Future Directions of Credentialing Research in Nursing: Workshop Summary (2015)

2

A New Framework for Credentialing Research in Nursing

Bobbie Berkowitz, chair of the planning committee, opened the workshop by welcoming all participants and thanking the workshop’s sponsors. The goal of the workshop, she said, is to advance the field of nursing credentialing—as it applies to individual nurses and to organizations—by rigorous examination of its impact on important outcomes for individual nurses, patients, and health care organizations. The workshop planners envisioned a forward-looking agenda, with a focus on developing a national agenda for research, identifying critical knowledge gaps, and sparking ideas to use existing research tools and databases and develop new ones.

UNDERSTANDING THE LANDSCAPE AND STATE OF SCIENCE IN CREDENTIALING RESEARCH IN NURSING

Robin Newhouse, University of Maryland School of Nursing

This presentation provided important background information and context to help frame the workshop’s goals. To lay the foundation for later discussions, Newhouse provided the International Council of Nurses’ definition of credentialing:

Conversations about nursing credentialing research often focus on voluntary programs. It is also important to recognize that credentialing research, while often focused on the credential itself, can also focus on a separate but related construct—standards—which are authoritative statements defined and promoted by the profession, Newhouse said.

An individual credential may reflect the holder’s desire to improve quality of practice, service, and education, and generally falls into one of three categories (Needleman et al., 2014):

1. Entry-level. Initial licensure by a state board of nursing, which affirms that basic skills are present.
2. Special skills or training. Voluntary certification within the scope of a basic professional license, such as in critical care nursing. This category may indicate a level of training well suited to improve access to, and quality of, care, particularly in special populations, said Newhouse.
3. Advanced practice. Licensure by state authorities for nurse practitioners, certified nurse midwives, clinical nurse specialists, and certified registered nurse anesthetists who also hold the appropriate specialty certification from a nurse credentialing organization.

In addition to credentialing of individuals, organizations can also seek credentialing. The American Nurses Credentialing Center’s (ANCC’s) Magnet Recognition Program (ANCC, 2014a) for hospitals and health systems and Pathway to Excellence program (ANCC, 2014b) for acute and long-term care settings are examples of organizational credentialing.

Current Salience of Credentialing and Credentialing Research

Credentialing is not an isolated activity within the U.S. health care system and, in fact, may be one answer to some of the nation’s most pressing health care questions, said Newhouse. For example, nursing credentialing may contribute to standardized care quality and promoting nurse participation in leadership roles. The issue of credentialing research in nursing complements some of the major initiatives to improve U.S. health care today, continued Newhouse, who cited as examples the Agency for Healthcare Research and Quality’s (AHRQ’s) 2013 National Healthcare Quality Report (2014) and the 2011 Institute

of Medicine (IOM) report, The Future of Nursing: Leading Change, Advancing Health. The AHRQ report summarizes quality metrics that the agency collects, and concludes that, overall, U.S. health care quality is suboptimal, Newhouse said, with about 70 percent of the population receiving necessary care (AHRQ, 2014).

The quality of care can vary widely from one geographic area to another and across various demographic parameters (IOM, 2013). Despite efforts to improve and standardize care under the Patient Protection and Affordable Care Act of 2010, access to care has worsened slightly, compared with previous reports, and is a challenge for approximately one in four Americans (AHRQ, 2014). Moreover, minority and low-income Americans still experience disparities in care. Nursing has the potential to ameliorate these problems, which represent both a national issue and a health care system priority.

Two specific recommendations from The Future of Nursing bear directly on the workshop (IOM, 2011). First, “nurses should be able to practice to the full extent of their education and training.” Second, nurses should be granted opportunities to partner with physicians and other health care professionals to redesign and improve health care. Credentialing programs can be used to help implement these recommendations by standardizing and validating specific knowledge sets, skills, and competencies to improve the quality of patient care. Furthermore, nurses with advanced leadership skills can help identify problems in health care access and quality, develop related solutions, and translate research to practice for specific populations, she said.

In addition to promoting a basic level of competence, Newhouse mentioned that credentialing also protects the public, provides some degree of professional accountability, and ensures quality of practices and services. Credentialing motivates health care providers to meet established standards for care quality. By standardizing care requirements, credentialing also reduces variations in care across settings, thereby improving overall care. Finally, credentialing serves as an evaluation tool for practice competencies. Individuals and organizations seeking a credential not only meet current standards, but may also be required to continually demonstrate appropriate knowledge, skill, and competency to meet credentialing requirements.

An Initial Theoretical Model

In recent years, the ANCC Research Council synthesized existing evidence about credentialing research in nursing and convened a national summit to review the findings (Lundmark et al., 2012). In its 2012 report, the ANCC Research Council proposed a conceptual framework (the ANCC Model) that identified categories of variables affecting the impact of credentialing research (see Figure 2-1).

Newhouse commented that the ANCC Model is intended to frame broad constructs. However, many variables affect not only an individual’s or an organization’s decision to seek credentials but also various outcomes affecting nurses, patients, organizations, communities, and populations. Standards and credentials have a dynamic, interactive relationship with each other and intervening variables. These intervening variables (e.g., work context or factors that affect a nurse’s ability to

FIGURE 2-1 ANCC Research Council Model for Credentialing Research.
NOTE: This figure has been updated by Hickey and colleagues (2014).
SOURCE: Lundmark et al., 2012.

deliver patient care) have a direct link to a range of outcomes and may include factors such as work context or organizational policies or structures. In addition, powerful environmental factors (e.g., organizational, social, economic, political, and technological) also influence credentialing seeking behavior.

Although the ANCC Model is a highly simplified model, it identifies the basic relationships among large categories of factors and reflects what has been learned to this point about these complex interactions. Research that elaborates on and clarifies the interactions in this conceptual model may lead to better understanding of the link between credentialing and outcomes, especially patient outcomes.

The State of the Science

Summaries of the science evaluating the impact of individual and organization credentialing were presented at meetings of the Standing Committee on Credentialing Research, said Newhouse (Aiken, 2013; Johantgen, 2013). For individual credentialing, the evidence is somewhat scant, ranging from whether credentials can be linked to prevention of specific adverse health outcomes (e.g., falls, pressure ulcers, infections, and even mortality), procedural outcomes (e.g., medication errors), or organization-level outcomes (such as length of stay, patient satisfaction) (Newhouse, 2014). Establishing a relationship between credentialing and patient outcomes has been even more difficult, with significant practical and methodological challenges. Many of the existing studies are descriptive and correlational, and their findings are inconsistent.

Evidence from studies of organization credentialing is more informative in terms of impact. For example, most (but not all) studies of Magnet hospitals find that Magnet status is correlated with improved work environments (e.g., reduced turnover and intent to leave). The evidence linking Magnet status to patient outcomes (e.g., mortality, failure to rescue, and hospital-acquired infections) generally, but not always, indicates positive relationships.

In individual and organization credentialing research, the high prevalence of observational studies (e.g., surveys or secondary data) that are relational and not comparative is problematic. The lack of operational and conceptual clarity, use of nonstandardized definitions, and differing interpretations of variables further complicate research efforts. Even with

an ideal study, it is methodologically challenging to attribute patient outcomes to a single nurse, much less to a particular credentialing status.

In short, credentialing research in nursing is at an early stage, with many aspects needing improvement to strengthen the science. But the research is evolving, and it is important, said Newhouse. The system—and the profession—need a solid research base for both individual and organizational credentialing to bring clarity to this endeavor.

DEVELOPING A NATIONAL AGENDA FOR CREDENTIALING RESEARCH IN NURSING

Jack Needleman, University of California, Los Angeles¹

Individual nurses and health care organizations invest in nursing credentialing because they believe it produces desirable outcomes for themselves and for the patients they treat. The justification for investing in nursing credentialing is rooted in this belief, yet the evidence remains inconclusive. As such, determining how and whether nurse credentialing affects institutional, nurse, and patient outcomes should be a primary goal of any national agenda for credentialing research in nursing, said Needleman. However, the complex nature of nursing work, the health care environment, and the credentials themselves, make demonstrating a causal relationship between credentials and outcomes difficult. Conceptual modeling of the causal pathways between credentials and outcomes offers a promising framework for addressing this challenge. In his workshop presentation, Needleman proposed such a model, compared conceptual models to other research frameworks, and discussed the relationship of research to funding as one of many key issues facing nursing credentialing researchers.

Complex Work in a Complex Environment

Nursing is intellectually, emotionally, and physically demanding work that often involves the simultaneous care of multiple patients by a single caregiver. For example, when a participating hospital in the Robert Wood Johnson Foundation’s Transforming Care at the Bedside Initiative

__________________

¹This presentation drew largely from the IOM Perspective paper Nurse Credentialing Research Frameworks and Perspectives for Assessing a Research Agenda (Needleman et al., 2014).

tracked nurse movement in a unit, it observed a single nurse making 23 stops during 50 minutes of a single shift (Rutherford, et al., 2008). This complexity of care places significant organizational demands on unit managers responsible for developing work assignments for staff members, which must be sensitive to how individual nurses interact with patients, one another, and other members of the care team to affect patient care.

It is important to note that the factors which individually contribute to the complexity of patient care may also interact with each other in a “non-linear fashion,” said Needleman, further complicating the health care environment and increasing the challenges for researchers trying to study factors, such as credentialing, which might influence the environment. For example, patient to nurse ratios and the quality of work environments affect patient outcomes individually and in combination (Aiken et al., 2011; McHugh et al., 2013).

The growing number of credentialing organizations and types of credentials further complicates the assessment of credentialing in the health care environment. The American Board of Nursing Specialties includes 34 nursing certification member organizations, of which 26 offer 88 credentials for basic practice, 14 offer 48 credentials for advance practice nursing, and 3 offer 8 credentials for non-registered nurse (RN) practice (Needleman, 2014; Needleman et al., 2014). Similarly, the ANCC offers 28 individual certification programs across a wide range of nurse practitioner, clinical nurse specialist, and specialty certifications (Needleman, 2014). For institutions, ANCC offers several credentials, including Magnet Recognition® and Pathway to Excellence®.

Moreover, credentialed staff or institutions can influence care in many direct and indirect ways. For example, a nurse with a specific credentialed competency may be better able to treat a patient with specific needs. But patients assigned to a nurse without that credential may still receive the benefits of the competencies associated with a credential if their nurse seeks guidance from a credentialed peer. To take this into account, research on credentialing and patient outcomes may need to examine both whether a patient was treated by a credentialed nurse and whether there were credentialed nurses on the unit.

Given the number of variables and interactions, tracking the effect of a single credentialed nurse or the concentration of credentialed nurses or an institution’s credential on patient outcomes constitutes a profound challenge for researchers. If a causal relationship exists between credentialing and outcomes, demonstrating it will depend on the development

of rigorous and carefully designed research studies using methods and approaches that can differentiate the effects of credentialing and other factors influencing patient care.

Potential Research Methods

Needleman posed the following question: “How do you design the research so that you have some reason to believe that the associations you are seeing are causal?” In terms of research design, the gold standard for demonstrating causality is the randomized controlled trial. However, randomized trials on credentialing are unlikely, and assessing the impact of credentials in observational studies has significant challenges. Individuals who seek voluntary credentials likely differ from their noncredentialed peers not only by credentialed status but also by personal characteristics that lead to credential-seeking behavior. In the same way, institutions that seek credentials may differ from those that do not. Needleman called this “the endogeneity problem,” and noted that other methods than randomization must be used to take these factors into account to assess the impact of the credentials themselves.

Other potential research methods for demonstrating causality include natural experiments² and statistical adjustment. However, there are a limited number of natural experiments available, and statistical adjustments may not be able to control for all variables. Causality may also be assessed using conceptual modeling and research that focus on specific causal pathways.

A New Framework for Credentialing Research

In the ANCC model, the relationship between credentials and outcomes is mediated by a set of “intervening variables.” These intervening variables, however, are not specified in detail, and the nature of the causal pathways linking credentialing and outcomes were not fully explained,

__________________

²Natural experiments occur when people or organizations differ in isolated and measurable ways that are outside their control. Needleman provided the example of a hypothetical case, where “some states have one set of credentialing requirements and other states have different credentialing requirements.” All else being equal, investigators can “compare the experience across states,” and conclude that differences in experience may be due to variability along this one parameter.

said Needleman. Therefore, research using this framework was often reductive. Often, the research on Magnet hospital status and patient and nursing outcomes regresses the outcome of interest on Magnet status and other potential confounders, but does not closely examine how the characteristics of Magnet hospitals or the Magnet journey contribute to the outcome. As a result, when conflicting research findings occur, the design of existing research cannot definitively link Magnet status to institutional performance. For example, research that compares Magnet and non-Magnet hospitals shows clear correlation between credential status and outcomes. However, research comparing Magnet hospitals to non-Magnet hospitals that are similar in size, teaching status, and other characteristics to Magnet hospitals has found weaker associations between Magnet status and outcomes of interest. Accordingly, this raises the “key question of whether it is the Magnet credential or whether it is something else about these institutions” that affects outcomes, said Needleman. Answering this question requires the development of a conceptual framework that eliminates causal ambiguities by mapping the mechanisms by which credentials and outcomes are potentially associated.

Needleman proposed an Expanded Conceptual Model (see Figure 2-2), which is a new framework that builds on the ANCC Model and attempts to map unique intervening variables to different types of outcomes in order to assess causality. Box 2-1 identifies important differences between the ANCC Model and the Expanded Conceptual Model.

The Expanded Conceptual Model includes three pathways: (1) Invisible Architecture, (2) Work Organization, and (3) Nursing Performance. Each of these pathways have three levels (from top to bottom), including competencies, a variety of intervening variables, and outcomes. Each pathway can directly or indirectly affect four types of outcomes: organization, nurse, patient, and population health outcomes.

The Invisible Architecture pathway (which includes factors such as leadership, culture, and climate of an organization) maps the associations between credentialing, the intangible characteristics of the workplace environment, and the outcomes of the organization, nurse, and patient. To illustrate this causal pathway, consider how the unique competencies of credentialed nurses may positively affect the culture and care expectations of the ward they staff in ways not mirrored by noncredentialed nurses.

FIGURE 2-2 The Expanded Conceptual Model of potential links among credentialing, patient and organizational outcomes, and intermediate activities.
NOTE: Models for research on organizational credentialing and individual credentialing might be modified to emphasize different components of the expanded framework.
SOURCE: Needleman et al., 2014.

The Work Organization pathway focuses on how the credential-related capabilities of nurses can affect their workplace, peers, and patients not under their immediate care. For example, a credentialed nurse may advise a noncredentialed colleague regarding the care of a patient; in this way, patients not directly treated by credentialed nurses may still benefit from their competency. Further, hospital administrators may organize staff schedules to capitalize on these secondary benefits, thereby extending the effect of an individual nurse’s credential to matters of work organization.

Finally, the Nurse Performance pathway represents the means by which an individual or organizational credential can affect the outcome of patients directly under the care of that individual nurse or organization. The Patient Care box adds a layer of complexity to this pathway by mediating the relation between the performance of credentialed nurse and the outcome of the patient. Thus, credentialed nurses leverage their special capabilities to improve their performance as nurses; in turn, this augmented performance leads by way of higher quality patient care to enhanced patient outcomes.

These three pathways are not isolated from one another. Rather, the pathways’ components interact to create myriad associations and causal mechanisms. Invisible architecture affects not only organizational outcomes, but also nurse outcomes, patient care, and work organization. Nurse outcomes interact reciprocally with patient care, and are affected by each of the three intervening variables. In this way, the Expanded Conceptual Model manages to capture the interactions among outcomes

and credentials in a simple but nonreductive way. The model encourages focused and theory-driven research, without downplaying the profound complexity of nursing and the health care environment.

The Importance of Funding

The complexity of the interactions depicted in the Expanded Conceptual Model highlight the difficulty of determining whether and how credentials are causally related to outcomes. Research that seeks to map causal mechanisms between nursing credentialing and outcomes is important to the larger nursing credentialing research agenda. The credentialing process is costly and “if people are being asked to spend money to get credentials … they want to know that they are paying for something that is actually producing the outcomes that they are getting,” said Needleman. The value of the Expanded Conceptual Model lies in its potential to direct and focus research, allowing it to convincingly demonstrate the means by which credentialing affects outcomes and, thereby, provide potential funders with justification for investment in nurse credentialing and related research.

Stakeholders obtain and, therefore, value credentialing for different reasons, Needleman explained. For example, nurses seeking a credential may do so because of the way it enhances “their feelings of competency, their interest in their work, [and] their sense that there is joy and pleasure in work.” Organizations may acquire credentials if they believe that it reduces cost or turnover. Consumers who believe that credentials signal safe and high quality care will preference the institutions and individuals with those credentials. Table 2-1 lists the value of credentialing as perceived by different stakeholders.

These stakeholders become potential funders in credentialing research when studies are designed to target specific stakeholder values. By indicating how a unique stakeholder can benefit from credentialing, it allows scientists to tailor research objectives to the interests of consumers, care providers, and health care organizations. The singular importance of funding to credentialing research is acknowledged in the “Business Model” box at the bottom of the Expanded Conceptual Model, into which all causal pathways feed. By explicitly acknowledging the role of funding within the context of nurse credentialing research, the Expanded Conceptual Model emphasizes the importance of framing stakeholder interest and opinion within the credentialing research agenda.

Just as all the pathways in the revised framework lead to the “Business Model” box (and the financial questions it implies), each of the key issues and challenges emerge from concerns about funding, suggested Needleman. To perform high-quality research and produce convincing conclusions, more rigorous methods and larger and more robust data samples are required. Researchers will also need to consider actual demand among stakeholders for credentialing research and the most effective means of research dissemination.

Neither a conceptual framework that identifies potential causal pathways connecting credentials to outcomes, nor the stakeholder-targeted research it is designed to inform, are comprehensive solutions to the questions and challenges of nurse credentialing research. Rather, they constitute significant and necessary steps along the road to a research agenda and the potential it holds for improved outcomes for patients, nurse, and institutions alike, concluded Needleman.

TABLE 2-1 Perceived Value of Credentialing by Stakeholder

SOURCE: Needleman, 2014.

QUESTIONS AND COMMENTS

Is attaining a credential more valuable (to an individual or organization) than simply having the standards and attributes associated with credentialing?

The questioner, Joanne Spetz, proposed that a credential may confer three additional benefits: transparency (i.e., greater assurance that the standards of the credentialing organization are followed), enforceability, and monitoring. She added that, if an organization’s human resources program wants to move its workforce in a particular direction, then support of credentialing clearly signals that desire, and individuals obtaining the credential show they comport with institutional goals.

Needleman agreed, responding that the credentialing process potentially provides a mechanism for identifying standards and for moving people toward achieving those standards. He continued, the equally important question of whether there should be an attempt to measure the value of the credentialing journey, as well as the end state, remains unanswered.

In addition to analyzing outcomes related to an individual nurse with a particular credential, are there outcomes associated with a team of nurses or interprofessional teams (e.g., including physical therapists, nurses) of providers, all of whom are credentialed?

Needleman said the authors of the framework paper generally recognized the growing importance of teams working together for the good of the patient. However, team practice and work organization patterns can complicate credentialing research. For example, patients at high risk for falls will continue to experience a greater number of falls (though lower than they would have had otherwise) even when they are cared for by a certified “fall experts,” simply because they are at an elevated risk from the outset. Therefore, a simple comparison of fall rates among patients of certified nurses with fall expertise and those without, could produce a misleading finding, as it may not capture the true impact of the certified team of experts.

Moreover, having special expertise on a unit allows the opportunity for informal or formal consultation and unit-based education that results in raising the expertise of all unit personnel, credentialed or not, said Needleman. Such knowledge sharing may decrease the observable performance differences between credentialed and noncredentialed staff members, again making it difficult to assess the true impact of credentialed providers.

As care becomes more interprofessional, should we move to credentialing teams—perhaps as a category between individual and organizational credentialing?

Needleman acknowledged that this is a vital question that deserves “substantial reflection.” In inpatient and outpatient care, group- or activity-specific credentialing may be possible for some distinct units and activities. These opportunities include increasingly specialized care.

Moreover, for research on the overall team performance to be feasible, unit staff would have to be relatively stable over time, continued Needleman. In reality, team compositions vary from week to week, depending on patient and institutional staffing needs. In some cases, care is provided by ad hoc groups, making team-based analyses nearly impossible. Large electronic health records systems may, in the future, develop the capacity to match human resources information on staff credentials with that individual’s respective assignments. However, current data systems do not capture this information.

Which data elements from electronic health records would be sought for this research?

Needleman suggested that, at a minimum, researchers would need to be able to identify specific credentials, including baccalaureate preparation, beyond credentials such as RN or licensed practical nurse. Furthermore, researchers would want to know which nurses are assigned to which patients, and to have enough patient-level data to understand outcomes that might be associated with credentialing. Finally, electronic records data might help build an understanding of team effects. Although obtaining these data may be somewhere down the road, Needleman advised starting to think about these information needs now.

Have you looked at the experience of other industries that require certification and credentialing, such as aviation or nuclear power?

The IOM Standing Committee on Credentialing Research in Nursing has had the benefit of public presentations regarding other industries’ credentialing programs, Needleman said. Understanding how much of the experience from these settings and the nature of that work is applicable to health care remains a challenge. However, some crew resource management techniques used in aviation have been used to improve communication patterns in health care.

This page intentionally left blank.