Evidence-Based Practice

Chapter 10


Evidence-Based Practice




Evidence-based practice (EBP) is defined as the conscientious, explicit, and judicious use of current best research-based evidence when making decisions about the care of individual patients (Sackett et al., 1996). It has evolved into a dominant approach for clinical decision making and a core competency for advanced practice nursing (see Chapter 3). Although components tend to overlap, three levels of this core competency can be identified: (1) interpretation and use of EBP principles in individual clinical decision making; (2) interpretation and use of EBP principles to determine policies for patient care; and (3) use of EBP to evaluate clinical practice.


Evidence-based practice is based on a four-step process: (1) formulation of a clinical question; (2) identification and retrieval of pertinent research findings based on literature review; (3) extraction and critical appraisal of data from pertinent studies; and (4) clinical decision making based on results of this process. Principles of EBP are used for clinical decision making for individual patients, constructing and applying clinical practice guidelines, and determining policies for delivering care to large groups (Gerrish et al., 2011; Stiffler & Cullen, 2010). Despite widespread acceptance of the concept of EBP, clinician understanding of this process and its application to direct patient care remains limited. For example, a cross-sectional analysis of 37 primary care practices revealed that when caring for patients with heart failure, 87% and 62% appropriately prescribed an angiotensin-converting enzyme inhibitor or beta blocker, respectively, but only 16% and 8% reached the target dose recommended by the clinical practice guideline for managing this condition (Peters-Klimm et al., 2008). Similarly, a cross-sectional survey of 720 primary care physicians revealed that only 26.9% and 4.3% adhered fully to evidence-based guidelines for managing acute lower back pain with and without sciatica (Webster et al., 2005). A number of factors are thought to influence clinician acceptance and application of this problem-solving approach to direct patient care, including a lack of knowledge of the principles of EBP. This chapter will define EBP, differentiate it from concepts of research and quality improvement, and define three levels of advanced practice nurse competency related to EBP: (1) use of evidence in individual APN practice; (2) use of evidence to change practice; and (3) use of evidence to evaluate practice (Table 10-1). Exemplars 10-1, 10-2, and 10-3 provide examples of each of these EBP-related competencies.




imageExemplar 10-1   Level I


Interpretation and Use of Evidence-Based Practice in Individual Clinical Decision Making


The most basic level of EBP competency is the application of the four steps for clinical decision making in an individual patient. This proficiency requires more than formulation of a clinical question and identification of pertinent studies needed to determine best available evidence. The advanced practice nurse (APN) must combine knowledge of best evidence with an assessment of individual patient factors likely to affect treatment effects, such as the presence of comorbid conditions, psychosocial and cultural factors such as locus of control, preference and impact on quality of life, and cost considerations.


Example: As an APN in a urology department, I am often asked by patients and physician colleagues whether cranberry juice or supplements (including cranberry capsules) should be prescribed to prevent urinary tract infection. This persistent query led me to formulate a clinical question, “Are cranberry juice or cranberry products effective in the prevention or management of urinary tract infection?” A systematic literature review based on current best evidence available in 2002 suggested that regular consumption of cranberry juice reduces the incidence of urinary tract infections in community-dwelling women and residents of long-term facilities but does not reduce the risk in patients who undergo intermittent or indwelling catheterization (Gray, 2002). However, more recent evidence has emerged that influences the conclusions I reached at that time. Specifically, two randomized controlled trials (RCTs) published in 2011 and 2012 found that cranberry juice was no more effective than antimicrobial therapy or cranberry flavored placebo drink for preventing urinary tract infection (UTI; Barbosa-Cesnik et al., 2011; Stapleton et al., 2012). On initial consideration, this evidence appeared to support discontinuing recommendations of consumption of cranberry for women seeking to prevent recurrent UTIs. However, additional evaluation of findings from one of the studies, a study using a placebo group (Barbosa-Cesnik et al., 2011), revealed that both groups experienced a considerably lower incidence of UTIs than anticipated. In a subsequent interview with one of the investigators, the researchers acknowledged a possibility that the placebo-flavored drink might have contained some of the ingredients hypothesized to exert an antimicrobial effect in the urine (Larson, 2010). In addition, I considered the fact that consumption of cranberry juice twice daily is not associated with any known harmful side effects. I also considered the fact that cranberry juice is relatively inexpensive compared with dietary supplement cranberry capsules. As a consequence of all these factors, cranberry juice is preferred as a natural means for preventing UTIs among many women in my practice.


This example of basing individual clinical decisions on an EBP process illustrates several important points. It points out the importance of remaining abreast of emerging evidence and the real possibility that newer evidence may significantly alter our understanding of the benefits or harmful effects associated with a specific intervention. In addition, this case illustrates the role of patient preference in clinical decision making. Clinical experience strongly suggests that a significant proportion of women prefer nonpharmacologic interventions for preventing UTIs, and regular consumption of cranberry juice tends to increase overall fluid intake and provide possibly beneficial effects without associated adverse side effects. Therefore, given the absence of harm, low direct cost, and mixed evidence of a possible preventive effect, I discuss consumption of cranberry juice with women as a possibly effective intervention, free from harmful side effects. However, I also counsel women to consider engaging in behavioral interventions for the prevention of UTIs, including adequate daily fluid intake based on recent recommendations from the Institute of Medicine, daily consumption of a dietary source of the probiotic lactobacillus, and consideration of avoiding use of a diaphragm and vaginal spermicide as birth control strategies (Salvatore et al., 2011).


This case also illustrates the time-consuming and rigorous demands of basing individual clinical decisions on the EBP process. Fortunately, APNs have access to various evidence-based resources. For example, the Cochrane Database of Systematic Reviews is the world’s largest single source of systematic reviews and meta-analyses addressing multiple aspects of APN practice (Cochrane Library, 2012). The U.S. Preventive Services Task Force has also produced a large number of systematic reviews linked to various clinical topics of special interest to APN practice, with its focus on primary and secondary intervention to promote optimal health (Trinite, Cherry, & Marion, 2009).


In addition to these resources, a growing number of professional societies have generated evidence-based clinical practice guidelines that address measurable clinical questions with thorough and extensive systematic reviews of existing evidence to formulate clinical recommendations covering comparatively broad topics such as heart failure, diabetes mellitus, chronic obstructive pulmonary disease, breast cancer, end-stage renal disease, osteoporosis, and other topics of special interest to APN practice. In addition to searching the resources of the appropriate professional association’s web page, the National Clearinghouse of Practice Guidelines, operated by the Agency for Healthcare Research and Quality (AHRQ), houses a large collection of evidence-based clinical practice guidelines that can be accessed at http://www.guideline.gov.



imageExemplar 10-2   Level II


Interpretation and Use of Evidence-Based Practice to Create Policies for Patient Care


For many APNs, the growing demand to formulate evidence-based policies and protocols needed to prevent the growing list of “never events” provides an opportunity to master the second level, interpretation and use of EBP to create policies for patient care.


Example: Fineout-Overholt and colleagues (2010a, b, c) have described the EBP process needed to answer a clinical question about whether a rapid response team affects the number of cardiac arrests and unplanned intensive care unit admissions in hospitalized adults. Based on this question, the authors described the process used to search the evidence for pertinent studies, code and extract data from these studies using a standardized protocol, and synthesize data to implement policies needed to launch a rapid response team at their facility. Based on this process, the team concluded that there is sufficient evidence to justify developing policies and committing the resources needed to form a rapid response team at their facility. In addition to providing an example of the EBP described in this chapter, this series of articles describes the processes required to implement such a program. Although a detailed discussion of this translation from research-based evidence to clinical practice is beyond the scope of this chapter, the authors identify and briefly review essential components of this step in the implementation process, including engaging stakeholders in their facility, securing administrative support, preparing a campaign to launch the rapid response team, including staff education and changes in care protocols, and measuring outcomes following implementation of the practice change.



imageExemplar 10-3   Level III


Evaluation of Evidence-Based Practice to Determine Standards of Care


Participation in an interdisciplinary team to evaluate and determine standards of care using EBP is the third and most advanced level of the EBP competency for APN practice. Generation of an evidence-based clinical practice guideline requires identification of a number of clinically measurable questions required for establishing and evaluating clinical practice in a broad area of patient care, along with an extensive systematic review of pertinent studies. This often encompasses major assessment strategies related to the management of a particular disorder and first-line and alternative interventions for management.


Example: An interdisciplinary group, including two APNs, urologists, urogynecologists, and a medical librarian with extensive experience in systematic literature reviews, was charged with producing an evidence-based clinical practice guideline for treating interstitial cystitis–painful bladder syndrome (IC-PBS) for the American Urological Association (AUA, 2011). This broad topic was broken into clinical questions focusing on assessment strategies and the interventions used to relieve the pain and associated lower urinary tract symptoms associated with this poorly understood syndrome. After the IC-PBS working group reached consensus on pertinent clinical questions, a systematic literature review was completed that covered studies and related articles published between January 1, 1983, and July 22, 2009. The initial electronic database searches retrieved approximately 1130 articles, from which 86 articles were identified for further review. Members of the IC-PBS working group were assigned to review articles for coding and data extraction and the APNs were charged with studies evaluating behavioral interventions, including fluid and dietary strategies for managing bladder pain and related lower urinary tract symptoms (e.g., use of dietary supplements such as calcium glycerophosphate), and pelvic floor muscle training. Because of the generally low levels of supporting evidence, the IC-PBS work group determined that meta-analysis was not feasible. Nevertheless, the group found sufficient evidence to identify and create recommendations for clinical practice for first- through sixth-line treatments (AUA, 2011). Conservative interventions were advocated as first-line treatments because of their combination of moderate benefit combined with few to no harmful side effects. Although the strength of recommendations for clinical practice generated by this interdisciplinary group varied from moderate to low, the production of this clinical practice guideline illustrates the process used to evaluate and create standards of care using the EBP process in an interdisciplinary setting.


The term evidence-based practice represents a blending of several related concepts, including evidence-based nursing and evidence-based medicine. The original term, evidence-based medicine, traces its historical roots to a strategy for educating medical students developed by the faculty at McMaster Medical School in Hamilton, Ontario (Rosenberg & Donald, 1995). Evidence-based nursing is defined as the process that nurses use to make clinical decisions using the best available research evidence, their clinical expertise, and patient preferences (Dicenso, Cullum, & Ciliska, 2002). The explicit inclusion of patient preference and clinical expertise are significant for APNs because they reflect the holistic approach central to nursing practice while maintaining the focus on current, research-based evidence.


EBP offers several advantages for clinical decision making when compared with previous models. For example, tradition-based practice is based on clinical and anecdotal experience, combined with received tradition from experienced clinicians, and expert opinion. Tradition-based clinical decision making is based on received wisdom, often provided by instructors or clinical preceptors and expert opinion from those perceived as experts in a given area of care. By substituting a standard of current best evidence for received wisdom or expert opinion, EBP encourages the advanced practice nurse to update and refine clinical practice continually as newer evidence is generated and published. EBP also offers distinctive advantages when compared with rationale-based clinical decision making. Rationale-based clinical decision making relies on identifying a rational explanation for an intervention (Gray et al., 2002). This form of clinical decision making relies on findings from a wide variety of studies, including the following: pathophysiologic research designed to identify the principal action of an intervention or the main reason it exerts a particular effect; and in vitro or in vivo research models that measure outcomes in animals, tissues, or individual cell lines. Although these types of studies are enormously valuable to our overall understanding of health, disease, and the reasons that interventions exert a particular effect, EBP limits its search for evidence to studies that directly measure the efficacy or effectiveness of a particular intervention, predictive power of diagnostic studies, and presence and severity of adverse side effects.


EBP differs from research and quality improvement (QI) projects. Research can be defined as a systematic investigation designed to generate or contribute generalizable knowledge to health care or advanced practice nursing in particular (Arndt & Netsch, 2012). EBP combines findings from multiple research studies that focus on the efficacy of a particular intervention or accuracy of a specific assessment technique to aid APNs and other clinicians when making decisions about the care of an individual patient or group of patients. EBP has been described as the study of studies; its goal is the synthesis of existing knowledge generated from multiple research studies, whereas the goal of an individual research study is to generate new knowledge about an intervention or assessment technique (Gray et al., 2002).


QI is a systematic activity guided by outcome data to achieve rapid improvements in health care delivery in a specific setting (Arndt & Netsch, 2012; Glasziou, Ogrinc, & Goodman, 2010). Both QI projects and research studies are driven by outcome data but QI is an intensely localized activity. The data generated during a QI project is designed to improve specific outcomes within a local facility, clinic, or community. Unlike the data generated by a research study, the results of a QI project can only be generalized to the specific patient population that comprised the project setting.


Despite these differences, the APN should remember that research, EPB, and QI projects share a common goal—improvement of patient care. Further, research, EBP and QI should be viewed as complementary and combined in a manner that improves individual clinical decision making and care processes affecting an entire facility, health care system, or larger community. For example, an acute care APN may observe that the ventilator-associated pneumonia (VAP) incidence in her facility’s critical care unit is higher than published benchmarks. As a result, the APN elects to complete a QI project designed at reducing the incidence of VAP. Initially, the APN should review the unit’s current prevention protocol to determine whether it is based on current best evidence, such as routine oral hygiene, regular evaluation for readiness to extubate, elevation of the head of the bed, and prophylaxis for peptic ulcer disease and deep vein thrombosis (Tablan et al., 2004). This review may incorporate principles of EBP and findings from individual research studies to answer two questions:



In reference to the first question, a review of current best evidence suggests that bundled interventions are effective for reducing the incidence of VAP (Ramirez, Bassi, & Torres, 2012; Wip & Napolitano, 2009). When examining individual interventions, the acute care APN may note that current best evidence supports regular oral hygiene that incorporates chlorhexidine as effective for preventing VAP (Wip & Napolitano, 2009). In contrast, limited evidence suggests that ongoing elevation of the head of the bed may not affect VAP incidence, even though it is associated with an increased likelihood of sacral pressure ulcer formation. Finally, the APN also may identify findings from an individual study, the NASCENT randomized clinical trial. This study demonstrated that a silver-coated endotracheal tube reduced the incidence of VAP (Kollef et al., 2008) in 9417 critically ill adults from 54 facilities in North America.


Thus, the APN has synthesized essential research-based knowledge using principles of EBP to provide a platform for a QI project. Depending on existing policies in the local critical care unit, the APN may collaborate with others to create a modified or novel prevention bundle and measure VAP incidence before and following implementation of this bundle. Findings of this process comprise a QI project; although these results cannot be generalized to every critical care unit, they can be used to evaluate care processes in the local critical care unit.



Evidence and Current Best Evidence: Historical Perspective


Although the concept of “best evidence” may appear transparent on initial consideration, a more careful analysis of the historical roots of evidence generation in health care is needed. The Oxford English Dictionary Online (2012) defines evidence as an object or document that serves as proof. Within the context of EBP, the objects that clinicians seek to establish evidence for the efficacy and safety of an intervention, or the predictive power of an assessment, are studies. Although the search for evidence can be traced back more than 2000 years, definitions for what constitutes sufficient evidence to reach these conclusions have evolved significantly over time.


The gold standard research design for generating evidence in the twenty-first century remains the RCT (Yoshika, 1998). This research design is based on three critical elements: (1) manipulation of an experimental intervention; (2) comparison of the experimental intervention to a control or comparison group that receives a placebo, sham device, or standard intervention, depending on ethical considerations; and (3) random assignment of subjects to the intervention or control group. Random allocation, advocated since the early 1930s, is an essential element of an RCT because it is the most effective technique for spreading potentially confounding factors evenly among treatment and control groups (Hill, 1937). A well-known RCT that compared streptomycin with standard care at the time (bed rest) is usually cited as the world’s first, large-scale, controlled trial (Streptomycin treatment, 1948). Randomization was achieved using a closed envelope system and subjects were blinded to treatment group. However, at least one trial was completed and published before this landmark study. Amberson and associates (1931) compared the antibiotic sanocrysin for the treatment of pulmonary tuberculosis with a placebo. In addition to random allocation of subjects by flipping a coin, they also blinded physician data collectors to group assignment to minimize bias, another important design feature of the modern RCT.


Based on this historical legacy and guided by the pioneering efforts of Archibald Cochrane, current best evidence is now defined as the best available studies evaluating the efficacy and safety of an active or preventive intervention or the predictive accuracy of an assessment (Gray et al., 2004; van Rijswijk & Gray, 2012). These studies must directly evaluate the effect of an intervention, compare the intervention with a placebo, standard care, or sham device, and document adverse side effects associated with the intervention. Studies used to establish current best evidence must be executed in human (rather than animal) subjects and must measure the most direct outcome of treatment, rather than relying on interim outcomes based on convenience. For example, a study of the efficacy of a topical wound therapy should measure wound closure rather than concluding efficacy based on the percentage of wound closure completed at an arbitrary point after the initiation of treatment (van Rijswijk & Gray, 2012).


This definition of current best evidence raises a corollary question: What criteria must be fulfilled to define an intervention as “evidence-based?” At least two major regulatory groups, the U.S. Food and Drug Administration (FDA) and European Medicines Agency (EMEA), have established specific criteria for labeling an intervention as evidence-based (Cormier, 2011). For a drug to receive an indication for clinical use, the FDA requires results from two well-designed RCTs with consistent results, both of which must compare the agent with a placebo- or sham-based control group. The EMEA criteria are similar (EMEA, 2000).


Although these groups provide well-defined criteria for defining an intervention (administration of a drug) as evidence-based, achieving this level of evidence is complex and enormously costly. For example, the total costs of achieving a new drug indication are estimated to vary from $55 million to more than $800 million (U.S. dollars; Adams and Brantner, 2006; DiMasi et al., 2003). Based on these rigid criteria, only a minority of interventions that APNs use to manage their patients would qualify as evidence-based, and limited research in this area has suggested that 40% of clinical decisions used in daily practice are unsupported by evidence (Gray, 2002; Greenhalgh, 2001). As a result, APNs often must search the literature and identify relevant evidence to support clinical decision making in a particular case, or retrieve this information from EBP resources, such as clinical practice guidelines or best practice documents. This chapter will describe the four steps of the evidence-based process and identify resources that the APN can identify and use when making clinical decisions.



Steps of the Evidence-Based Process



Step 1: Formulate a Measurable Clinical Question


Clinical decision making using the EBP process begins with the formulation of a measurable clinical question. Questions arise from various sources. For example, many APNs will formulate their first clinical questions as part of an EBP process when planning a capstone project as part of their Doctor of Nursing Practice degree. Individual clinical practice provides another rich source for clinical questions.


The APN also may be tasked with formulating, and answering clinical questions in response to the need to formulate evidence-based policies for prevention of never events. For example, in 2008, the Centers for Medicare and Medicaid Services (CMS) began a list of “never events” to reduce potentially preventable facility-acquired conditions that included VAP; this list continues to be updated (Chicano & Drolshagen, 2009). For example, in 2007, Gastmeier and Geffers reported on the work of a group of clinicians, including APNs, who were asked which interventions were effective for the prevention of ventilator-associated pneumonia. Their original search for current best evidence identified clinical practice guidelines first released in 2003 and published in 2004 (Tablan et al., 2004). However, they also queried whether newer evidence warranted changes or additions to the preventive interventions recommended in these guidelines. They specifically questioned whether two interventions—oral decontamination via topical chlorhexidine and drainage of subglottic secretions—delay the onset of or prevent VAP. A subsequent literature search revealed recent evidence supporting these interventions that was not included in the 2003 clinical practice guidelines.


Clinical questions may arise from a need to identify evidence supporting the role of the APN in response to a specific disease or disorder. For example, Evans (2010) asked about the effect of adding a follow-up telephone intervention by an APN on blood glucose control when compared with standard treatment. This question was used to identify and retrieve evidence from five RCTs and one systematic literature review to design and implement a protocol that resulted in clinically relevant and statistically significant reductions in fasting blood glucose levels in patients receiving the APN-directed intervention.


After identifying the general topic to be scrutinized, the APN must formulate a measurable question that can be meaningfully addressed using evidence-based clinical decision strategies. Results of several studies have suggested that application of the PICO model aids nurses in formulating clinically relevant and measurable questions (Balakas & Sparks, 2010; LaRue, Draus & Klem, 2009; Smith-Strom & Nortvedt, 2008; Table 10-2). PICO stands for:




Thus, as shown, the P in PICO indicates patient or population (Sackett, et al., 2000), although the P is sometimes expanded to include primary problem (Balakas & Sparks, 2010). This element of the formula alerts the APN to define the population to be studied and the nature of the problem to be scrutinized carefully. The population may comprise a subgroup of patients in a facility, such as critically ill patients receiving artificial ventilation or all patients with an indwelling urinary catheter, but it often incorporates much larger populations, such as any individual with a wound or any patient recently diagnosed with diabetes mellitus. As these examples illustrate, identification of the primary problem is closely tied to the population under scrutiny. Examples of primary problems may be a disease such as sinusitis, a disorder such as chronic osteoarthritis, or a predisposition to a potentially preventable condition such as a pressure ulcer.


The I in the PICO model represents the main intervention to be considered. In many cases, the APN will examine a single intervention such as the follow-up telephone intervention for reduction of the fasting blood glucose level in patients with diabetes mellitus described earlier (Evans, 2010). In contrast, the combined effect of more than one intervention used to prevent or treat a specific disorder will be evaluated. In this case, the APN will identify a protocol or bundle of interventions and analyze their effect on a given outcome. Searching for evidence that evaluates the combined effect of multiple interventions is clinically useful, but presents unique challenges. For example, Hagiwara and coworkers (2011) studied whether decision support tools decrease the time to receive definitive care in acutely ill or trauma patients prior to hospital admission. They operationally defined decision support tools as active knowledge systems that use two or more items to generate case-specific advice. They further classified these tools as electronic or nonelectronic. However, their literature search retrieved only 2 of 33 studies that specifically addressed this clinically relevant question. Despite the use of a well-accepted definition for decision support tools, the authors observed that a number of studies were excluded because it was not possible to classify the study intervention as a decision support tool.


Nayan and coworkers (2011) faced a similar challenge when studying whether smoking cessation rates were higher in oncology patients who receive smoking cessation interventions as compared with usual care. Their initial search identified a meta-analysis of data from eight RCTs that detected no differences in self-reported cessation rates when these interventions were compared with usual care. However, subclassifying smoking interventions into pharmacologic, behavioral, and combined interventions suggested that cessation protocols that combine pharmacologic and behavioral interventions appeared to increase cessation rates when compared with usual care or single-intervention protocols.


The C in the PICO model represents the approach used as a basis for comparison to the intervention undergoing scrutiny. This approach, sometimes called a bundled intervention, is frequently described in research reports as standard care or usual care. Although these terms are descriptive, it is essential that the APN specifically define the intervention(s) that comprise standard care and ensure that the studies retrieved enable adequate differentiation of this standard care from the intervention under scrutiny, especially when evaluating the effect of a bundled intervention or protocol.


The O in PICO represents the outcome, or intended goal of the intervention. When determining the outcome, it is important to identify and evaluate the most direct result indicating clinical efficacy and avoid reliance on indirect outcomes that are more easily measured. Careful consideration of the most direct and clinically relevant outcome is essential when constructing a clinically relevant question. For prevention studies, the most direct outcome is generally a reduction in the incidence of the disease or disorder under scrutiny. For example, an APN evaluating the effect of a prevention protocol on surgical site infection rates should base conclusions of efficacy on incidence rates, rather than on interim outcomes, such as differences in a nurse’s knowledge after education on prevention or self-reported changes in practice following in-service training.


A final element, T, indicating time, may be added to the PICO conceptual framework. The time frame is meant to indicate the relevant observation period for outcomes; it may be short, such as the first 24 to 48 hours following surgery, or long, such as years to decades following the onset of a chronic condition such as dementia or diabetes mellitus (Balakas & Sparks, 2010; LaRue, Draus, & Klem, 2009; Smith-Strom & Nortvedt, 2008).



Step 2: Search the Literature for Relevant Studies


Evidence-based clinical decision making relies on identifying research-based evidence. Therefore, it is essential for the APN to develop expertise in searching the literature to identify and retrieve appropriate studies. Fortunately, the development of modern electronic databases has revolutionized our ability to search the published literature rapidly and access pertinent research reports. A number of electronic databases are now available to the APN (Table 10-3). Although full access to these databases usually requires a paid subscription, APNs may access these electronic databases via a facility-based subscription. Specifically, the vast majority of health system, university, or college libraries maintain institutional subscriptions to OVID, ensuring access to multiple electronic databases such as MEDLINE or CINAHL. In addition, access to PubMed, a service of the MEDLINE database, is available without charge on the Internet.



imageTABLE 10-3


Electronic Databases for Identifying and Retrieving Pertinent Research



































Name Description URL
MEDLINE Largest online database for nursing, medical, and allied health journals http://medline.cos.com/
PubMed Freely accessible online version of MEDLINE database; lacks the robust Boolean features of MEDLINE http://www.ncbi.nlm.nih.gov/sites/entrez?db=PubMed
Cumulative Index for Nursing and Allied Health Literature (CINAHL) Largest database for nursing and allied health literature; includes multiple nursing journals not indexed in the MEDLINE database http://www.ebscohost.com/biomedical-libraries/the-cinahl-database
Education Resource Information Center (ERIC) Linked to more than 320,000 articles from 1966 to the present; focuses on educational literature, including undergraduate and graduate nursing http://www.eric.ed.gov/
PsycINFO Contains more than 3 million resources dating back to 1888; excellent resource for the APN who specializes in providing mental health care http://www.apa.org/pubs/databases/psycinfo/index.aspx
MD Consult Incorporates approximately 80 health care journals; includes access to patient education handouts and clinical practice guidelines; accessible via smartphone; administered by Elsevier http://www.mdconsult.com/php/327720868-2/home.html
Web of Science Includes journals in the basic and clinical sciences drawn from approximately 9300 journals with impact factors; administered by Thomson-Reuter http://thomsonreuters.com/products_services/science/training/wos/


MEDLINE and PubMed


Administered by the U.S. National Library of Medicine, MEDLINE is the world’s largest electronic database of health-related research and literature (U.S. National Library of Medicine, 2012). There are articles from a number of professions, including medicine, nursing, dentistry, veterinary medicine, and associated disciplines such as physiology, pharmacology, and molecular biology. Approximately 5600 journals are indexed. MEDLINE can be accessed via various strategies, including keywords. However, the MEDLINE database is primarily organized around MESH terms (medical subject headings). Entering a MESH term, such as “coronary artery disease” or “osteoporosis,” will trigger a number of subheads that are potentially useful to identify evidence for answering a clinical question, such as “diagnosis,” “drug therapy,” “diet therapy,” and “nursing.” The MEDLINE database may also be searched using various keywords that are not official MESH terms; these searches retrieve articles that include the keyword in its title, abstract, or list of identifying keywords, but they will not provide the subheads available when a MESH term is accessed. The MEDLINE database includes articles published in 39 languages; 91% are printed in English and 83% of those published in other languages have English language abstracts, greatly increasing access for English-speaking searchers.


MEDLINE has particularly robust Boolean functions, allowing the APN to focus or narrow a search based on combining two or more MESH terms or keywords using the functions “AND,” “OR,” and “NOT” (U.S. National Library of Medicine, 2012). For example, an APN might pose a question about the effectiveness of administering an angiotensin-converting enzyme (ACE) inhibitor for the prevention of mortality and disease progression in patients with heart failure. In this case, the APN might initially select the MESH term “heart failure” along with the MESH term “angiotensin-converting enzyme inhibitors.” By using the “AND” Boolean function, the database will retrieve articles that merge the intervention (ACE inhibitor agents) with the primary patient problem under scrutiny (heart failure).


A second Boolean function, “OR” allows the searcher to retrieve articles that contain either of two keywords or MESH terms. This function is useful when terms that are recently coined or historically relevant differ from the corresponding MESH term. For example, an APN may be seeking information about patients who experience chronic lower urinary tract pain not associated with bacterial infection. The MESH term for this condition is “interstitial cystitis.” However, a more recent term (painful bladder syndrome) has been increasingly used to describe this condition; combining the MESH term “interstitial cystitis” with the keyword “painful bladder syndrome” retrieves more citations that entering either term alone.


A third Boolean function, “NOT,” allows the APN to limit a search by eliminating articles that do not address the intervention, assessment, or patient population under scrutiny. For example, an APN interested in prevention of central line infections might enter the MESH term “indwelling catheters,” which will retrieve studies focusing on infections associated with multiple types of catheters, including urinary and peritoneal dialysis catheters. Use of the “NOT” Boolean function will enable the APN to eliminate articles about various types of catheters not pertinent to a clinical question focusing on hospital-acquired central line infections.


The MEDLINE database allows searches via multiple alternative fields, including author, journal, publication type (e.g., review article), language, experimental approach (human, in vivo, or in vitro), gender, age range, and publication year. These options are useful for focusing searches based on the parameters specified in the clinical question.


The PubMed web page provides free access to the MEDLINE database; it can be accessed at http://www.ncbi.nlm.nih.gov/pubmed. The basic search engine will retrieve articles based on keywords. Clinicians searching PubMed can click on an advanced search icon and access a site that allows combination of keywords or keyword and author or journal using the Boolean function “AND.” However, the PubMed database does not have the robust search functions characterized by MEDLINE. In addition, although a limited number of articles can be downloaded directly from the PubMED site, access to most articles is restricted to the complete citation and abstract.

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Oct 19, 2016 | Posted by in NURSING | Comments Off on Evidence-Based Practice

Full access? Get Clinical Tree

Get Clinical Tree app for offline access