Nursing management minimum data set: cost-effective tool to demonstrate the value of nurse staffing in the big data science era.
The development of information systems increased the opportunity to acquire standardized data sets, which enable descriptions and comparisons of patient care and discovery of nursing knowledge to enhance decision making. From an administrative perspective, such data analytics could evaluate nursing services effectiveness for cost and quality. Administrators are expected to quantify the value of provided services within the context of care. However, standardized nursing administrative and management data are needed to support analysis and understanding of nursing management within and across care settings. The NMMDS is a standardized data set to support management analytics.
The Nursing Management Minimum Data Set
The NMMDS builds on the clinical data focus of the NMDS. The NMMDS is a minimum collection of core variables needed by nurse managers to make decisions and compare nursing practice across institutions and geographic areas (Simpson, 1993). The NMMDS is a profession-specific data set, developed through grounded theory, expert review panels, national Delphi surveys, focus groups, cross-sectional descriptive surveys, and a national consensus working conference (Huber, Schumacher, & Delaney, 1997). It identifies common elements to represent nursing care delivery or the context of nursing care at the unit level or service line in any setting. The NMMDS provides managers and administrators standardized administrative data that can be compared within and across settings to understand how the context of care and nursing care resources can influence patient and staff outcomes. The NMMDS focuses on the context of care to describe the overall setting or experience in which care takes place, such as home care, acute care, long-term care, and ambulatory or alternative care sites, as well as the nursing care resources within these environments.
Over the past 7 years, the NMMDS was updated, normalized with other national standards, and coded within the national standard--the Logical Observation Identifiers Names and Codes (LOINC) and is publicly available for use (Delaney, 2015). LOINC applies universal code names and identifiers to health terms related to laboratory and other clinical observations; it is now extended to include management data. LOINC is one of the standards recommended by the U.S. Government for use in electronic exchange of clinical health information. The availability of the NMMDS coding via LOINC ensures access to coded management data, capturing the essential context of care and nursing resources, and enables associating clinical data available through the NMMDS to the context of care. The NMMDS coding and definitions are readily available without charge to support universal adoption (http://search.loinc.org).
The NMMDS adoption and implementation in information systems, such as human resources and scheduling systems linked with electronic health records (EHRs) and other data sources, will support measurement of health outcomes in a timely manner and examination of the relationship of administrative and management data on these clinical outcomes, including the environment where care is delivered, nursing resources, scheduling, and acuity of care data. For instance, nurse staffing identifiers in EHRs with date/time stamps and locations can link to staffing data in scheduling and human resource systems as well as to patient identifiers and dates of service. In turn, patient addresses can be linked to geocodes, which provide connections to neighborhood income, education, and crime rate reports. As a result of this integration, the NMMDS provides useful data to examine the influence of the context of care and the value of nursing to achieve people's health.
Nursing Big Data Science
Big data are not new; however, harnessing the power of big data through standardized data and processes, as well as using a variety of methods for demonstrating the value of nursing, is emerging. Big data are typically defined by the "4Vs:" Volume, Velocity, Variety, and Veracity (Bellazzi, 2014). Volume is the shear amount of data, while velocity is the rate at which data accumulate. Variety refers to the type of data, such as structured and semi-structured nursing documentation, monitoring devices, genomics, imaging studies, scheduling and human resource data, and patient-generated data. Veracity is the uncertainty of the data, either in terms of accuracy of the data for the original purpose for which the data were collected or appropriateness for secondary use of the data (Bellazzi, 2014; Brennan & Bakken, 2015). The veracity for secondary use of nursing big data can be enhanced by applying standardized terms, definitions, and codes, such as the NMMDS. The scholarship of working with large data sets is known as data science, which draws on knowledge and theories from nursing and multiple disciplines (Brennan & Bakken, 2015). Big data science uses a variety of methods for analyzing data from traditional statistics to visualization techniques, data mining, and natural language processing (Stanton, 2012; Wang & Krishnan, 2014). Use of big data is assuming a critical role in health care analytics as the complexity and variety of data available are increasing and advanced computational methods and tools to analyze these data are available.
While there are numerous studies of nurse staffing using large data sets, the majority rely on original data collection which can be expensive. For example, Lake and colleagues collected original data to describe the variation in neonatal intensive care units (NICUs) in acuity-adjusted nurse staffing levels, and to link staffing and nurse practice environments to very low birthweight infant outcomes and outcome disparities (Lake et al., 2015; Rogowski et al., 2013). Two studies were conducted by this research team with data collected via web-based surveys of 7,038 NICU nurses assigned to 15,191 infants at 101 hospitals, and a second study in which the researchers collected prospective data on one shift once per quarter from 70 hospitals. To fulfill one aim, the researchers examined infant acuity; nurse education, certification, and experience; and additional personnel to determine their association with nurse workload. Infant acuity was the only determinant of nurse workload. Funding for the two studies totaled $2.6 million. With the advancement of data analytic methods, there is an increasing emphasis for researchers to use existing data sets, including big data generated in health care settings. Moreover, reuse of data is cost effective as shown by comparing two research approaches. Another research team led by Westra (Bliss, Westra, Savik, & Hou, 2014; Westra, Bliss, Savik, Hou, & Borchert, 2013) used secondary analysis of the Outcome and Assessment Information Set (OASIS) and staffing data to identify the influence of a wound, ostomy, continence certified nurse on agency-level and patient-level outcomes (surgical wounds, pressure ulcers, decubiti, urinary incontinence, bowel incontinence, and urinary tract infections). Data were collected from 785 home care agencies' EHRs resulting in analysis of 449,243 episodes of care. The cost of the study was $200,000. Research using big data can be more cost effective; however, standardization of both the data definition and measurement as well as the process for data capture in EHRs and other information systems is essential to support nursing big data science.
Value of Implementing NMMDS: C-Suite Perspective
Nursing is essential to delivering quality, cost-effective, and safe health care. The contribution of nurses to health care has been documented through the rigors of research and the anecdotes of thousands of patients and consumers. The evidence linking nurse staffing with patient outcomes has been established; however, incorporating this evidence about staffing into decision making for practice is still lacking (Anderson, Frith, & Caspers, 2011). The delay of translating research into practice continues today even with increasing data sources tying payment to performance (Welton & Harper, 2015) and increasing pressure to provide nursing care across the care continuum in a manner that minimizes nursing costs and promotes quality. The need for evidence-based, outcome-driven nurse staffing is a strategic priority for today's health care enterprise. Decision support tools for nurse administrators, such as dashboards, provide key information to inform operational decision making (Dowding et al., 2015). Implementing the NMMDS can be the foundation for excellence and evidence in enterprise staffing by providing administrative decision support for timely determination of health outcomes, cost effectiveness, and the quality of nursing care.
The value of the NMMDS is that it is an evidence-based tool that defines and measures the context of care within and across all enterprise settings (Huber et al., 1997). NMMDS analytics provide data about variables impacting the delivery of timely, cost-effective care. This knowledge is critical to achieving and maintaining clinical and operational excellence. The data set supports staffing excellence and enterprise performance improvement initiatives, which nurse executives and their C-suite colleagues are routinely requested to provide for data-driven decisions. The NMMDS provides management data that can be linked with clinical data to understand how the environment and processes of care can influence staff outcomes as well as patient outcomes. These data are instrumental in supporting effective workforce management practices and in developing effective staff recruitment and retention strategies. The data elements in the NMMDS are already captured in most health care enterprise information systems and can readily be transferred to enterprise clinical data repositories (Evans, Lloyd, & Pierce, 2012). The capture of the NMMDS data set linked to EHR data provides enterprise-wide essential data about care delivery and outcomes.
Health Information Technology and NMMDS Implementation
Involving health information technology (HIT) vendors and personnel is essential for successful implementation of the NMMDS. Big data science and technologies can support integration of relevant staffing data like the NMMDS from health system human resource, scheduling, time and attendance, productivity management, case management, and pharmacy information systems. However, if nursing leaders do not request, advocate for, and insist HIT vendors and their HIT staff design and implement data standards, measuring the value of nursing within and across systems can be compromised. Communication, collaboration, decision making, and resource use of the NMMDS and all health data are optimized when standardized and integrated across information systems. To support implementation of the NMMDS, HIT must be widely accessible and used to ensure timely access to the data set. C-suite leaders must partner with technology vendors to innovate and leverage technology to advance staffing excellence across the enterprise and create a competitive market advantage (Douglas, 2008).
Driven by the current value-based health ecosystem, the exploding use of HIT, and the digitalization of health information, the evolution of big data and big data nursing science is emerging and becoming a key competitive strategy for large, integrated health systems. This phenomenon is raising long-standing questions about how to define and measure the value and contributions of nurses to patient, family, and community health outcomes and raising questions about staffing excellence. New questions are emerging about how to define and measure the value of nursing.
NMMDS and Its Value for Sharable and Comparable Staffing Data
The value of the NMMDS was identified by an expert panel of chief nurse executives at the "Nursing Knowledge: Big Data Science" conference convened by the University of Minnesota School of Nursing. The purpose of this expert panel was to examine emerging issues related to big data and nursing knowledge development aimed to define and measure the tangible products of nursing value (University of Minnesota School of Nursing, 2015). The panel recommended changing the view of nursing from that of a staffing model to one where each nurse is viewed as a unique provider of care. This ongoing work drives review of nursing care based on best performance and outcomes (Welton & Harper, 2015).
Asserting nursing is essential to health care value, nursing value can be defined as the function of health outcomes divided by cost (Pappas & Welton, 2015). This definition drives the need to better define the measures and analytics for patient-level costs and outcomes of nursing care. This new thinking propels measuring the value of nursing away from tasks, ratios, and staffing levels to an outcomes-driven staffing model. This model for staffing excellence, capturing the patient or consumer outcomes of nursing care, presumes measurement of the clinical and financial impact of nursing care. The result is a measurement of true nursing value (Pappas & Welton, 2015). Integration of NMMDS elements into efforts to define nursing value enhances emerging models of staffing excellence and evidence. The data set provides a common, consistent set of measures for sharing and comparing nursing impact on health outcomes across the care continuum and across the health care enterprise. Implementation of the NMMDS elements in EHRs is foundational to evolving evidence in staffing.
One clinical resource management model for best value nursing care integrates the four core clinical operations concepts of productivity, cost, acuity, and patient outcomes with six points of care technology-enabled, standardized business processes (Caspers & Pickard, 2013). Building on evidence and decision support at the point of care, the underlying assumption is integration and management of these four ideals result in clinical and operational excellence. The end result is robust patient-level data delivered real time at the point of care. These data enable nurses to assign the right caregiver to the right patient at the right time and to implement staffing patterns that produce the most effective resource utilization and optimal patient outcomes (Caspers & Pickard, 2013). Integrating data elements of the NMMDS with current point-of-care technologies promotes staffing excellence by advancing sharable and comparable staffing analytics across the care continuum and across all health care enterprise settings. The NMMDS provides an established, recognized infrastructure for organizing workforce data (Garcia, Caspers, Westra, Pruinelli, & Delaney, 2015).
Practical Use of NMMDS Elements for Staffing
There is a growing body of evidence of the relationship of nurse staffing to patient, nurse, and financial outcomes. Douglas (2009) observed, "If staffing were only a question of numbers, this would be so much easier to address. But staffing is extremely complex with many things at play..." (p. 418). Depending on the care setting, scheduling and staffing based on evidence requires information from each element of the NMMDS (Garcia et al., 2015) shown in Figure 1.
Nurse leaders have struggled to assemble and translate the data elements in a meaningful way, as the data sets are large, the report formats are difficult to align, and the systems are rarely integrated. Nurse leaders have worked with industry to identify gaps and develop systems to gather and transform data into new information. Capacity Management, a Cerner Clairvia component, is an example of a system that integrates elements of the NMMDS from multiple disparate sources to provide actionable, real-time, decision support for nurse scheduling, staffing, and patient throughput.
Clairvia is mapped at the interface level to integrate data and generate new information from eight of the NMMDS elements. Figure 2 is a conceptual model of Clairvia, illustrating the typical data sources for each of the elements. The nursing schedule (described as the supply of nurses) draws data related to Elements 2: Type of Unit, 4: Volume of Service, and 13: Staffing. The patient demand for care includes acuity, the index of admissions, discharges and transfers, and realtime census. This uses data found in the NMMDS data element 1: Unique Patient Identifier, 3: Patient Population, and 5: Care Delivery Structure and Outcomes. Staffing occurs when an individual nurse is assigned to an individual patient or group of patients. It is critical to know the licensure and skills of the individual, and this draws data related to the NMMDS data element 10: Accreditation, Licensure, and Certification and 19: Nursing Demographics.
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
Data for each element may come from one or more systems, often in different departments. Data related to the NMMDS data element 10: Nurse Accreditation/Certification/Licensure, is typically found in disparate systems. Human resources often maintains licensure and years of experience, while nurse competencies may be in an e-learning system, and certifications on a noncentralized spreadsheet in a nurse manager's computer. It can be difficult to identify a nurse with a specific combination of qualifications quickly, such as chemotherapy certification and advanced cardiac life support certification or language proficiency. Ideally, this information would be visible while building a schedule, assigning individual nurses to individual patients, and when analyzing patient outcomes.
Integration of data from multiple sources helps nursing leaders balance workloads, improving patient outcomes and nurse satisfaction. Figure 3 illustrates transparency of data from multiple sources. In this example, hovering over Karen Arn's name provides visibility to her competencies. Each patient has a real-time measure of the workload associated with acuity and transitions, so these are compiled into assigned hours and percent assigned. The nurse leader is easily able to consider continuity of care, tasks like narcotics counts, and relief for breaks and meals. The system is integrated with communication and alerting devices to enable a smooth flow of information. The resulting workloads are translated into hours per patient day or ratios.
Assembling data from multiple sources can be a time-consuming task for nurse leaders. Technology that integrates elements of the NMMDS for staffing or other complex management decisions can return valuable hours to the nurse leader. Not every facility has technology like Clairvia, but many do have data warehouses. Unless data are integrated into and extracted from the warehouse in a meaningful way, nurse leaders will struggle to achieve better patient, nurse, and financial outcomes.
[FIGURE 3 OMITTED]
The NMMDS provides standard administrative data elements, definitions, and codes to measure the context where care is delivered and, consequently, the value of nursing. With the advent of big data science and developing big data analytics in nursing, data science with the reuse of big data is emerging as a timely and cost-effective approach. With the scarcity of research funding opportunities, future research should consider this time and cost-effective approach when building and conducting research with the reuse of data. Finally, the integration of the NMMDS elements in the current health system provides evidence for nursing leaders to measure and manage decisions that lead to better patient, staffing, and financial outcomes, and enables the reuse of data for clinical scholarship and research.
Anderson, E.F., Frith, K.H., & Caspers, B. (2011). Linking economics and quality: Developing an evidence-based nurse staffing tool. Nursing Administration Quarterly, 35(1), 53-60. doi:10.1097/NAQ.0b013e3182047dff
Bellazzi, R. (2014). Big data and biomedical informatics: A challenging opportunity. Yearbook of Medical Informatics, 9, 8-13. doi:10.15265/IY-2014-0024
Bliss, D.Z., Westra, B.L., Savik, K., & Hou, Y. (2014). Effectiveness of wound, ostomy and continence-certified nurses on individual patient outcomes in home health care. Home Healthcare Nurse, 32(1), 31-38. doi:10.1097/NHH.000000 0000000012
Bodenheimer, T., & Sinsky, C. (2014). From triple to quadruple aim: Care of the patient requires care of the provider. Annals of Family Medicine, 12(6), 573-576. doi:10.1370/afm.1713
Brennan, P.F., & Bakken, S. (2015). Nursing needs big data and big data needs nursing. Journal of Nursing Scholarship, 47(5), 477-484. doi:10.1111/jnu.12159
Caspers, B.A., & Pickard, B. (2013). Value-based resource management: A model for best value nursing care. Nursing Administration Quarterly, 37(2), 95104. doi:10.1097/NAQ.0b013e318286 9e17
Centers for Medicare & Medicaid Services. (2009). Medicare program; prospective payment system and consolidated billing for skilled nursing facilities for FY 2010; minimum data set, version 3.0 for skilled nursing facilities and Medicaid nursing facilities. Final rule. Federal Register, 74(153), 40287-40395.
Delaney, C. (2015). Implementation guide for the Nursing Management Minimum Data Set. Retrieved from http://www. nursing.umn.edu/prod/groups/nurs/ @pub/@nurs/documents/content/ nurs_content_498749.pdf
Delaney, C., Reed, D., & Clarke, M. (2000). Describing patient problems & nursing treatment patterns using nursing minimum data sets (NMDS & NMMDS) & UHDDS repositories. Proceedings AMIA Annual Symposium, 176-179.
Douglas, K. (2009). Getting real on evidence-based staffing. Nursing Economics, 27(6), 416-418, 421.
Douglas, K. (2008). Report out from the excellence and evidence in staffing roundtable. Kansas City MO: Concerro, Inc.
Dowding, D., Randell, R., Gardner, P., Fitzpatrick, G., Dykes, P., Favela, J., ... Currie, L. (2015). Dashboards for improving patient care: Review of the literature. International Journal of Medical Informatics, 84(2), 87-100. doi:10.1016/j.ijmedinf.2014.10.001
Evans, R.S., Lloyd, J.F., & Pierce, L.A. (2012). Clinical use of an enterprise data warehouse. Proceedings AMIA Annual Symposium, 189-198.
Garcia, A., Caspers, B., Westra, B., Pruinelli, L., & Delaney, C. (2015). Sharable and comparable data for nursing management. Nursing Administration Quarterly, 39(4), 297-303. doi:10.1097/ NAQ.0000000000000120
Huber, D., Schumacher, L., & Delaney, C. (1997). Nursing management minimum data set (NMMDS). The Journal of Nursing Administration, 27(4), 42-48.
Huber, D.G., Delaney, C., Crossley, J., Mehmert, M., & Ellerbe, S. (1992). A nursing management minimum data set. Significance and development. Journal of Nursing Administration, 22(7-8), 35-40.
Lake, E.T., Hallowell, S.G., Kutney-Lee, A., Hatfield, L.A., Del Guidice, M., Alan Boxer, B., ... Aiken, L.H. (2015). Higher quality of care and patient safety associated with better NICU work environments. Journal of Nursing Care Quality, 31(1), 24-32. doi:10.1097/NCQ.000000 0000000146
Murray, J.E., & Barnes, B.A. (1971). Organ transplant registry. JAMA, 217(11), 1546-1547.
Pappas, S., & Welton, J.M. (2015). Nursing: Essential to health care value. Nurse Leader, 13(3), 26-29, 38.
Rogowski, J.A., Staiger, D., Patrick, T., Horbar, J., Kenny, M., & Lake, E.T. (2013). Nurse staffing and NICU infection rates. JAMA Pediatrics, 167(5), 444-450. doi:10.1001/jamapediatrics. 2013.18
Simpson, R.L. (1993). Technology: Nursing the system. A nursing management minimum data set. Nursing Management, 24(4), 24-25.
Stanton, J. (2012). An introduction to data science. Syracuse, NY: Syracuse University.
Stoker, J. (2003). CMS announces OASIS-B1 (12/2002) data set approved. Home Healthcare Nurse, 21(4), 216.
University of Minnesota School of Nursing. (2015). Proceedings of the Nursing Knowledge: Big Data and Science for Transforming Health Care. Retrieved from http://www.nursing.umn.edu/ prod/groups/nurs/@pub/@nurs/ documents/content/nurs_content_504 042.pdf
Wang, W., & Krishnan, E. (2014). Big data and clinicians: A review on the state of the science. JMIR Medical Informatics, 2(1), el. doi:10.2196/medinform.2913
Welton, J.M., & Harper, E.M. (2015). Nursing care value-based financial models. Nursing Economics, 33(1), 14-19, 25.
Werley, H.H., & Lang, N.M. (Eds.). (1988). Identification of the nursing minimum data set. New York, NY: Springer Publishing Company.
Westra, B.L., Bliss, D.Z., Savik, K., Hou, Y., & Borchert, A. (2013). Effectiveness of wound, ostomy, and continence nurses on agency-level wound and incontinence outcomes in home care. Journal of Wound, Ostomy, and Continence Nursing, 40(1), 25-53. doi:10.1097/WON.0b013e31827bcc4f
LISIANE PRUINELLI, MSN, RN, PhD(c), is Research Assistant, School of Nursing, University of Minnesota, Minneapolis, MN.
CONNIE W. DELANEY, PhD, RN, FAAN, FACMI, is Professor and Dean, School of Nursing, University of Minnesota, Minneapolis, MN.
AMY GARCIA, MSN, RN, is Director and Chief Nursing Officer of Workforce Capacity Management, Cerner Corporation, Kansas City, MO.
BARBARA CASPERS, MS, RN, PHN, is Owner and Health Care Management Consultant, Barbara Caspers Associates, Greater Minneapolis-Saint Paul, MN.
BONNIE L. WESTRA, PhD, RN, FAAN, FACMI, is Associate Professor, School of Nursing; Co-Director, Center for Nursing Informatics; and Core Faculty, Institute for Health Informatics, University of Minnesota, Minneapolis, MN.
|Printer friendly Cite/link Email Feedback|
|Author:||Pruinelli, Lisiane; Delaney, Connie W.; Garcia, Amy; Caspers, Barbara; Westra, Bonnie L.|
|Date:||Mar 1, 2016|
|Previous Article:||Connie Curran--continuing the legacy: Scholarships Planned in Connie's Honor.|
|Next Article:||Cost-effective staffing for an EHR implementation.|