The Uses of Artificial Intelligence in Hospital Medicine

Artificial intelligence (AI) is one of the hottest topics in health care. As one writer put it,?

“Machine learning (ML) and artificial intelligence (AI) have already taken center stage for complex problem-solving in many industries. They are becoming one of the most discussed and exciting topics in medicine.?Healthcare systems are developing AI-powered predictive models using features like age, gender, social determinants of health, patients’ comorbidities, previous hospitalizations, ER visits, and other clinical risk factors to determine the risk of readmission and make better decisions.”[1]

AI will certainly have an impact on hospital medicine.??The purpose of this article is to suggest some of the ways in which the AI revolution is likely to affect hospital medicine in general, and telehospital medicine in particular.

The challenge

One of the keys to maintaining a strong in-patient hospital system is to have hospitalists available 24/7 to admit, plan care and treatment and ensure transfer of patients to the medical floor are as efficiently as possible.??Access to a hospitalist who can cover the night shift at a hospital has been shown to be extremely beneficial for the hospital staff, nighttime covering nurses and patient care as a whole.

Many hospitals, especially in rural??areas, are finding it increasingly challenging to hire and retain hospitalists.??Telehospitalists are a cost-effective alternative way for these hospitals to meet their hospitalist needs with remote, “virtual” hospitalists.

Growing hospital use of AI

Artificial intelligence (AI) is already contributing to making aspects of health care more efficient, such as robot-assisted surgery and image analysis.????The COVID-19 pandemic accelerated the adoption of AI technologies, as hospitals faced with staff shortages and overwhelming caseloads rolled out a range of AI tools to help them triage COVID patients and care for them effectively.?

Hospitals’ use of AI will continue to expand as they find ways to use AI to increase patient insights and improve their services, while reducing costs and proactively detecting waste and fraud.??AI will affect hospital medicine in a number of ways, including the following.

Reducing??the total number of hospitalists needed

The need for hospitalists currently exceeds the supply.???In the long term, AI can contribute indirectly to solving this problem,??by reducing the population of patients who require hospital care in the first place.?

By increasing understanding of the totality of factors that affect people’s health, AI can enhance healthcare systems’ ability to provide predictive, proactive care.??For example, Marpai, Inc. has launched??an AI-powered proactive health service called Marpai Cares.??Marpai Cares uses proprietary, predictive deep learning models to identify near-term health events related to chronic illness (including cardiovascular disease, diabetes, and COPD) and major procedures (such as knee surgery)??and identify appropriate courses of action that can help to avoid worst-case outcomes.??Using such a system,??healthcare providers increasingly will be able to anticipate when a person is at risk of developing a chronic disease, for example, and suggest preventative measures before their condition reaches the point where they require hospitalization.

Another way that AI could reduce the number of hospitalists that hospitals need to hire is by making it possible for in-house hospitalists to care for a larger number of patients, by streamlining diagnostic and treatment processes.??This could reduce pressure on the available pool of hospitalists.

Improving hospitals’ ability to deploy and manage hospitalist resources

AI can help hospitals to manage their personnel resources, including hospitalists, as efficiently as possible.??AI-enabled “smart workforce management systems” using natural language processing or NLP (a branch of computer science that is concerned with giving computers the ability to understand text and spoken words as humans do) , deep learning and machine learning can facilitate the assignment??of hospitalist and other medical staff resources, making the process easier and more efficient.??

These systems can allocate resources across the hospital or system, based on need, personal preferences and expertise.??They can also supplement the in-house workforce with outside clinicians as needed, opening the door to more “hybrid” hospitalist models that employ a mix of in-house and virtual hospitalists that can be adjusted in real time according to need.

Reducing hospitalist “burnout”.

One of the reasons why some hospitals have trouble retaining hospitalists is that their hospitalists want a better quality of work and work-life balance and leave their jobs when they feel unable to cope with the stress, work hours, billing and paperwork. Clinical autonomy and control over their work environment have been shown to be another driving factor in physician job satisfaction.?

There is growing evidence that AI can help in reducing physician burnout.??For example, AI can help to simplify the documentation process, allowing physicians to focus more on care delivery than data entry tasks.????According to a 2022 American Society of Anesthesiologists study, AI -based scheduling tools result in physicians having one or two mornings and one or two afternoons off a month to allow for increased work-life balance.

Improving hospitalist service quality?

The benefits of AI are not limited to increasing hospitalists’ productivity, but also extent to improving the quality of the care they provide.???AI’s most powerful use is to enhance human capabilities, not replace them.??

AI will provide hospitalists with predictive analytics and clinical decision support tools that alert them to problems before they might otherwise recognize the need to act.??AI can provide early warning of conditions like seizures or sepsis, which often require intensive analysis of highly complex datasets.

AI can also inform decisions about the most appropriate care.??AI is capable of providing real-time, data-driven insights that hospitalists can alter and implement based on their personal expertise, enhancing the overall quality of the care they provide.??

AI can increase the time and attention that hospitalists can devote to actual patient care by reducing the need for time-consuming data energy and administrative tasks.??AI can help the hospitalist to prioritize patients and tasks according to priority, e.g., with?real-time reminders and a progress tracker.

Conclusion

AI should relieve some of the pressures that are currently helping to drive the demand for telehospitalists, by reducing the number of patients that hospitalists have in their care, and addressing certain root causes of hospitalist burnout.??However, the fact remains that telehospitalists will continue to be the only solution for some hospitals, especially in low-income or rural areas, that have difficulty hiring and retaining in-house hospitalists.

Even in some large, well-funded urban hospitals, AI-enabled smart workforce management systems may create addition opportunities for telehospitalists as part of the flexible, need- driven approach to staffing they enable.??For all hospitalists, in-house as well as virtual, AI has the potential to significantly improve the quality of the care they provide to their patients.

[1]?Sareer Zia, “Artificial Intelligence in Tackling Hospital Readmissions”, https://vhospitalist.com/ai-readmissions/