writing 3

skimsibra

required.pdf

Home >Nursing homework help >writing 3

www.nursingmanagement.com Nursing Management • March 2019 15

NURSING INFORMATICS

The synthesis of nursing knowledge and predictive analytics

By Whende M. Carroll, MSN, RN-BC

A s healthcare organizations enter the maintenance and optimization phases of electronic health record

(EHR) implementation, the time has come for us to leverage the vast amounts of data generated by the EHR and associated tech- nology to improve information sharing and deliver excellent clin- ical care and patient experience. The evolution from simple data collection to aggregating, track- ing, trending, and analyzing big data to enhance care is in flight. Now, the ability to use even more advanced data manipulation tech- niques for care planning and de- livery is, in many cases, required to meet the needs of modern nursing practice.1 Through the ap- plication of emerging technolo- gies, such as predictive analytics and machine learning, nurses can add tremendous value to the fu- ture of care delivery and opera- tions.

Nurses as knowledge workers Nurses are knowledge workers, performing highly variable, fo- cused work that involves a signifi- cant amount of information.2 In our daily work, we use our spe- cialized nursing skills to compile, sift through, and find actionable solutions using disparate data sources and large datasets. With explicit knowledge of clinical sci- ence and by applying the nursing

process and critical thinking, nurses instinctively take discrete data elements and organize them into information to use in every patient experience. The applica- tion of our nursing knowledge and experience, married with suc- cessful data handling, allows us to make critical decisions at the point of care. The result is nurses disseminating wisdom and the

improved application of evidence- based practice, adding immeasur- able value to the clinical setting and moving toward improving the health of populations and communities. Through advanced data analytics, we can use this in- formation to our advantage and distribute the subsequent wisdom with greater impact.

Studies have shown that nurses spend upwards of 50% of their time recording and managing this assimilated information.3 By using acquired patient data, nurses gain information and apply knowledge to guide practice.4 Nursing knowledge identifies information and creates relationships so it can be synthesized and formalized.2

These relationships leverage the nurse’s ability to apply inferences to information and make a judg- ment to determine patient prog- ress toward expected outcomes or identify nursing problems and in- terventions appropriate for the challenge. A set of vital signs is information; however, the inter- pretation of that information as abnormal indicates knowledge.5

Increasingly, new ways of using data enhance the clinical experi- ence by allowing nurses to make informed, data-driven decisions.

Today’s nurses need constant involvement in technical innova- tion to stay current and forward- thinking in care delivery.6 To that end, technologic advances en- abled through the EHR, medical claims, patient prescription his- tory, and digital sensor data now allow nurses to provide more pre- cise, higher quality, and safer care. The application of emerging tech- nologies enables nurses to reap the benefits of data manipulated though nonhuman processing, ac- celerating and expanding nursing

16 March 2019 • Nursing Management www.nursingmanagement.com

NURSING INFORMATICS

knowledge generation and priori- tizing care based on patient needs.

Applied predictive analytics Advanced computational analysis of healthcare data, particularly pre- dictive analytics, can help nurses unearth unidentified trends within multiple sources of data. Predictive analytics is the statistical science of data analysis that discovers vari- ous patterns.7 By applying compu- tational models and analysis, nurses can draw on historical,

present, and simulated future data to provide actionable insights into real-world clinical and operational problems.8 Predictive analytics al- lows a machine approach to refine these data and extract hidden value from the newly discovered patterns to dynamically inform data-driven decision-making so we can know what will happen in healthcare settings, when, and what to do about it.9

Further robust exploration of data is needed to harness the power of prediction in clinical care. The addition of advanced algo- rithms through machine learning is a way to guide and standardize best practices and expedite treat- ment. Machine learning is the study of computer algorithms that improve automatically through ex- perience.10 It’s a form of artificial intelligence that enables software applications to become more accu- rate in predicting outcomes with- out being explicitly programmed.11

Machine learning methods take historical data and compare them with current data to predict what will happen in the future. With every refresh of new data from designated sources, the machine learns how to be more precise in predicting.10

Predictive analytics and machine learning in clinical care function as “assistive intelligence.”12 Nurses’ critical thinking is still needed to assess the clinical situation, synthe- size the derived information to

make the best decision, and put the decision into action. Although human judgment is paramount to the success of predicting trends and identifying variation, the use of algorithms is promising in at- taining the best outcomes, expound- ing on existing clinical decision support systems, and adding a helpful layer of precision. Look- ing toward the future, nurses can count on advanced technologies to drive cutting-edge, enhanced practices and research-based evi- dence to the point of care to help make the most complex clinical decisions with a higher degree of confidence.13

Using data for prediction Nurses have the influence to pro- actively adopt and expertly apply emerging technologies, adding value to care delivery by making the best data-driven decisions to improve outcomes and patient ex- perience. Using the assistive intelli-

gence of predictive analytics and machine learning along with nurs- ing knowledge can keep patients from: • rapid deterioration. Predictive analytics can help nurses identify when a patient is declining by sending a warning or risk score based on patient-specific data, such as vital signs and lab or ra- diology results, along with exter- nal data sources from sensors and remote devices.14 A machine- assimilated risk score, in addition

to patient assessment and presen- tation, quickly enables nurses to determine if the patient’s status is indeed declining, which allows us to begin immediate care, prevent further deterioration, and move the patient to a higher level of care if needed. • staying in the hospital for too long or not long enough. An aggre- gate of the patient’s demographics, comorbidities, number of medica- tions, and lab and vital signs val- ues derived from the EHR can de- termine the risk of readmission. Understanding a patient’s risk of rehospitalization powered by ad- vanced analytics such as machine learning will better enable nurses to personalize care, discharge plan- ning, and outpatient care needs earlier—all factors that can prevent rehospitalization.15 Conversely, with predictive analytics, nurses can recognize what may inappro- priately lengthen a patient’s stay, such as ineffective medication

By applying computational models and analysis, nurses can draw on historical, present, and simulated future data to provide actionable insights into

real-world clinical and operational problems.

www.nursingmanagement.com Nursing Management • March 2019 17

NURSING INFORMATICS

management, missed treatments and procedures, and not meeting discharge criteria. • failing to receive the best op- tions at end of life. Predicting mortality using machine learning is also on the horizon. Machine learning algorithms interpret mul- tiple data sources, including the EHR, medical claims, and geo- graphic data, to discover patterns indicating imminent mortality in patients.16 Predictive analytics can help nurses lead data-driven criti- cal conversations to ensure that patients receive appropriate care. These knowledge-derived discus- sions help patients and family members consider the best care options approaching death, in- cluding palliative and hospice care. Using analytics can aid nurses to engage patients and families with end-of-life choices to improve quality of life.17

Into the future The value of nursing knowledge synthesized with predictive analyt- ics enables the provision of evidence- based care and the promotion of safety, quality, and appropriate pa- tient outcomes—the end goal of using all health information tech- nology. Emerging technologies, such as predictive analytics and machine learning, will strengthen our ability to collect data, assimi- late these data into information, apply newly discovered knowl- edge, and gain wisdom to improve care delivery. Moving forward, we’ll use these technologies to en- hance EHR clinical decision sup- port tools and help optimize oper- ational workforce issues such as inadequate staffing through more precise scheduling. We’ll also de- crease inefficiencies that hinder caregiver satisfaction, such as breakdowns in multidepartmental processes and patient throughput,

and become key players in solving the challenges of transitional care. Harnessing the power of using data to extract valuable patterns to inform better decision-making gives nurses an edge in healthcare. We’ll collectively add influence as we provide appropriate, evidence- based care and advance the nurs- ing profession. NM

REFERENCES 1. Lutfiyya MN, Schicker T, Jarabek A,

Pechacek J, Brandt B, Cerra F. Gener- ating the data for analyzing the effects of interprofessional teams for improv- ing triple aim outcomes. In: Delaney C, Weaver C, Warren J, Clancy T, Simp- son R, eds. Big Data-Enabled Nursing: Education, Research and Practice (Health Informatics). New York, NY: Springer; 2017:136.

2. McGonigle D, Mastrian K. Nursing Informatics and the Foundation of Knowledge. 4th ed. Burlington, MA: Jones & Bartlett Publishers; 2017.

3. Ommaya AK, Cipriano PF, Hoyt DB, et al. Care-centered clinical docu- mentation in the digital environment: solutions to alleviate burnout. NAM Perspectives. 2018. https://nam.edu/ wp-content/uploads/2018/01/Care- Centered-Clinical-Documentation. pdf.

4. McGonigle D, Mastrian K. Nursing In- formatics and the Foundation of Knowledge. 2nd ed. Burlington, MA: Jones & Bartlett Publishers; 2012.

5. Matney SA, Maddox LJ, Staggers N. Nurses as knowledge workers: is there evidence of knowledge in pa- tient handoffs? West J Nurs Res. 2014;36(2):171-190.

6. Herron J, Sampson C. The knowledge worker in healthcare. Royer Maddox Herron Advisors. 2016. www.royer maddoxherronadvisors.com/Blog/2016/ April/The-Knowledge-Worker-in- Healthcare.aspx.

7. Bari A, Chaouchi M, Jung T. Predictive Analytics for Dummies. 2nd ed. Hobo- ken, NJ: John Wiley & Sons, Inc.; 2017.

8. Thornton C, O’Flaherty B. Improving customer centric design for self-service predictive analytics. In: Donnellan B, Helfert M, Kenneally J, VanderMeer D, Rothenberger M, Winter R, eds. New Horizons in Design Science:

Broadening the Research Agenda. New York, NY: Springer; 2015:230-245.

9. Carroll WM, Hofmeister N. Predictive analytics and the impact on nursing care delivery. HIMSS18 Conference & Exhibition. 2018. http://365.himss.org/ sites/himss365/files/365/hand- outs/550229529/handout-NI2.pdf.

10. Mitchell TM. Machine Learning. New York, NY: McGraw Hill; 1997.

11. Posadas S, Kunkel K. Machine learn- ing: a brave new world. Valve Maga- zine. 2017. www.valvemagazine.com/ web-only/categories/technical- topics/8687-machine-learning-a-brave- new-world.html.

12. McGrane C. Health tech podcast: how AI is making humans the “fundamental thing in the internet of things.” Geek- Wire. 2018. www.geekwire.com/2018/ health-tech-podcast-ai-making-hu mans-fundamental-thing-internet- things.

13. Simpson RL. Technology enables value-based nursing care. Nurs Adm Q. 2012;36(1):85-87.

14. O’Dowd E. IoT devices significantly lower nurse response times. HITIn- frastructure. 2017. https://hitinfra structure.com/news/iot-devices- significantly-lower-nurse-response- times.

15. Nelson JM, Rosenthal L. How nurses can help reduce hospital readmis- sions. Am Nurse Today. 2015;10(5) (suppl):18-20.

16. Ahmad MA, Eckert C, McKelvey G, Zolfaghar K, Zahid A, Teredesai A. Death versus data science: predicting end of life. The Thirtieth AAAI Confer- ence on Innovative Applications of Ar- tificial Intelligence. New Orleans, LA. February 4-6, 2018.

17. Ferguson R. Care coordination at end of life: the nurse’s role. Nursing. 2018;48(2):11-13.

Whende M. Carroll is the founder of Nurse Evolution and the senior editor of the Online Journal of Nursing Informatics.

The author has disclosed no financial relationships related to this article.

DOI-10.1097/01.NUMA.0000553503.78274.f7