Artificial Intelligence in Curbing the Nursing Shortage

Background of the Clinical Problem 

Forming the leading sector of health care practitioners with nearly six million functional professional members, registered nurses take part in a crucial role in the provision of safe and quality medical care within the U.S health care system (Jones, 2017). Their enhanced expertise, interventions and compassion, observational skills, and the transformational influence that is portrayed by nurses within the nation, societies, and in the lives of patients as well as their families is unrivaled. If the U.S. is to continue being a leader in the delivery of care, a constant of highly qualified and trained nurses’ professionals is required. The United States Bureau of Labor Statistics estimated that the shortage of nurses is estimated to increase by 30 percent in 2030. Given the complexity of modern-day care, achieving the triple aims of health care largely depends on highly trained and experienced college-educated nurses. 

The shortage of nurses is exacerbating at unprecedented rates due to retirements and attrition. Accompanied by an aging workforce, improved health outcomes due to Medicare and Medicaid, and people above the age of 65 living for longer is putting a strain on the already limited workforce (Jones, 2017). Other factors contributing to the nursing shortage include nurse burn out, an increase in violence in healthcare settings, technology, staffing ratios, and nurses leaving the workforce to take care of their families. The nursing shortage has a fundamental impact on the delivery of care and could lead to medical errors, thus resulting in the loss of life. 

Clinical Problem Statement 

Across the United States, the nursing shortage is putting a strain on healthcare workers, thus jeopardizing patient safety. The hospitals in the United States were already understaffed before the outbreak of the novel coronavirus (Gennaro, 2020). Across the United States, healthcare institutions do not have enough ICU-trained nurses, and they are finding it hard to cope with existing patient loads. Also, there is a shortage of registered nurses to meet the current demand for patients and the already overstretched workforce is being asked to pick up extra shifts (Jones, 2017). 

The number of surgeries in the U.S. is increasing as Americans grow older, and chronic illnesses proliferate. It is estimated that by 2030, one in five Americans will be sixty-five years and older. It has been estimated that more than 40 million inpatient surgeries are conducted annually in the U.S., with an additional 50 million outpatient procedures conducted in ambulatory sites (Snavely, 2016). As the number of surgeries grows, the need for baccalaureate-educated health care professionals is to maintain patient safety is growing. Therefore, the nursing shortage in the country must be addressed to prevent burnout, which compromises both the safety of the patient and healthcare workers (Snavely, 2016). According to recent data by the National Institute of Health, there is a positive correlation between staffing and improved patient and population outcomes across experience, safety, and quality. 

Purpose of the Change Proposal 

The above statistics not only provide information on the amplitude of the problem but also indicate the widening gap between supply and demand. The health care workforce is experiencing a crisis due to the shortage of nurses, the increased need for chronic care, the aging population, and burnout from the physicians themselves (Snavely, 2016). Health care institutions can leverage the use of technology in collaboration with nurses to fix the nursing shortage. For decades, the nursing shortage has impacted the delivery of services, especially during emergencies (Gennaro, 2020). The implementation of artificial intelligence in the delivery of care could accelerate more accurate diagnoses, decision-making, facilitate the development of new treatments, and reduce medical errors. Smart algorithms could be deployed in hospitals to take over repetitive tasks while allowing nurses and physicians can prioritize other tasks than fighting the treadwheels of bureaucracy. 

Literature Research Strategy 

The literature research strategy employed was in-depth research through various databases such as JSTOR, ScienceDirect, EBSCOHost, and ProQuest in the search for materials to be used for this project. A deep internet search in Google was also used to find the relevant literature to support the concepts and terminologies used in this research. Keywords such as Artificial intelligence in health care, nursing shortages in the United States, nursing shortages in healthcare, nursing shortages and patient safety, nursing shortage solutions, and artificial intelligence integrations in healthcare were employed in the search for the relevant materials to be used in this paper. After a comprehensive analysis of the literature, the report was prepared. 

Literature Review 

In recent years, the United States is experiencing difficulties in the supply and demand of nursing professionals. In 2011, the Institute of Medicine (IOM) called for an increase in the number of baccalaureate-prepared nurses by 80 percent in the workforce and doubling the number of healthcare professionals by 2020 (Snavely, 2016). Although the number of nurses graduating from graduate programs has increased, the current workforce is yet to meet the recommendations by the IOM, with only 55 percent of registered nurses prepared at the graduate or the baccalaureate degree level. The nursing shortage can be attributed to the natural consequence of aging and attrition by retirement (Gennaro, 2020). To keep up with the demand, the IOM estimated that the United States would need an additional 1 million nurses in 2020. 

The implications for the nursing shortage in the United States healthcare system is far-reaching at local, regional, and national levels. These shortages directly contribute to increasing healthcare costs. According to Gennaro (2020), healthcare institutions with inadequate nurses are positively correlated with higher readmission rates. Another study conducted by Jones (2017) found that practices with high patent to nurse ratios correlated with increased urinary tract and surgical site infections. 

PICOT Statement 

PICOT question: How effective can artificial intelligence compared to creating new nursing education system be in addressing nursing shortage to achieve better healthcare? 

P (Problem)-Nursing shortage 

I (intervention)-Artificial intelligence 

C (comparison)-Creating new nursing education system 

O (Outcome)-Better healthcare 

T (Time) – Time required for practices to realized improved patient outcomes and reduced burnout. 

Change Theory 

Today, the health care environment continues to adapt and evolve to meet the complex health care demands of the population. In the current climate, nurses must adopt technology in practice to address complex health care challenges and deliver optimal care to patients (Hussain et al., 2018) . The quality of care dispensed relies on the ability of nurses to access comprehensive and accurate health care information. In deploying artificial intelligence in clinical settings, this paper will explore the use of Kurt Lewin's theoretical framework that will help nurses recognize the need for change, traverse through the change process, and finally achieve the desired strategic objective – improve patient care and promote positive population outcomes (Hussain et al., 2018) . 

The first step in the "unfreezing" process where healthcare institutions will identify the need for nurses and prioritize the driving and restraining forces in the institution. This can be achieved through brainstorming, questionnaires, interviews, and collaborations with nurses and administrative staff (Hussain et al., 2018) . The second step involves the implementation of the change plan identified in step 1. The change nurse leader must be in constant communication with other nurses and the management to ensure that they are part of the process and continue to acknowledge their suggestions and opinions (Hussain et al., 2018) . The change nurse leader acts as an advocate by educating other nurses on the importance of artificial intelligence and be involved in the planning and implementation process (Kunie et al., 2017). 

Once the organization has achieved full adoption of artificial intelligence, the practice can proceed to the final stage of Lewin's model, which is refreezing. At this stage, the organization will be using machine learning in helping the institution improve on intake predictions and prevent overdiagnosis, which could help prevent more than 15 percent hospital readmissions. Nurses can leverage this technology for better internal communication, and absences can be managed more easily (Hussain et al., 2018) . Thus, the refreezing process ensures that the change implemented is "frozen" and becomes part of the standard working procedures. 

Implementation and Outcome Measures 

In implementing the change, business processes must be designed and redesigned and adapted to specific clinical settings. The nursing workforce and all stakeholders must be retrained and ready to incorporate the changes in their daily routines (Kunie et al., 2017). Effective change processes begin with leaders as they play a critical role in organizing and maintaining a favorable climate for growth within the organization by communicating the vision, inspiring others to act on the vision, and committing new approaches (Kunie et al., 2017). 

The following outcomes will be measure after six months of implementation. 

The use of artificial intelligence in reducing information overload for physicians and nurses. 

The use of artificial intelligence in designing treatment plans for patients. 

How artificial intelligence has shaped the nursing workflow system in reducing repetitive tasks so that nurses and physicians can spend more time on tasks that matter. 

Using Evidence-Based Practice 

The University of California-San Francisco has implemented artificial intelligence in their ICU department to be used in improving medical care for patients. In traditional ICU care, nurses and physicians respond to an alarm every 90 seconds, of which two-thirds will prove to be false alarms which they do not' signify real danger (Opel et al., 2017). According to the FDA, these alarms contributed to more than 500 deaths between 2005 and 2008. With artificial intelligence, the patient monitoring system can identify hours in advance, thus making work much faster and efficient than responding to false alarms (Kunie et al., 2017). 

Evaluation Plan 

The first step will involve the measuring of organizational performance by evaluating the speed of execution in responding to incidents in the ICU and how first were the diagnosis of patients during peak and off-peak hours. 

The second step involves measuring the Individual performance of the nurses by evaluating their understanding and awareness of the change. This will include employee satisfaction surveys on their view about the change process; this will also include employee feedback on the blocker and enablers in the change management plan. 

Potential barriers and Solutions 

Artificial intelligence is expensive to deploy, and this could lead to delays in its implementation (Kunie et al., 2017). Health care institutions that do not have financial incentives to implement AI tools and services will push them behind in improving healthcare. This can be achieved by setting apart funds for selected healthcare institutions by the federal government for at least three hospitals per state (Opel et al., 2017). 

In conclusion, AI is a powerful tool that eases the nursing staffing crisis in healthcare. However, this does not mean AI the human element in care or medical professionals but rather support the lack of a healthcare workforce in achieving the triple aim objectives – reducing the cost per capita of care for populations, improving outcomes of communities, and improving the experience of care. 

References 

Gennaro, S. (2020). 2020: The Year of the Nurse as Seen Through a Coronavirus Lens.  Journal of Nursing Scholarship ,  52 (3), 231. 

Hussain, S., Lei, S., Akram, T., Haider, M., Hussain, S., & Ali, M. (2018). Kurt Lewin's change model: A critical review of the role of leadership and employee involvement in organizational change. Journal of Innovation & Knowledge , 3 (3), 123-127. https://doi.org/10.1016/j.jik.2016.07.002 

Jones, S. (2017). Succession Planning: Creating A Case for Hiring New Graduates.  Nursing Economic$ ,  35 (2), 64–87. 

Kunie, K., Kawakami, N., Shimazu, A., Yonekura, Y., & Miyamoto, Y. (2017). The relationship between work engagement and psychological distress of hospital nurses and the perceived communication behaviors of their nurse managers: A cross-sectional survey. International Journal of Nursing Studies, 71 , 115-124. Doi: 10.1016/j.ijnurstu.2017.03011. 

Opel, E.M., Winter, V., & Schreyong, J. (2017). Evaluating the link between human resource management decisions and patient satisfaction with quality of care . Health care management review, 42 (1), 53-64. 

Snavely, T. M. (2016). Data Watch. A Brief Economic Analysis of the Looming Nursing Shortage in the United States.  Nursing Economic$ ,  34 (2), 98–100.