience (Table 1). Unlike the national average,31 64% of nurses held bachelor's degrees. Most nurses sampled (n = 47; 56.6%) identified themselves as proficient, with only 2 nurses describing themselves as novices and 10 describing themselves as experts. Of the 10 experts, 8 worked part-time, mirroring a national trend that older, more experienced nurses are working less.31 Two experts held associate degrees, 5 held bachelor's degrees, and 3 held diplomas. There was no difference in nurses' level of expertise between nurses practicing at the Magnet and non-Magnet hospital.
Significant moderate correlations existed between expertise and the demographic variables of age (r = 0.46; P < .000), years on the unit (r = 0.51; P < .000), years experience (r = 0.53; P < .000) (Fig. 1), and years at the facility (r = 0.51; P < .000). These findings support the theoretical notion that experience is a component of expertise. Significant moderate correlations were also found between expertise and the following items: "How often are you selected to be a preceptor for another nurse?" (r = 0.40; P < .000); "How often do nurses come to you for clinical judgment on a difficult clinical problem?" (r = 0.55; P < .000); and "Which of the following categories would you say best describes your level of clinical nursing expertise?" (r = 0.52; P < .000), adding to the validity of the CNE scale.
Perception of Work Environment
The PES-NWI scores were in the moderately high range (mean, 2.83 [SD, 0.35]), indicating a favorable perception of the practice environment.28 Cronbach [alpha] for the composite PES-NWI was.92. The ICC (1,k) of the composite score for the Magnet hospital was 0.94 and 0.84 for the non-Magnet hospital, indicating a high level of interrater agreement in both hospitals, although higher in the Magnet hospital. No significant relationships existed between the PES-NWI scores and demographic variables or expertise. There were no differences across units on PES-NWI scores; however, there was a difference in scores between hospitals (t83 = 2.45; P = .016), with the Magnet hospital having a higher PES-NWI score.
Evidence-Based Pain Management
Nurses' scores on the SPMDRS ranged from 2.5 to 6.5 (SD, 0.77). Pain management documentation scores formed a normal distribution, with a tail toward poorer PMD ratings. There was no relationship between PMD scores and demographic variables, expertise scores, or PES-NWI scores. Analysis of PMD patterns revealed that only 42.1% (n = 143) of the entries met EBPM guidelines. A quantifiable pain assessment could be identified in 44.6% (n = 151) of the entries; 81% (n = 275) of entries contained an intervention when required; reassessment documentation was found in 55% (n = 187) of the record entries.
Predictors of Evidence-Based Pain Management
Standard multiple regression was conducted to determine the effect of CNE and PES-NWI on EBPM. Results indicated that the model significantly predicted 5.2% of EBPM (F2,80 = 3.27; P = .043). Clinical expertise explained 4.4% (partial r = 0.219, P = .48) of the variance, whereas the perception of the practice environment did not contribute any variance (partial r = 0.192, not statistically significant). A higher rank on expertise was associated with a poorer PMD score.
DISCUSSION
The findings of this study indicated that nurses' clinical expertise contributed 4.4% of the variance in PMD quality, whereas there was no contribution of practice environment perception. The contribution of the expertise was in the negative direction such that the greater the clinical expertise, the lower the PMD quality. Expertise, a barrier to quality PMD in this study, supports the findings of others that demonstrated the prevalence of individual practice patterns.20 Perhaps educated, experienced, and proficient professionals, such as the nurses in this sample, relied on their individual experience and knowledge when documenting and managing pain. Experts may have more difficulty incorporating new knowledge, such as pain management guidelines, into their practice than less experienced nurses. The findings in this study corroborate the survey of Pravikoff et al,20 which found that nurses, most of whom graduated before 1995, have difficulty accessing information needed to update their practice.
Although only a small percentage of the EBPM was explained, the complexity of EBPM suggests that these results may be clinically significant. Barriers consistently identified by nurses, such as healthcare provider pain management practices, time management pressure of a busy surgical unit, and patient variables, all contribute to pain management and can influence the adoption of EBPM.
Although seemingly contradictory, the finding that relatively higher levels of clinical expertise contributed to relatively poorer EBPM is consistent with Benner's32 conceptualization. Benner32 theorized that expert practice deteriorated when bound by rules, whereas novices flourished when guided by structure. Implementing new evidence may not be aligned with previous practice experiences. Those with more expertise may need to unlearn to relearn. Discarding previously accepted behaviors requires significant energy. The holistic and unconscious decision making of the expert may be devoid of appropriate reflection especially if clouded by disillusionment or burnout,33 which may impact the integration of new practice behaviors.
It is possible that clinical expertise does not translate into pain management expertise expected of surgical nurses. A critical lack of nursing pain management knowledge undermines the development of true expertise. A lack of knowledge coupled with inadequate information retrieval skills, lack of access to guidelines, and little ability or time to reflect also challenges the expertise development. The CNE instrument does not include an assessment of professional development. Reconceptualizing expertise as an ability to integrate evidence and continued learning throughout a nursing career seems warranted.
The practice environment perception of the nurses sampled in this study did not contribute to PMD quality. The high ICC of the Magnet hospital nurses may not have provided the variability required for regression analysis, which challenges the notion that the PES-NWI may be useful at the individual level of measurement. System-wide analysis using the PES-NWI may be a useful parameter to predict system-wide outcomes such as morbidity and mortality. However, usefulness of the PES-NWI at the unit level to predict EBPM, a patient-specific and practice-specific outcome, appears questionable.
The SPMDRS as a measure of EBPM determined relationships previously undetected. The normal distribution and skew of the PMD indicated poor compliance with EBPM. Interventions to increase the implementation and documentation of pain management guidelines must be a healthcare priority.
IMPLICATIONS FOR CLINICAL SPECIALISTS
Implementing EBP into acute care settings across differing levels of nurse expertise is challenging. Findings in the current study suggest that interventions must be individualized with respect to expertise level. Assisting the novice integrate evidence into practice requires different strategies than their expert colleagues. Expert nurses, acting as preceptors for new employees and new graduates, should demonstrate pain management skills and knowledge consistent with current guidelines. New nurses, although inexperienced in patient care, may actually be better equipped with research literacy and EBPM guidelines. Providing opportunities for less experienced nurses to present at pain management case conferences and in-services may be a mechanism for blending new knowledge with past experience.
Lecture methods may not be as effective as case study analysis or clinical discussions when trying to teach the expert or proficient nurses.17 Using the clinical record as a learning tool may encourage the dialogue needed to help expert practitioners escape from EBPM barriers and traditional patterns. Alternatively, novice and advanced beginner nurses may learn better with structured templates. Decentralizing pain management educational efforts to the bedside may work for all levels of nursing expertise.
Finally, it is time to require individual as well as organizational accountability for pain management outcomes. Patients deserve the benefits of current practice. Counseling and reinforcement may be required to help practice behaviors align with current evidence. With the overwhelming amount of evidence readily available, clinicians failing to implement EBPM or managers allowing old practices to continue are complicit in their responsibilities to patients.
LIMITATIONS
The findings in the study are limited by a high ICC on the PES-NWI, resulting in the inability of the PEW-NWI to detect variability at the individual level. The study findings are limited to a sample of general surgical nurses from 2 hospitals and cannot be generalized.
FURTHER RESEARCH
Clinical expertise as measured by the CNE instrument may be a barrier to EBPM. Further research is needed to refine quantitative measurements of nursing expertise. With only 4.4% of the variance in EBPM identified, further study is needed to uncover predictors. Knowledge, attitude, and skills were not measured in this study and may contribute SPMDRS scores. Specific environmental characteristics such as nurse-patient ratio, leadership style, and professionalism also need further investigation. Further study on different samples from nonteaching hospitals is needed. The PMD quality of acknowledged experts in pain management may provide insight into EBPM practices.
References