Nursing Knowledge: Big Data Research for Transforming

Nursing Knowledge: Big Data Research for Transforming

University of Minnesota, School of Nursing, Center for Nursing Informatics

August 12 – 13, 2013

Developed by: John Welton, PhD, RN

Nursing and the Value Proposition:
How information can help transform the healthcare system

The Wisdom of Nightingale 150 years later:

In attempting to arrive at the truth, I have applied everywhere for information, but in scarcely an instance have I been able to obtain hospital records fit for any purposes of comparison. If they could be obtained, they would enable us to decide many other questions besides the one alluded to. They would show subscribers how their money was being spent, what amount of good was really being done with it, or whether the money was not doing mischief rather than good; they would tell us the exact sanitary state of every hospital and of every ward in it, where to seek for causes of insalubrity and their nature; and, if wisely used, these improved statistics would tell us more of the relative value of particular operations and modes of treatment than we have any means of ascertaining at present. They would enable us, besides, to ascertain the influence of the hospital with its numerous diseased inmates, its overcrowded and possibly ill-ventilated wards, its bad site, bad drainage, impure water, and want of cleanliness - or the reverse of all these - upon the general course of operations and diseases passing through its wards; and the truth thus ascertained would enable us to save life and suffering, and to improve the treatment and management of the sick and maimed poor.

Florence Nightingale (1863)1

A.  Building a better healthcare system

How do we build a better healthcare system to achieve patient-centered, effective, safe, timely, efficient, and equitable care?2 Extending this vision, how do we build a better nursing care delivery system and how do we build a better nurse to meet the needs of 21st century healthcare?

The current challenge is different from the one Nightingale faced – we now have the data available in a plethora of electronic health record systems to answer the questions above. The compelling issue is how can we use these data to:

1. Develop efficient, effective, productive, data-driven and value-driven nursing care systems?
2. Compare and benchmark nursing care across many different settings and identify and emulate best performance?
3. Achieve exceptional outcomes and results of nursing care?
4. Provide the best nursing care at the least costs and best price?

Nurses represent the largest number of healthcare professionals in the nation and hospital-based registered nurses alone represents 25% of all hospital expenditures and a quarter trillion dollars in healthcare costs annually.3 Small changes in the amount or quality of nursing care produces a large effect in the overall cost structure of healthcare.4 If we are to achieve a more productive, efficient, effective, higher performing, lower cost, and higher quality nursing care delivery system, better and timelier information are needed to guide these efforts.

Table 1 Problems and Solutions

Problem / Potential Solutions
·  Optimum nursing care is unknown
·  Nursing system and individual nurse performance, efficiency, effectiveness, etc., unknown relative to a patient or healthcare setting
·  Lots of data about patients and diseases, little information about the delivery and results of nursing care
·  The "true" cost of nursing care for each patient is unknown, at best, nursing hours and costs are averaged across all patients (nursing hours/costs per patient day)
·  No relationship between the cost of nursing care with billing and payment for services to patients
·  Nursing has little if any economic value and added value of individual nurses to each patient is unknown – there is no price for nursing care / ·  EHR: Link nursing, clinical, operational, financial, outcomes data systems (operational informatics)
·  Identify each nurse caring for each patient (assignment data)
·  Develop nursing business intelligence and analytic tools
·  Allocate actual nursing hours and cost of care to each patient (true nurse costing)
·  Link costs of nursing care to billing and reimbursement = best payment for optimum nursing care
·  Incorporate nurses and nursing care into Accountable Care Organizations and bundled payment
·  Develop nursing pay-for-performance models
·  Restructure nursing delivery models to achieve better value

B.  Recommendation 1: Link nurses to patients in the EHR

The missing link to achieve the broader goal of better nursing care is to understand how individual nurses affect the care and outcomes of each patient. Prior studies have relied on aggregate measures of nursing care such as nurse to patient ratios or average nursing hours and costs per patient day. 5 While these studies have demonstrated an association of these levels and patient outcomes, more precise measures of the direct effect of nurses are needed.

Nearly every healthcare setting has a method to track which nurses are caring for patients. For example in hospitals, the nursing assignment is the primary method to organize care. These data are ubiquitous and collected in essentially the same way across hospitals as well as other healthcare settings. With the emergence of the EHR, there is a potential to capture the nurse-patient assignment in the electronic database along with other clinical and operational data will allow further analysis of individual nurses and patients. More important, the assignment alone can account for most the actual hours of care and costs of that nursing care consumed by each patient.6 For example, having data about the nursing assignments it will be possible to ascertain the actual hours of nursing care delivered to each patient. These data can be used to determine productivity, efficiency, performance, effectiveness, costs of nursing care, and ultimately the value of nursing care at a more refined level than traditional measures (Table 2).

Table 2 Nursing Value Dimensions

Productivity / ·  Maximize nursing time/$ dedicated to patients/patient care
·  See Appendix A
Efficiency / ·  Minimize waste of nursing time/$
·  Compare patient need with efficiency and effectiveness (Appendix B)
Patient Acuity / ·  Identify need for nursing care
Nursing/Nurse Performance / ·  Optimize the process of nursing care
·  See Appendix C
Effectiveness / ·  Optimize the practice of nurses and nursing care to meet patient needs
Quality/Safety / ·  Minimize “defects” of nursing care
Assignment and Staffing / ·  Assign right nurse for the patient
·  Provide the necessary nursing resources to meet changing patient needs
·  Compare individual nurses and aggregate unit performance
·  (Appendix D)
Nursing Finance / ·  Assure best nursing care at the best price
·  Identify nursing added value that decrease patient costs of care
Outcomes / ·  Promote superior results of nursing care

Patient level nursing time and costs derived from the nurse-patient assignment has several advantages over traditional methods that aggregate nursing across many patients. First, every patient has different nursing care needs across a care episode. Using traditional average costing and intensity[§] methods hides this natural variability. Second, costing nursing care based on the actual nursing resources provided each patient is a more precise measure of "true" nursing costs and intensity. Third, linking each patient in existing EHRs using a unique nurse identifier allows further data about the nurse to be allocated to the patient. For example, a nurse's experience, academic preparation, etc. can be used to better understand the effects of nurse characteristics on patient outcomes. Last, the ability to measure direct nursing time and costs for each patient opens a new potential to measure and monitor nursing productivity, efficiency, performance, effectiveness, and finance in ways that have not been possible.

C.  Recommendation 2: Develop nursing business intelligence and analytic methods and tools

Rita Zielstorff advocated future nursing information systems be designed to collect data once and then use many different times and in many different ways.7 Such a system would combine data from primary clinical, operational, billing, outcomes, etc., systems for secondary use to better understand and improve the healthcare system. For example, electronic medication administration and bar coding is used to streamline the delivery of drugs to patients by nurses. The performance of how well this nursing activity is carried out is measured in part by the actual versus planned time for the medication. The actual versus scheduled time difference can be used to examine overall nurse or unit performance (Appendix C). Such secondary use of clinical and operational data can inform managers and other nursing leaders of the overall unit performance. For example if average time lags begin to increase, this could be due to workload factors, changes, in patient acuity, etc. Out of range medication administration such as giving aminoglycoside antibiotics too soon or too late could lead to toxicity or ineffectiveness and are indicators of potential poor quality of nursing care.

A second example is using nursing time and costs derived from clinical and operational data, e.g. nursing assignment, acuity based on Nursing Outcomes Classification (NOC) and clinical indicator mapping, and the new information is used for both real-time management of nurse staffing and assignment and secondary purposes of nursing performance monitoring. Catholic Health Initiatives has prototyped and deployed such a system, named Care Value System (CVS) across several hospitals.8, 9 Real-time status of patient nursing care needs is used by charge nurses, managers, supervisors, and nursing and hospital leadership teams to understand how nursing resources are used for patient care. The NOC data provides patient level nursing care needs and the nurse-patient assignment shows the actual nursing intensity for each patient. These data are used to optimize both staffing and the assignment of nurses to patient. Two examples of using such data to determine the relationship between nursing efficiency and effectiveness and the performance of a single nurse are shown in Appendix B and Appendix D respectively.

One additional worth of mention is the true costs of nursing care. Summary methods using NHPPD and NCPPD cannot detect the individual variability and costs of nursing care for each patient. This has been a longstanding problem that has roots in the 1930's when modern hospital accounting and billing practices were developed.10 Hospitals and other healthcare settings that are able to determine the actual or true costs of nursing care per patient will have a competitive advantage and possibly position themselves to more efficient and effective in the coming years as healthcare reform takes hold. Better information about the amount and cost of nursing care compared to new benchmarks of performance and effectiveness will be needed to meet the vision and goals of a better healthcare system. The data derived from information systems about nursing care can and should be used in innovative ways to facilitate a better nursing care delivery system.

D.  Summary

Nurses collect a substantial amount of data about patients in the course of their daily practice. Until recently, these data have been difficult to process and analyze because they were bound to paper records. With widespread adoption of Electronic Health Records (EHR), there is an emerging capability to use these data in combination with other clinical, operational, billing, and outcomes data to form a more complete picture of patient care.

E.  References

1. Nightingale F. Notes on Hospitals. London: Longman, Green, Longman, Roberts & Green; 1863. Available at http://books.google.com/books?hl=en&lr=lang_en&id=FJhN-SqxUawC&oi=fnd&pg=PA171&dq=florence+nightingale&ots=8eOKJ9M4xL&sig=4qLze0RK8rugdzA4A29CxjyUZf4

2. Institute of Medicine. Crossing the Quality Chasm: A New Health System for the Twenty-first Century. Washington: National Academies Press; 2001.

3. Welton JM. Hospital nursing workforce costs, wages, occupational mix, and resource utilization. J Nurs Adm. 2011;41(7-8):309-314.

4. Kane NM, Siegrist RB. Understanding Rising Hospital Inpatient Costs: Key Components of Cost and the Impact of Poor Quality. BlueCross BlueShield Association [serial online]. 2002. Available at: http://www.bcbs.com/blueresources/cost/4_Inpatient_Qual_Assess.pdf. Accessed August 1, 2009.

5. Kane RL, Shamliyan TA, Mueller C, et al. The Association of Registered Nurse Staffing Levels and Patient Outcomes: Systematic Review and Meta-Analysis. Med Care. 2007;45(12):1195-1204.

6. Welton JM, Sermeus W. Use of Data by Nursing to Make Nursing Visible: Business and Efficiency of Health Care System, and Clinical Outcomes. In: Weaver CA, Delaney C, Weber P, et al., eds. Nursing and Informatics for the 21st Century: An International Look at Practice, Trends and the Future. Chicago, IL: Healthcare Information and Management Systems Society; 2010.

7. Zielstorff RD, Hudgings CI, Grobe SJ. Next-Generation Nursing Information Systems: Essential Characteristics for Professional Practice. Washington, D.C.: ANA PUBL NP-83; 1993.

8. Caspers BA, Pickard B. Value-based resource management: a model for best value nursing care. Nurs Adm Q. 2013;37(2):95-104.

9. Pickard B, Warner M. Demand management: A methodology for outcomes-driven staffing and patient flow management. Nurse Leader. 2007;5(2):30-34.

10. Thompson JD, Diers D. Nursing resources. In: Fetter RB, Brand DF, Gamache D, eds. DRGs. Their Design and Development. Ann Arbor: Health Administration Press; 1991:121-183.

Appendix A Unit level measure of productivity (actual patient care hours vs total nursing hours)

Note: productivity is defined as the actual nursing care hours delivered to patients derived from the assignment (assigned vs. total hours of nursing care).

Appendix B Nursing efficiency (simulated data)

These data are derived from two sources, the first is an acuity measure that indicated patient need for nursing care in total hours per shift/patient day, the second is the actual hours of nursing care provided to patients based on the assignment (see Caspers, et. al. for further discussion8). The difference between the actual nursing hours versus required is termed the efficiency-effectiveness measure (EEM). An EEM of 1.0 means that patients are getting their needed hours of nursing care (balance between available vs. needed hours, for example if a patient needs 6 hours of care and was given 6 hours = 6/6 = 1.0). An EEM higher than 1.0 indicates more nursing care hours, e.g. in January EEM was 1.04 or about 4% more hours than actually needed. From an organizational standpoint, this small amount may be needed for organizational slack, e.g. to keep a bed open for anticipated or unexpected admissions.